Silver by Fire, Silver by Mercury: A Chemical History of Silver Refining in New Spain and Mexico, 16th to 19th Centuries [Illustrated] 9004343822, 9789004343825

Citation preview

Silver by Fire, Silver by Mercury

European Expansion and Indigenous Response Editor-in-Chief George Bryan Souza (University of Texas, San Antonio)

Editorial Board Cátia Antunes (Leiden University) João Paulo Oliveira e Costa (cham, Universidade Nova de Lisboa) Frank Dutra (University of California, Santa Barbara) Kris Lane (Tulane University) Pedro Machado (Indiana University, Bloomington) Malyn Newitt (King’s College, London) Michael Pearson (University of New South Wales)

volume 25

The titles published in this series are listed at brill.com/euro

Silver by Fire, Silver by Mercury A Chemical History of Silver Refining in New Spain and Mexico, 16th to 19th Centuries

By

Saul Guerrero

leiden | boston

Cover illustration: Patio de la Hacienda de Beneficio de la Mina de Proaño, Zacatecas, by Pietro Gualdi (1840). Illustration reproduction authorised by the Instituto Nacional de Antropología e Historia, Museo Nacional de Historia, Ciudad de México.

Library of Congress Cataloging-in-Publication Data Names: Guerrero, Saul (Saul Jose), author. Title: Silver by fire, silver by mercury : a chemical history of silver refining in New Spain and Mexico, 16th to 19th centuries / by Saul Guerrero. Description: Leiden ; Boston : Brill, 2017. | Series: European expansion and indigenous response, issn 1873-8974 ; volume 25 | Based on author's doctoral thesis, McGill University (Montreal, Quebec), 2015. | Includes bibliographical references and index. Identifiers: lccn 2017029863 (print) | lccn 2017032359 (ebook) | isbn 9789004343832 (e-book) | isbn 9789004343825 (hardback : alk. paper) Subjects: lcsh: Silver–Refining–Mexico–History–16th century. | Silver– Refining–Mexico–History–17th century. | Silver–Refining–Mexico–History–18th century. | Silver–Refining–Mexico–History–19th century. | Silver–Metallurgy– Mexico–History. | Metals–Refining–History. | Extraction (Chemistry) Classification: lcc tn760 (ebook) | lcc tn760 .g84 2017 (print) | ddc 669/.23–dc23 lc record available at https://lccn.loc.gov/2017029863

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. issn 1873-8974 isbn 978-90-04-34382-5 (hardback) isbn 978-90-04-34383-2 (e-book) Copyright 2017 by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill nv incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, ma 01923, usa. Fees are subject to change. This book is printed on acid-free paper and produced in a sustainable manner.

To Adriana

∵

Contents General Series Editor’s Preface Acknowledgements xiv List of Illustrations xvi Guide to the Text xxvi Introduction

xi

1

1 The Genesis and Nature of Silver Ores 9 Why Spain? 9 To Have and Have Not 18 Old World Silver Ores 19 New World Silver Ores 22 Silver Sulphide and Silver Chloride 26 Argentiferous Galena 30 A Red Herring 32 The Other Chemical Keys 37 The Immoveable Object and the Unstoppable Force The Table is Set 42 2 The Dry Refining Process: Smelting of Silver Ores 44 Deceitful Mercury 44 Smelting of Silver Ores: The Human Context 48 The Chemistry of Smelting and the Nature of the Ore The Architecture of Smelting in New Spain 58 The Infrastructure of Smelting in New Spain 65 Plata de fuego (Silver by Fire) 75

40

53

3 The Dry Refining Process: Its Impact on the Environment 78 Lead: The Nature of Its Consumption 79 Lead: The Directionality of Its Loss 86 Lead: Its Source 92 Charcoal and the Scale of Depletion of Woodland 93 The Local Environmental Impact of Smelting 96 A Straightforward Decision 98

viii

contents

4 The Wet Refining Process: The Chemistry of the patio Process 102 Plus ça change 102 The Alchemy of Mercury 104 The Gold Connection 111 The Complex Mechanism of a Mercury-Based Refining Process 118 The correspondencia: The Key to the Fate of Mercury 123 The Loss of Calomel 131 The Stages in the Use of Mercury to Refine Gold and Silver Ores 133 The Twists in the Trail 143 Mercury-Based Refining of Silver Ores: The Human Factor 146 Plata de azogue (Silver by Mercury) 151 5 The Physical Infrastructure of the patio Process 153 The patio Process 153 Milling the Ore 154 The patio Reactor 158 Planillas 161 The azoguería, or Mercury Room 163 Desazogaderas or capellinas and the Recycling of Mercury The Architecture of the patio Process 169 The Environmental Impact Vectors of the patio Process 187 A Unique Industrial Effort 189 6 The Hacienda Santa María de Regla 192 The Nineteenth Century 192 The Adventurers in the Mines of Real del Monte 195 The Hacienda de Regla 200 Main Process Areas 203 Main Gate and Storage of Incoming Ore 203 Stamp Mills, Chilean Mills and arrastres 204 The patio Reactor 208 Furnace Area a: Covered Storage, hornos castellanos and capellinas 211 Furnace Area b: English Blast Furnaces 216 Final Comments on the Architectural Layout of Regla 219 The Mass Balance of the Silver Refining Processes at Regla, 1872 to 1888 220

164

contents

ix

7 The patio Process and Smelting at Regla 224 The Keys to an Efficient patio Process at Regla 224 Coordinated Milling and Supply to the patio Reactor 224 An Intelligent and Interactive Application of the patio Process Recipe 229 Long Term Planning of Inventory 232 Optimal Use of the patio Reactor Area 232 Low Use of Fuel in the patio Process 234 The Challenges of the Smelting Process at Regla 235 An Unpredictable Supply of Ore 235 Low Lead Content in the Ores 237 However, a Very Efficient Fuel Consumption 237 The Efficiency of Extracting Silver at Regla 237 The Labour Force at Regla 239 The Mass Balance for the patio Process at Regla 245 The Mass Balance for Smelting at Regla 245 The Environmental Loss Vectors in the Period 1872 to 1888 248 A Snapshot of a Refining hacienda 251 8 The Economies of Refining Silver 255 Roads to Riches 255 Refining Costs in New Spain, as Reported 257 The Refining Costs at Regla 268 The Macroeconomic Context in the Nineteenth Century 269 The Variable Costs of the patio Process at Regla, 1872–1888 276 The Variable Costs of Smelting at Regla, 1875–1886 280 Refining Cost as a Function of Silver Content 281 A Window to the Past: Extrapolating the Results from Regla 287 The False Positives of the patio Process 293 Apples and Pears 294 ‘ganando indulgencias con escapulario ajeno’ 296 The Succour to Refiners 299 Barrel Process 304 Silver in the Context of Other Commodity Trades 310 The Bottom Line 312

x

contents

9 The Environmental Impact of Silver Refining: A Shift of Paradigm 316 The Base Line 316 An Estimate of the Breakdown of Silver Production by Refining Process by Caja 321 Caja of Zacatecas 321 Caja of Guanajuato 323 Caja of México 323 Caja of Durango 325 Caja of San Luis Potosí 326 Caja of Guadalajara 328 Caja of Pachuca 329 Caja of Sombrerete 330 Caja of Bolaños 332 Caja of Rosario 333 Caja of Zimapán 333 Caja of Chihuahua 333 Aggregate Totals for New Spain 335 Aggregate Totals for Mexico, 1820 to 1900 338 Environmental Impact Vectors, Sixteenth to Nineteenth Century 339 The Environmental Impact on New Spain and 19c Mexico: The Modern Legacy 346 A Change of Paradigm 347 What Did They Know and When Did They Know It? 350 Was Mercury the Indispensable Key to Silver in the New World? 358 Epilogue

361

Appendix a: The Accounting Books of Regla 367 Appendix b: Sensitivity Matrix for Refining Costs 370 Appendix c: Estimates of Silver Production by Caja and Refining Process, Including Balance of Mercury Consumption and Physical Losses 374 Glossary of Technical Terms in Spanish 396 Archival Sources 399 Bibliography 401 Index 422

General Series Editor’s Preface Over the past half millennium, from circa 1450 until the last third or so of the twentieth century, much of the world’s history has been influenced in great part by one general dynamic and complex historical process known as European expansion. Defined as the opening up, unfolding, or increasing the extent, number, volume, or scope of the space, size, or participants belonging to a certain people or group, location, or geographical region, Europe’s expansion initially emerged and emanated physically, intellectually, and politically from southern Europe—specifically from the Iberian peninsula—during the fifteenth century, expanding rapidly from that locus to include, first, all of Europe’s maritime and, later, most of its continental states and peoples. Most commonly associated with events described as the discovery of America and of a passage to the East Indies (Asia) by rounding the Cape of Good Hope (Africa) during the early modern and modern periods, European expansion and encounters with the rest of the world multiplied and morphed into several ancillary historical processes, including colonisation, imperialism, capitalism, and globalisation, encompassing themes, among others, relating to contacts and, to quote the euro series’ original mission statement, “connections and exchanges; peoples, ideas and products, especially through the medium of trading companies; the exchange of religions and traditions; the transfer of technologies; and the development of new forms of political, social and economic policy, as well as identity formation.” Because of its intrinsic importance, extensive research has been performed and much has been written about the entire period of European expansion. With the first volume published in 2009, Brill launched the European Expansion and Indigenous Response book series at the initiative of well-known scholar and respected historian, Glenn J. Ames, who, prior to his untimely passing, was the founding editor and guided the first seven volumes of the series to publication. George Bryan Souza, who was one of the early members of the series’ editorial board, was appointed the series’ second General Editor. The series’ founding objectives are to focus on publications “that understand and deal with the process of European expansion, interchange and connectivity in a global context in the early modern and modern period” and to “provide a forum for a variety of types of scholarly work with a wider disciplinary approach that moves beyond the traditional isolated and nation bound historiographical emphases of this field, encouraging whenever possible non-European perspectives … that seek to understand this indigenous transformative process and period in autonomous as well as inter-related cultural, economic, social, and ideological terms.”

xii

general series editor’s preface

The history of European expansion is a challenging field in which interest is likely to grow, in spite of, or perhaps because of, its polemical nature. Controversy has centered on tropes conceived and written in the past by Europeans, primarily concerning their early reflections and claims regarding the transcendental historical nature of this process and its emergence and importance in the creation of an early modern global economy and society. One of the most persistent objections is that the field has been “Eurocentric.” This complaint arises because of the difficulty in introducing and balancing different historical perspectives, when one of the actors in the process is to some degree neither European nor Europeanised—a conundrum alluded to in the African proverb: “Until the lion tells his tale, the hunt will always glorify the hunter.” Another, and perhaps even more important and growing historiographical issue, is that with the re-emergence of historical millennial societies (China and India, for example) and the emergence of other non-Western European societies successfully competing politically, economically, and intellectually on the global scene vis-à-vis Europe, the seminal nature of European expansion is being subjected to greater scrutiny, debate, and comparison with other historical alternatives. Despite, or perhaps because of, these new directions and stimulating sources of existing and emerging lines of dispute regarding the history of European expansion, Souza and the editorial board of the series will continue with the original objectives and mission statement of the series and vigorously “… seek out studies that employ diverse forms of analysis from all scholarly disciplines, including anthropology, archaeology, art history, history (including the history of science), linguistics, literature, music, philosophy, and religious studies.” In addition, we shall seek to stimulate, locate, incorporate, and publish the most important and exciting scholarship in the field. Towards that purpose, I am pleased to introduce volume 25 of Brill’s euro series, authored by Saul Guerrero, which is entitled Silver by Fire, Silver by Mercury: A Chemical History of Silver Refining in New Spain and Mexico, 16th to 19th Centuries. The author is a Venezuelan born and trained professional chemist and recently Anglo-Canadian trained historian. He has produced a work of seminal importance. It deals with interactions between Europeans, indigenous and enslaved peoples in establishing a mining industry that produced a commodity in the New World and Mexico that wrought major changes in exchanges around the world and has been credited as a key element in fuelling the historical process of globalisation—silver. Silver by Fire, Silver by Mercury, as a consequence, by truly engaging in a multi-disciplinary effort and combining history and chemistry produces a work that constitutes a major contribution to the history of science and technology in general and essentially re-writes the history of silver refining in colonial and

general series editor’s preface

xiii

early national Mexico in a revisionist and a truly global context, by comparing developments there with those in Europe, parts of Asia, and highland South America. Although some of the work found here has been touched upon by a series of other distinguished authors, but to the best of our knowledge, mine and most importantly from peer review expertise, this is the first study of colonial silver refining to engage the chemistry that was practiced in more than a general way. And it has yielded some intriguing and important findings. Guerrero demonstrates that of all Mexican silver bearing ores some 60 % were processed by refining with mercury, and the rest by smelting, which used lead as flux. The resulting impact and damage to the environment has been examined and explained in a manner that all specialist and general readers will understand and appreciate. His examination of the world’s greatest and longest silver production boom and its consequences to the participants and the societies that were involved, as well as, to mankind’s habitat and environment will contribute positively to emergent and growing discussions and debates on these themes among historians, archaeologists, environmental scientists, and others. George Bryan Souza

Acknowledgements This work has its roots in the History Department of the University of Warwick, United Kingdom, who accepted a retired ex-chemist into their m.a. program in Global History. Without the encouragement and guidance provided by Maxine Berg, Giorgio Riello, Anne Gerritsen and Rebecca Earle, I would have never attempted the transition between chemistry and history. I am also most grateful to the Department of History and Classical Studies of McGill University, Montreal, Canada, who accepted my application to their doctoral program, and offered the financial support of a Fellowship from the Cundill Foundation. My thanks to Catherine Desbarats and Daviken Studnicki-Gizbert for their help and guidance during the various stages of my study and research at McGill. Pamela Welbourn, Queen’s University, Canada, went well beyond the call of duty as the external supervisor of my doctoral thesis. She provided much-needed support, timely corrections, encouragement and a very valued friendship that made it possible for me to structure ideas into a thesis. Dennis Flynn of the University of the Pacific, u.s.a., through his generous enthusiasm and belief in my work made this book a reality. Along the way I have been helped by experts in the field. Richard Sillitoe (u.k.), never failed to answer my requests for clarifications on geological issues. David Johnson, formerly of the Open University (u.k.), unearthed his old lab notes to provide me with additional critical information. Herbert S. Klein (Columbia University) provided his Excel files with data from the Cajas of New Spain. Daniel Engstrom (St. Croix Watershed Research Station, u.s.a.) and Colin Cooke (Dept. of Environment and Sustainable Resource Development, Alberta, Canada) opened a window to a wider audience among environmental scientists. Three editors have enriched my work with their comments and corrections, and then taken the risk of publishing my mix of chemistry and history: Carmen Giunta (Bulletin for the History of Chemistry), Martin Collins (History and Technology) and George Souza at Brill. The Faculty of Arts and the Department of History of McGill provided financial support for my research in Mexico, where I have been made to feel most welcome at every archive and region I had the great fortune to visit. Mi más sincero agradecimiento a Guadalupe Salazar González, Rafael Morales Bocardo, Thomas Hillerkuss, Eréndira María Guadalupe Guzmán Segoviano, Ada Marina López Meza, María Guevara S., to the proud owners who opened their old haciendas to my visits, and others who helped a total stranger along the way, too many to mention. Two people changed the course of my archival research: the Director of the Museo de Minería y Archivo Histórico de Pachuca, Asociación

acknowledgements

xv

Civil, Belem Oviedo Gámez, and her assistant, Aracely Monroy Pérez, who from the first email strongly encouraged me to visit their archives in Pachuca, and then offered all the help I could have wished for. To those of the above who have read parts of the following chapters, and to the unknown reviewers of articles and this manuscript, who have pointed out much needed corrections, my thanks. I bear the sole responsibility for the use and interpretation of the information I have been allowed to access. Finally, my grown-up family has been my constant support: Cristina and Juan Cristóbal, Saul Ignacio and Claire, Mariana and Juan Carlos, Carlos Pedro and the sky, and the joy of all our lives, my grandchildren Manuela, Olivia, Leo, Tomás, Julieta and Lya. As to my wife Adriana, she has had to endure an unexpected late twist to our life, yet has never complained nor failed to help me through our many recent wanderings.

List of Illustrations Figures 1 2

3 4 5

6 7

8 9 10 11

12

13

14

15

Location of the world’s 40 major silver deposits, as discovered until the year 2009 11 Approximate spheres of European colonial presence in the New World, and the location of all major silver and mercury sources exploited in the Early Modern period 12 Black smoker 13 The spreading ocean floor subducts under the continental crust of the Americas 14 Subduction is the underlying link between the unique geographical concentrations of silver along the Andes, North American Cordillera and Japan 17 Location of main historic silver mining regions in Europe by mid sixteenth century 23 A simplified representation of Sillitoe’s interpretation of the effect of weathering on an original (hypogene) non-weathered sulphidic silver deposit 28 The main historical silver, lead and salt deposits of New Spain / Mexico 31 Salient traits of the silver ore deposits under extraction in the Early Modern period in the Hispanic New World and in Europe 34 Schematic diagram of two stage refining by smelting of silver ores using lead 56 The solid white lines encapsulate the minimum area that can be clearly identified with each smelting hacienda, the dotted line the minimum area of the extant dumps of grasas. The Hacienda Santa María lies at 22°12’10” n 100°44’27” w, and the Hacienda hmc1 at 22°12’31” n 100°44’47”w. 61 ‘La Noble y Leal Ciudad de S. Luis Potosí dividida en Quarteles de Orden Superior del Exmo. Señor Virrey Marqués de Branciforte. Diciembre 15 de 1794.’ 62 Location of the main mines, smelting haciendas, charcoal production, agricultural and cattle rearing areas around the town of San Luis Potosí (slp) 66 Top left: Molino mill stones, Monte Caldera, San Luis Potosí. The diameter can reach 2m. Top right: A modern reconstruction of a molino in Zacatecas, at the exit of the El Edén mine. Bottom: Drawing of a molino. 68 Tuyère, or alcribís, date unknown, found in the ruins of the hacienda hmc2 in Monte Caldera 70

list of illustrations 16

17

18 19

20 21 22 23 24 25 26 27 28 29

30 31

32a

xvii

Exterior of smelting furnace at the ruins of the Hacienda Santa María in Monte Caldera. The arched port would have been used to feed ore or fuel to the furnace. Approximate height of chimney is around 6 metres. 72 Ruins of the Hacienda de Aranzazu in Guadalcázar. Top left: Front view showing archways under two smelting furnace chimneys with a height of approximately 7m. Top right: Back view of chimneys, showing possible aperture for drive shaft of bellows. Bottom left: Fields of grasas. Bottom right: satellite view of hacienda grounds. 74 Mounds of grasas in Monte Caldera, Hacienda Santa María, the chimney of Figure 16 in the background 76 Scheme of the mass balance for lead during the smelting of silver ores. Letters in bold indicate mass input, letters in capitals indicate mass output. 80 Photograph by Charles Waite titled ‘Mexican Adobe Smelter Taxco Guerrero’, 1905 88 Silver smelter, Zimapán, 1941, lithograph by Ira Moskowitz 89 Parallel alignment of arrays of furnaces from three different smelting haciendas in the area of Monte Caldera 91 An illustration of the sulphur-mercury alchemical explanation for the formation of metal deposits 105 Amalgamation of gold with mercury, as published in 1574 116 Sensitivity of the Hg/Ag weight ratio to fb and fa values of ores refined by mercury 127 Mercury recovered from the soil of the patio area of a historic silver refining hacienda in Mexico 130 Graph of mercury, lead and silver airborne depositions measured at Laguna Lobato, near Potosí 133 The chemical evolution of the recipe for the refining of silver ores with mercury 140 Main milestones and directionality in the proposed transmission of experiential knowledge on the use of mercury to refine gold and silver, 12c to 19c 145 The main stages of the patio process as practised in New Spain / Mexico 155 Top: A horse-powered molino in Mexico, late nineteenth century. Middle: Example of the pit of a molino, from where the finer grains are withdrawn via the arched tunnel and taken to the tahonas/arrastres. Photo taken in the ruins of the Hacienda San Juan Nepomuceno, Marfil, Guanajuato. Bottom: Drawing of a tahona /arrastre, showing its four voladoras. 156 Patio of the Hacienda de Salgado, Guanajuato, 1839 158

xviii 32b 32c 33 34 35 36

37 38

39 40 41 42 43 44 45 46 47 48 49

50 51 52

53 54

list of illustrations Photograph titled ‘People and mules in the patio of a mining refining unit’, archived under Charles B. Waite/W. Scott 159 Patio of unidentified hacienda in Guanajuato 160 Top: Cross section of a planilla. Bottom: Photo of planilleros at work. 162 Photo of an azoguería (mercury room) 164 Equipment used at the end of the sixteenth century to recycle mercury from the amalgam 165 Top: Cross-section of capellina used by nineteenth century in Mexico. Bottom left: The use of pulleys to manoeuvre capellina into place. Bottom right: Indirect heating of capellina. 167 Cross section of a capellina building in Guanajuato 169 Perimeter walls, clockwise from upper left: Hacienda La Escalera, Guanajuato, Hacienda San Juan Nepomuceno, Marfil, Hacienda Santa Ana, Marfil, Hacienda, name unknown, Pánuco, Zacatecas and Hacienda Las Mercedes, Zacatecas 171 Monthly silver production for four haciendas de patio in Guanajuato 173 Plan of the Hacienda Las Mercedes, Zacatecas 177 Main process areas in the Hacienda Las Mercedes 178 Plan of the Hacienda Casas Blancas, Marfil, 1885 179 Gallery of arrastres, Hacienda de Salgado, Guanajuato 180 Plan of the Hacienda Nueva de Fresnillo, Zacatecas 181 Another perspective of the Hacienda Nueva de Fresnillo, Zacatecas 183 Main loss vectors of calomel and mercury 189 Production of silver in metric tons, from 1493 to 1900, in New Spain then Mexico 194 The Hacienda de Regla in relation to Real del Monte, Pachuca and Ciudad de México 197 The locations of three of the historical silver refining haciendas operated by the Compañía de Minas del Real del Monte y Pachuca in the second half of the nineteenth century, on the outskirts of present day Huasca de Ocampo, Hidalgo State, Mexico 201 Satellite image of Regla 203 Reconstruction of main functional areas at Regla 204 Top: Detail from Figure 50 indicating deemed location of pit for a water wheel to drive a stamp mill. Bottom: View of the sixteen circular vats of the arrastres. 206 Photograph of water-driven arrastre, from unnamed hacienda in Mexico 207 Water distribution channels and overflow outlet in the southern area of Regla 208

list of illustrations 55 56 57 58 59 60 61

62

63 64

65a 65b 66a 66b 67 68a 68b 68c 69a

xix

The Hacienda de Regla, unknown artist and date 210 Histogram of the distribution of the sizes of tortas, 1872–1888 211 Original furnace area, dry storage vaults and high security process area in the southern section of Regla 212 Left: Vaulted storage areas (va). Right: Corridor of Furnace Area a (fa), with Spanish furnaces to the left 213 Tentative assignment of base foundations for a capellina, showing the channel and pool for cooling water (March 2013) 214 Photo of capellina in the entrance to the Hacienda Santa María de Regla 215 Top left: Two of the extant blast furnaces. Top right: Trough in front of furnaces, possibly for water. The photo was taken in March 2013. By November 2014 this trough had been demolished as part of major alterations in the original structures in the north area of Regla by the current owners. Bottom left: Furnace stacks, with opening to charge the furnace. Bottom right: Close-up of inscription on top of opening of right-hand stack, indicating a probable date of construction (1854). 218 Original photos taken in March 2013 in the oc area. By November 2014: Top left: This crenelated wall had been half demolished. Top right: The area has been converted to a set of pools and toilets. Bottom: This arch was being filled in using stones from the demolished interior walls. 221 Main transit corridors for the process stages of the patio process and smelting at Regla 223 Average monthly transit of materials and silver produced at Regla in the period June 1872 to December 1873 and June 1875 to August 1888 (patio process) and June 1875 to January 1886 (smelting) 223 Reception of silver ore for the patio process, and inventory levels, 1872–1888 225 Ore milled for the patio process, inventory levels and ore refined by the patio process, 1872–1888 225 Histogram of silver content (ley) as measured on raw ore used in the patio process, 1872–1888 226 Silver content (1872–1888) according to silver extracted 226 Silver produced by the patio process, 1872–1888 227 Weight ratios (1872–1888) of salt to silver refined by the patio process 230 Weight ratios (1872–1888) of copper sulphate to silver refined by the patio process 230 Weight ratios (1872–1888) of mercury to silver refined by the patio process 230 Monthly variations (1872–1888) in the inventory of mercury 231

xx 69b 69c 70a 70b 71 72a 72b 73 74a 74b 75 76 77 78 79a 79b 79c 79d 80 81 82 83 84 85 85 86 87

88

list of illustrations Monthly variations (1872–1888) in the inventory of salt 231 Monthly variations (1872–1888) in the inventory of copper sulphate 231 Histogram of days employed in the patio process (October 1872–December 1873) 233 Percentage extraction of silver as a function of the refining period, (October 1872–December 1873) 233 Reception of silver ores for smelting and inventory 235 Histogram of the silver content (ley) of ores for smelting, 1875–1886 236 Percentage of silver in the ores, as extracted by smelting, 1875–1886 236 Production of silver by smelting, 1875–1886 236 Weight ratio of charcoal to silver smelted, 1875–1886 238 Inventory of charcoal, 1875–1886 238 Levels of unextracted silver recorded for each refining method at Regla, 1872–1888 238 Labour force assigned to major areas of activity at Regla, in the four-week period ending on May 29th 1877 244 Main loss vectors of waste material, monthly average at Regla in the period 1872/73 and 1875/88 (patio process) and Jun 1875 to Jan 1886 (smelting) 249 The evolution of the silver to gold ratio from 1690 to 1900 269 Yearly average expense on salt 272 Yearly average expense on copper sulphate 272 Yearly average expense on charcoal 273 Yearly average expense on litharge 273 Yearly average expenditure on mercury, in pesos per pound 274 Monthly production costs of silver refined by the patio process at Regla (1872–1888) 275 Percentage breakdown of the total variable cost of the patio process at Regla (1872–1888, excluding 1874) 277 Monthly production costs of silver refined by smelting at Regla (1875–1886) 280 Percentage breakdown of the total variable smelting cost at Regla (1875–1886) 282 Projected margins at Regla, as a function of silver content: the patio process 286 Projected margins at Regla, as a function of silver content: smelting 286 Variable refining cost as a function of silver content 287 Refining costs of the patio process and smelting in the context of the second half of the sixteenth century as a function of the silver content of the ore 290 Refining costs of the patio process and smelting in the context of the period

list of illustrations

89 90

91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

xxi

from the mid-seventeenth to mid-eighteenth century as a function of the silver content of the ore 291 Prices in Spain and price of mercury in New Spain 302 Annual average of cargas of silver ore processed at the refining haciendas of the Compañía de Minas del Real del Monte y Pachuca in the period 1853 to 1873 306 The fraction of plata de azogue registered at the Caja of Zacatecas in the period 1670 to 1820 322 Estimate of silver registered at the Caja of Zacatecas according to refining process 322 The fraction of plata de azogue registered at the Caja of Guanajuato in the period 1679 to 1816 323 Estimate of silver registered at the Caja of Guanajuato according to refining process 324 The fraction of plata de azogue registered at the Caja of México in the period 1786 to 1816 324 The fraction of plata de azogue registered at the Caja of Durango in the period 1696 to 1813 325 Estimate of silver registered at the Caja of Durango according to refining process 326 The fraction of plata de azogue registered at the Caja of San Luis Potosí in the period 1713 to 1806 326 Estimate of silver registered at the Caja of San Luis Potosí according to refining process 327 The fraction of plata de azogue registered at the Caja of Guadalajara in the period 1691 to 1804 328 Estimate of silver registered at the Caja of Guadalajara according to refining process 329 The fraction of plata de azogue registered at the Caja of Pachuca in the period 1667 to 1820 329 Estimate of silver registered at the Caja of Pachuca according to refining process 330 The fraction of plata de azogue registered at the Caja of Sombrerete in the period 1680 to 1820 331 Estimate of silver registered at the Caja of Sombrerete according to refining process 332 The fraction of plata de azogue registered at the Caja of Bolaños in the period 1753 to 1804 332 The fraction of plata de azogue registered at the Caja of Rosario in the period 1770 to 1813 333

xxii 108 109 110

111 112 113a 113b 113c 113d 113e

list of illustrations The fraction of plata de azogue registered at the Caja of Zimapán in the period 1729 to 1806 334 The fraction of plata de azogue registered at the Caja of Chihuahua in the period 1788 to 1813 334 Approximate areas in black correspond to Cajas where on average, over the colonial period, silver was refined mainly by smelting, those in grey by the patio process 336 Registry of silver by refining process, as projected for New Spain 337 The fraction of total silver produced in Mexico, 1876 to 1892, according to refining process 338 Ranking of Cajas by the magnitude of the environmental impact vector corresponding to loss of lead in lead fume 341 Ranking of Cajas by the magnitude of the environmental impact vector corresponding to woodland consumed 342 Ranking of Cajas by the magnitude of the environmental impact vector corresponding to mineral waste voided into waterways 342 Ranking of Cajas by the magnitude of the environmental impact vector corresponding to salt voided into waterways 343 Ranking of Cajas by the magnitude of the environmental impact vector corresponding to mercury in calomel voided into waterways 343

Colour Plates i ii

Patio de la Hacienda de Beneficio de la Mina de Proaño, Zacatecas, by Pietro Gualdi, Museo Nacional de Historia, Ciudad de México. (1840). 184 ‘Die Hacienda von Regla’, by Johann Moritz Rugendas (1832), Ethnologisches Museum, smb, Berlin 185

Tables i ii iii iv v

Range of lead to silver weight ratios from individual smelting runs carried out in the region of Vetagrande, Zacatecas, in 1718 83 Range of percentage values for net lead losses during smelting of lead ores 84 Lead content in slags from different smelting sites and periods 85 Range of values of percentage of mercury converted to calomel, and lost via physical pathways to the environment, as a function of the Hg/Ag ratio 126 List of debtors on supplies provided by the Hacienda La Escalera,

list of illustrations

vi vii viii

ix x xi xii xiii xiv xv xvi xvii xviii xix xx xxi

xxii xxiii xxiv

xxiii

Guanajuato, with their estimate of selected assets offered as collateral to the debt, as of 1791 174 A comparison of some of the main spatial and operational features of selected haciendas de patio in Mexico, nineteenth century 186 Monthly amounts, in cargas, of milled ore for the patio process at Regla 228 Breakdown of labour and its costs for the various refining stages carried out at Regla based on data for the four-week period ending on May 29th 1877 241 Mass balance for the patio process as practised at Regla between 1872 and 1888 246 Mass balance for the smelting of silver ores as practised at Regla between June 1875 and January 1886 247 Interpretation of Villaseñor’s working examples and method that sustained his argument against decreasing the price of mercury 259 A sampling of costs reported for the patio process and smelting in New Spain/Mexico 265 Mining and other costs for Real del Monte mines in the period 1849 to 1854 276 The percentage breakdown of the main variable costs of the patio process at Regla, excluding the cost of ore at the plant gate 278 The percentage breakdown of the main variable costs of the smelting process at Regla, excluding the cost of ore at the plant gate 281 Matrix to calculate the variable cost of refining by the patio process at Regla as a function of silver content in the ore 283 Matrix to calculate the variable cost of smelting at Regla as a function of silver content in the ore 284 Selection of historical costs of refining in New Spain / Mexico, from the sixteenth to nineteenth century, and those for Regla from 1875 to 1888 288 Values of projected refining costs of silver in the second half of the sixteenth century in New Spain 289 Values of projected refining costs of silver from mid 17c to mid 18c 292 Comparison of major refining parameters for the various haciendas of the Compañía de Minas del Real del Monte y Pachuca, in the period 1853 to 1873 (except for 1856–1858 and 1866–1868) 307 Average costs required to refine 1kg of silver using the patio and barrel processes 308 Silver production in Mexico, nineteenth century 319 Silver production by Caja, according to refining processes, over the whole colonial period in New Spain 335

xxiv xxv

xxvi xxvii b-i b-ii b-iii b-iv c-i

c-ii

c-iii

c-iv

c-v

c-vi

c-vii

c-viii

c-ix

c-x

list of illustrations Weight ratios of consumables to silver refined by the patio process (ores with 0.2% silver content) and smelting (ores with 1.9 % silver content), for New Spain 340 Estimated magnitudes of main environmental impact vectors by Caja in New Spain 341 Comparative total magnitudes of main environmental impact vectors for New Spain and nineteenth century Mexico 345 Mercury-based refining, sixteenth century context 370 Smelting, sixteenth century context 371 Patio process, seventeenth and eighteenth century context 372 Smelting, seventeenth and eighteenth century context 373 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Zacatecas 378 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Guanajuato 380 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of México 381 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Durango 382 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of San Luis Potosí 384 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Guadalajara 386 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Pachuca 388 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Sombrerete 390 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Bolaños 392 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Rosario 393

list of illustrations c-xi c-xii

xxv

Estimate of the breakdown of silver production by refining process for the Caja of Zimapán 394 Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Chihuahua 395

Guide to the Text Chemical Symbols Ag Cl Cu Fe Hg Na O Pb S Sb

silver chloride copper Iron mercury sodium oxygen lead sulphur antimony

Units of Measure kg lb oz t m m2 m3 d Ma mo. y

kilogram pound ounce ton (metric) metre square metre cubic metre day million years month year

Equivalence of Units of Measure 1 arroba = 11.5kg 1 carga = 12 arrobas = 138kg 1 fanega of salt = 55kg 1lb = 0.454kg 1 mark of silver (weight) = 8oz = 0.23kg 1 mark of silver (value) = 8.5 to 8.75 pesos

guide to the text

xxvii

1 montón (at Regla) = 10 cargas = 1,380kg 1 quintal = 46kg 1 t = 1,000kg 1 troy oz = 0.031kg 1 vara = 0.84m

Monetary Units All calculations involving monetary units of pesos, reales and tomines have been rounded off to pesos only.

Translations All translations are by the author unless otherwise indicated. Non-English words in the text and footnotes are inserted in italics, except for proper names and institutions.

Illustrations and Drawings All graphs, tables, and drawings by the author, unless otherwise indicated. Drawings in Figures 35 and 37 by Francisco Schutte. Where permission to reproduce has been requested and provided, the acknowledgement is indicated in the captions to the figures.

Cover Illustration Patio de la Hacienda de Beneficio de la Mina de Proaño, Zacatecas, by Pietro Gualdi (1840), courtesy Instituto Nacional de Antropología e Historia (inah), Museo Nacional de Historia, Ciudad de México.

Photographs Where permission to reproduce has been requested and provided, the acknowledgement is indicated in the captions to the figures. All photographs of modern day Regla are by the author, except for Figure 60, which was taken by Josue

xxviii

guide to the text

Soto, at the request of the author. Satellite images from Google Earth© are reproduced under Google guidelines.

Other An early version of the Epilogue was published by the author as the section ‘Back to the basics’ in Saul Guerrero, “The History of Silver Refining in New Spain, 16c to 19c: back to the basics,” History & Technology, 32 (2016): 20–23.

Introduction Hands and eyes judged the silver ore cradled in their fingers. Sometimes they would take a mouthful of water to squirt over the lump of rock, to bring out the darker veins against a lighter background. The luckier ones would see skeins of silver threaded around the islands of quartz, or fat globs of native metal promising the easiest of wealth. For most, whether the hands belonged to a local worker digging away at an underground vein, a woman sifting carefully through discarded tailings, a soldier judging the worth of a find, an African slave applying his artisan skills honed on iron ores or gold, a priest making good his knowledge of metallurgical texts, a German technical mercenary from MittelEuropa, or an uncouth Spanish colonist seeking the quickest route to riches, the weight of the ore would be the first indication of its nature. Then its colour, under the flicker of a candle underground, or as it splayed in myriad directions the rays of the sun, would become the guide. Each of these hands represented the first stage in a journey that would see the silver refined in the Hispanic New World provide over 70% of total world production in the seventeenth and eighteenth centuries. New Spain alone would account for more than half of the world’s new silver in the latter part of the colonial period. Republican Mexico in the nineteenth century would then supply as much refined silver as the total colonial output from New Spain over the previous two and a half centuries. None of this motley crew of refiners could have known, or would have given much thought, that they were the individual artificers of a pivotal moment in global history. Their only aim was to become rich. The very success of their aggregate efforts, the game-changing nature of the silver each one pumped into a global circuit that tied Chinese consumers to European suppliers, makes it difficult to appreciate the magnitude of the technical challenge they had to overcome. The massive scale of the silver deposits, and the isolation of their technical prowess from the European circles of knowledge during most of the colonial period in question, can lead to the doubt whether all this wealth was produced despite of, rather than thanks to, the effort of the decidedly mixed-bag of miner-refiners. Nature was extremely generous to the Spanish conquerors of the New World, but it was never inevitable that they, or any other European of the time, would be able to extract from the ores the amount of silver wealth that came to change the course of world history. The problem for the historian who wants to establish a technical empathy with the challenge faced by the refiners in the Hispanic New World, is how to avoid the impression that the end result of this effort had been a foregone

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_002

2

introduction

conclusion from the start. In the following pages I focus on their challenges, to understand why somehow they managed to get it technically right for all the wrong reasons. I aim to recreate the pragmatic decisions taken in the field, answers found in desolate places to many novel problems, without my analysis undermining the anxiety each must have felt every day they woke to a new batch of ore.1 I will reconstruct the technical landscape of this period, including its alchemical mirages, within which refiners had to find a way to apply the initial experiential knowledge brought from Europe, and to adapt by trial and error said knowledge to the massive amounts of a new type of silver ore that awaited them in the deeper levels of the mines. For much of the time they had no guide to this novel technical terrain, no blue-print or well established practice they could blindly follow. They would have been the first to scorn the idea that the path before them was clear, predictable and certain to succeed. Failure was ever present, personally and financially painful, with no safety net from the State to soften the fall. A haphazard and inefficient refining operation would never have extracted silver at a profit, nor have guaranteed a continuous output over such a long period, irrespective of normal cycles of peaks and troughs in production. Waste there would be, but if the majority of refiners in the New World had not arrived at a common set of best operational practices for extracting silver from its ore, much of the silver wealth of the New World would have remained underground, in wait for a more experienced suitor. The research presented in the following chapters began initially as an exploration of the silences in the historiography: why was Spain, and no other, to be so favoured as to dominate the world’s production of silver for more than two centuries? How could so much mercury be consumed in the New World with so little reported on cases of mercurialism among the population? Why was so much attention paid to refining with mercury, when only smelting could process the argentiferous lead ores known to be present in many mines of the New World? What made the colonial refining activity profitable, since all costs had to be covered solely by the extraction of silver, when in Europe it was the baser metals that helped the refiner make a profit from the refining of silver? Why was refining based on the use of mercury always a fringe process for European ores? What role did European centres of refining expertise play in the New World? 1 More than three hundred years after the first Spanish miners began refining silver in the New World, the same sense of uncertainty echoes from the mines of Nevada, in the u.s.a.: ‘None of those early Comstockers knew anything about silver mines, and the ablest of them could only guess what the next day’s work would disclose … [they] had no background; they had to learn by doing’, in Grant H. Smith, The History of the Comstock Lode (Reno: Nevada Bureau of Mines and Geology and University of Nevada Press, 1998), 45.

introduction

3

How was experiential knowledge shared and transmitted within the refining centres of the Hispanic New World? A major part of the challenge was to recreate the hit and miss process that led to the final success of refining routes implemented on an industrial scale. A more detailed picture emerged of how technology was forced to evolve independently in the colonial periphery, and gave birth to an autochthonous industrial architecture and practice, peerless and unknown in Europe during this Early Modern period. As a technical narrative fleshed itself out, new facets appeared. First was the opportunity to re-establish some basic facts regarding the environmental legacy of historical silver refining, since the modern emphasis on copious historic air emissions of mercury did not match the fundamental chemical nature of the processes involved, nor the historical record left by first-hand observers. In addition, in the case of New Spain and Mexico, lead and its compounds had an important role in this environmental history, given the mines of argentiferous lead around San Luis Potosí, among other important mining regions with similar ores. Furthermore, I found it possible to use a combination of written records and chemical theory to estimate quantitative levels of historical environmental impact with a precision on dates, down to a month, that could not be matched even by modern instrumental methods of soil or water sampling. This fruitful marriage of history and science also allowed me to attempt a quantitative accounting of the minimum gross magnitudes of the main environmental impact vectors from each refining process, both for New Spain and nineteenth century Mexico. Finally, the analysis of a unique set of long-term, continuous accounting records of the Hacienda de Regla, a major refining centre in newly republican Mexico, was a step forward in the understanding of how profit could be made at the many levels of silver refining activity, that could meet the spread of financial expectations that ranged from the small individual refiner to those of the industrial-scale hacienda de beneficio (silver refining centre).2

2 A useful discussion on the origin of the term and architectural evolution of the hacienda in New Spain can be found in Guadalupe Salazar González, Las haciendas en el siglo xvii en la región minera de San Luis Potosí: Su espacio, forma, función, material, significado y la estructuración regional (San Luis Potosí: Universidad Autónoma de San Luis de Potosí, 2000), 83–103. The word hacienda in Spanish still refers to ‘wealth and finances, either private or public’, so that in modern Spain and Latin America the Ministerio de Hacienda is the Ministry of Finance. It also had ‘a secondary connotation as work, activity, business … in the first usage when it is associated with a rural property, such as hacienda de minas o de beneficio [or hacienda for short], it is referring not to the rural aspect but to a productive unit’. The evolution of the original refining hacienda into ‘houses in the country that belong to the

4

introduction

The basic tool in my approach to the subject matter of this book is what I term the chemical history of silver refining, the use of chemical theory to guide the historian in the interpretation of the historical records, in the same manner social science or economics have been applied to the discipline of History. I urge the reader with little or no scientific background to bear with me during those passages where I have no option but to present the chemical foundations to the technical processes that took place. The underlying chemistry provides very solid guidelines in deconstructing the extant texts, and for the mathematical treatment of historical data. It will also lead me to quantitative conclusions both on the cost of production and on the environmental consequences of silver refining that would have been impossible otherwise. It is in many ways unorthodox, and its results are a radical departure from much of the modern historiography on this subject, which has mainly centred around the deemed ‘poor’ silver ores of the New World, and not on the inflexible consequences of their distinctive chemical composition. Most of the current mainstream narrative has posited that the richer superficial silver ores of the New World had been quickly exhausted by mid sixteenth century, and that smelting was incapable of refining at a profit the deeper ores with a low silver content. It is then argued that the implementation of mercury-based refining was therefore the only viable technical and economic choice that allowed Spain to reap the wealth of silver from the ‘other’ poorer type of ore prevalent in the New World. As a corollary to this narrative, smelting had to be relegated to a minor role in the production of silver, since ‘rich’ ores were assumed not to be the average norm in the American continent. Thus, mercury came to dominate the historical narrative on colonial silver refining, a modern yet unfiltered echo of the voices of colonial miners complaining of lower silver yields and clamouring for more and cheaper mercury as the key to silver production in the New World.3

wealthiest people, with lands, horses and sheep, as well as pastures and agricultural land’ was complete by the late eighteenth century. Original quotations from Shell and Patiño cited in ibid., 30. See also further work on these haciendas in Alejandro Galván Arellano, Arquitectura y urbanismo de la ciudad de San Luis Potosí en el siglo xvii (San Luis Potosí: Editorial Universitaria Potosina, 1999), 211–213. 3 It is difficult to single out any specific origin to this narrative, since it merges well with some of the primary documents from the colonial era. The Spanish historian Modesto Bargalló is a very useful source for extracts from early historical sources for mining and refining in New Spain and Peru. He presents in his book published in 1955 one of the first modern versions of this narrative, in Modesto Bargalló, La minería y la metalurgia en la América española durante la época colonial (México: Fondo de Cultura Económica, 1955), 240–245. The same

introduction

5

By returning to the basic chemistry involved in the refining of silver ores, I have been able to fuse many nuggets of valuable but disperse information in the historiography into a novel narrative. The thread that guides my argument is the fundamental role that geochemistry and geography played to first allow, and then shape, the power of Spain to produce silver in the New World, and the relations of knowledge and technology between the imperial core and the colonial periphery. The distinct character of a large fraction of these ores from those known in Europe, forced the development of a local path of experiential knowledge. The chemistry of the silver ores of the New World also determined the choice of technology that could be applied to refine them, the scope for profit to each refiner, and the nature and extent of the collateral environmental impact from the production of silver at every scale. Furthermore, the fact that the Spanish Empire left the refining of silver in the hands of private individuals, makes this story a challenging mix of technical hurdles overcome by many anonymous actors, and the economic consequence of a multiplicity of individual decisions as to what constituted an acceptable balance of risk and profit. I will argue that Europe never managed to appropriate the technology of silver refining that matured in the Hispanic New World, much less contribute any truly novel developments to it. The silver refining technology based on mercury that was applied to the rich silver ore deposits of the Comstock Lode in Nevada, u.s.a., in the second half of the nineteenth century was only possible thanks to the accumulated experience and developments of the colonial and then Mexican silver refining industry. Silver from the Hispanic New World shares with cotton, sugar or spices a major role in the way commodity trades accompanied the exercise of European power and empire building of the Early Modern period. However, as I will point out at certain junctures in the following narrative, it follows its own quite distinctive path, and presents interesting variations in the relationships between the distant colonial centres of production and the imperial core in Europe. Some of the issues I will draw attention to, in contrast to the other major commodities of the Early Modern period, are: a fixed geographical sourcing that could not be displaced at the will of the European powers; technology

narrative is voiced by more recent historians, for example: Peter J. Bakewell, A History of Latin America, 2nd ed. (Oxford: Blackwell Publishing, 2004), 184–189; D.A. Brading and Harry E. Cross, “Colonial Silver Mining: Mexico and Peru,” The Hispanic American Historical Review 52, no. 4 (1972): 552–556; Richard L. Garner and Spiro E. Stefanou, Economic Growth and Change in Bourbon Mexico (Gainesville: University Press of Florida, 1993), 111.

6

introduction

developed and diffused within the colonial periphery, making the European core eventually mostly irrelevant; a truly enviable instant liquidity at the colonial source; and in the case of mercury, a more balanced (though still negative) environmental impact on both sides of the Atlantic. What silver does share with the other commodities traded globally of the Early Modern period, is the hidden cost to all those involved in the production process, in terms of its impact on the environment, and the historical price paid by the indigenous communities of the New World, for the priming of the global economic pump with silver from their lands. My interest in how the technical challenges were met and overcome has not inured me from the social price paid in the New World due to the success of each technology, and I hope the quantification of the environmental impact will help future studies to establish the human cost of each process. My area of research will be geographically circumscribed to New Spain, and then to republican Mexico. This was the main source of silver from the Hispanic New World from the latter part of the seventeenth century onward, and this region saw the use of two quite distinct refining techniques in a very balanced manner during the colonial period. I will cover the period from the mid sixteenth to the end of the nineteenth century, because from a chemical and process point of view it represents a continuum, only interrupted in the late nineteenth century by the displacement of mercury by the cyanide process. To provide body and depth to my textual interpretations I have illustrated wherever possible the unique industrial architecture that was created in New Spain/Mexico for the refining of silver in major quantities. At times an image can best help drive home the undeniable risk of smoke emanating from a furnace, or the way water dominated a critical part of a process. I begin by addressing in the first chapter the most basic of questions in the mind of a reader approaching this subject for the first time: why Spain? What confluence of geological factors singled out Spain amongst all the ocean trading nations of Europe, to gift its mines in the New World a commanding presence in the silver market for nearly three hundred years? What was the difference between the new silver deposits of the New World and the deposits of Europe? Once the geochemical characteristics on each side of the Atlantic are established, I proceed in chapters 2 to 5 to follow the possible trails of experiential knowledge of silver refining, from Mitteleuropa and Venice to the first mining ventures in the Americas. Once the seeds of the know-how of smelting and the amalgamation of gold were planted in the New World, the latter would evolve independently, to become an industrial scale patio process that would only be displaced three centuries later by the cyanide process. It would in turn create a new industrial architecture based on very efficient milling

introduction

7

equipment that fed batch chemical reactors of great processing flexibility, the hacienda patios. An important key to interpret the historical trail of silver refining was found in the impressive architecture and extant production records of the Hacienda de Santa María de Regla on the outskirts of Pachuca (modern state of Hidalgo, Mexico). Chapters 6 and 7 provide a close-up view of the historical operation of this major silver refining hacienda, as interpreted through the mass balance of materials entering and leaving its walls. They serve to illustrate the operational care required in the running of the process, the integrated handling of inventories, recipe, processing times and careful recycling of mercury, without which it would have been impossible to refine at a profit the steady flow of silver from the New World. These accounts also offer an alternate method to calculate the historical magnitude of the environmental impact of its operations. In Chapter 8, the economic data from Regla give a very rare opportunity to analyse in depth the comparative production costs and structure of the patio process and smelting. In turn these insights are used to explain why refiners in New Spain could have applied in a profitable way both the patio process and smelting, subject only to the nature of the silver ore being refined. The results clarify the role of production costs in the choice of refining process by a refiner, since chemistry was always the first gatekeeper to the two quite different paths of refining. The mass ratios for the main environmental impact vectors, calculated for each of the processes, complemented by the data on silver produced and mercury sold obtained from primary and secondary sources, are the basis for the calculations in Chapter 9. They are used to project over the main mining districts of New Spain, or in total over Mexico, a gross estimate for the mass balances of the major waste products issued to the environment as a result of the historical refining of silver. The different chemical compositions of the ores being refined lead to totally different environmental impact vectors, thus highlighting the importance of the geochemistry of the silver ore deposits to the environmental history of silver refining in the New World. The results propose a major departure from the narrative that until now has dominated the studies on this topic. The projections indicate it was lead compounds from smelting, not mercury from the patio process, that was the only heavy metal to be issued to the air in major quantities. Calomel (mercurous chloride) that exited as solid waste was the main reason behind the consumption of mercury during the patio process. I will argue that the chemical and physical nature of the refining reactions mitigated the impact of mercury on the workers and communities, while exacerbating the environmental impact of lead and its compounds. Human choices were made in full knowledge of the toxicity of mercury, and only much later of lead, and the final environmental consequences

8

introduction

owed less to human foresight than to a fortunate attenuation of effects, due to the balance of refining processes influenced both by geochemistry and fiscal reasons. Finally, a word of caution. My conclusions apply only to the refining stage of silver ores in New Spain and Mexico, though some of my general comments also apply to refining of silver ore carried out in Upper Peru. The history of colonial mining of silver ore, and the production of mercury from cinnabar, together with their environmental consequences, lie completely outside the scope of this work.

chapter 1

The Genesis and Nature of Silver Ores It seemeth to me a thinge undecent to reade so much of golde and sylver and to knowe so lyttle or nothinge of the naturall generation thereof. richard eden (1555)

∵ Why Spain? The immediate answer springs from Figure 1, the stark contrast between the huddle of major primary silver deposits close to the Pacific coast of the New World, and their near total absence within the huge landmass of Eurasia and Africa.1 This geological concentration of silver in the areas of the New World first conquered by Spain had two consequences. It excluded de facto all other European powers who came to carve out colonies in the Americas from accessing these deposits (Figure 2). It also explains why China and India never had a major domestic source of silver, so that their demand for this precious metal could only be satisfied by trade. It was therefore not only the case that Spain, and through it the rest of the European economies, benefited from a ‘gift’ of silver, as Pomeranz has described it.2 Spain, by pure chance, had effectively shut out the rest of the world from directly accessing a new major deposit of silver, for the next two and a half centuries. Of the European maritime powers of the Early Modern period, Portugal’s name would be mostly associated with African and Brazilian gold, whilst neither England nor the Dutch Republic ever came across a major deposit of silver.3 Outside of Europe, neither the Chinese,

1 Richard H. Sillitoe, “Supergene Silver Enrichment Reassessed,” in Supergene Environments, Processes and Products, ed. Spencer R. Titley (Society of Economic Geologists, 2009), Special Publication 14, 16. 2 Kenneth Pomeranz, The Great Divergence. China, Europe, and the Making of the Modern World Economy (Princeton, New Jersey: Princeton University Press, 2000), 269–274. 3 Gold would be found on both sides of the Tordesillas Line, though geologists have remarked on the fact that silver and gold deposits tend to be dominated by one or the other: Frederick T. Graybeal, Douglas M. Smith, and Peter G. Vikre, “The Geology of Silver Deposits,” in

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_003

10

chapter 1

Ottoman, Russian or Mughal Empires would have the opportunity the New World offered to the Spanish Empire. Only the luck of the draw kept France from joining Spain as a major producer of silver of the Early Modern period. Its colonists of New France were tantalisingly within reach of a source of silver in the Cobalt area of modern Ontario, that equalled over 80 % of the total silver extracted from Potosí by Spain.4 By conquering the mountainous spine of the New World, Spain would have a unique access to a source of silver of a magnitude beyond the powers of imagination of any miner of the fifteenth century. The silver flows from the Hispanic New World would swamp production from Europe, and only Japan for a few decades at the turn of the sixteenth century would offer a source of supply comparable to the American mines.5 Not all the potential silver of the New World became available at the time: the major silver deposits in the region of Cobalt (present day Canada) and of Nevada (present day u.s.a.) within the North American Cordillera would remain hidden from the European powers, until their discovery in the late nineteenth and early twentieth century. There Handbook of Strata-Bound and Stratiform Ore Deposits. Part iv, ed. K.H. Wolf (Amsterdam: Elsevier 1986), 159; Walter Pohl, Economic Geology: Principles and Practice: Metals, Minerals, Coal and Hydrocarbons—Introduction to Formation and Sustainable Exploitation of Mineral Deposits (Chichester, West Sussex; Hoboken, nj: Wiley-Blackwell, 2011), 227. Colombia, Brazil and Venezuela are known more for their gold than for their silver. 4 Over 600 million ounces of silver was produced from mines around the Cobalt area in modern Ontario, according to Graybeal, Smith, and Vikre, “Geology Silver Deposits,” 15. This amount represents approximately 80% of the silver mined from Potosí up to the end of the eighteenth century (22,170 t), as reported in John Jay TePaske and Kendall W. Brown, A New World of Gold and Silver (Leiden; Boston: Brill, 2010), 184. In monetary terms this would have provided a windfall of over 4,000 million livres to the French economy of the Early Modern period. Annual values of sugar arriving in France from its Caribbean plantations went from 15 to 75 million livres per year from 1730 to 1790 (Robert Louis Stein, The French Sugar Business in the Eighteenth Century (Baton Rouge: Louisiana State University Press, 1988), 103), so just the silver value of these deposits represented around 60 years of sugar imports from the French Caribbean to France at the highest range of sugar prices. If the French mining and refining establishment of the eighteenth century had the expertise to mine and smelt the Cobalt silver ores, assisted by the ample fuel sources of New France, the impact of this silver throughout the French colonial economy of the New World, and on the overall French strategy towards its empire in the Americas, would have been substantial. The deposits would only be discovered in 1903. 5 Between 1520 and 1780, just New Spain would pour into the global economy an equivalent amount of silver to the total European stocks of this metal accumulated to the year 1500, according to data in TePaske and Brown, Gold and Silver, 113; Andre Gunder Frank, ReOrient: Global Economy in the Asian Age (Berkeley: University of California Press, 1998), 142.

figure 1

Location of the world’s 40 major silver deposits, as discovered until the year 2009 illustration reproduction from footnote 1, courtesy of the society of economic geologists

the genesis and nature of silver ores

11

12

figure 2

chapter 1

Approximate spheres of European colonial presence in the New World, and the location of all major silver and mercury sources exploited in the Early Modern period

is a sense of wonder at the unique triangulation of geological gifts that resulted from the trans-oceanic voyages due west of the Spanish mainland. These would result in a Spanish geopolitical sphere of power that encompassed all the resources it would require, for the refining of approximately 85,000 metric tons (t) of silver for nearly three centuries.6 The chemical context that defined the technical options for refining silver ores in the New World, and consequently the cost of their production and its environmental history, belongs to the longest of the Braudelian cycles, that of geological time. Geology did not determine European imperial policy, but it did favour the imperial ambitions of the Spaniards, who stumbled across the vast silver deposits of the Andes and New Spain, to complement their own vast deposits of mercury of Almadén in Spain, and later Huancavelica in Upper Peru, together with all the salt (sodium chloride), lead and copper ores that were indispensable for the refining of silver ores. It is geology that can address Richard Eden’s complaint, and in the process explain the imbalance of the world’s distribution of silver deposits.7

6 TePaske and Brown, Gold and Silver, 78. 7 Richard Eden (1555) as quoted by Cyril Stanley Smith in his introduction to Vannoccio Biringuccio, The Pirotechnia, trans. Cyril Stanley Smith and Martha Teach Gnudi (Cambridge, Mass.: The m.i.t. Press, 1966), xxii. Richard Eden translated into English Biringuccio’s metallurgical text.

the genesis and nature of silver ores

figure 3

13

Black smoker photograph by chris german and the jason group, courtesy woods hole oceanographic institution

This narrative thus begins in the depths of the Pacific Ocean, where the mid-ocean ridges are to be found, extended gashes on the sea-floor where magma rises from the deeper layers of the Earth’s mantle, to solidify and create the tectonic plate of the ocean floor. In the vicinity of these ridges metal rich solutions are spewed into the ocean from conical ‘black smokers’ (Figure 3), and precipitated minerals fall on the newly created oceanic crust that creeps slowly away from the ridge, as it is pushed aside by new pulses of emerging magma.8 When the eastern portion of this tectonic plate in perpetual motion reaches the continental crust of South America, instead of a headon-collision, a subduction of the oceanic crust takes place at the site of the ocean trench, as it slides under the thicker continental crust, a very slow motion version of the last step of a mechanical escalator sliding continuously under the more static landing stage (Figure 4). In so doing it entrains with it all the metallic deposits vented onto its surface. As it slides back into the mantle, complex geothermal mechanisms liberate sequentially its entrained

8 L.J. Robb, Introduction to Ore-Forming Processes (Malden, ma: Blackwell Pub., 2005), 179.

14

chapter 1

figure 4

The spreading ocean floor subducts under the continental crust of the Americas illustration reproduction of earth courtesy google earth© data sio, noaa, u.s. navy, nga, gebco. depiction of subducting plate based on sillitoe, footnote 9

metals.9 Hydrothermal processes will play a major role, through the action of saline water to transport metals under the Earth in concentrated solutions, akin to the strong metal-rich soup of a hot mineral spring. Epithermal is the term used for the resulting hydrothermal ore deposits formed at shallow depths (less than 1,500 meters) where mining can take place.10 In Fortey’s powerful and succinct summation, whenever the Earth moves, metals are indeed concentrated.11

9

10 11

R.H. Sillitoe, “Relation of Metal Provinces in Western America to Subduction of Oceanic Lithosphere,” Geological Society of America Bulletin 83, no. 3 (1972). See also B. Lehmann, A. Dietrich, and A. Wallianos, “From Rocks to Ore,” International Journal of Earth Sciences 89, no. 2 (2000): 287–292. Robb, Ore-Forming, 7. There is an interesting account of the paradigm change brought about by plate tectonics, and a discussion in terms accessible to the non-geologist as to why deposits of silver arise from movements within the Earth, in Richard A. Fortey, Earth: An Intimate History (London: Harper Perennial, 2005), 151–192, 253.

the genesis and nature of silver ores

15

What I have summarised above for the section of the Andes where the major silver deposits are found (Potosí, Oruro), is representative of the action of tectonic plates under the Pacific Ocean subducting over millions of years against the western seaboard of the Americas (Figure 5).12 In the Pacific coast of the Americas, subduction started in the Mesozoic [201 to 66 Ma] and early and middle Cenozoic [66 to 2.5 Ma] and is still very much active beneath Central and South America.13 The Andes have been described by modern geologists as the single most important concentration of metallic ore deposits to be found in the world. According to plate theory, the reason for this distinction is the ‘singular longevity of the convergent plate boundary of the eastern Pacific rim’, so that even as these words are being read, a section of oceanic crust that was formed some 55 million years ago at the mid-ocean ridge is now sliding continuously under the South American plate in an on-going process of subduction.14 Nowhere else on earth has such persistent and active subduction occurred so close to a large and inhabitable landmass.15 Nowhere else on the planet is there a similar concentration of world-class silver deposits readily accessible from so many centres of population. Long-term and active subduction is what sets apart the silver mines of the New World from the virtual 12

13

14

15

William R. Dickinson, “Anatomy and Global Context of the North American Cordillera,” in Backbone of the Americas: Shallow Subduction, Plateau Uplift, and Ridge and Terrane Collision, ed. Suzanne Mahlburg Kay, Víctor A. Ramos, and William R. Dickinson (Boulder, Colo.: Geological Society of America, 2009), 2. Sillitoe, “Metal Provinces in Western America,” 815. Ages in parenthesis throughout are expressed in million years (Ma) and correspond to the classification as published in the 2011 version of the International Stratigraphic Chart (www.stratigraphy.org). Pohl, Economic Geology, 222; A.H.G. Mitchell and M.S. Garson, Mineral Deposits and Global Tectonic Settings (London; New York: Academic Press, 1981), 186; F.T. Graybeal and D.M. Smith Jr, “Regional Distribution of Silver Deposits on the Pacific Rim,” in Silver— Exploration, Mining and Treatment (London: Institution of Mining and Metallurgy, 1988), 7. The cornucopia of New World metals, including silver, ‘may’ be related to its proximity to the East Pacific Rise, as metals keep being added to the spreading oceanic crust. ‘The whole Eastern Pacific continental margin arcs from British Columbia to Chile are well endowed with major silver deposits (deposits in which silver forms over 50 % of the value of the ore) while the Western Pacific continental margin has virtually none’. Anthony M. Evans, Ore Geology and Industrial Minerals: An Introduction (Oxford; Boston: Blackwell Scientific Publications, 1992), 331. According to Graybeal and Smith, the ‘enrichment of silver in young ocean crust at spreading centres’ would explain why the older subducting crust of the western areas are silver-poor in comparison to the younger crust subducting in the eastern areas of the Pacific Rim. Graybeal and Smith Jr, “Regional Distribution of Silver Deposits on the Pacific Rim,” 3,7.

16

chapter 1

dearth of silver in other continents, as observed in Figure 1. Subduction of the floor of the Pacific Ocean is the geological link that binds the history of silver in Japan with the history of silver from the Hispanic New World, and explains the emergence of the United States of America as the major exporter of silver in the late nineteenth century. Nine of the ten largest sources of primary silver ore known to humankind (those where at least 50% of their revenue is or was derived from the production of silver) lie on the mountainous spine of the New World.16 The combined silver endowment of Mexico (smaller now than New Spain) and modern Peru / Bolivia (historical Alto Perú, to be referred to in this text as Upper Peru) continues to be the greatest known deposit of silver on Earth, with a total amount of silver (produced to date and remaining) to the year 1994 estimated at approximately 440,000 t.17 Little did Charles v suspect there was no hyperbole in the motto granted under his reign to Potosí: ‘I am the rich Potosí, treasure of the world, Lord of all the mountains and the envy of the Kings’.18 At present Potosí remains what Laznicka terms the world’s ‘largest “Ag-supergiant” ’ ore reserve, a distinction all the more remarkable since the search for new deposits has continued since the sixteenth century, with the support of ever increasing technical sophistication.19 The Spanish employed the word bonanza to denote a ‘spectacularly rich precious metal zone’, but it could as well describe the whole geological panorama that opened up to them in the New World.20

16

17 18

19

20

Graybeal, Smith, and Vikre, “Geology Silver Deposits,” 2–32. Their listing published in 1989 includes only deposits that had yielded over one million ounces to that date. At present the majority of silver produced is a by-product from the mining and refining of other metals (for example, copper), which explains why the Lubin Kupferschiefer district in Poland is now listed as the single largest deposit of silver, even greater than Potosí, even though the silver content of the ores is a paltry 4 grams per ton, Peter Laznicka, Giant Metallic Deposits: Future Sources of Industrial Metals (Berlin: Springer, 2006), 54. D.A. Singer, “World Class Base and Precious Metal Deposits; A Quantitative Analysis,” Economic Geology 90, no. 1 (1995): 91. Table 2. ‘Soy el rico Potosí, del mundo soy el tesoro, el rey de todos los montes y la envidia de los reyes’ quoted in Pedro Cunill Grau, “El paisaje andino: Punas, Salares y Cerros,” in Potosí: plata para Europa, ed. José Villa Rodríguez (Sevilla: Universidad de Sevilla, 2000), 81. Laznicka, Giant Metallic Deposits, 144. The terms giant and supergiant are based on a relative scale determined by the scarcity of each metal and the size of the endowment in each deposit, thus the more scarce a metal the lower the threshold in endowment size required to classify it as a giant or supergiant deposit. See ibid., 38–54. Graybeal, Smith, and Vikre, “Geology Silver Deposits,” 1. According to the Diccionario de la lengua española (http://www.rae.es), bonanza was a Spanish term originally applied to sailing, and denoted a calm (favourable) sea.

figure 5

Subduction is the underlying link between the unique geographical concentrations of silver along the Andes, North American Cordillera and Japan illustration reproduction from footnote 12, courtesy of the geological society of america

the genesis and nature of silver ores

17

18

chapter 1

It has been estimated that of the total 337 ‘giant metal deposits’ discovered until now, only 13 were known prior to 1492, and only an additional 12 would be found up to the year 1800.21 Spain would find three of the new dozen in the first fifty years after starting their conquest of the New World, including the two major epithermal deposits of primary silver ore. A total of six of the top ten major silver ore deposits known to man would come to be under the control of the Spanish Crown for a space of some 250 years. The scale of the bonanza found in the Spanish silver mines in the New World is highlighted by the fact that Pachuca [in present day Mexico] ‘is the world’s second largest epithermal silver accumulation after Potosí’.22 In its westward thrust of conquest and expansion, Spain had stumbled without knowing across ‘the Andean and Cordilleran orogens of the western Americas [that] contain the greatest concentration of metals on Earth, and are pervasively mineralised from one end to another’.23

To Have and Have Not To have had the luck to come across this unique concentration of major silver deposits was a necessary but not sufficient condition to guarantee the extraction of that wealth, as the case of gold shows. The production and export of gold by the Spaniards from the Caribbean islands have been interpreted as being somehow deficient: The search for sources of both metals [gold and silver] carried the Spaniards far and wide across the Americas … on the promise of gold they first settled the Caribbean; finding little in the islands they were lured on by golden visions to the Isthmus.24

21

22 23 24

Peter Laznicka, “Discovery of Giant Metal Deposits and Districts,” in Proceedings of the 30th International Geology Congress: Energy and Mineral Resources for 21th Century: Geology of Mineral Economics ed. Pei Rongfu (Utrecht: vsp, 1997), 356–357. Giant Metallic Deposits, 137. Robb, Ore-Forming, 338. See also Evans, Ore Geology, 331. Peter J. Bakewell, “Mining in Colonial Spanish America,” in The Cambridge History of Latin America, ed. Leslie Bethell (Cambridge: Cambridge University Press, 1984), 108; ‘Columbus first established bases on the islands … though they yielded … but little gold’ in Leslie Aitchison, History of Metals (London: Macdonald & Evans, 1960), 360.

the genesis and nature of silver ores

19

However, the hard data indicate otherwise. Gold exports just from Hispaniola between 1492 and 1555 reached a total of 23.4 t, and in the peak decade of 1501 to 1510 they averaged over 1.3 t per year.25 This represents roughly half the total annual gold produced by amalgamation from Rhineland gold workings in Europe from 1460 to 1485.26 Columbus had not exaggerated on the wealth of gold he found on Hispaniola, equivalent to any single European source of gold known up to that time. The same geological deposits of gold found by Columbus continue to place the modern-day Dominican Republic seventh among the top ten gold sources of the world, and third as to yearly production of gold in 2014 at Pueblo Viejo.27 This vast wealth of gold that Spanish efforts could not extract from Hispaniola, indicate the limits to the mining and refining technology brought from Europe by Columbus to the New World. The size of a deposit was never a guarantee that its wealth could be tapped in a technical and profitable way in every historical period. What happened to the gold of Hispaniola could have happened to the major deposits of silver found by Spain in the New World: after all, it was the same level of refining technology and skills. The unextracted gold of Hispaniola is a timely reminder of the scale of the technical challenge of unlocking the silver wealth of the New World. To understand how experiential knowledge initially imported from Europe managed to unlock this wealth, contrary to the case of Hispaniola’s gold, I need to begin by explaining the difference in the chemistry of the silver ores on both sides of the Atlantic.

Old World Silver Ores The silver ore of the New World was to be refined initially thanks to the accumulated experience of Europe and Asia of mining and metallurgy, dating back at least two thousand years.28 In Europe, subduction processes had given rise to the Variscan orogen, the geological process of rock deformation and mountain building that took place during the Carboniferous to Permian, 359 to 252 Ma. It gave rise to silver-bearing polymetallic ore deposits stretching

25 26 27

28

TePaske and Brown, Gold and Silver, 33, 56. Ian Blanchard, Mining, Metallurgy and Minting in the Middle Ages. Continuing Afro-American Supremacy 1250–1450. Vol. 3 (Munich: Franz Steiner Verlag Stuttgart, 2005), 1160. Data from Global Gold Mines & Deposits 2012 Ranking, published by Natural Resource Holdings, http://www.nrh.co.il/i/pdf/NRH_Research_2012%20World_Gold_Deposits.pdf. Data on yearly production from http://www.mining.com/the-worlds-top-10-gold-mines/. An overview of European experience and results is presented in Blanchard, Mining, Metallurgy and Minting in the Middle Ages.

20

chapter 1

from Devon/Cornwall to Spain and the modern day Czech Republic.29 The first major difference on both sides of the Atlantic is therefore age: the European silver ores were deposited over 250 Ma ago while the silver ores in the mines of the New World are as young as 12 Ma.30 The significantly different metallogenic epochs led to a fundamental change in the way silver ore deposits were generated, thus in their chemical make-up. This is a critical difference that sustains some of the main arguments of this work, so I will quote at length on this topic: The principal historic source of silver in Precambrian and Paleozoic mineral deposits [European ores] has been as a byproduct or coproduct of base metal ores. Most of this silver was concentrated by syngenetic [concurrent] or diagenetic [transformation in time of existing deposits] processes … the principal historic source of silver in Mesozoic and Cenozoic mineral deposits [New World] has been as a coproduct or major economic component of ores [emphasis added].31 In other words, in Europe the main silver source known to generations of miners and refiners up to the early sixteenth century were ores in which silver was secondary to metals such as lead or copper. The first silver deposits to be exploited in Europe were argentiferous galena (lead sulphide). The silver content can range from 0.01% to over 1%.32 These are the lead-silver sources that funded the Athenian Empire, provided the silver and lead of Rome, and from the Middle Ages the silver of the Harz and some of the Erzgebirge mines of

29 30

31

32

‘Variscan and Hercyninan orogenies are essentially synonymous terms’ Robb, Ore-Forming, 334–338. There is an interesting graph that shows the clear division in time between the formation of the silver deposits at Erzgebirge and those of the Americas in Graybeal, Smith, and Vikre, “Geology Silver Deposits,” 39. Ibid., 163–164. ‘It’s not just the large age difference between European and western American silver deposits that dictates their mineralogical differences; it’s also their different formational conditions: the European veins were formed relatively deeply (mesothermal) whereas in the western Americas many deposits are shallowly formed (epithermal). The greater age of the European veins provided the time required to exhume them. Also the humid European climate during the late Tertiary and Quaternary resulted in relatively rapid erosion rates and shallow water tables, which combined to limit the depth of supergene profiles. In contrast, the arid and semi-arid conditions in much of western North and South America slowed erosion, favoured deep water tables and, consequently, thick supergene profiles rich in silver halides.’ Richard Sillitoe, private communication. Pohl, Economic Geology, 195.

the genesis and nature of silver ores

21

Germany, of Kutná-Hora and the lead and silver of England.33 Lead would gift to European silver refiners an in-built key for the extraction of metallic silver via smelting, but lead would also place a barrier to any attempt to use mercury in the refining of these ores. When from the fifteenth century the argentiferous lead ores of Europe began to be exhausted or required deeper mines subject to flooding, it was the expertise in a metallurgy based on lead smelting that was now adapted to extract secondary silver from copper ores found at Erzgebirge (Joachimstahl), in the Tyrol (Schwaz) and in Hungary (Neusohl). This was the new generation of silver bearing ores that would make the Fuggers an extremely rich banking family based on their large-scale approach to refining operations.34 In both cases silver is not the primary economic target in the ore, in total contrast to the silver-bearing ores found in the Spanish mines of the New World. The metallurgy of silver in Europe cut its teeth on the smelting of argentiferous lead and copper ores. Figure 6 situates the main silver historic mining areas within Europe. The earliest mining in Central Europe took place in the Harz region around Goslar, including Rammelsberg and Freiberg, known for their argentiferous galena as source of its silver. The Erzgebirge, the Ore Mountain region straddling Germany and the modern day Czech Republic, rose to prominence as a major source of silver rich copper ores such as made Joachimstahl (present day Jachymov) famous. The English deposits of Devon and Cornwall have been more important for tin and lead than for silver but played a historic role in supplying lead to the silver smelters of Europe and New Spain.35 The argentiferous copper ores of the Hungarian mines at Neusohl would supply the major silver and copper refining centres units set up by the Fuggers both at Neusohl

33

34 35

For example, in the sixteenth century Agricola wrote: ‘In the same region is found Goslar, where one finds so much galena from which lead is extracted that one could say the whole mountain is made of galena’ Georgius Agricola, Bermannus, trans. Robert Halleux and Albert Yans (Paris: Belles Lettres, 1990), 18. All translations by author unless otherwise stated. J.U. Nef, “Silver Production in Central Europe, 1450–1618,” The Journal of Political Economy (1941): 578–585. Mitchell and Garson, Global Tectonic Settings, 280; Ian Blanchard, “England and the International Bullion Crisis of the 1550s,” in Precious Metals in the Age of Expansion: Papers of the xivth International Congress of the Historical Sciences ed. Hermann Kellenbenz (Stuttgart: Klett-Cotta, 1981), 87–93. For silver mining in England see Stephen Rippon, Peter Claughton, and Chris Smart, Mining in a Medieval Landscape: The Royal Silver Mines of the Tamar Valley (Exeter: Univ of Exeter Press, 2009).

22

chapter 1

(Hungary) and also at Vilbach to process the ores from the Tyrol region.36 The saigerhütten indicated in the map were the refining centres where the saigerprocess, the use of liquefied lead to extract silver from argentiferous copper, was applied. Finally, the Rio Tinto and Guadalcanal mines in Spain were the sources of jarositic and argentiferous galena ores, refined by smelting.37 The map is important because the history of silver refining in New Spain, and then Mexico, is woven from threads of technical experience spooled from each of these areas. German smelting know-how based on lead would migrate from the mines of MittelEuropa to New Spain in the early sixteenth century, as I will narrate in Chapter 2. Venice will play a pioneering role in the development of the mercury-based refining process applied to silver ores, as I will explain in Chapter 4. The Fuggers will lend unwillingly their considerable wealth obtained from smelting European argentiferous copper ores to the initial supply of mercury to New Spain, as I will argue in Chapter 8. Cornish miners expert in dressing tin ores and English smelters raised on lead ores will migrate to Mexico in the nineteenth century, where I will catch up with them in Chapter 6.

New World Silver Ores There is a distinct geological genesis to the silver ores in the Andes compared to those of New Spain, even if both are the ultimate product of subduction processes. It has long been evident that the two silver refining industries proceeded along different historical lines, and the cause has been attributed to factors such as differences in the rate of exhaustion of the mines, the supply of mercury, the structure of capital and labour, the tax burden imposed upon refiners, and the relative efficiency of each structure of production.38 Here I would like to draw attention to the difference in the mix of the geochemistry of the silver ores at each location. In geographical terms the silver in the Andes has been concentrated in few and large deposits at very high altitude, such as Potosí. In the case of New Spain, the geographical dispersion of silver deposits 36 37

38

Nef, “Silver in Europe,” 584. ‘The jarositic silver ores of Rio Tinto (Spain) have a supergene origin, in common with the shallow ores of the western Americas, but the silver there is in solid solution in jarosite (the mineral argentojarosite) rather than being present mainly as halides or in native form. This class of silver ore remains refractory to this day so I suspect the Romans must have recovered minor associated native silver’. Richard Sillitoe, private communication. Brading and Cross, “Colonial Silver Mining: Mexico and Peru.”

figure 6

Location of main historic silver mining regions in Europe by mid sixteenth century adapted from blanchard in footnote 35. inset not to scale

the genesis and nature of silver ores

23

24

chapter 1

is much more pronounced, and none has reached at a single location the magnitude of Potosí. As to the geological difference between the two: It should be noted that the Bolivian silver deposits are parts of tin systems formed in a back-arc setting by chemically reduced magmas dominated by a crustal source. In contrast, the main Mexican silver deposits are accompanied by lead, zinc and subordinate gold and are associated with chemically oxidized magmas of combined mantle plus crustal origin that were formed in an arc setting above a shallowly inclined subduction zone. Hence, the two regions differ substantially in both their tectonic and magmatic settings [emphasis added].39 There is yet no work that relates directly the geological difference of both areas to the geographical concentration, dispersal and size of silver deposits. However, the greater probability of finding both lead and gold together with silver in the deposits of New Spain (Mexico) compared to the Andean locations is extremely relevant in the light of the importance of smelting in New Spain compared to the Andes. The presence of lead is a necessary condition for smelting, and the added cash flow from gold would assist in making this process economically viable. I will return to both issues in the chapters that follow. It is virtually impossible to obtain a wide sampling of the ores found by Europeans at each of the mining sites developed during the first centuries of silver mining in the New World.40 In one location even the historical mines have been physically erased from the face of the earth by modern day mining techniques in search of parts per million of gold.41 The problem is compounded

39

40

41

Richard Sillitoe, private communication. The mention of crustal or mantle sources of magma indicate the origin of the magma from where metals and their compounds are ultimately leached from by hydrothermal processes. Chemical reduction or oxidation will determine the final chemical profile of these metals and compounds. Subduction processes of a different nature are indicated, back-arc setting (inland side of the Andes), or shallowly inclined subduction zones in the case of Mexico. ‘Almost no specimens of silver minerals survive from the original mines [of Potosí]’. T.C. Wallace, M. Barton, and W.E. Wilson, “Silver & Silver-Bearing Minerals,” Rocks & Minerals 69, no. 1 (1994): 35. While there are reports of samples collected and sent back to Spain from the New World, the fate of these samples is unknown to the author. The whole Cerro San Pedro on the outskirts of San Luis Potosí is at present being levelled to the ground through open-pit mining operations by the Canadian mining company New Gold. The historical mines dating from the late sixteenth century onwards have disappeared literally into thin air.

the genesis and nature of silver ores

25

by the fact that silver is extracted from a greater number of different ores than any other metal: over 200 varieties of silver ore have been reported.42 The complexity of the multiple sources of silver is captured in its bewildering visual array by the early historical observers: it is common to find some [silver] clean and purified, that does not need to be refined … sometimes as glitter; other times, wrapped around a stone like a thin piece of string made of fine silver … it is a marvellous thing to see in how many ores it is nurtured … because some are black; others, yellow, dark-brown, light-brown, light-coloured or of many shades of colour; some, extremely hard and thus stubborn, and others soft, tender … some ores are earthy, others leaden, others are laced with iron pyrites; and others are mixed with gold, copper, tin, lead, caparrosa [ferrous or copper sulphate]; to sum it up, there is hardly any [silver] that can be found that is not mixed in various ways.43 In the absence of physical evidence it is necessary to reconstruct the nature of the deposits found by the first Spanish miners, using the current state of knowledge on the chemical transformations that are known to take place in deposits of silver ore. I will expressly avoid addressing each of the historical silver deposits of New Spain in the formal terminology of geology that categorises ore deposits of different genetic types according to all the metals present, the host rocks that ultimately make up the gangue or waste mineral material, the mechanisms by which these metal compounds were deposited in the host rocks, and even by the environmental consequences of the chemical compounds they contain. This would require detailed historical knowledge of the geological genetic type of silver ore deposit for each mining district (Real de Minas) of New Spain, which according to Humboldt numbered around 500 in the 1800s.44 On the contrary, I will base my analysis of colonial silver refining on the one factor common to all the Reales, that the silver produced in New Spain came from basically just two chemical groups of silver compounds distributed among these deposits: 1) silver sulphide compounds (simple to complex), that via

42 43 44

Claudia Gasparrini, “The Mineralogy of Silver and its Significance in Metal Extraction,” cim Bulletin 77, no. 866 (1984): 99. Bernabé Cobo (s.j.), Historia del Nuevo Mundo, ed. Francisco Mateos (s.j.), Biblioteca de Autores Españoles (Madrid: Atlas, 1964), 141. Alexander von Humboldt, Essai politique sur le royaume de la Nouvelle-Espagne (Paris: Chez F. Schoell, 1811), Tome iii, 310.

26

chapter 1

weathering as described below could give rise to surface concentrations of silver halides (mainly silver chloride) and also metallic silver and 2) argentiferous galena (lead sulphide, PbS, that contained silver compounds).45 Silver Sulphide and Silver Chloride Nearly one hundred years ago, Emmons explained that a silver ore deposit is not an inert mountain of material but a constantly evolving chemical reactor in which over time the composition of the original (hypogene) silver compounds is constantly changing.46 Because the surface layer of the deposit is a zone where oxygen plays an important role, the term ‘oxidised zone’ has led to the erroneous statement in some of the historiography that silver oxide is generated in this surface layer.47 It is not silver that is oxidised but other chemical species present in the ore deposit: the ferrous ions to ferric ions, or the sulphide ions to sulphate. Quite the opposite, any silver ion will be reduced to elemental silver, or be solubilised by the sulphate ions and later precipitated as chlorides, and other halides.48 Sillitoe has published an extensive review of the research up to the year 2009 on the changing nature of the chemical composition of silver ore deposits.49 He states there is a fundamental difference between copper deposits, where 45

46 47

48

49

In contrast to gold, the main primary source of silver was not the native metal but the chemical compound silver sulphide (Ag2S). V.M. Goldschmidt, Geochemistry (Oxford: Clarendon Press, 1954), 189. Silver is classified as a chalcophile metal, one that prefers to react with sulphur, as anyone eating a boiled egg with a silver spoon can quickly ascertain. Silver sulphide presents itself at ambient temperature as the mineral acanthite, the most common of all silver ores, so soft that it can be cut with a knife, with a colour range ‘from black to mirror white silver’. Acanthite is the low temperature (< 173 ° c) form of argentite, and in some of the literature both terms appear to be readily interchanged. Wallace, Barton, and Wilson, “Silver-Bearing Minerals,” 16–38. Acanthite would be found in most mines of New Spain and Peru in huge quantities, as the main hypogene (original) silver mineral. It can also be found as part of the silver content in galena, the sulphide of lead. William Harvey Emmons, The Enrichment of Ore Deposits (Washington: United States Geological Survey, 1917), 251ff. ‘silver oxide could not accumulate in oxidising zones, because it is soluble in acid and also [to a limited extent] in water … it is unknown as a natural mineral [of silver]’ in ibid., 255. See also Sillitoe, “Supergene Silver Enrichment,” 22. Although it is an older paper, Emmons provides detailed chemical reactions that take place in the oxidation zone, and serves as a guide to a correct interpretation of the chemical changes in this region. Emmons, The Enrichment of Ore Deposits, 252–263. Sillitoe, “Supergene Silver Enrichment,” 21–23. In this paper Sillitoe provides a more detailed geological description of major silver deposits in the world, including an estimate

the genesis and nature of silver ores

27

enrichment has been studied and confirmed even down to levels below the water table, and the behaviour of deposits of silver sulphide, which are the ones that predominate in the New World. The revisionist proposal by Sillitoe that ‘supergene sulphide enrichment is an economically unimportant process’, being absent in most examples of major extant silver deposits he reviews in his paper, leads to a conclusion of great relevance in the interpretation of the early subjective reports by miners in the New World. Because indigenous mining of silver was not a major activity prior to the Spanish conquest, the first generation of Spanish refiners in the New World found a text-book example of an undisturbed deposit of silver sulphide, in which superficial levels had undergone weathering to silver chloride and silver in the zone above the water table. The absence of supergene enrichment means that on average the silver content found at the more superficial levels is indicative of the silver content as a whole for the deposit. In other words, the major change as extraction proceeded within most mines in the New World was not so much in total silver content as in the nature of the silver compounds within which it was to be found (Figure 7).50 Not all the silver sulphide deposits or even all the veins of a major deposit necessarily undergo any process of chemical change at all. When it does take place, the first segment a Spanish miner would have found close or at the surface would have provided him initially with an ore mainly in the form of native silver and silver chloride (chlorargyrite), for the most part generated by oxidation-reduction reactions above the water table.51 As the Spanish miners continued to extract silver ores at deeper levels down to and below the water table, these would revert to the original and primary (hypogene)

50

51

of their reserves of silver. An earlier overview of silver sources based on the type of deposit in the Americas is provided in D.M. Smith Jr, “Geology of Silver Deposits along the Western Cordillera,” in Silver—Exploration, Mining and Treatment (London: Institution of Mining and Metallurgy, 1988). There is no doubt that miners came across pockets of extremely rich silver content, as in this description: ‘I have seen in the mine of Zacatecas such a rich vein of silver that, on placing it in the fire, it spit out pieces of silver the size of a broad bean’, in Agustín de Sotomayor, as quoted in Julio Sánchez Gómez, De minería, metalúrgica y comercio de metales: la minería no férrica en el Reino de Castilla, 1450–1610 (Salamanca: Universidad de Salamanca: Instituto tecnológico geominero de España, 1989), 34. Blanchard has identified silver chloride as the main silver compound that differentiates silver ores of the New World from those known in Europe up to the sixteenth century. Ian Blanchard, Russia’s “Age of Silver”. Precious-metal Production and Economic Growth in the Eighteenth Century (London; New York: Routledge, 1989), 3.

28

chapter 1

figure 7

A simplified representation of Sillitoe’s interpretation of the effect of weathering on an original (hypogene) non-weathered sulphidic silver deposit adapted from sillitoe, footnote 49

silver sulphide, native silver, and more complex sulfosalts such as pyrargyrite that contain antimony, an element that interferes in all silver refining processes.52 The earliest historical texts that reflect the mining practices of the Spanish conquerors of the sixteenth century reflect from the first moment the main classes of silver compounds indicated in the discussion based on Sillitoe’s review. A useful guide is found in the dictionary by García De Llanos from 1609, which describes three main visually distinct classes of silver ores:53 a. b.

52 53

Very rich silver ore or native metal that could be worked directly with a hammer was called respectively tacana or machacado. Ores found from the surface down to the water table were called colorados in New Spain and pacos in Peru. In some areas, the presence of reddish iron oxide-hydroxide minerals such as limonite-haematite in this zone (gossan is the term used in geology texts) or the mixture of oxidised

Sillitoe, “Supergene Silver Enrichment,” 22–30. García de Llanos, Diccionario y maneras de hablar que se usan en las minas y sus labores en los Ingenios y beneficios de los metales (1609) (La Paz, Bolivia: musef, 1983), 79–80, 86.

the genesis and nature of silver ores

c.

54

55 56 57

29

iron pyrites and enriched silver minerals found near the surface gave rise to the term of ‘coloured’ ores in Spanish. By inference these ores would be made up primarily by native silver and silver chloride, other silver halides, and some silver sulphide.54 The depth of the oxidised zone will vary according to each deposit and its climatic conditions. Burkart stated that the colorados in certain mines of New Spain reached down to 150 metres, in others the negrillos (see below) reached the surface, testimony to the fact that the degree of weathering is not necessarily the same in every silver ore deposit.55 The greater the aridity, the greater the amount of silver chloride (cerargyrite or chlorargyrite) to be expected in a deposit.56 Darker and deeper ores found above and below the water table called negros or negrillos, the silver sulphide and sulfosalt ores.57

For a description of how gossan is formed see Pohl, Economic Geology, 85. Silver chloride, AgCl, is found as the mineral cerargyrite (chlorargyrite, horn silver). Cerargyrite is very soft, white or transparent when fresh; on exposure to light, it immediately darkens and becomes opaque. Typically, specimens are brown. Wallace, Barton, and Wilson, “SilverBearing Minerals,” 28. Silver halides would allow the mines of Catorce, in the province of San Luis Potosí, to apply Alonso Barba’s cazo process on a major scale at the end of the eighteenth century, as will be explained in the following chapters. J. Burkart, “Du filon et des mines de Veta-Grande, près de la ville de Zacatecas, dans l’ état du même nom, au Mexique,” Annales des Mines, 3eme série 8 (1835): 66. Emmons, The Enrichment of Ore Deposits., 256; Pohl, Economic Geology, 222. These are the ‘blasted blue stuff’ described by the first miners of the Comstock Lode, Nevada (u.s.a.) in the late 1850s, the as-yet unrecognised major deposit of silver sulphides that was initially discarded as waste in their search for gold, as described in Dan De Quille, The Big Bonanza (London: Eyre & Spottiswoode, 1947), 19. Another sulphur containing hypogene silver mineral of commercial importance is pyrargyrite, Ag3SbS3, a sulphide containing both silver and antimony: ‘“dark ruby silver” is much more common than proustite [see below] and is an important ore of silver … a deeper red than proustite and … less sensitive to light. In mining lore, high grade pyrargyrite ore is known as “blood mining” in reference to the colour and texture of a freshly excavated face’. The presence of complex silver sulphide salts, together with other metals, would lead them to be branded as rebellious, since they did not respond in a straightforward manner to mercury-based refining or smelting. Another sulphide hypogene silver mineral is proustite, Ag3AsS3, which contains both silver and arsenic. ‘Known as the “ruby silvers” due to their translucent red color when fresh … perhaps the most vivid color in all the mineral kingdom is the scarlet-vermilion of proustite … [though] exposure to light causes proustite to darken’. Wallace, Barton, and Wilson, “Silver-Bearing Minerals,” 29.

30

chapter 1

As the knowledge of the nature of the different silver ores advanced in tandem with chemical theory, the information on silver ores in the historiography converges on the average chemical profile of a silver sulphide ore deposit as described in Sillitoe’s summary above.58 The lack of scientific knowledge of the indigenous workers at the mine was more than compensated by the skill to pursue a vein underground based on the visual and tactile evidence under the tenuous flicker of a yellow flame. However, what would have posed a greater challenge to the miner-refiner aboveground was the ability to recognise that ores could contain the same overall silver content, but require completely different levels of refining skills to extract the same amount of silver from them. The historical consequence was the waste of good silver ores from silver sulphide deposits, since the absence of adequate refining skills resulted in the complaint by Spanish miners of the apparent sudden onset of poor silver content relatively soon after mines started to be exploited.59 The miners and authorities did not have the knowledge base at the time to recognise that the real poverty lay not in the silver content of the deeper ores, but in the range of refining skills they could bring to bear on silver sulphide ores during the early years of silver production. Argentiferous Galena Argentiferous galena (lead sulphide) was the silver ore in the New World that Europeans in the sixteenth century were most familiar with. The fact that both galena and lead could be found in many places in New Spain is an important part of the history of silver refining in the region. The evidence comes from modern geological mappings of present-day Mexico. The Mexican geologist Guillermo Salas classified the locations of deposits of Mexico into six metallogenic provinces. Many of the historic silver mining and refining locations are 58

59

Humboldt identifies by name silver sulphides, horn silver (silver chloride) and antimony/arsenic compounds of silver among the main silver ores of Mexico, in Humboldt, Essai politique, Tome iii, 354–361; by mid nineteenth century silver chloride is identified as one of the main components in pacos or colorados, together with descriptions of silver and antimony sulphides, among others in Edward Pique, A Practical Treatise on the Chemistry of Gold, Silver, Quicksilver and Lead, Tracing the Crude Ores from the Mines Through the Various Mechanical and Metallurgic Elaborations, Until the Pure Metal is Obtained (San Francisco: Towne & Bacon, printers, 1860), 81–84; by the turn of the century Emmons was publishing his research on the chemistry of silver ore deposits. In Chapter 4, I will comment in detail on this issue, since the early use of mercury in a refining process made its inroads on the ore recovered from mountains of tailings. In the mid eighteenth century, women would still be combing through 200-year-old tailings in search of useful silver ore.

the genesis and nature of silver ores

figure 8

31

The main historical silver, lead and salt deposits of New Spain / Mexico (based partly on Salas, footnote 60). All locations approximate.

found in the Sierra Madre Oriental (Eastern Sierra Madre), an extension of the North American Rocky Mountains that is the link between the Cordilleras of the north to the Andes of the south, the Mesa Central and its border with the Sierra Madre Occidental (Western Sierra Madre), and to a lesser extent within the Eje Neo-Volcánico (Neovolcanic Axis). These regions are reported to contain deposits of silver (in galena or silver sulphide ores) and lead. Gold-silver and argentiferous lead deposits are to be found in Zacualpan, Sultepec and Taxco, as well as in Pachuca and Real del Monte, all areas historically known as pioneers in the mining and refining of silver. As well as the better known deposits in Cerro San Pedro of San Luis Potosí (see next chapter), other argentiferous lead deposits are reported in Guanajuato, Zacatecas and Fresnillo (see Figure 8).60 60

Guillermo P. Salas, Carta y Provincias Metalogenéticas de la República Mexicana (México: Consejo de Recursos Minerales, 1980), 45–84, 107–125. The work includes a detailed foldout map with the location of known metal deposits in Mexico. A useful summary of the classification made by Salas can be found in Atlántida Coll-Hurtado, María Teresa Sánchez-Salazar, and Josefina Morales, La minería en México: geografía, historia, economía y medio ambiente (México: unam, 2002), 18–22. A metal province has been defined as an

32

chapter 1

A Red Herring Argentiferous galena was the only familiar silver landmark in the geological landscape that confronted the European miners and refiners arriving in the New World. The geological youth of the New World silver deposits is reflected in the altitude at which they are found, compared to the historical silver deposits of Europe. It also meant that in the case of the Andean deposits, most were at or above the treeline, which would have immediate consequences on the sourcing and pricing of fuel: [Silver] is generated usually in rough and sterile lands, in páramos and punas [Andean highlands with very scarce vegetation, usually above the tree line] of severe cold, hills and snowy ranges … the most highly regarded are the mines in high mountains and places … the mountains with mines are bare, treeless, with no vegetation.61 Figure 9 visualises some of the major differences between deposits of silver on both sides of the Atlantic.62 What is significant is that silver content does not figure in it at all. The current narrative on the history of colonial silver refining

61 62

area ‘characterized by an abnormal concentration of large deposits of a particular metal or metals, by numerous occurrences of a metal, or both’. Mitchell and Garson, Global Tectonic Settings, 5. Cobo (s.j.), Historia del Nuevo Mundo, 141. Sources: treeline range (Christian Körner, “A Re-Assessment of High Elevation Treeline Positions and Their Explanation,” Oecologia 115, no. 4 (1998): 446–447); altitude Potosí (Cunill Grau, “Paisaje andino,” 79); Porco (C.G. Cunningham et al., “Relationship between the Porco, Bolivia, Ag-Zn-Pb-Sn deposit and the Porco caldera,”Economic Geology 89, no. 8 (1994): 1833); Pasco (Laznicka, Giant Metallic Deposits, 127); Oruro (José De Mesa and Teresa Gisbert, “Oruro. Origen de una villa minera” in La minería hispana e iberoamericana. Ponencias del i coloquio internacional sobre historia de la minería. (León: Cátedra de San Isidro, 1970), 560); Zacatecas (Burkart, “Mines de Veta-Grande,” 60); Real del Monte (José J. Galindo y R, El distrito minero Pachuca-Real del Monte ([Pachuca?]: Cia. de Real del Monte y Pachuca, 1957), 2.); Catorce (Rafael Montejano y Aguiñaga, El Real de Minas de la Purísima Concepción de los Catorce, slp (San Luis Potosí: Editorial Universitaria Potosina, 1993), 169); Cerro San Pedro (Álvaro Sánchez-Crispín, Eurosia Carrascal, and Alejandrina de Sicilia Muñoz, “De la minería al turismo: Real de Catorce y Cerro de San Pedro, México. Una interpretación geográfico-económica,” Revista Geográfica, no. 119 (1994): 85); Guanajuato (Yann René Ramos-Arroyo, Rosa María Prol-Ledesma, and Christina SiebeGrabach, “Características geológicas y mineralógicas e historia de extracción del Distrito de Guanajuato, México. Posibles escenarios geoquímicos para los residuos mineros,” Revista Mexicana de Ciencias Geológicas 21, no. 2 (2004): 273); Parral (Robert C. West, The

the genesis and nature of silver ores

33

has been built upon the notion that the average silver content of the ores in the New World was very poor, which in turn is claimed to have limited their refining to just one viable technique, the use of mercury to refine its silver ores.63 The major exception is Blanchard, who bases his historical analysis on the different types of ore found on either side of the Atlantic, and not on any deemed difference in silver content. He explains the use of mercury as a refining method not as a consequence of a deemed poverty of the ores but rather on their chemistry (silver halides in abundance), their lack of lead content (‘dry’ ores), together with expensive lead and the appearance of new supplies of mercury.64 The roots of the ‘poor silver content’ line of argument can be traced to the earliest years of silver refining in the New World, when for example in the 1550s in New Spain it was claimed that: ‘the ore that only had three marks [1.5% by weight] was considered poor’.65 On the other hand, early claims can also be found for a high level of silver in these ores, as stated in the early 1600s: we understand, that minerals are mined in the provinces of Germany and metals are refined from them though with little benefit, and we have been told … [that] ours have much more silver content.66

63

64 65 66

Mining Community in Northern New Spain: the Parral Mining District (Berkeley and Los Angeles: University of California Press, 1949), 9); average estimate for central European mines (Humboldt, Essai politique, Tome iii, 333). Among the most recent examples in the modern historiography are the following: ‘the vast quantities of ore with a low to medium silver content’ in Bernd Hausberger, “El universalismo científico del Barón Ignaz von Born y la transferencia de tecnología minera entre Hispano américa y Alemania a finales del siglo xviii,” Historia Mexicana 59, no. 2 (2009): 607; ‘low-yield ore, characteristic of South America’ in Mervyn F. Lang, “Silver Refining Technology in Spanish America (patio y fundición)” in 5th International Mining History Congress, ed. James E. Fell, P.D. Nicolaou, and G.D. Xydous (Milos Island: Milos Conference Center-George Eliopoulos, 2001), 140; ‘the low [silver] content of the silver deposits’, Peter J. Bakewell, “La transferencia de la tecnología y la minería hispanoamericana, siglos xvi y xvii: algunas observaciones,” in Hombres, técnica, plata: minería y sociedad en Europa y América, siglos xvi–xix, ed. Julio Sánchez Gómez, Guillermo Mira Delli-Zotti, and Francisco A. Rubio Durán (Sevilla: Aconcagua Libros, 2000), 365. Blanchard, Russia’s “Age of Silver”. Precious-metal Production and Economic Growth in the Eighteenth Century 3–4. As quoted in Henry R. Wagner, “Early Silver Mining in New Spain,” Revista de Historia de América, no. 14 (1942): 61. Marcos Jiménez de la Espada, Relaciones Geográficas de Indias—Perú, ed. José Urbano Martínez Carrera vol. ii (Madrid: Atlas, 1965), 132.

34

chapter 1

figure 9

Salient traits of the silver ore deposits under extraction in the Early Modern period in the Hispanic New World and in Europe. For sources see footnote 62.

In the early nineteenth century Humboldt stated that Mexican silver ores were deemed much richer than European ores. However, ‘it is not thus … by the intrinsic richness of the ores, but rather by the great abundance in which they are found in the ground, and by the ease of their exploitation, that distinguishes the mines of America’.67 In the late nineteenth century Burkart would claim from his own first-hand experience that Mexican silver ores were in no way inferior in silver content to European ores.68 Historical judgements 67 68

Humboldt, Essai politique, Tome iii, 371. Johann Burkart, “Memoria sobre la explotación de minas de los distritos de Pachuca y Real del Monte de México,” Anales de la Minería Mexicana (Revista de Minas) i (1861): 96–100.

the genesis and nature of silver ores

35

of this nature need to be judged with care. Even by the nineteenth century the assaying of silver ores in Mexico prior to refining was the exception and not the rule.69 This problem of the absence of analytical information does not seem confined to the New World, since Burkart complains of the difficulty of finding sufficient data to calculate the silver content of European ores.70 In addition, sampling of large ore masses was a major challenge, and silver content when reported was based on the silver that could be extracted, which was never necessarily the silver actually present in the ore.71 There are three ways to address this issue. Brading and Cross question the statements on operational grounds: various problems confront the historian who attempts to come to a closer view of colonial refining … on many occasions both miners and royal officials claim that ore levels had fallen; they then provide an average figure, let us say, of one ounce of silver per hundredweight of mineral. The historian then has to decide: did these refiners know how much silver their ores really contained?72 To their observation I would add the following question: could the inability to extract silver from an ore that has suddenly changed its chemical structure, so as to place it outside the scope of the refining method being used, have been misinterpreted as a sudden impoverishment of the silver content of the ore? Finally, what exactly is meant by a ‘low’ silver content? There is no atemporal and absolute threshold that determines when an ore is low or high in silver.73 What exists is a location-specific set of human skills, the chemical nature of the ore being refined, a total production cost as a function of silver content

69

70 71 72 73

Saint Clair Duport, De la production des métaux précieux au Mexique, considérée dans ses rapports avec la géologie, la métallurgie et l’économie politique (Paris: Firmin Didot Frères, 1843), 138–139. By mid-nineteenth century no assaying of ores was carried out in refining centres of Catorce, Sombrerete, Fresnillo, Zacatecas and Guanajuato, visited by M.P. Laur, “De la metallurgie de l’argent au Mexique,” Annales des Mines, 6th series, 20 (1871): 182. Burkart, “Memoria Real del Monte,” 92. Laur, “De la metallurgie de l’argent au Mexique,” 49. Brading and Cross, “Colonial Silver Mining: Mexico and Peru,” 555. To highlight how relative the terms ‘poor’ or ‘rich’ are in relation to the silver content of silver ores, in modern times the typical range reported for silver ore deposits is between 0.001% to 0.1%, as stated in Wallace, Barton, and Wilson, “Silver-Bearing Minerals,” 30. Until the mid-eighteenth century, ores in New Spain with a silver content just below 0.1 % were not even accepted for refining.

36

chapter 1

(with or without environmental and social consequences factored in), and a price for silver in each historical period, that determine how much silver can be extracted at a profit from a given ore. The option to market any metallic coproducts from the refining process (gold, or base metals like lead and copper), can also make a major impact on the final feasibility of the refining operation. Therefore, the only number that is relevant is the minimum value of extracted silver that is required to meet the cost of production, a value that changes with location and historical period. An orphaned number designating a silver content, devoid of context, can mislead rather than assist in deciphering the events that determined the history of silver refining in the New World. The sparse and scattered quantitative data on silver content of ores from the New World as reported in historical sources has to be interpreted with caution. First of all, the source of much of this information is sometimes unclear, whether it was hearsay, assumptions or actual assays. Second, there is no histogram available up to the end of the nineteenth century that segments total quantities of ore in New Spain or Mexico according to their silver content. Thus, the statistical significance of each data point varies and cannot be quantified. At the most, the data reported in the historiography can be grouped approximately by period, and a strong aggregation of values may be representative of the average of silver content during a certain period. Such qualified data points to average values grouped around 0.25% from the seventeenth century onwards, on both sides of the Atlantic.74 This is as much as can be said on this matter, without incurring the mistake that the silver content of ores of the New World could be compared simply on absolute value with those of the ores of Europe. As will become more evident in the subsequent chapters, a silver ore containing 0.25% of silver, together with major amounts of lead or copper in Europe, cannot be compared on the same basis as a silver ore containing 0.25 % of silver in the form of silver sulphide, with no base metal that will add revenues to the refiner, in New Spain. In the latter case, it is the value of silver alone that must provide a net profit to the refiner. The course of the history of silver refining in the New World was determined by the chemistry of the ores, and not by their deemed silver content.

74

For a collection of historical data see Saul Guerrero, “The Environmental History of Silver Refining in New Spain and Mexico, 16c to 19c: A Shift of Paradigm” (PhD diss., McGill University, 2015), 90–96.

the genesis and nature of silver ores

37

The Other Chemical Keys Silver deposits by themselves do not explain the dominance and choices exercised by Spain over silver production. Geology would also provide the Spanish Crown with the ownership of vast deposits of the three chemical substances that subsequent chapters will show were indispensable for the refining of the silver ores: mercury, salt and lead. Mercury: During the Early Modern period there were only three sources of mercury in the world for all practical purposes, of which the two most important ones would be under the direct control of the Spanish Crown, one on each side of the Atlantic (see Figure 2). Nearly 80 % of all the mercury produced in this period was owned by Spain. The mercury mine in Almadén, Spain, remains the doyen of the group even after over one thousand years of production. In Europe only Idria achieved important and sustained levels of exploitation, but always inferior to those of Almadén. Spain would find in the New World the second most important source of mercury of the world of that time at Huancavelica, in present day Peru. China never materialised as a viable supplier of mercury to the New World, in spite of repeated attempts by the Spanish authorities to find Chinese supplies.75 There is a sense of incredulity at the scale and variety of geological deposits at the service of the Spanish Crown during this period. Not only had Spain conquered exclusive access to the main silver depository on Earth, it was already in possession of a mercury deposit that even modern geologists address in mystified terms: Almadén … is a most enigmatic mineral system … it is the largest Hg [mercury] “supergiant” that stores close to 30–40 % of mercury of the world’s endowment. It is also the number 1 deposit in terms of geochemical magnitude of accumulation, of all metals. Despite this, there is no satisfactory explanation where this mercury had come from, and why it had accumulated in this geologically almost “normal” setting.76

75

76

S.M. Cargill, D.H. Root, and E.H. Bailey, “Resource Estimation from Historical Data: Mercury, A Test Case,” Mathematical Geology 12, no. 5 (1980): 492–493; Lars D. Hylander and Markus Meili, “500 Years of Mercury Production: Global Annual Inventory by Region until 2000 and Associated Emissions,” Science of The Total Environment 304, no. 1–3 (2003): 7. Laznicka, Giant Metallic Deposits, 356–357. A similar sentiment, ‘possibly the largest geochemical anomaly on the planet’, is expressed in Cris M. Hall et al., “Dating of Alteration Episodes Related to Mercury Mineralization in the Almadén district, Spain,” Earth and Planetary Science Letters 148 no. 1–2 (1997): 287.

38

chapter 1

Up to 1977 the Almadén mine had produced about one-third of the world’s mercury, with extraction commencing prior to the arrival of the Romans in Spain.77 Spain would be further awarded in the New World with the mercury mine at Huancavelica, which would provide most the supply of mercury to the refining of silver ores in Upper Peru until its exhaustion at the beginning of the nineteenth century. Salt (sodium chloride): ‘Without salt there was no silver. Without silver another would have been the history of New Spain’.78 Salt was as critical an ingredient for the refining of the silver ores in the New World as mercury. The size of the natural deposits of salt that were gifted by geology to Spain, at locations close to the main silver deposits, could only have strengthened their belief in Divine Intervention. The largest salt deposit in the Americas, Uyuni, lies in the Andes, and the other nearby salt deposit of Yocalla was harvested by the refiners of Potosí, who were thus spared the cost and logistics of bringing salt from the sea.79 While New Spain did not have a single deposit of salt of the scale of the Andean Uyuni, Peñón Blanco (Figure 8) among others satisfied the industrial demand for salt: (from the 16c to the 19c) Southern and Central Mexico relied mainly on their own inland resources, besides importing salt from either coast. Colima salt came to the altiplano via Guadalajara … sold as far away as Guanajuato … from Campeche to Veracruz [then] pack trains [to] the capital and the neighbouring mining districts. The salinas del Peñón Blanco supplied all the mines from northern Central Mexico up to Durango … and Pachuca … free market economy ruled trade.80

77 78 79

80

Cargill, Root, and Bailey, “Resource Estimation from Historical Data: Mercury, A Test Case,” 492. Juan Carlos Reyes, “Introducción,” in La sal en México ed. Juan Carlos Reyes (Colima: Universidad de Colima, Consejo Nacional para la Cultura y las Artes, 1995), vii. Cunill Grau, “Paisaje andino,” 77. The Uyuni salt deposits are clearly seen as a white patch on satellite images taken from a height of 64,000km, as appear on Google Earth© images of the central Andes. Apart from Peñón Blanco, other salt mines are mentioned as belonging to Ocotlán, Piaxtla, Chila, Tehuacán, Cuzcatlán, Zapotilo, Nuevo Santander, Colima, Yucatán coast, Ixtapan de la sal, Omitlán, Chiautla, Acatlán, among others. See Ursula Ewald, The Mexican Salt Industry, 1560–1980: A Study in Change (Stuttgart; New York: G. Fischer, 1985), 20, 202– 203; Miguel Othón Mendizábal, La minería y la metalurgia mexicanas (1520–1943) (México: Centro de Estudios Históricos del Movimiento Obrero, 1980), 80. An illustrative case study

the genesis and nature of silver ores

39

Lead: lead was present in important quantities in New Spain. The argentiferous lead mining districts are (among others) Sombrerete, Vetagrande in Zacatecas; Catorce, Matehuala in San Luis Potosí; Taxco in Guerrero; Santa Eulalia, Hidalgo del Parral in Chihuahua; Temascaltepec and Sultepec in Mexico, Zimapán close to Pachuca. Lead deposits are found in Chihuahua (Santa Eulalia among others), Coahuila (Sierra Mojada), Nuevo León (Sombrerete, Ojo Caliente) and Zacatecas.81 Mendizábal draws attention to the contrast between the colonial knowledge of the existence of lead deposits in New Spain and the lack of initiatives to develop them: Lead … was present in great abundance … as well as in argentiferous galena … it seems that lead was never the aim of any special exploitation, except in the mines of El Cardonal and Zimapán (Lomo del Toro) in the state of Hidalgo, which produced some fifteen thousand cargas annually (4,140 tons), sufficient quantity to meet the industrial demand for lead … at the end of the eighteenth century lead mines are exploited in Sultepec.82 The location of the principal lead resources in the northern provinces that took longer to pacify may have delayed their discovery (for example the deposits at Santa Eulalia were only discovered in 1704). The lead endowment of Mexico as calculated to 1994 was of 11,062,188 t, just below that of Germany at 12,150,180 t.83 Up to the Early Modern period both Germany and England had based their refining of silver ores on the use of their endogenous lead deposits and the application of smelting technology. From a geological point of view there was no lack of local deposits of lead in New Spain.

81 82 83

of the interaction between local communities, the fixing of salt prices by the Crown, and the interests of the silver refiners, is given in Margarita Menegus Borneman, “Las comunidades productoras de sal y los mercados mineros: los casos de Taxco y Temascaltepec,” in Minería regional mexicana. Primera reunión de historiadores de la minería latinoamericana, ed. Dolores Avila Herrera and Rina Ortiz (México: Instituto Nacional de Antropología e Historia, 1994), 21–31. Teodoro Flores, Yacimientos minerales de la República Mexicana, con algunos datos relativos a su producción (México: Instituto Geológico de México, 1933), 15–25. Mendizábal, La minería mexicana, 71. Singer, “Precious Metal Deposits,” 94.

40

chapter 1

The Immoveable Object and the Unstoppable Force The Early Modern period encompasses the expansion of the major European naval powers over the globe, in the pursuit and manipulation of sources of commodities. For many historians, this was the necessary and sufficient condition for the rise of a nascent capitalism, that had to be nurtured by an everexpanding geographical supply of resources and new markets.84 A tangible web of trade now tied the future of distant communities and environments to the consumer whims and needs of a militarily powerful, and technologically gifted, imperial Europe.85 Its network could be traced by following the history of the most iconic of the commodities of this period. Spices, cotton, sugar and tea, among others, became as closely identified with the imperial aspirations of the individual European maritime powers, as the flags of the ships that bore them across the world. In effect, European empires played them as chess pieces in a strategic game of power, growth, capital and profits over a global geographical board, only limited by climate, soil and the supply of labour from the infamous slave trade from Africa. Some of these commodities had been camp followers of previous expansive empires. In the case of sugar, Arabs had first spread its cultivation around the warmer regions of the Mediterranean basin, and then the centre of gravity of production shifted under Spain and Portugal to the Atlantic islands under their control, from where it would continue to their colonies in the New World. The Dutch in their flight from Brazil would then introduce sugar plantations in the Caribbean, to be joined by the English and French in the early seventeenth century.86 Cotton mirrors the geographical displacement shown by sugar. European manufacturers had required increasing supplies of cotton and searched for their own sources, so that the most southerly of the Portuguese Atlantic islands (Cape Verde) became production centres by the mid-sixteenth century. Brazil, then the Caribbean colonies in

84 85

86

For example, the concluding arguments in Ellen Meiksins Wood, The Origin of Capitalism. A Longer View. (New York: Verso, 2002), 193–198. Frank has pointed out the Eurocentric ambiguity in the use of terms such as core and periphery, since Europe was neither central or core to any ‘world-embracing economy or system’. He further points out that if any single country in this period was at the centre of a world economy, it was China. Mindful of the danger of implying a Eurocentric approach in my analysis, I will use sparingly terms such as imperial core and colonial periphery, limited to only indicating the relation between Imperial Spain and its colonial presence in the New World. Terms such as metropolis and satellite would apply just as well. Frank, ReOrient, 4–5, 29. Sidney W. Mintz, Sweetness and Power (New York: Penguin Books, 1986), 23–46.

the genesis and nature of silver ores

41

the seventeenth century saw cotton being cultivated for export to Europe, until the southern states of the u.s.a. became the main supplier to British mills in the nineteenth century.87 The most callous displacement of all was the trade in African slaves that made possible the mass production of sugar and cotton in the New World, over 12 million souls, including young children, transported in European ships, climbing to its peak volumes at the same time that the Enlightenment was sweeping the salons of Europe.88 Even spices and porcelain, a commodity and a finished technological product respectively, strongly tied to their original places of origin, ultimately were displaced to other regions by European forces. Cloves and nutmeg ended up being spirited away from the Spice Islands, to be cultivated in Mauritius and Reunion Islands in the Indian Ocean, and then on to Madagascar, Zanzibar, Martinique and Grenada by the end of the eighteenth century.89 The human skill, and the clays and kaolin of Jingdezhen and Qimen in China, were not unique enough to avoid the manufacture of porcelain to be replicated in factories in Meissen, Sèvres and by Wedgwood in England. These newcomers would become the leading suppliers of porcelain products to the world, displacing China from the monopoly it had held, based on ‘more than two millennia of Chinese craft expertise and technology’.90 Such is the power of this image of Early Modern Europe pulling all the commodity strings to meet its own interests, that such a perceptive historian of this period as Fernand Braudel came to wonder if ‘Europe, because of her very expansion, was acting as if she had decided to delegate the trouble of handling the mining and metallurgical industries to dependent regions on her periphery’, and cited the case of gold and silver in America.91 For once, however, the apparently unstoppable force emanating from Europe, that displaced at will the sources of its commodities and kept at arm’s length any collateral environmental damage created as a consequence of their

87 88

89 90 91

Giorgio Riello, Cotton. The Fabric that Made the Modern World (Cambridge: Cambridge University Press, 2015), 187–210. A visual guide to the scale and geographical reach of the trade can be found in Davis Eltis and David Richardson, Atlas of the Transatlantic Slave Trade (New Haven, London: Yale University Press, 2010). Arturo Giraldez, The Age of Trade. The Manila Galleons and the Dawn of the Global Economy (Lanham: Rowman & Littlefield, 2015), 173–174. Robert Finlay, The Pilgrim Art. Cultures of Porcelain in World History (Berkeley, Los Angeles: University of California Press, 2010), 9,13. Fernand Braudel, The Wheels of Commerce, trans. Sian Reynolds (London: Collins/Fontana Press, 1988), 325.

42

chapter 1

exploitation, had met its match in the immoveable object of the silver deposits of the New World. Imperial Spain had as little influence on the location of its silver mines, or the siting of the environmental consequences of mining, as King Canute over the tides. In contrast to cotton and sugar, it was geology, and not a strategy implemented wilfully at the imperial core, that situated its mining and refining activities at the colonial periphery. Had it only been the static nature of these deposits that set silver apart from the other commodities, there would be little to add to a straightforward geological reality. As I will argue in the following chapters, the independence imposed by geology would be accompanied by peerless technological innovation created solely in the New World, quite unusual for a colonial commodity of this magnitude and period. What is certain is that Europe, through its own deposits before 1492, and then through the conduit of the Spanish mines in the New World, virtually monopolised the major sources of silver in the world until the closing decades of the nineteenth century, thus gaining a unique head-start in global trade that no other region would ever have in the Early Modern period. This was no mean advantage, and it allowed Europe ‘to profit from the predominant position of Asia in the world economy. Europe climbed up on the back of Asia, then stood on Asian shoulders’.92

The Table is Set Portugal had enforced a no-sail zone down the Atlantic seaboard of Africa by the end of the fifteenth century, all the while whetting the appetite of Europe by returning with African gold loaded onto its ships. The only unopposed expansion route by sea for other European powers lay to the West, and Spain was the first to back the leap into the unknown. As a result of this decision by the Reyes Católicos, Spain stumbled onto a continent with an active and extended subduction zone along all its western coastline. It colonised first the narrowest portion of what is now North America, endowed with a very rich metallogenic zone, and would thread the rest of its conquest and colonisation along the metal-rich spine of the Andes. Spain thus came to control a unique monopoly of mineral resources that would allow its mines to corner the market of silver production for two and a half centuries, a geological-political conjunction that has not been repeated for any other metal up to the present. It would only be in the nineteenth century that the new United States of America would join Spain

92

Frank, ReOrient, 5.

the genesis and nature of silver ores

43

in profiting from the silver and gold bonanza of its own politically controlled subduction area. Spain would own within just 50 years the two major deposits of primary silver ore known to man, together with the major mercury deposit of Huancavelica to complement its Spanish mine of Almadén, the salt deposits of Uyuni and the polymetallic deposits that contained lead and gold in New Spain. The diversity of this geological bonanza meant that Spain had at its disposal all the raw materials needed to make possible the refining of the silver ores of the New World. The term ‘silver ore’ however is too generic, and implies a uniformity that does not exist. The silver ores of Europe, the Andes of Upper Peru and the Silver Belt of New Spain do not share an identical geological genesis, though all are the product of subduction processes. The different roots in geological time may be the reason why silver ores found in Europe were refined at a profit thanks to their content of lead or copper, with silver as a collateral benefit. In contrast, in the New World it was only the silver content that sustained all mining and refining production costs, in the chemical form of silver sulphides or argentiferous galena. This was the fundamental and relevant difference on both sides of the Atlantic, not the silver content in each ore. Once geology determined the embarrassment of mineral riches of the New World, it was up to the Spanish miners and authorities to make a conscious choice as to which technical route to follow in their pursuit of silver. The history of the refining of silver in the New World was not the unavoidable consequence of ores poor in silver content. The proposed absence of major supergene enrichment above the water table in sulphidic silver deposits, shores up the strong suspicion that the early decrease in the production of silver, as mines clawed deeper into the mineral veins of the New World, was a technical issue unrelated to silver content. I have argued in this chapter that it was the consequence of the change in the chemical profile of these deposits, from superficial elemental silver and silver halides to a deeper and much more intractable silver sulphide ore. Both argentiferous lead and superficial silver halides were easy to refine for the early Spanish miners, yet the deeper silver sulphides posed a much greater technical challenge. If unmet, it would have left a major part of the silver deposits of the New World undisturbed and underground, along the lines of the gold left in Hispaniola. The narrative of how refiners sorted the problem of extracting silver from its ores in the Hispanic New World, by importing European experiential knowledge in the sixteenth century, then proceeding virtually on their own for the next three centuries by adapting it to local conditions through innovation, takes us now into the next four chapters.

chapter 2

The Dry Refining Process: Smelting of Silver Ores The most general and most proper way, better adapted to the nature of metals, to separate them from the earth and stones where they are raised, and to reduce them to their purity and perfection … is through the fire of the furnaces, which to this end are called smelting ovens. alvaro alonso barba, Arte de los metales (1637)

∵ Deceitful Mercury The Spanish priest Alvaro Alonso Barba (born 1569) is a singularity amongst the early Spanish refiners of silver ores in the New World. He wrote the only extant metallurgical text of the early colonial period that is sourced in the practice of the New World, thus providing a first-hand guide to the mind-set and skill level of the time. He also proposed the last original refining method for silver ores based on mercury, the cazo process that will be described in Chapter 4. As a historical figure he is firmly entrenched in the historiography on mercury-based refining of silver ores in the Americas. Thus at first sight it might seem odd that he would exalt smelting as the ‘most general and proper way’ to extract silver from its ores.1 In fact, the major part of Alonso Barba’s much cited book, Arte de los Metales, is dedicated to the smelting of ores, longer than his discussion on what he terms ordinary mercury-based refining or even the description of his new cocimiento (cooking) process. And yet forty years before this manuscript was sent to Madrid for printing (1637), the practice of smelting silver ores in New Spain had suffered a fate similar to Mark Twain’s news of his early demise: 1 Alvaro Alonso Barba, Arte de los metales (Barcelona: Editorial Labor, 1977), 130. The metallurgical term ‘smelting’ is confusingly close to the word ‘melting’. The latter implies the use of heat to change the physical state of a crystalline substance from solid to liquid. The former is only applied to metallic ores, and is an operation that requires both heat and chemical reactions to bring about the extraction of a metal from its ore. For a definition of smelting see Manuel Eissler, The Metallurgy of Argentiferous Lead (London: Crosby Lockwood & Son, 1891), 33.

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_004

the dry refining process: smelting of silver ores

45

and with respect to smelting, I say that it is very much forgotten since 35 years have passed since it was last used, so if new ores with sufficient silver tenor for smelting were to be discovered, no man would still be alive who knew how to smelt them, nor would there be a smith to make the tools required for smelting.2 The author of this premature obituary for smelting was Gómez de Cervantes, a commentator at the end of the sixteenth century on events unfolding in New Spain, and not a minero.3 He was claiming that smelting had long since disappeared in New Spain, a victim of the success of mercury to refine silver ores. The notion that the new process swept its way past smelting to become virtually the sole colonial refining process still percolates its way through the modern historiography.4 Yet the view expressed at the time by Gómez de Cervantes contains two important errors of fact. First, towards the end of the sixteenth century not enough mercury was being imported into New Spain to refine all its silver ores. Second, the important deposits of argentiferous galena discovered towards the end of the sixteenth century in the mines of the Cerro San Pedro, close to the town of San Luis Potosí north of Ciudad de México, could only be refined by smelting. In fact, the recurrent theme throughout Alonso Barba’s text is that mercurybased refining is never taken for granted as being a better method than smelting. Time and again he cautions his contemporary readers against being led astray by the limitations of mercury, which could not even be used to provide a true assay of silver in an ore. ‘He runs a great risk … whoever deals with ores 2 Gonzalo Gómez de Cervantes, La vida económica y social de Nueva España al finalizar el siglo xvi (México: Antigua Librería Robredo de José Porrúa e hijos, 1944), 156–157. 3 The term minero is applied in the early documents both to the person that owned and/or operated the mines and the person who owned and/or operated the refining haciendas. The use of the masculine does not mean women were excluded from the business of mining and refining, but they do not appear in documents as much as the men. The following extract shows an interesting exception since women are involved on both sides of the business: ‘Doña Francisca de Paz minera of this village [San Luis Potosí] declares before Your Eminence one hundred and fifty cargas of ore from my mines to be processed in the refining hacienda of Doña Ysabel de Adriansen which she owns in Los Pozos’, ahslp, Fondo Alcaldía Mayor 1635.3, expediente 19, 11 July 1635. Many widows undersign documents relating to sales or rental agreements of refining haciendas in New Spain. 4 For example, the claim that more than 95% of all silver was produced by refining with mercury, in Manuel Castillo Martos, “Alquimia en la metalurgia de plata y oro en Europa y América” in Informes para obtener plata y azogue en el mundo hispánico (Granada: Universidad de Granada, 2008), xxiv.

46

chapter 2

without knowing how to assay them by fire to learn correctly their silver content … do not trust the assay by mercury, which is very deceitful’.5 Deceitful mercury is certainly not a part of the modern mainstream narrative on colonial silver refining. Alonso Barba knew first-hand that refining with mercury was not a process that would extract silver efficiently from every ore. In his words: a certain minero … extracted a lot of very rich ore, but did not realise this; he assayed it by mercury and measured four or five pesos per quintal [0.24 to 0.3%] … he abandoned the mine, because he deemed it without profit … [later Alonso Barba] assayed it by fire [smelting with lead] and it had nine hundred pesos per quintal [approx. 54 %].6 Alonso Barba’s tale of reaping windfall profits was the result of a confusion still found to the present day, the assumption that the extraction of silver from ores through the use of mercury was exactly the same as the amalgamation of gold. Mercury amalgamation was one of the tested methods to assay gold ores in Europe in the sixteenth century, and it did not require the skills of assaying with lead using a cupel and a furnace.7 The clue to explain the great discrepancy between assaying with mercury or by smelting can be deduced from the similar experience of another priest that was a friend of his: In the Cerro de Santa Juana … ores like Soroches [galena, lead sulphide] were extracted, that when assayed with mercury showed little or no silver; they were thrown away by the mineros … [then Alonso Barba assayed them by smelting] and found they had sixty or more pesos per quintal [3.6% silver] … on my advice he collected many … [and] extracted much wealth from them.8

5 Alonso Barba, Arte de los metales, 151. 6 Ibid., 70–71. 7 L. Ercker, Treatise on Ores and Assaying, trans. Anneliese Grünhaldt Sisco and Cyril Stanley Smith (Chicago: University of Chicago Press, 1951), 57–60, 96–97. Any of the metallurgical texts of this period provides details on how to make cupels from crushed bones, hollow receptacles in which ores and lead could be assayed in a furnace. 8 Alonso Barba, Arte de los metales, 71. The fact there is a similar tale in the historiography, of a skilled priest buying cheaply an ore wrongly classified as poor in silver and then refining its true worth to his profit, underlines the ubiquity of technically proficient Spanish priests with a good eye for wealth in the mining landscape of the New World. The source is Juan de Peralta, late sixteenth century New Spain, as quoted in Wagner, “Early Silver Mining in New Spain,” 61–62.

the dry refining process: smelting of silver ores

47

The chemical explanation to both priests’ profit-taking is that lead is the main metal present in galena or soroche (as it was called in the Andes) and it forms an amalgam with mercury, thus competing and interfering with the extraction of silver (further details in Chapter 4). Contrary to the claim that mercury-based refining was best suited for ‘poor’ silver ores, the lower the silver content in galena, the more useless mercury becomes as a refining agent. Argentiferous lead could never be refined efficiently by mercury. Its limitations for assaying also extend to non-lead silver ores rich in silver sulphide or sulfosalts, the negrillos. Mercury alone will not reduce silver sulphide to silver metal, as I will explain in Chapter 4, so the assaying of negrillos using the steps that apply to gold-bearing ores would have given false low values. As long as deposits of argentiferous lead were mined in the New World, smelting could never disappear.9 The attention paid by Alonso Barba in his metallurgical text on the technical merits of smelting indicates this was a topic of current interest to the mining community from whose experience this text was sourced. The handful of German, Siennese or Spanish authors who wrote texts on mining and refining in the sixteenth and seventeenth century, reflected the most relevant practices of the mining district from where the authors drew their experience.10 It is to be expected that Agricola does not address the technique of refining silver ores by the use of mercury. After all, he was writing about techniques applicable to the silver ores found in the Erzgebirge and Harz mountains, argentiferous copper or lead, for which mercury was not an option.11 In contrast, both Agricola and Ercker mention in detail the 9

10

11

This perception is shared by other historians, for example: ‘All descriptions and accounts of South American mining emphasise the amalgamation technique giving the impression that smelting was totally discarded. This was not true’ in Lang, “Silver Refining Technology in Spanish America (patio y fundición)” 140. Earlier, Garner had proposed that smelting and refining with mercury shared production until the eighteenth century, a deduction based on the relative rates of increase of silver production and mercury imports during the colonial period, as set forth in Richard L. Garner, “Long-Term Silver Mining Trends in Spanish America: A Comparative Analysis of Peru and Mexico,” The American Historical Review 93, no. 4 (1988): 918. ‘Agricola’s … outlook is severely local to Germany and the topics considered are almost completely restricted to the activities current within regions in and around the Harz mountains and the Erzgebirge … De Re Metallica gives a picture of the best practices of the age, practices that had made Germany lead Europe in non-ferrous metallurgy and which caused the services of German metallists to be sought by the rulers of many other countries’. Aitchison, History of Metals, 373. An alternate interpretation advanced in the historiography is that Agricola’s silence on

48

chapter 2

amalgamation of gold ores, a reflection on the importance of this new process as of the fifteenth century in the Rhennish workings of Europe.12 Biringuccio, a Siennese and not a German, is the only metallurgist and author of the sixteenth century to mention in detail the use of mercury to refine silver ores by the use of mercury, in a book that was published in Venice, because he was addressing a Venetian mining audience already conversant with the mercury-based refining of silver ores, as I will explain in Chapter 4.

Smelting of Silver Ores: The Human Context The early history of smelting silver in the New World is defined by the skills, or lack of them, of the Spanish soldiers, priests and colonists who arrived in the newly conquered lands in search of a wealth they could never attain back in their homeland. When Spain reached the New World it turned its attention very early to mining activities. It is claimed that some ten percent of the 1,500strong contingent that came on the second voyage of Columbus was made up of ‘workers … to take gold out of the mines’ on Hispaniola.13 Precious metals were not the only target, and as early as 1505 King Ferdinand dispatched equipment and one hundred African slaves to mine for copper in Hispaniola. Five years later the King would add another 250 slaves destined to mining. ‘A few black slaves had been sent to the New World in the first years of the sixteenth century … in twos and threes, never as many as a hundred. So 1507 marked a new phase in the history of the Indies, of Africa, of Europe, and of human population’.14

12

13 14

the use of mercury to refine silver ores proves that Biringuccio’s instructions were quickly forgotten and never put in practice. Bargalló, Minería y metalurgia colonial, 111. ‘Agricola was the first writer to give a comprehensive account of the metallurgy of gold, and his most extensive recordings deal with that subject [including amalgamating gold with mercury]’. Aitchison, History of Metals, 385. According to Cyril Stanley Smith ‘there are no books on metallurgy among the incunabula’. The most prominent mining texts published in the sixteenth century and early seventeenth centuries are: the anonymous German Probienbuchs (early sixteenth century), the Siennese Biringuccio’s De la Pirotechnia (1540), the German Agricola’s De Re Metallica (1556) and Ercker’s Treatise on Ores and Assaying (1574) and the Spaniard Alonso Barba’s Arte de los Metales (1637). See the introduction to Biringuccio, The Pirotechnia, x–xix. J.E. Pérez Sáenz de Urturi, “La minería colonial americana bajo la dominación española,” Boletín Millares Carlo, no. 7 (1985): 55. Hugh Thomas, Rivers of Gold: The Rise of the Spanish Empire (New York: Random House, 2005), 256–257, 291.

the dry refining process: smelting of silver ores

49

The Spanish contingent that swarmed over New Spain on the heels of the conquest was described by Bernal Díaz, one of the original band of conquistadores of New Spain under Hernán Cortés, as ‘poor and greedy, curs hungry for wealth and slaves’.15 Oviedo describes well the predatory, rags-to-riches mentality of many of these mineros: And in particular, those in these parts that have no intention of remaining nor wish anything from this land other than to enjoy it and return to their homes, turn to trading or to the mines … or any other activity that will allow them to get rich quickly and leave … for most who are here treat this land as a step-mother, even though many have fared much better here than in their own motherland.16 Part of the problem faced by Spain was the very small pool of home-grown talent from where the early miners could be sourced, with no generational experience on any type of silver ore: This manner of extracting silver [smelting] was not learnt from the indians, nor did men go from here who knew about it, because they did not know how to smelt, and they were also ignorant of refining over a bed of ash … previously they used to disinter the dead and burn their bones, so as to benefit from the ash alone to make the cupels in which they refined, and in a similar manner there were other primitive actions that show the ignorance of that time.17 Even by the early seventeenth century Alonso Barba was sufficiently worried by the overall lack of skills he observed in Peru to propose that only those who passed an official exam to demonstrate their ability to assay ores by smelting should be allowed to refine silver ores, such had been the waste incurred

15

16 17

Bernal Díaz del Castillo, Historia verdadera de la conquista de la Nueva España: Manuscrito “Guatemala” (México: Colegio de México: Universidad Nacional Autónoma de México, 2005), 834. Gonzalo Fernández de Oviedo y Valdés, Historia general y natural de las Indias (Madrid: Ediciones Atlas, 1959), 80. From a letter from Agustín de Sotomayor to the King, dated 20 April 1573, as quoted in Tomás González, Noticia histórica documentada de las célebres minas de Guadalcanal, desde su descubrimiento en el año 1555, hasta que dejaron de labrarse por cuenta de la Real Hacienda vol. ii (Madrid 1831), 409. See also Fernández de Oviedo y Valdés, Historia natural de Indias, 47.

50

chapter 2

by ignorant refiners in the past.18 A modern Mexican historian shares the judgement on the poverty of metallurgical skills among the Spaniards: the number of expert miners that were available was very small and their knowledge very rudimentary, since except for the iron mines of Vizcaya, and those of mercury at Almadén, exploited by the German bankers the Fuggers, there was no working of important mines at that time in Spain, given that the famous silver mines of Guadalcanal, in the Sierra Morena, [some 110km north of Seville] were not discovered until 1555.19 The Spanish historian of mining, Julio Sánchez Gómez, argues that from Roman times up to the sixteenth century it became common practice in mining deposits of argentiferous lead (galena) in Spain, to only work the most easily accessible surface deposits, abandoning the mine as soon as it required deeper levels of mining.20 By the middle of the sixteenth century the contribution of Spanish mines, with the exception of Almadén, to the revenues of the Crown was on average less than 0.4%, which gives an indication of how little mining and refining know-how would have figured among its subjects.21 From a metallurgical point of view the working of iron ores in Asturias and the Basque country (part of the Kingdom of Castile at the time of the conquest of the New World) would have provided some background on furnaces and smelting but little in the way of providing prior experience for the silver sulphide ores to be found in the New World. The technology employed in the iron works of the Basque country is claimed to have been too simple to be of use in the refining of silver in the New World.22 18 19

20

21 22

Alonso Barba, Arte de los metales, 70. Mendizábal, La minería mexicana, 19. Silver ore from the mines at Guadalcanal was smelted, since it was argentiferous galena as is evidenced in the following extract: ‘in this month [January 1559] three thousand six hundred and eighty eight marks of silver were extracted; refining goes well, with the care that has been taken in assaying the lead’. González, Noticia histórica minas de Guadalcanal, ii 17. Sánchez Gómez, Minería, metalúrgica y comercio de metales, 47. The Mexican historian Mendizábal shares the view that ‘due to the imperfection in mining techniques and metallurgy, the miners abandoned their works at the first sign of impoverishment [of the ore]’ in Mendizábal, La minería mexicana, 21. Sánchez Gómez, Minería, metalúrgica y comercio de metales, 269. E. Fernández de Pinedo y Fernández, “Influencias recíprocas de las técnicas extractivas entre la minería vasca y la americana en la Edad Moderna,” Areas, no. 16 (1994): 38. Other historians have pointed out the lack of experience of early miners in the New World, such as Bakewell, “Colonial Mining,” 111; R.C. West, “Early Silver Mining in New Spain, 1531–1555,”

the dry refining process: smelting of silver ores

51

The best expertise on silver refining in Europe in the sixteenth century was embodied in the refiners of MittelEuropa. German technical artisans begin to be reported in New Spain from the 1520s.23 The presence of German miners in the New World came about either through individuals circumventing the initial restrictions on emigration for non-Castilians, or later in the sixteenth century as part of technical teams sent by either the Welsers to assist in the extraction of gold, or by the Fuggers at the request of the Spanish authorities, once travel was allowed after November 15, 1526 to subjects of the Holy Roman Empire (Germans, Flemish) under Charles v.24 The need for technical assistance in smelting the ores of New Spain had arisen quite soon after mines started to be exploited. According to a document dated June 6th 1571 by the Cabildo de México to the Concejo de Indias, less than ten years had passed between the start of mining in 1532 and the claim that by 1542 ores had declined in silver content and in ease of smelting, ‘minas comenzaron a perder la ley y la buena fundición’. This is a very significant pairing of concurrent events in a sixteenth century document. It strengthens the line of argument that the initial unskilled contingent of Spanish miners had benefited from the ease of smelting of silver chloride ores and native silver found on the surface layers of weathered deposits (see next section). However, the deeper and more intractable silver sulphides could not be smelted easily, and the drop in silver production when refining the same amount of ore triggered the conclusion that silver content had suddenly decreased. The arrival of German know-how introduced German

23

24

in In Quest of Mineral Wealth. Aboriginal and Colonial Mining and Metallurgy in Spanish America, ed. Alan Craig and Robert C. West (Baton Rouge: Geoscience Publishers, 1994), 122. For further information on the role of German refining know-how within Europe, see Marie-Christine Bailly-Maître, L’argent: du minerai au pouvoir dans la France médiévale (Paris: Picard, 2002), 147; Martin Lynch, Mining in World History (London: Reaktion Books, 2002), 17; John U. Nef, The Conquest of the Material World (Chicago: University of Chicago Press, 1964), 12. Demetrio Ramos, “Ordenación de la minería en Hispanoamérica durante la época provincial (siglos xvi, xvii y xviii),” in La minería hispana e iberoamericana. Ponencias del i coloquio internacional sobre historia de la minería. (León: Cátedra de San Isidro 1970), 381–382; Hugh Thomas, The Golden Empire. Spain, Charles v, and the creation of America (New York: Random House, 2010), 113. The ill-fated expedition of eighty miners sent by the Welsers came mainly from the Erzgebirge area, where the main mining activity was focused on argentiferous copper ores that were being refined by the liquation process with lead in the early sixteenth century. All except a few would die in the New World, with little support for their endeavours. Juan Friede, “La introducción de mineros alemanes en América por la compañía Welser de Augsburgo,”Revista de Historia de América, no. 51 (1961): 99–104.

52

chapter 2

smelting techniques using lead and litharge.25 Bargalló places the arrival in New Spain of the German smelters Juan Enchel and others in 1536, sent by the factors of the Fuggers in Seville, with tools and techniques to smelt metals from silver mines that until then had not been understood, and they set up grinding and refining facilities … from where came great benefit to the republic and great service to your Majesty.26 The Germans were not the only source of smelting skills to aid the initial wave of Spaniards. The technical role of African slaves within the initial smelting haciendas is an intriguing but unknown facet to the history of smelting in the New World. They are mentioned as of the mid-sixteenth century by Bartolomé de Medina.27 The seventeenth century records of Zacatecas make frequent mention of the African slaves that are sold or rented together with the physical assets of a refining hacienda. The level of skills of some of these slaves is indicated by the following notations: ‘a black called domingo smelter from the land of angola; another black called juanito … smelter from the land of angola’.28 Since all the slaves in the inventory are identified according to their place of origin in Africa, probably as an indicator of their behavioural and physical nature, the fact that the smelters came from Angola may not necessarily indicate a premium being placed on the level of skills originally brought by the slave. Slaves would persist in silver mining duties, and Mendizábal has pointed out the great number of slaves that worked in refining haciendas of Zacatecas, though according to the quote provided from the Bishop Alonso de la Mota y Escobar (1602), the indigenous workforce was better skilled than the African slaves or the Spaniards for both smelting and refining with mercury.29

25 26 27 28

29

Henry R. Wagner, “Early Silver Mining in New Spain,” ibid., no. 14 (1942): 69–70; Bargalló, Minería y metalurgia colonial, 91. At least one Spaniard, Alonso Carreño, is said to have established an ‘ingenio de fundición’, a smelting refinery, in Sultepec by 1543. Minería y metalurgia colonial, 58, 95. As quoted in Manuel Castillo Martos, Bartolomé de Medina y el siglo xvi (Santander: Servicio de Publicaciones de la Universidad de Cantabria, 2006), 112. In a rental agreement for a smelting and refining hacienda between Pedro de Medina and Andrés Pereira, 20 March 1608, ahez, Notaría-Colonia, Número 01 (Pedro Venegas, 1608), expediente 1. Mendizábal, La minería mexicana, 32, 36. See also Peter J. Bakewell, Silver Mining and Society in Colonial Mexico: Zacatecas, 1546–1700 (Cambridge, u.k.: University Press, 1971), 122–124.

the dry refining process: smelting of silver ores

53

A picture emerges of a stage from early to mid sixteenth century, with a very limited resource of refining skills in the New World, given the possible exception of some literate clergy who refined silver ores, together with pockets of autochthonous Andean and imported German and African know-how of smelting methods. Bargalló refers to a proto-smelting period in New Spain where the lack of knowledge of German smelting techniques led to many trial and error processes.30 In Chapter 4 I will return to this scenario, since it may have been an important factor in the fast adoption of an alternate refining process, one more amenable to such a motley crew looking to refine major amounts of silver ores.31

The Chemistry of Smelting and the Nature of the Ore As will become more evident in the course of this section, the smelting of silver ores in general is a complex operation that requires a high degree of skill to coax the silver in the ore through at least two major changes before isolating it in a pure state, without losing most of it through the chimney stack or in the slag. However, the early history of silver refining in the New World hinges on the fact that the only exception to this statement is for an ore that contains silver chloride, like those found in the superficial and weathered section of a silver sulphide deposit. For most silver ores the initial technical hurdle is

30

31

Bargalló, Minería y metalurgia colonial, 23. It is interesting to note the assertion that at the other end of the Early Modern period: ‘the dependence on Spanish mining methods was so great on the American [United States of America] frontier that as late as the 1880s mining men were literally relying on Spanish techniques hooked up to a steam engine’. The author states that the Spaniards had a strong mining and refining experience before arriving in the New World. Otis E. Young, “The Spanish Tradition in Gold and Silver Mining,” Arizona and the West 7, no. 4 (1965): 299. As will become clear in the following chapters, the refining technology based on mercury applied north of the Mexican border in the late nineteenth century was the next stage of evolution of three centuries of silver refining in the New World, with little or no contribution from Europe to the patio process. Even by the eighteenth century in Europe, the high level of skill required for an efficient use of smelting to refine silver ores is made evident from the following criticism to French refining practice found in the preface to Schlutter’s classic textbook by M. Hellot. He points out that among the lead mines in France were those that gave six ounces of silver per quintal [approx. 0.4%], but even though ‘their smelting and refining [was] very easy … [they] have been abandoned due to a lack of intelligence and [management]’. ChristopheAndre Schlutter, De la fonte des mines, des fonderies, etc., trans. M. Konig, Tome Second (Paris: Jean-Thomas Herissant et Jacques-Noel Pissot, 1753), xiii.

54

chapter 2

the temperature required to extract the metal by fire. The highest accessible temperature range to a person with no special training is that of an ordinary camp fire, in the range of 600 to 650° c.32 Native silver melts at 962° c, silver sulphide at 825 o c, so this temperature would not be high enough to cause these compounds, trapped in a mineral matrix, to flow and be recovered as a molten puddle in the ashes of the fire. However, silver chloride, with a melting point as low as 455° c, could flow under these conditions.33 Furthermore, in the presence of charcoal (carbon) lying at the base of the fire, it will be reduced to elemental silver at these relatively low temperatures.34 The weathering that chemically transformed the more superficial veins of silver sulphide deposits in the New World, therefore gifted the early mass of untrained Spanish mineros with silver chloride, the easiest of silver compounds to refine by smelting, with virtually no skills required. All they needed was to avoid overheating the silver ores and losing silver via volatilisation, and even this level of care may have been beyond many of them. The downside to such an easy operation was that no effort was made to refine the silver to its highest possible level.35 It is no surprise that as soon as these silver halides were depleted, and more silver sulphide was present in the ore, the gaggle of Spanish dilettante refiners would find less and less silver coming out of their primitive operations, regardless of the true silver content of their ores. For all other silver compounds usually found in ores, the process requires a much greater level of skill, such as was supplied through the German artisans in the New World. Europe by mid fifteenth century had been refining its silver mainly from argentiferous galena, lead sulphide ores.36 In contrast to most 32 33 34

35 36

Aitchison, History of Metals, 38. S.A. Cotton, Chemistry of Precious Metals (Bristol: Blackie Academic & Professional, 1997), 277. Wallace, Barton, and Wilson, “Silver-Bearing Minerals,” 17. The term reduction in this work, except when it appears within a quotation, refers to the modern chemical term as applied to redox equations, whereby the metal in its cationic state gains electrons from another element that in turn is being oxidised (losing electrons), so that the metal is transformed to its elemental state. The use of the term in alchemical texts was in accordance with the Latin original of the word, reducere—“to lead back” to an original state. Thus, ‘reduction had the more specific sense … of the isolation or extraction of a metal from … an ore’. William Royall Newman, Atoms and Alchemy: Chymistry and the Experimental Origins of the Scientific Revolution (Chicago: University of Chicago Press, 2006), xiii. ‘it is not likely that for a long time anything but very impure silver was shipped’ in Wagner, “Early Silver Mining in New Spain,” 61. P.T. Craddock, Early Metal Mining and Production (Washington, d.c.: Smithsonian Institution Press, 1995), 211.

the dry refining process: smelting of silver ores

55

metal ores, silver will not form silver oxide on heating in air, and heating silver ores to high temperatures over 1,000° c can simply lead to major physical losses of silver in the smoke of a furnace. It is estimated that at some point prior to 2500 bce, a major breakthrough in smelting occurred, possibly invented by tribal people on the southern shore of the Black Sea: silver could be processed as a by-product of smelting lead from galena (lead sulphide, PbS).37 The process from a chemical point of view is the same whether lead is present in the ore (as in argentiferous galena) or whether lead is added to the ‘dry’ silver ore.38 The difference only lies in the economics of production, as will be analysed in Chapter 8. Figure 10 represents a simplified version of the two main stages involved in the refining of silver from either of these two starting minerals. The diagram is based on the description by Craddock of a two-step refining process, where the first step is a smelting process under high temperatures and reducing conditions ( fundición) that creates a lead bar rich in silver (plomo rico).39 When argentiferous galena is heated to around 1,000° c in the presence of charcoal within a furnace (an horno castellano), the lead sulphide is reduced to produce lead, and any silver compounds present (silver sulphide, silver sulfosalts, argentojarosite) can also be reduced to metallic silver. Molten lead absorbs all this metallic silver, and also any gold and certain other metals (copper, zinc, arsenic, antimony and bismuth) that may be present. All the rest of the components of the ore, plus any additives used to assist the smelting process, separate as waste slag. This waste material can comprise quartz (silicon oxide) and iron compounds which were used as fluxing agents in the furnace. Once the slag is physically isolated by the refining workers from the molten silver-rich lead, the latter is cast into bars, called pigs in the English texts.

37

38

39

Galena is one of the easier ores to smelt, since at temperatures as low as 800° c lead oxide (litharge) is formed from the lead sulphide at the upper, more oxidising zone, while the reduction of the oxide to metallic lead (which melts at 327° c) takes place in the presence of unburnt charcoal or carbon monoxide. In addition, a further quantity of lead is produced by the reaction between the lead oxide and the lead sulphide. N.H. Gale and Z.A. Stos-Gale, “Cycladic Lead and Silver Metallurgy,” The Annual of the British School at Athens 76(1981): 178. It was known by the sixteenth century that to avoid losses via volatilisation even native silver or silver chloride benefit from the addition of lead prior to smelting. Georgius Agricola, De re metallica, trans. Herbert Clark Hoover and Lou Henry Hoover (New York: Dover Publications, 1950), 400. Craddock, Early Metal Production, 221–231.

56

chapter 2

figure 10 Schematic diagram of two stage refining by smelting of silver ores using lead. See text for explanation.

The second step is sometimes called the cupellation stage, or afinación in Spanish texts. The silver-enriched lead is re-melted in a cupel placed within a reverberatory furnace at over 1,000° c. In this stage heat is applied indirectly via a reverberatory furnace, designed so that there is no contact between a reducing agent such as charcoal and the litharge being generated. To isolate the silver from the lead it is necessary to blow air (oxygen) onto the surface of the molten lead so that litharge (lead oxide) is formed as a surface scum that entrains with it the oxides of the majority of the other metals present. Silver and gold do not form oxides under these conditions.40 The litharge (greta in Spanish texts) is skimmed off to be recycled. The litharge is also absorbed by 40

I return to reverberatory ovens later in this chapter. A detailed physical-chemical analysis of a silver smelting process dating from the 11th and 12th centuries can be found in F. Ströbele et al., “Mineralogical and Geochemical Characterization of High-medieval Lead-Silver Smelting Slags from Wiesloch near Heidelberg (Germany)—An Approach to Process Reconstruction,” Archaeological and Anthropological Sciences 2, no. 3 (2010). Figure 16 (p. 212) in this article is a very useful flow diagram, a visual guide to an earlier yet more complex smelting process than the one I describe.

the dry refining process: smelting of silver ores

57

the material of the cupel (porous crushed bone in many cases) which can also be recycled (cendrada in Spanish texts). Recycling did not require reducing lead from litharge, and had the added advantage of recovering any silver entrained in the litharge. What remains in the bottom of the cupel is refined silver, that may also contain gold originally present in the ore.41 Simply heating efficiently and uniformly to the right temperature (in itself a major technological breakthrough) is not enough: in the first step, it must be done under oxygen-depleted conditions that reduce the silver compounds (chlorides, sulphides or more complex sulfosalts) to metallic silver, to assure maximum recovery of the silver content of the ore. Silver has to be absorbed by the lead and not lost in the slag or volatilised if the temperature is too high. The second step requires quite the opposite, a careful oxidation of the surface of the molten lead avoiding spatter or other physical loss of silver into the litharge that is being continually scraped off, within a furnace that has to be kept at or over one thousand degrees centigrade.42 As a historian of silver refining in the New World and mining engineer has written, after describing smelting of silver ores: litharge fluxed everything … smelting furnaces fell apart in a week; refining furnaces … lasted a little longer … This [his description of the process] is so over simplified that it may appear anyone could do it. If so, blame this narration, because it was a very complicated and difficult process which taxed the skill of experienced furnacemen.43 While smelting was never a simple method, it had the major advantage of being applicable in theory to any silver ore, if enough fuel, lead flux and skill was available. When dry, polymetallic ores are smelted, the two-stage process of Figure 10 can become an iterative sequence of multiple stages of smelting, as has been reconstructed for the way jarositic earths containing silver of the Rio Tinto mines in Spain were refined by smelting in Phoenician times, using lead imported from Carthage.44 The silver sulphide and complex silver sulfosalts of the silver ores in the New World represent the intermediate spectrum of smelting complexity between the beginner’s luck of silver chloride and the challenge

41 42 43 44

Biringuccio, The Pirotechnia, 172. Ibid., 163. Alan Probert, “Bartolomé de Medina: the Patio Process and the Sixteenth Century Silver Crisis,” Journal of the West 8, no. 1 (1969): 96. Craddock, Early Metal Production, 220.

58

chapter 2

of the jarositic silver compounds of Rio Tinto. Smelting was always the refining workhorse for ores that could not be refined by the patio process in the New World, such as those containing lead or antimony.45 The level of smelting skill developed in Europe allowed its refiners to extract silver from ores with approx. 0.04% silver content.46 This technical threshold should not be confused with the profit threshold determined by the balance between production costs and total revenues obtained from the refining operation. The presence of other metals, such as gold, copper and/or lead, which could also be marketed, would allow the refiner to operate as close as possible to the technical threshold.

The Architecture of Smelting in New Spain The region around the town of San Luis Potosí epitomises the use of smelting to refine silver ores in New Spain, so its haciendas are a prime example of the physical structures of this genre. The whole State of San Luis Potosí has at present 36 ruins of refining haciendas that correspond to the sixteenth and seventeenth centuries, all either in an advanced state of decay or with substantial changes to their original architecture. Salazar González has pioneered the architectural study and reconstruction of some of these haciendas.47 In the historical legal

45

46

47

Antimonial ores were being efficiently smelted together with iron or copper pyrites by 1604 in New Spain, according to West, The Parral Mining District, 30; ‘when smelting ores from hard sulphidic veins, or with antimony, it would have been enough to mix them with lead ores, or greta, or cendrada’ in Bargalló, Minería y metalurgia colonial, 91; argentiferous lead, copper or zinc ores rebellious to refining with mercury and required smelting according to Thomas Egleston, The Metallurgy of Silver, Gold, and Mercury in the United States (London; New York: J. Wiley & Sons, 1887), 40–41. For detailed technical descriptions of how smelting was carried out up to the nineteenth century see for example Laur, “De la métallurgie de l’argent au Mexique,” 240–255; Bargalló, Minería y metalurgia colonial, 92–98, 249–251; La química inorgánica y el beneficio de los metales en el México prehispánico y colonial (México: Universidad Nacional Autónoma de México, 1966), 109–110. Approximately 12oz of silver per short ton of ore, Agricola, De re metallica, 388. A threshold of less than 0.02% silver content for smelting as practised in Saxony in the late eighteenth century is mentioned in Francisco Javier de Sarriá, Ensayo de metalurgia (México: D. Felipe de Zúñiga y Ontiveros, 1784), 107. In Poland argentiferous lead ores containing approx. 0.04% were smelted, according to Danuta Molenda, “La métallurgie du plomb en Pologne au moyen age et aux xvie–xviiie siècles” in Mines et métallurgie ed. Paul Benoit (Villeurbanne: Programme Rhône-Alpes recherches en sciences humaines, 1994), 52. Salazar González, Las haciendas de San Luis Potosí, 83–103. See Figure 13, this chapter, for a map of the region.

the dry refining process: smelting of silver ores

59

documents, the refining haciendas are not characterised by their nameplate silver production output, or their capacity to handle a certain throughput of ore per month or year. Rather they are described by the number of smelting furnaces (ornos or hornos) and cupelling furnaces (vasos) they possessed, which would indicate that furnace size was standard and their capacity so well known it did not merit a special mention in the legal documents pertaining to the sale or rent of these haciendas. A non-exhaustive survey of textual sources points to a range of smelting furnaces per hacienda between one and sixteen, while the number of cupelling furnaces is in general just the one, very rarely two.48 The only textual clue as to the capacity of these smelting furnaces is found in a document dated 1620 which states that a total of 22 smelting haciendas produced in a year 150,000 marks of silver from 100 furnaces.49 If I assume as a working hypotheses that the size of these furnaces is standard, then on average one smelting furnace could produce at least 1,500 marks of silver (345 kg) per year, though ultimately this depends on the quality of the ore being smelted and not on the size of the furnace. The data can also be interpreted to mean that an average smelting hacienda had installed between 4 to 5 furnaces, and produced in that period approximately 1.5 tons of silver per year. There is an account book of smelting operations in the Valle de Pozos that gives support to this estimate. It dates from a later period, 1660 to 1661, and was presented as evidence for the state of business agreements between Juan López de la Madriz and the deceased Miguel de Santibáñez. As such it is not an account book based on operational data, but a set of prepared accounts of expenditures and revenues from sales of silver and gold produced between May 1660 and December 1661 that are presented in defence of its author, López de la Madriz. It includes a series of entries that detail the amount of silver and gold produced. but there is no guarantee that it reproduces all the production data for the period covered. The aggregate amount of silver reported from May to November 1660 was approximately 500 kg, which prorated to the

48

49

Sale by Juan Domínguez de Sequera to Cristobal del Castillo of an hacienda in Monte Caldera with one smelting furnace, ahslp, Fondo Alcaldía Mayor 1653.1, expediente 8, 14 March 1653; sale by Nicolás Peralta de Pimentel to Cristóbal Zapata of an hacienda within the town of San Luis Potosí with two smelting furnaces, ahslp, Fondo Alcaldía Mayor 1667.3, expediente 6, 18 July 1667; rental by Mathias Pardo to Sánchez and Rodríguez of an hacienda with four smelting furnaces in the valley of Pozos, ahslp, Fondo Alcaldía Mayor 1629.3, expediente 24, 31 March 1629; the Hacienda de Briones in Monte Caldera is reported as having eight smelting furnaces in 1628, and up to sixteen furnaces have been reported according to ibid., 90, 97. As quoted in ibid., 80.

60

chapter 2

whole year corresponds to approximately 860 kg, ignoring production fluctuations according to the season. A similar aggregate amount (approximately 880kg) is reported for the whole year of 1661. The maximum monthly production registered for July 1660 was approximately 150 kg of silver. All these amounts, if they correspond to the production from a single hacienda, fall within the ranges estimated in the preceding paragraph. The other important finding to come out of this account book is the role played by the revenues from gold refined from these silver ores. In 1660 and 1661, gold contributed 25 to 30% of the total revenues reported by its author, López de la Madriz.50 This is a very relevant percentage, which confirms that the presence of gold in a silver ore would have played a major role in meeting the production costs of smelting in New Spain, an argument to which I will return in Chapter 8. As to the area of the footprint of a smelting hacienda, a document from 1772 details the sale of a smelting hacienda in Monte Caldera between Cristóbal Pardo and Juan Nieto for 640 pesos, with a dimension of ‘140 varas de oriente a poniente, 85 varas de norte a sur’, approximately 110 m east to west by 70m north to south, for an area of 7,700m2. It included a small reservoir (‘tanquesito’) to collect water in the rainy season.51 An idea of the balance between functional areas and waste areas can be gained from modern satellite images. Figure 11 shows satellite images of the area around the village of Monte Caldera, and the location of ruins of smelting haciendas. The approximate area for the Hacienda hmc1 is 3,000m2, of which approximately 2,000 m2 are taken up by the waste grasas. In the case of the Hacienda Santa María, approximately 3,500m2 correspond to operational areas including the water reservoir, and approximately the same area for the grasas. In each of these satellite images the area occupied by the waste grasas either matches or surpasses the extant footprint of the historical hacienda, thus doubling the size of the industrial plot required overall for silver production.52 The environmental implications are important, since these wastelands of grasas inadvertently helped to isolate

50 51 52

ahslp, Fondo Alcaldía Mayor 1650.3, expediente 8, selected accounts of the years 1660 and 1661. ahslp, Fondo Alcaldía Mayor 1772.2 expediente 18, 7 November 1772. Satellite images do not necessarily show the whole extent of the hacienda or the grasas, the former due to a blurring of physical perimeter walls with time, the latter due to cover from trees or displacement with time by agricultural land, disposal as gravel to line tracks in the countryside around Monte Caldera, or even by the stream when in flow. Even with these limitations the images are a useful tool to appreciate the relative dimensions of each area.

the dry refining process: smelting of silver ores

figure 11

61

The solid white lines encapsulate the minimum area that can be clearly identified with each smelting hacienda, the dotted line the minimum area of the extant dumps of grasas all satellite image reproductions courtesy google earth © 2014 digitalglobe. the hacienda santa maría lies at 22°12’10” n 100°44’27” w, and the hacienda hmc1 at 22°12’31” n 100°44’47” w

62

chapter 2

figure 12 ‘La Noble y Leal Ciudad de S. Luis Potosí dividida en Quarteles de Orden Superior del Exmo. Señor Virrey Marqués de Branciforte. Diciembre 15 de 1794.’ illustration reproduction courtesy mapoteca manuel orozco y berra, servicio de información agroalimentaria y pesquera, sagarpa, mexico

the hacienda from agricultural or cattle grazing activities and from human dwellings, and by desolating their stretch of land they acted as a sink of further chemical depositions issued from the smelting chimneys.53 Some of the smelting haciendas of the state of San Luis Potosí were set up within town limits.54 It would not have been possible for each smelting hacienda within the town to have occupied a space or to have spread its mineral waste in the same manner as in the countryside, but further information is required on how the disposal of grasas was legislated by the town authorities. Significant mounds of grasas from previous or existing smelting haciendas 53 54

The large area taken up by grasas and other solid wastes is another reason to include them in sale contracts that involved the ownership of land. For the location of haciendas in the region close to the city see Salazar González, Las haciendas de San Luis Potosí, 396.

the dry refining process: smelting of silver ores

63

inside the city were sufficiently part of the urban landscape that they merited being included in a map of San Luis Potosí from as late as 1789 (Figure 12). When the streets of present day San Luis Potosí are dug up for major road works or to lay pipes, workers come across an underground layer of grasas, as if a historic volcano had at one time covered the area in ash.55 The historical documents that record the assets sold or rented comprise not only furnaces but at times mines, charcoal making facilities (carboneras), slaves, indigenous work squads, livestock and mounds of solids of varying magnitude. Prevalent market conditions dictated the value of these production facilities. This would be consistent with a valuation based on expected future cash-flows rather than on the cost of construction. Thus an hacienda with four smelting furnaces is to be sold for 20,000 pesos in 1628, while others with just one smelting furnace were sold for 1,700 pesos in 1653, and for 700 pesos in 1667. The near halving in price per furnace as the century advanced correlates well with a decrease in the silver refining activity in the area.56 I have only come across one reference to the actual cost of building a smelter within an hacienda in the Valle de Bledos being rented by Juan de Sandoval in 1607. It cost 525 pesos to build, but the cost excluded both wood and certain basic equipment (bellows and other accessories) furnished by Sandoval. The contract calls for construction to be completed within budget in 40 days, which reflects a simple structure. It does not state how many furnaces are covered by the contract.57

55

56

57

Personal communication from Dr. Rafael Morales Bocardo. The presence of smelting haciendas within city limits in New Spain is illustrated in the following extract: ‘Unhappy would be the city of Zacatecas if it consented that in its centre be built similar haciendas, that all would live either sick, or bothered by their smoke: but what am I saying? Can my adversaries deny, that they have right in the middle of the city four haciendas surrounded by many houses?’, arguments in favour of building a smelting hacienda presented by Manuel Correa, 31 March 1761, ahez, Serie Civil c37–005. Proposed sale of the hacienda of the deceased Juan Pérez Basurto by Antonio de Arismendi Gogorrón for the proposed sum of twenty thousand pesos, photocopy of original document dated 30 May 1628, folio 52 r, ahslp, Colección Miguel Iwadare; sale by Juan Domínguez de Sequera to Cristóbal del Castillo of an hacienda with one smelting and one refining furnace in Monte Caldera for 1,700 pesos, ahslp, Fondo Alcaldía Mayor 1653.1, expediente 8, 14 March 1653; sale by Nicolás de Peralta Pimentel to Cristóbal Zapata of an hacienda behind the convent of St. Francis in the town of San Luis Potosí with one smelting and one refining furnace for 700 pesos, ahslp, Fondo Alcaldía Mayor 1667.3, expediente 6, 18 July 1667. The builder, Juan de Vargas, is asking for the pending amount of 232 ‘pesos de oro común’ that have not yet been paid since Sandoval is not satisfied with the work carried out. ahslp, Fondo Alcaldía Mayor, 1607, expediente 1, 9 October 1604 and 24 April 1608.

64

chapter 2

Some, but not all, rental contracts of smelting haciendas also specify the number of furnaces being rented, as well as a wide menu of additional services, such as workers, or access to mines or mounds of grasas, which in turn determine a range in rents between 250 and 1,000 pesos per year.58 The most intriguing contract is for the use of a smelting furnace at 5 pesos per day, and for mines and two indigenous workers ‘that belong to me’ at 50 pesos/y, since it shows a significant degree of entrepreneurship within the silver refining business in this region.59 Overall these contracts show an active market for the outsourcing of refining services, that did not require a capital investment by those trying their luck as miners and refiners. The level of rent in this market was the equivalent of approximately 30 to 120 marks of silver per year, at the most some 1% of the average silver production per hacienda registered in 1620. One strong incentive to rent must have been the provision of ores and skilled workers, more than the relatively simple infrastructure that could be built in five weeks. Towards the end of the eighteenth century, smelting would be overshadowed by the cazo process (see Chapter 9) in the refining of silver presented to the Caja (Treasury) of San Luis Potosí. The boom in silver production shifted to the mines of Catorce in the northern part of the state, where the nature of the ore (mainly silver halides, poor in lead) made the cazo process the best refining option. The difficulties faced by some smelting haciendas during this period is reflected in the nineteen surviving weekly accounts from the Hacienda de

58

59

Two refining haciendas, furnaces not given, and their mounds of grasas, at 500 pesos/y each, between Gerónimo de León and Palomo y María de Mendoza. ahslp, Fondo Alcaldía Mayor 1627.5, expediente 27, 24 December 1627; one hacienda with four smelting furnaces at 500 pesos/y between Mathias Pardo and Sánchez and Rodríguez (document damaged). ahslp, Fondo Alcaldía Mayor 1629.3, expediente 24, 31 March 1629; one of four furnaces within a smelting hacienda, plus use of refining furnace once a month, and partial use of existing workforce, plus supply from existing mounds of grasas, at 1,000 pesos/y between de Fraga and Rodrigo de Aldana Chávez. ahslp, Fondo Alcaldía Mayor 1631.3, expediente 38, 27 December 1631; one smelting and one refining furnace but without fuel at 250 pesos/y between Gaspar de Villanueva and Fernando de Mesa Godines. ahslp, Fondo Alcaldía Mayor 1641.1, expediente 9, 25 June 1641; an hacienda with two smelting and one refining furnace for 400 pesos/y between Gerónimo Días and Alonso de Borja. ahslp, Fondo Alcaldía Mayor 1658.1 expediente 4, 7 January 1658; an hacienda with three smelting furnaces for 750 pesos/y between Francisco Días del Campo y Diego Sánchez. ahslp, Fondo Alcaldía Mayor 1658.1, expediente 18, 18 March 1658. Rental agreement between Francisco Días del Campo and Hernán Vásquez, ahslp Fondo Alcaldía Mayor 1635.5, expediente 28.

the dry refining process: smelting of silver ores

65

Nuestra Señora de los Dolores, corresponding to the year 1773. An analysis of the records that tracked the silver refined per week indicate that on average 2.4 kg of silver were produced on a weekly basis, approximately 0.12 tons per year. This is one third of the average value per furnace in 1620. If the number of cargas of ore set down in these weekly accounts were the source of the silver refined, the values indicate a silver content for the ore being smelted around 0.6 %. The data show a persistent operational loss in the accounts being rendered, based only on silver revenues.60

The Infrastructure of Smelting in New Spain Though smelting was carried out in many locations and different periods within New Spain, the region around the town of San Luis Potosí became a major colonial producer of silver initially on the strength of its smelting of silver ores. The silver deposits in the vicinity of San Luis Potosí began to be exploited by the Spanish conquerors of the northern Chichimeca territory by the end of the sixteenth century (1592). The most prominent were the deposits of the Cerro San Pedro, but due to the lack of sufficient water at the site some of the smelting haciendas were built in Monte Caldera, some 7 km from the mines and 25km from the town of San Luis Potosí (Figure 13).61 Others were in the town itself. The presence of lead in most of the ores found near the town of San Luis Potosí made smelting the only viable refining process right from the start.62 In the absence of technical documents written by refiners of this region, it is the historical legal documents of San Luis Potosí that provide an indirect guide to the way smelting was practised, in one of the regions that became its

60 61 62

Weekly accounts signed by Lorenzo Mata that cover, with major gaps, the period from 27 December 1772 to 28 November 1773, ahslp, Fondo Alcaldía Mayor 1773.2. Salazar González, Las haciendas de San Luis Potosí, 396. Even by 1744, by which time the use of mercury in some haciendas of the region has been documented, a committee of miners including the local mayor (Alcalde Mayor) was set up to test a new refining recipe based on mercury at the request of the Crown. The report stated that ‘what is generally processed are ores by fire [smelting], and at this time none by mercury … the ores for mercury are of low silver content, with no stability [of supply] … but to demonstrate their obedience to the Crown they will set up the test’ ahslp, Fondo Alcaldía Mayor 1744.1, expediente 35, dated 13 December 1744. Not all the ore was rich in lead, at least by 1622, when there are references both to metal plomosso (lead ores) and to metal seco or sequillo (dry or dryish ores) in ahslp, Fondo Alcaldía Mayor 1622.6 expediente 15, receipts for ores dated 1622 and 1623.

66

chapter 2

figure 13 Location of the main mines, smelting haciendas, charcoal production, agricultural and cattle rearing areas around the town of San Luis Potosí (slp) adapted from salazar gonzález in footnote 61

major exponent in New Spain. The operational stages of the smelting process have been described at length in the historiography.63 I will only focus on those areas where some further light can be shed: The introduction of molinos: a common sight at present around San Luis Potosí are the large circular stones from the molinos (Chilean mills), Figure 14, used to crush the raw ore placed in the path of the stone within the circular

63

For sixteenth century practices in the New World, Alonso Barba, Arte de los metales, 130– 170; for smelting practices in mid-nineteenth century, see Francisco de Paula Hermosa, Manual de Laboreo de Minas (Besanzon: Librería de Rosa Bouret y Cia, 1857), 250–260. In English, the nineteenth century texts are the most useful, for example John Arthur Phillips, The Mining and Metallurgy of Gold and Silver (London: E. and F.N. Spon, 1867), 497–503.

the dry refining process: smelting of silver ores

67

trough.64 In this example the stone was powered by a mule tethered to the arm leading to the axle of the wheel. Smelting does not require the flour-like consistency demanded by the patio process (see Chapter 4), so breaking down the ore by hand was also a viable option, especially for small refining haciendas set up during the early years of refining.65 It is not clear when and to what extent these molinos were used in the seventeenth century smelting haciendas of San Luis Potosí. The legal documents of the period are ambiguous, since they mention a wheel (rueda), but in such a way that it could also apply to the assembly required by a bellows and not only the rueda of a molino.66 However the claim by West that Chilean mills were not introduced in New Spain until the nineteenth century does not seem correct.67 The role of dressing ores: the importance given to the requirement of water for the smelting haciendas at first sight is surprising, since water seems a more critical issue for the patio process (Chapter 4) than for smelting: Antonio de Espinoza … townsman and miner of the mines of San Luis Potosí states that as is notorious and public [knowledge] the ores that are extracted from this hill [Cerro San Pedro] cannot be processed or smelted without washing. And due to the great lack of water that has been and still is … some haciendas have ceased to process [these ores].68

64 65

66

67 68

Manuel Amador, Tratado práctico y completo de trabajos de minas y haciendas de beneficio (México: E. Sánchez, Editor, 1901), 66–68, Lámina 5a Figura 1. The final size should range between a grain of rice and that of a ‘Grueso de garvanzo’, the thickness of a chick pea, and the more docile the ore, the bigger the particle, according to Joseph Garcés y Eguía, Nueva teórica y práctica del beneficio de los metales de oro y plata por fundición y amalgamación (México: D. Mariano de Zúñiga y Ontiveros, 1802), 63–64. The bean size had already been pointed out in Biringuccio, The Pirotechnia, 153. For example, a typical inventory will list ‘wheel, pinion, shafts, cross-beam and bellows’, as in the sales contract between Nicolás de Peralta and Cristóbal Zapata, ahslp, Fondo Alcaldía Mayor 1667.3, expediente 6, 18 July 1667. I have used the English translation for the Spanish terms as reported in Appendix i of Bakewell, Silver Mining in Zacatecas, 267. A guide to these terms in Spanish appears in Francisco Xavier de Gamboa, Comentarios a las Ordenanzas de Minas, Madrid, Oficina de Joaquín Ibarra (1761), 399–401. The drive wheel is a fundamental element that transforms the horizontal circular motion imparted by mule or water power via a crankshaft into the reciprocating movement that drives the bellows for the furnace. West, The Parral Mining District, 113. In Chapter 5, I cite evidence that places them in Zacatecas earlier. ahslp, Fondo Alcaldía Mayor 1635.1, expediente 6, 16 January 1635.

68

chapter 2

figure 14 Top left: Molino mill stones, Monte Caldera, San Luis Potosí. The diameter can reach 2 m. Top right: A modern reconstruction of a molino in Zacatecas, at the exit of the El Edén mine. Bottom: Drawing of a molino. illustration reproduction from footnote 64, courtesy of rare books, the latin american library, tulane university

the dry refining process: smelting of silver ores

69

The need to have water for the workers and animals is evident.69 What is interesting is the need for water in order to wash the ores.70 This meant that silver ores in San Luis Potosí were dressed, concentrated for silver and lead content prior to smelting, a procedure based on a differential sedimentation rate in water depending on the density of the mineral particles, as explained by Agricola in the sixteenth century.71 According to Alonso Barba, washing ores prior to smelting was not normal practice in Upper Peru.72 The dressing of ores, an important part of the initial strategy of the English investors of the nineteenth century in Pachuca, was in fact never implemented (Chapter 6). It is important therefore that the documents indicate that for San Luis Potosí, dressing was part of the operational practice in smelting haciendas. Lavadores, workers who carried out the washing of ores, are mentioned amongst the labour force of a smelting refining hacienda in this area as late as 1773.73 The chimneys of the smelting furnaces: smelting was carried out in an horno castellano.74 Alonso Barba defines them as simply the same type of furnace currently in use in most of his known world to smelt all sorts of ores, as described by Agricola. They were initially very simple structures built from stone and lime mortar (‘piedra y cal’) in the form of a square pillar up to 2 metres tall and with an internal square cross section of under one square metre, or built in the shape of an inverted sectioned cone.75 It was an inexpensive construction made from local materials that could be easily rebuilt as required

69

70 71 72 73 74 75

Where water power was deficient, animal power was required to drive mills and bellows, as well as to haul ore from the mines to the haciendas. As late as 1778 a new invention was being touted in San Luis Potosí that would not require ‘abundant water for the beasts’. Claim by Don Juan Martín de Irrazu, ahslp, Fondo Alcaldía Mayor 1778.2, 30 October 1778. The physical infrastructure required for the washing of ores in the haciendas around San Luis Potosí has been identified in Salazar González, Las haciendas de San Luis Potosí, 91. Agricola, De re metallica, 300–310. ‘at great length Agricola teaches how to wash ores prior to smelting; it is not much used in these Kingdoms’ in Alonso Barba, Arte de los metales, 148. Weekly accounts of the Hacienda de Nuestra Señora de los Dolores, signed by Lorenzo de Mata for the year 1773, ahslp, Fondo Alcaldía Mayor 1773.2. This is translated to English as a Castilian furnace. ‘erect these furnaces plumb to the ground, in the shape of a square pillar somewhat taller than wider at the cavity. Their height is one vara [approx. 80 cm], some nearly two and some less … at the back they have a small window … the [alcribis] where the nozzles of the bellows are placed … others make these furnaces round, wider at the top than at the bottom’. Alonso Barba, Arte de los metales, 139.

70

chapter 2

figure 15 Tuyère, or alcribís, date unknown, found in the ruins of the hacienda hmc2 in Monte Caldera

by the constant wear and tear of smelting. An inventory of a rented smelting hacienda in Zacatecas, dated 1608, describes well the precarious nature of the early structures: ‘two chimneys of stone and lime one good and the other with openings in three or four places … four furnaces of stone and lime with chimneys made from adobe, all open in many places and propped up’.76 Not many remain in the area of San Luis Potosí, their intangible presence amongst the ruins of these haciendas is inferred by the more evident remains of a tuyère, an alcribís in Spanish texts, the element of the furnace lower back wall fitted with an orifice, that allowed the bellows to pump air through a nozzle (cañón) into these furnaces (Figure 15).77 Because a smelting furnace required

76

77

In the inventory drawn up by Pedro de Medina and Andrés Pereira for an hacienda ‘de refinar y afinar’ (to smelt and cupellate) rented from Doña Margarita de Cobarrubia in Fresnillo, 20 March 1608, ahez, Notaría-Colonia, Número 01 (Pedro Venegas), expediente 1, 4r. The weight of the alcribís in legal documents is an important factor, possibly as an indication of its quality. For example, the rental agreement between Fraga Gorbaran and

the dry refining process: smelting of silver ores

71

a set of adjacent bellows, its architectural footprint required an additional area contiguous to the furnace to fit both the bellows, the mules turning round in a circle, or a waterwheel fed by hydraulic power. Thus for example: ‘there are two joined mules that move the wheel of said ingenio [machinery] that makes the blast provided by a set of bellows that are placed on the side of one of two furnaces that are found in this hacienda’.78 In the nineteenth century the description had hardly changed: ‘the blast of air is given by two bellows … one mule for each furnace’.79 The space required by this additional area shared by up to a pair of furnaces, where the effective power to drive the bellows was generated, is not always possible to identify in the present reconstructions of smelting haciendas. The low heights of the early chimneys would have exposed the workers and immediate surroundings to the lead fumes issued from the furnace.80 The surviving pyramidal chimney that characterises the ruins of the smelting haciendas around San Luis Potosí (Figure 16) is not the original from the early period but acquired its present shape and height following the mining laws of the eighteenth century.81 By the nineteenth century ‘ordinarily two of these furnaces are placed one beside the other, under the same pyramidal chimney

78

79 80

81

Rodrigo de Aldana, in which it is stated that the new alcribís that weighs 37 lbs must be returned at the end of the rental period. ahslp, Fondo Alcaldía Mayor, 1631.3, expediente 38, 27 December 1631. In a similar vein the cost of three sets of bellows (at 70 pesos each) and an alcribís said to weigh 48 lbs and a nozzle of 13 lbs costing together 60 pesos, are among the few fixed costs listed by Juan López de la Madriz in his book of accounts, ahslp, Fondo Alcaldía Mayor 1650.3, expediente 8. From the report of 1593 written by Juan López de Riego on the smelting hacienda of Captain Miguel Caldera, as cited in Galván Arellano, Arquitectura de San Luis Potosí, 60– 61. Duport, Métaux précieux au Mexique, 70. The chimneys of these early furnaces were sometimes raised in height not because of the fumes but to capture any silver entrained in the flue gas. Alonso Barba, Arte de los metales, 167. Alonso Barba’s dimensions and the height of the extant pyramidal chimneys measured in Monte Caldera and Guadalcázar are significantly lower than Bakewell’s account of a mining edict implemented by Vice-Roy Toledo in 1574 in Upper Peru whereby lead smelting had to be carried out an enclosed building with chimneys some 7 meters tall (4 estados). Peter J. Bakewell, Miners of the Red Mountain. Indian Labor in Potosí, 1545–1650 (Albuquerque: University of New Mexico Press, 1984), 150. Salazar González, private communication. Hermosa describes a ‘German furnace’ in the nineteenth century with a height between 3 and 6 varas, under 5 m, which is more in line with extant chimney heights. Hermosa, Manual de Laboreo de Minas, 254–255.

72

chapter 2

figure 16 Exterior of smelting furnace at the ruins of the Hacienda Santa María in Monte Caldera. The arched port would have been used to feed ore or fuel to the furnace. Approximate height of chimney is around 6 metres.

the dry refining process: smelting of silver ores

73

without a roof on its top’.82 The only example I found of two standing chimneys (Figure 17) was in the ruins of the Hacienda de Aranzazu situated in Guadalcázar, a historical mining and refining district some 100 km north-east from the town of San Luis Potosí.83 The architectural layout of this hacienda has not yet been reconstructed. By the nineteenth century, blast furnaces were installed in Mexican smelting haciendas, with the blast of air driven either by water or steam engines. These furnaces required top-loading of the smelting charge, via an opening to the furnace situated at an upper floor to the level of the furnace hearth.84 Photographs of these furnaces as installed at the Hacienda de Regla, near Pachuca, are provided in Chapter 6. Cupellation of the silver-enriched lead bars (barras in Spanish texts) took place in a separate reverberatory oven, where heating is by indirect reflection from a curved roof and the ore is not in direct contact with the carbon fuel.85 The whole set, furnace and prepared bed, is named a vazo, baso or vaso (literally a vessel, a tumbler) in the legal documents of San Luis Potosí and Zacatecas.86 A single vaso could serve to refine the enriched lead bars from up to four smelting furnaces.87 Because of the attrition of the processes

82

83

84 85

86

87

Duport, Métaux précieux au Mexique, 70. There is an illustration of the side by side arrangement of furnaces under a common chimney, a stamp mill driven by a water wheel, and a reverberatory furnace with a removal metal dome on chains, in the paper on silver smelting by Václav Vaněk and Dalibor Velebil, “Staré hutnictví stříbra,” in Stříbrná Jihlava 2007. Studie k dějinám hornictví a důlních prací (Jihlava: Archaia Brno / Muzeum Vysočiny Jihlava, 2007). The illustration can be accessed via http://www.velebil.net/clanky/ hutnictvi-stribra/stribrna-hut-4. In Guadalcázar I was led to the ruins of the Hacienda de Aranzazu by Doña María Esther, who is also in charge of the colonial museum of the church. For further general information on the area see Galván Arellano, Arquitectura de San Luis Potosí, 70–71. Robert H. Lamborn, The Metallurgy of Silver and Lead: A Description of the Ores; their Assay and Treatment, and Valuable Constituents (London: C. Lockwood, 1878), 125. The low domed roof of these ovens reflects the heat from the wood fire onto a separate chamber where the material to be heated is placed. For details of early sixteenth century reverberatory ovens used in the New World see Alonso Barba, Arte de los metales, 136–138; Carlos Sempat Assadourian, Zacatecas, conquista y transformación de la frontera en el siglo xvi: minas de plata, guerra y evangelización (México, d.f.: Colegio de México, Centro de Estudios Históricos, 2008), 151–152. For later periods see for example Phillips, Metallurgy Silver, 448–455. By the nineteenth century they are still called vasos: ‘le vaso ou fourneau du coupelle’, the biggest width reaching up to 1.4m, and a depth of 15 cm; these ovens do not have an enclosed vent to the roof. Duport, Métaux précieux au Mexique, 70–72. These lead bars enriched with silver were an asset worth stealing from haciendas, as

74

chapter 2

figure 17 Ruins of the Hacienda de Aranzazu in Guadalcázar. Top left: Front view showing archways under two smelting furnace chimneys with a height of approximately 7 m. Top right: Back view of chimneys, showing possible aperture for drive shaft of bellows. Bottom left: Fields of grasas (slag). Bottom right: Satellite view of hacienda grounds. image reproduction courtesy google earth © 2014 digitalglobe, 22°37’21” n 100°24’9” w

on these structures it is not usual to find whole vasos from the first two centuries of refining activity. The fields of grasas: the smelting furnaces produced a solid waste that was dumped close to the refining hacienda. Historically, smelting refineries have had no compunction about soiling their own nests, and major slag heaps still abound around the husks of smelting haciendas in San Luis Potosí, giving the landscape the desolate look and crunchy step of an old lava field

reported in the claim by Rodrigo de Noriega against Juan Rodríguez for allowing two indigenous workers to refine three stolen bars of lead using his set of bellows. ahslp, Fondo Alcaldía Mayor 1650.3, expediente 1, 13 June 1650.

the dry refining process: smelting of silver ores

75

(Figure 18).88 The lead-rich slag from smelters are termed grasas (literally ‘greases’) in Spanish texts of the period.89 Other solid waste products are called granzas and mazamorras (discarded broken-down ore with very low silver content) or lamas (fine waste from washing the ores during dressing).90 None of these solid wastes were considered truly final and worthless products, and the sale contracts for smelting haciendas carried a stock phrase along the lines of explicitly including ‘grasas granzas lamas mazamorras y desechaderos’ among the tangible assets of the hacienda being sold.91 Furthermore, idle land that was suspected of containing waste from previous refining activity was sought after and dug up to recover these potential sources of recoverable amounts of silver and lead.92

Plata de fuego (Silver by Fire) The refiners working with the ores of the mines around San Luis Potosí, implemented the European technique of smelting argentiferous lead ores, since they had no other alternative. They also developed in the process an active local service industry that came to offer a varied menu that covered rentals of smelting

88

89

90 91

92

The heaps of slag at the ancient lead-silver smelting sites of Laurion in Greece were of the order of several million tons, of which approx. 1.5 million were dumped inland and an equivalent amount into the sea, according to Hans-Gert Bachmann, “Archäometallurgische Untersuchungen zur antiken Silbergewinnung in Laurion. ii. Charakterisierung von Bleiverhüttungsschlacken aus Laurion= Archaeometallurgical Investigation on Ancient Silver Smelting at Laurion. ii. Characterisation of Lead Smelting Slags from Laurion,” Erzmetall 35, no. 5 (1982): 246. I have not found an etymology for this use of the word ‘grasas’. A clue may lie in the description by Alonso Barba of molten slag from lead ores: ‘when the slag is well melted, and liquid like oil’. The visual similarity to oil from grease may have led to the word ‘grasas’. Alonso Barba, Arte de los metales, 152. Ricardo N. Alonso, Diccionario Minero. Glosario de voces utilizadas por los mineros de Iberoamérica (Madrid: Consejo Superior de Investigaciones Científicas, 1995), 116, 132, 144. Contract for sale of smelting hacienda by Juan Domínguez de Sequera to Cristóbal del Castillo in Monte Caldera, ahslp, Fondo Alcaldía Mayor 1653.1, expediente 8, 14 March 1653; by Nicolás de Peralta Pimentel to Cristóbal Zapata, ahslp, Fondo Alcaldía Mayor 1667.3, expediente 6, 18 July 1667; by Pedro de la Perna to Captain Juan Manuel Rendón, ahslp, Fondo Alcaldía Mayor, 1696.3, 6 November 1696. Request by Juan López de Meza to dig up a site previously occupied by a smelting hacienda near the Convent of San Agustín in San Luis Potosí in order to process its content of grasas and other solid waste. ahslp, Fondo Alcaldía Mayor 1672.1, 10 February 1672.

76

chapter 2

figure 18 Mounds of grasas in Monte Caldera, Hacienda Santa María, the chimney of Figure 16 in the background

the dry refining process: smelting of silver ores

77

haciendas, mines and workers, and even furnaces by the hour. A newcomer to the business was thus offered the option, or more probable chimera, of becoming extremely rich through the smelting of silver ores in easy installments. The output from smelting registered in the Caja of San Luis Potosí remained fairly constant from the early seventeenth century until the late eighteenth century, when the mines of Catorce would dominate production. The presence of gold was without doubt of great help in allowing the refiners to meet their smelting costs. The production of this plata de fuego both at San Luis Potosí and other smelting regions in New Spain came however with a high environmental cost, that until now has not received much attention in the historiography. Many metallurgical texts up to the nineteenth century mention the high loss of lead products during smelting and cupellation, and the great consumption of charcoal during smelting. The next chapter will attempt to estimate a magnitude for both, as a function of each unit of silver refined by smelting.

chapter 3

The Dry Refining Process: Its Impact on the Environment The strongest argument of the detractors is that the fields are devastated … woods and groves are cut down … then are exterminated the beasts and birds … when the ores are washed, the water which has been used poisons the brooks and streams. agricola, De Re Metallica (1556)

∵ The smelting process presents two environmental impact vectors of special importance: the emission of lead products and the consumption of woodland to produce charcoal.1 I treat these environmental impact variables as vectors, since they not only have a magnitude but also a directionality that is relevant to any environmental analysis. In modern environmental impact studies, the quantification of the level of emissions is determined by in situ field measurements of the chemicals being tracked. For historical estimates of chemical emissions of lead and other chemicals involved in the refining of silver ores, I will use the principle of the conservation of matter. The mass of all the chemical reagents and ore that entered a refining hacienda has to equal the silver produced, any by-products sold, and the total emission of chemicals and mineral waste to the environment. I have not come across any similar approach to estimate historical environmental impacts, but cannot affirm the method is original across all disciplines. If sufficiently detailed historical accounting records are available, the method provides a very good order-of-magnitude estimate, and a precision in time equal to the chronological degree of resolution of the original record, which can be a month or even less. This is a precision 1 Of these two vectors, Agricola’s quote in the epigraph to this chapter mentions only one, the loss of woodland. Agricola’s text refers to lead ores in many places, to the point it compares favourably the profit from lead mines to the profit from agriculture. Lead fume, which figures prominently in the present chapter, is not identified by Agricola as a hazard or poison to merit mention, a point to which I will return in Chapter 9. Agricola, De re metallica, 8.

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_005

the dry refining process: its impact on the environment

79

impossible to achieve using modern dating techniques of soil samples from past centuries. For the aims of this chapter, the method allows me to calculate the quantitative ratio of the consumption/emission of each compound of interest, and the amount of energy consumed per kg of silver refined. This ratio in turn will serve to project over a whole region the quantitative environmental impact for each major emission or energy source, based on the amount of silver produced either by smelting or the patio process (Chapter 9).

Lead: The Nature of Its Consumption The sources of lead to the smelting process are the ore itself and/or any fresh lead compound (poor lead, greta or cendrada) that is added to compensate for the dryness (absence of lead) of the ore. Recycling of lead, greta, cendrada or the recovered accretions from the furnace walls does not enter the gross mass balance equation. The consumption of lead during the smelting process is due to four causes (Figure 19). 1. Loss to the atmosphere of lead fume, spread via the chimney flue gas or within the area around the smelting and refining furnaces. The heating of lead or lead ores creates what is known as a lead fume, an aerosol of particles composed mainly of lead sulphide, lead sulphate, lead oxide, lead carbonate and metallic lead.2 The exact chemical composition and the size distribution of the particles varies according to the temperature in the furnace, the presence of oxygen, and 2 ‘fume is the general name given to the usually greyish white, feathery, partially crystalline, partially dusty deposit … which adheres onto the sides of chimneys and other flueways along which gaseous material from the furnaces pass … its composition was predominantly lead sulphate and oxides, and … silver, arsenic and zinc compounds’, as described in Lynn Willies, “Derbyshire Lead Smelting in the Eighteenth and Nineteenth Centuries,” Bulletin of the Peak District Mines Historical Society 11(1990): 12–13. Modern studies on the composition of lead aerosols from smelting lead indicate the presence of Pb (lead), PbS (lead sulphide), PbSO4 (lead sulphate), PbO (lead oxide), PbCO3 (lead carbonate) and others. usepa, Air Quality Criteria for Lead epa/600/r-5/144aF, vol. i (2006), 2_8. Lead fumes analysed in the nineteenth century, when furnace conditions corresponded more closely to the period of this study, conform to this profile, but the source of the fume (ie furnace conditions and presence of other compounds) determine which specific lead compound predominates. See examples in John Percy, The Metallurgy of Lead Including Desilverization and Cupellation (London: J. Murray, 1870), 451–458. In the discussions that follow I will refer to ‘lead and its compounds’ to encompass all the lead speciation present in lead fumes that are produced during smelting.

80

chapter 3

figure 19 Scheme of the mass balance for lead during the smelting of silver ores. Letters in bold indicate mass input, letters in capitals indicate mass output.

the other compounds present in the furnace. This fume is so rich in lead and lead products that in Europe it was recovered in horizontal flue traps to extract its lead products as of the end of the eighteenth century.3 The toxicity to humans of these lead compounds varies substantially, both with chemical nature, particle size and type of exposure (inhalation, ingestion, contact with the skin). Because of the many combinations of factors possible, a very approximate scale of high to low toxicity spans lead oxide, lead carbonate, small particles of lead inhaled, to the least toxic lead sulphide.4 The toxic effects of the smoke from a smelting furnace, and the lead fume that lodges on the furnace walls, on the workers who were assigned to end a smelting run, is identified by de Gamboa, in his description of the smelting of silver ores at the end of the eighteenth century in New Spain.5 The phrase ‘lead poisoning’ or ‘lead toxicity’ will refer not only to the metal itself but to its many toxic compounds. 3 Metallurgy of Lead, 434–451. 4 The estate of research on the toxicity of lead and its compounds is extensively covered in usepa, Air Quality Criteria for Lead epa/600/r-5/144aF, vol. i, ii (2006). 5 de Gamboa, Comentarios Ordenanzas de Minas, 403.

the dry refining process: its impact on the environment

81

2. Loss via solid particles spread by the wind from stockpiles of ore, greta or cendrada within the hacienda compound (fugitive lead). Again the size, chemical nature and type of exposure of the particles would determine their toxicity. Ingested greta or cendrada (containing lead oxide) would be the most toxic, gross particles of the ore (lead sulphide) the least. Oral ingestion by children of the workers of this fugitive lead would be a major problem.6 3. Loss of lead contained in the solid grasas dumped alongside the haciendas. This lead (metallic lead, lead sulphide, other lead compounds) is encased in a solid matrix, either porous like a lava stone or glassy. Leaching of the lead into the soil would be expected under mostly acidic conditions.7 4. Sales of poor lead and/or greta as a by-product of the process. The last group is contingent on the manner in which each individual smelting hacienda managed its business. Lead-poor ores would require a maximum recycling of any greta or poor lead to minimize the purchase of fresh additions of lead to reach the required lead to silver ratios in the smelting recipe. Lead-rich ores would create a surplus of greta that would be thrown away as waste or else offered to other mining localities that were deficient in lead. In San Luis Potosí legislation was in place that regulated this export of greta, forcing potential sellers to offer it locally during nine consecutive public offerings (pregones) at a set price before it could be sold out of the jurisdiction.8 Penalties for contraband of lead in 1678 were high: confiscation of the greta, the cart, fines, jail for the Spaniards, and 200 lashes for any indigenous workers or African

6 The topic is addressed for modern cases of workers acting as transport vectors for lead compounds from the workplace to the home, or of children poisoned by ingestion of lead contaminated soil or dust, in usepa, Air Quality Lead, i 3_17, 3_27, 3_28. See also footnote 50 at the end of this chapter. 7 One study that reports an enhanced mobility of lead in soils of historic smelting sites, with increasing acidity, is reported in J.E. Maskall and I. Thornton, “Chemical Partitioning of Heavy Metals in Soils, Clays and Rocks at Historical Lead Smelting Sites,” Water, Air, and Soil Pollution 108, no. 3–4 (1998). 8 Request by Antonio Maldonado Zapata to sell in Sombrerete and Guanajuato 200 quintales of greta, ahslp, Fondo Alcaldía Mayor 1674.3, expediente 11, 31 August 1674; request by Dionissio de Rojas y Valdez to export 30 quintales of greta, ahslp, Fondo Alcaldía Mayor 1674.4, expediente s/n, 18 September 1674; request by Fernando de Vaca y Castro to offer locally or export 600 quintales of greta or of lead obtained therefrom, Fondo Alcaldía Mayor 1680.2, expediente 10, 5 October 1680.

82

chapter 3

slaves caught participating in the act.9 The fact that greta could be offered with no takers locally, or that contraband export was attempted, would confirm indirectly that lead was not an issue for the haciendas of San Luis Potosí. As for poor lead, it will figure in my mass balance of Chapter 7 for the case study of the Hacienda de Regla in Pachuca. To calculate the emissions of lead and its compounds into the environment, per kg of silver refined by smelting, I will base my mass balance calculation on the estimated net loss of lead during smelting, in whatever chemical form or physical avenue (slag, emissions to the air) this may take. This allows me to arrive at a mass ratio without having to know exactly the profile of lead compounds involved for each specific furnace condition. A more detailed study on the toxic effects of this net loss of lead will require knowing the speciation of lead products issued during smelting. Since both lead rich and lead poor silver ores were smelted, using added greta or recycled lead to make up any deficiency in lead, the first step is to establish what is the average total lead to silver ratio required for a smelting run. The problem is that historic smelting recipes do not always provide sufficient information on the lead content of the ores, or the quality of the greta, to allow a calculation of this ratio. From scattered data in the historiography, that provides both direct ratios of lead to silver, or recipes with greta and ores of different silver content, a picture emerges of a range of lead to silver from 300 to 1 to 100 to 1, that applies on both sides of the Atlantic.10

9

10

Capture of contraband of 27 cargas of greta destined for Guanajuato by Antonio Veles de la Torre, Alcalde Ordinario of San Luis Potosí. An added incentive was the share of the proceeds from the sale of the impounded contraband greta between the judge and the person who informs and/or captures the contraband. ahslp, Fondo Alcaldía Mayor 1686.1, expediente 11, 14 March 1686. a) Alonso Barba proposes a ratio of lead to silver ore between 2:1 and 5:2 for ores rich in silver, and for smelting silver sulphides he recommends a ratio approaching 4:1. Alonso Barba, Arte de los metales, 152–161 b) A ratio of 1.5:1, subject to the lead and silver content of the ores. Gómez de Cervantes, Nueva España siglo xvi, 158–159 c) West quotes from a 1539 document that in Taxco (New Spain), 25 hundredweight of litharge were required to refine 75 to 126 ounces of silver. Robert West, “Aboriginal metallurgy and metalworking in Spanish America: a brief overview,” in In Quest of Mineral Wealth. Aboriginal and Colonial Mining and Metallurgy in Spanish America, ed. Alan Craig and Robert West (Baton Rouge: Geoscience Publications, 1994), 127 d) The ratios could range from just over 1:1 to 4:1 according to J. de Oñate, Nuevas leyes de las minas de España: 1625 edición de Juan de Oñate: con tratado de re Metalica de Juan de Oñate (Santa Fe, New Mexico: Sunstone Press, 1998), 81–85 Other ratios can be inferred from de Sarriá, Ensayo de metalurgia, 105.; Hermosa, Manual de Laboreo de Minas, 254; Bruno Kerl, William Crookes, and Ernst Otto

the dry refining process: its impact on the environment table i

Year 1718

Range of lead to silver weight ratios from individual smelting runs carried out in the region of Vetagrande, Zacatecas, in 1718. Source data from footnote 11. Lead added

Ore

Silver smelted Silver in ore Lead to silver

Quintales Arrobas Quintales Arrobas Marks 16 August 21 August 26 August

01 September 05 September 15 September 19 September

83

8 2 5 2 2 3 2 5 7 5 8 4

2 3 1 3 2 3 1 2 1

4 3 0

3 4 4 2

2 5 6 6 9 5 1 9 1 1

9 4 8 10 7 4 3 10 19 19 14 3

Oz

% Minimum

Ratio

2 3

1.03 % 1.75 % 1.33 % 3.33 % 2.46 % 0.89 % 1.20 % 1.58 % 2.38 % 4.22 % 1.68 % 0.78 %

184 126 131 55 68 188 150 107 76 53 112 229

3

2

2 4

There is an operational record dated 1718 that provides confirmation of this range. It forms part of a bundle of accounts rendered by Andrés de Soliz related to his management of a ‘fuelle’ (literally, bellows) in Vetagrande (Zacatecas), belonging to Captain Don Salvador de Inostrosa. ‘Fuelle’ in this context is a smelting furnace situated close to the mine head. The specific account that is of interest is signed by Marcos Alcay and begins with an invocation to ‘Jesús, María y Joseph’, followed by the title: ‘Book of charges and discharges of lead and ore that were received in this smelter of captain Don Salvador de Inostrosa … August 16 1718 year’. It contains a record of individual smelting runs carried out from August 16 to September 15 in the year 1718, registering the amount of lead added to a specified quantity of silver ore, and the total amount of silver obtained from the operation.11 I have included in Table i the

11

Röhrig, A Practical Treatise on Metallurgy Adapted from the Last German Edition of Prof. Kerl’s Metallurgy (London: Longmans, Green, and Co., 1868), 219; Eissler, The Metallurgy of Argentiferous Lead, 355. ‘Libro de cargo y descargo del plomo y metal que resivo en este fuelle del capitán Don Salbador de Inostrosa, Agosto 16 de 1718 años’, ahez, Serie Civil c15-e08, 13r, 17 r,v.

84

chapter 3

table ii

Range of percentage values for net lead losses during smelting of lead ores. Sources in footnote 12.

Total % loss

In lead fume

10 to 25 6 21

5.4 5 to 20

7 7 to 17 30

up to 12

Location, period

Source

England, 14c England, 19c Utah, u.s.a., 19c England, 19c u.s.a., 19c Pontgibaud, France, 19c Poland, 16c–17c

a, 10–11 b, 53 c, 365 d, 13 e, 59 f, 480, 489 g, 53

information provided in the document, except for two runs where the data are not clear, and my calculations of the minimum silver content of the ore being smelted, together with the resulting lead to silver weight ratio for each run. The results show that the richer silver ores required less lead for smelting, and that a ratio of 100 to 1 can represent the operational range of the lead to silver weight ratio used to smelt silver ores with approximately 2 % silver content. Estimates of the types of losses of lead during smelting can be found in the historiography of the period, reported as percentages of the total lead employed in the smelting run (Table ii).12 That lead was lost during each smelting run, sometimes in major quantities, is unquestioned. According to Biringuccio it was preferable to lose a certain amount of lead through losses to the atmosphere as lead fume in the last step of cupellation, rather than to lose silver entrained during the removal of the last traces of litharge.13 The values in Table ii nearly all reflect European smelting operations. Some of the higher values correspond to earlier periods, and it could be argued that by the nineteenth century there was a greater economic incentive to control losses

12

13

a) Ian Blanchard, “Technical Implications of the Transition from Silver to Lead Smelting in Twelfth Century Britain,” in Boles and Smeltmills Seminar, ed. Lynn Willies and David Cranstone (Reeth, Yorkshire: Historical Society, Ltd., 1992) b) Michael C. Gill, “Analysis of Lead Slags,” ibid. c) Eissler, The Metallurgy of Argentiferous Lead d) Willies, “Derbyshire Lead Smelting” e) Pique, A Practical Treatise on Silver f) Phillips, Metallurgy Silver g) Molenda, “La métallurgie du plomb en Pologne au moyen âge et aux xvie–xviiie siècles”. Biringuccio, The Pirotechnia, 164–165.

85

the dry refining process: its impact on the environment table iii

Lead content in slags from different smelting sites and periods. Sources in footnote 15.

Average lead in slag (%) Location

Source

3 217

approximate total labour cost

Subtotal

40 3,195

Miscellaneous work paid by unit or varas such as horseshoe fitting, stone laying and wood sawing has not been included since no information is provided except total cost of labour, and it represents a minor fraction of the total wage cost.

I have visually partitioned the total reported labour force, as indicated in Table viii, according to each refining process in Figure 76. In those areas where both processes would have shared the same support, I simply indicate the total number of workers assigned to that activity. The main conclusions derived from this breakdown are: 1.

2.

3.

Repairs make up around 20% of total labour costs, an important amount second only to labour during the patio phase of the process, and above the combined cost of milling and arrastres, or the labour cost for the smelting section. Though the machinery at Regla was not complex, it involved many moving parts and constant contact with abrasive powders, furnaces and water. The labour cost of the planilleros belies the amount of personnel expended in the search for amalgam and mercury that was entrained in the slurry from the washing of the tortas. In number of workers it would have represented a group of people three-quarters the size of the whole patio reactor work-force. This underlines the effort applied to the recovery cycle of amalgam and mercury entrained in the water used to wash the slurries. In the smelting process, the labour related to the furnace smelting of the ores represent 60% of costs, and around 70% of the workforce. There is no equivalent stage in the patio process that absorbs so much labour power. Regla may be an anomaly in this respect, since it possessed six blast furnaces at one time, and was clearly overdesigned for the amount of ore it received for smelting. In addition, the need to maintain a trained

244

chapter 7

figure 76 Labour force assigned to major areas of activity at Regla, in the four-week period ending on May 29th 1877 (based on Table viii)

group of smelting craftsmen and support workers regardless of the actual amount of ore being delivered converted their wages into a virtual fixed cost.21 What the data does not provide is a more three-dimensional picture of the lives of this labour force, their relation to the surrounding communities, if there was rotation between the different refining haciendas of the Company, their training, the other options, if any, in the local area for work, their relation to outlaw bands, to foreign workers, tensions with the oversight applied to control theft and pilfering, and the effects on their health and their life expectancy, and that of their families, from the environmental impact vectors that I will analyse in the following sections.22 During the same period that writers such as Dan de Quille and Mark Twain were providing the human context to the silver refining activity in Nevada, even if only from a white man’s perspective, there is no equivalent for this period in Mexico, other than foreign visitors’ diaries, where the local population barely figures, if at all.

21 22

Since during the late 1870s Regla was operating its patio process to its full installed capacity, the same consideration did not apply to the workforce assigned to this process. A window on work and life at Regla from the viewpoint of the English contingent is provided in Todd, Cornish Miners in Mexico.

the patio process and smelting at regla

245

The Mass Balance for the patio Process at Regla Table ix summarises the magnitude of the main environmental impact vectors from the patio process as carried out at Regla between 1872 and 1888. Based on the analysis presented in Chapter 4, I have calculated that on average 85% of the total mercury consumption, equivalent to 1.8 t/mo., was in the form of insoluble calomel washed away, entrained in the water used to wash the tortas, and deposited over an undetermined area in the water basins downstream from Regla. The remaining 15% corresponded to a physical loss of mercury, an average of 0.3 t/mo., of which the largest share would be by entrainment in water used for washing tortas or in the condensation channel underneath the capellina, together with seepage into the soil in the patio reactor area. Of the total physical loss, I estimate an average of 0.02 t/mo. was due mainly to volatilisation of residual mercury during the casting of the silver bars, and during the heating of the capellinas, 0.6 % at the most on average. Under this scenario, for every kg of silver refined by the patio process at Regla, approximately 1.1kg of mercury was lost as calomel, 0.2 kg was lost as liquid mercury in the waste water or by seepage to the soil at Regla, and just over 10g were lost as volatile mercury. In terms of mass, these losses pale in comparison to the 620kg of solid mineral waste and salts in solution that were washed downstream from Regla for every kg of silver refined by the patio process, from an ore with 0.19% silver content. The yearly average amounts to approximately 12,000 t of mineral waste and salts, 20t of mercury in the form of calomel, 4 t as losses of liquid mercury and 0.2 t of mercury via air emissions.

The Mass Balance for Smelting at Regla The average monthly mass balance for lead can be expressed as: L + Lith = la + ls + p, where L denotes lead in the silver ore, Lith is the amount of lead in the litharge consumed, la indicates lead in compounds issued to the atmosphere, ls indicates lead in solid slag and p is poor lead. The problem lies in that the lead content of the silver ores was not reported in the monthly ledger accounts at Regla, nor was the lead content in the slags. To arrive at a first approximation on the amount of lead lost as air emissions,

table ix

Mass balance for the patio process as practised at Regla between 1872 and 1888. Data compounded from different sections of this chapter. The numbers in italic denote a calculated number, not directly derived from the accounting data.

246 chapter 7

table x

Mass balance for the smelting of silver ores as practised at Regla between June 1875 and January 1886. Data was compounded from the sections on smelting of this chapter. The numbers in italic were calculated and not obtained directly from the accounting data.

the patio process and smelting at regla

247

248

chapter 7

I will assume L is nil (this can be adjusted as more data becomes available), and that the lead content in the slag is on average 3% (see Chapter 3). Furthermore, I assume that all the non-silver content in the ore is lost as slag, and ignore any added fluxes. Based on an average of 5.4 t/mo. of litharge at 92.8 % lead content, 24.7 t/mo. of ore with 1.9% silver and sales of 0.15 t/mo. of poor lead, the equation gives the following working figure: la = 4.1 t/mo. This value corresponds to a 10 to 1 ratio with respect to the average monthly production rate of silver by smelting of 0.4 t/mo. It correlates with the upper level of the range deduced in Chapter 3, for the ratio of losses of lead and lead compounds to the atmosphere for every kg of silver refined by smelting. Based on these projections, Table x summarises the mass balance for smelting as practised at Regla from early 1875 to early 1886. For every kg of silver, 59kg of solid waste would be generated. An ore with zero lead content would produce a loss of lead compounds in the fumes of 10 kg of total lead per kg of silver refined by smelting. Any additional lead in the ore or any loss of waste ore by volatilisation in the blast furnace, instead of as slag, would increase this baseline level. On a yearly average, at Regla some 50 t of lead in lead fume would have been issued to the atmosphere during this period, and some 300 t of solid waste in the form of slags.

The Environmental Loss Vectors in the Period 1872 to 1888 Four areas of the environment were impacted by the refining activities at Regla (Figure 77). a

The Stream Flowing Past the South and Eastern Perimeter Walls of Regla The stream to the side of Regla becomes a tributary of the Río Metztitlán, which in turn flows until it reaches the natural dam of the Laguna de Metztitlán, approximately 60km downstream from Regla. Every year some 12,000 t of solids would have been discharged into the stream, to join the waste generated by the haciendas upstream of Regla (San Antonio and San Miguel). In terms of heavy metals, it would be the conduit to dispose of around 90 % of all mercury losses, mainly in the form of calomel, encapsulated within the fine solid mineral silt from the milled ore. The loss of lead compounds via the stream would be minor since there is no textual evidence for the dressing of

the patio process and smelting at regla

249

figure 77 Main loss vectors of waste material, monthly average at Regla in the period 1872/73 and 1875/88 (patio process) and Jun 1875 to Jan 1886 (smelting)

ores at Regla. Losses of liquid mercury would have been entrained far downstream from Regla.23 b The Soil within the Perimeter Walls of Regla The soil of the patio reactor area would have been impregnated with mercury that percolated through the tortas and through the wood slabs that covered the patio. In contrast to the waste continually removed from Regla by the stream, the soil within Regla would become a depository of accumulated liquid mercury, down to an unknown depth. How much of this mercury would have been entrained in the skin and clothes of the workforce and transported to their homes and families is unknown. Fugitive losses of lead compounds, including litharge, would also have accumulated within the grounds of Regla, constrained by the very high perimeter walls, but I am not able to quantify this amount.

23

Modern studies on the presence of mercury downstream from artisanal centres of gold extraction using mercury are detecting it as far away as 600 km from the site of mining. Sarah Diringer et al., “Mercury Biogeochemistry and Artisanal and Small-Scale Gold Mining in Madre de Dios, Peru,” (Poster, 11th International Conference on Mercury as a Global Pollutant, Edinburgh 2013).

250

chapter 7

c

Airborne Emissions and Deposition of Airborne Particles in and around Regla Air was the domain of lead fume emissions, not of mercury. I project that up to 200 times more lead (as lead fume) was issued to the air than mercury during the whole period covered by this chapter. Taking into account that smelting was not carried out during certain years, and that smelted silver only corresponds to 14% of all the silver produced during the period, this is a significant ratio. From the point of view of the workers, lead fume posed a much more concentrated threat in time than the constant but much lower level of mercury emissions. Within Regla, high ambient levels of lead compounds would permeate the areas around the hearths of the blast furnaces and in the load floor during the charging of the stack. Other areas within and without Regla would be subject to airborne diffusion of lead compounds in particles, and to the deposition of these airborne particles from the furnace stacks. The exact footprint of these lead-rich depositions can only be established with knowledge of the historic prevailing wind direction in the gorge. The isolation of the site and its location within a gorge would have concentrated the impact of its air emissions to the workforce and local dwellings, while minimising it on the surrounding area. d Loss of Woodland Smelting was a much greater cause of deforestation than the patio process at Regla. Even with much more efficient blast furnaces at the end of the nineteenth century, charcoal consumption was still thirty times greater for smelting than for the patio process, based on the heating energy requirements to produce 1kg of silver. If the whole period is considered, the patio process at Regla would have required the equivalent of approximately 2,000 t of charcoal, and approximately 10,000 t of charcoal for smelting. The total amount of woodland that would have been deforested in the period 1872 to 1888, estimated as required to supply the industrial fuel for Regla, would have been in the range of 5,000 hectares (over 12,000 acres), of which more than 80 % was due to the smelting process. These projections are within the order of magnitude of data reported by Buchan in 1855 for Regla, which covered the total requirement for timber for mining and wood for charcoal for all the operations of the Company, where steam engines and the barrel process were also major consumers of energy: our consumption of wood is not less than 60,000 tons … per annum … for the supply of this fuel we hold … some 25,000 acres, for the most part forest … the nearest of these woods have already, during the last twentyfive years, been much diminished; but we have lately acquired others …

the patio process and smelting at regla

251

and with due care of the young trees which are reproducing in those portions already many years cut, even our nearest forests are not likely soon to fail; while from a distance … the supply is inexhaustible.24

A Snapshot of a Refining hacienda Regla smelled not of riches, but of rich animal manure and the sweat of overworked men and animals, overlaid with the earthy overtones of the dark mud of the silver ore slurry spread out over the patio. The smelting runs punctuated with their acrid sulphur the background levels of wood smoke and lead fume from the reverberatory furnaces. Gritty dust must have coated all surfaces, dust from the stamp mills, from the piles of ore and charcoal and litharge, dust whirling impotently, imprisoned within those imposing perimeter walls. The basalt columns of the gorge resonated with the cacophony from the daily pounding of the stamps, the whirring and scraping of the arrastres, the braying from the mule-trains, the neighing of the horses and the shouts of the workmen. In the background water gurgled continuously through channels, splashed from spouts, shoved against water-wheels, lubricated the ground ore, held together slurries and wetted the tortas, washed away mountains of unwanted waste, out of sight and out of mind, leaving the grounds of the Hacienda free from the eyesore of hills of slag or useless ore. It is no surprise that the first Conde de Regla chose the less intense Hacienda de San Miguel as his residence, instead of the fortress at Regla. Over this ‘Babylonia’ of back-breaking hard labour hung the mistrust and eternal vigilance of its management, overseeing the production of the small but heavy silver bars with the same obsessive attention to detail that a miser pays to his hoard. A cat and mouse game of accounting and production versus pilferage of ore, mercury, amalgam and silver, played every day of the year under the gaze of guards and overseers, regardless of barred gates and underground storage chambers. The mercury legacy of silver refining at Regla lies dormant underground, as calomel entombed along river beds or as mercury impregnating the soil within the hacienda, but only a minor fraction as mercury dispersed long ago in the air. The air at Regla was the domain of dust from the morteros or the lead fume from the Satanic furnaces and reverberatory ovens. The former would tend to be confined within the compound of the hacienda, the latter would heavily contaminate the work areas adjacent to the furnaces, or come to ground after

24

Buchan, Report Real del Monte, 19.

252

chapter 7

having been spewed from chimneys with no long flues or bag houses to protect their immediate surroundings. The English managers and workers had brought pasties and football from England to Pachuca, but there is no evidence of an attempt to control and collect the lead set free in the furnaces. Regla had a most wasteful metabolism, expelling as useless over 99.6 % of what it ingested. In terms of sheer weight, it was the volume of solid waste that would have had a major impact on the immediate environment, had it not been washed downstream, away from its source. The impact of large amounts of fine mineral silt, calomel and increased salt and copper levels in the water used by other communities far from the three refining haciendas polluting this stream of water. remains to be studied and quantified. As to the impact on woodlands, all metallurgical activity created the need for firewood and charcoal. The data from the end of the nineteenth century show that improvements in furnace efficiency had brought down the consumption of charcoal for smelting to nearly a sixth of the level reported for Pachuca one hundred years earlier. Even so, the forests around Regla were being cleared for wood, at a rate that would have become a major obstacle had more ores been destined for smelting. The calculation of the mass balance of these processes, based on historical accounting records, has proved to be a very powerful tool. The method can quantify the environmental impact of a historic industrial process, with a historical precision only limited by the time spans in the accounting ledgers. In the case of Regla, it would be possible to determine magnitudes of consumption and emission of one week or one month, of events taking place nearly 150 years ago. This level of precision in the chronology and measurement of historical environmental events is impossible to achieve with other methods, such as those that determine the historical deposition of elements through the dating and analysis of cores of soil samples. The method I have applied is based on the principle of the conservation of matter: the weight of all materials consumed in the hacienda is equal to the weight of all materials produced at the hacienda, silver and waste products. Regla produced a mix of silver from the patio process and smelting that mirrors the nineteenth century more than the colonial period, when smelting was more prevalent in New Spain (Chapter 9). On average the quality of the ore refined by the patio process was 0.19% silver by weight, and 1.9 % silver by weight for smelting, if anything within the higher range of silver ores in Mexico. The ores sent for refining by the patio process at Regla were mainly silver sulphide ores, as were most of the ores of Mexico, with the presence of native silver. There was no help from silver chlorides in these ores. The consumption of mercury, salt and copper sulphate at Regla lies on the lower range of the historical scale, but not outside the expected parameters. The total

the patio process and smelting at regla

253

silver output from Regla during the period covered in this chapter reached a total of 350 t, which represents 0.7% of all the silver refined in New Spain from the sixteenth to the end of the eighteenth century. The operations at Regla at the end of the nineteenth century were as chemically representative as any other that took place from the end of the sixteenth century. Even iron, the additive that first appears in Upper Peru in the late sixteenth century, continues to make its presence felt at Regla. Only the efficiency of the process could have improved during this whole period, as shown by the marked reduction in the use of charcoal for smelting, or the late switch to faster turn-around times for the tortas in the patio reactor, but not its chemical underpinning. How representative is Regla of the overall environmental impact of silver refining in New Spain or Mexico? From a chemical point of view, it is as good a case study as one can hope for. As to the quantities involved, the environmental impact of silver refining in Mexico from 1850 to 1900 would have been more intense on a yearly basis than at any other period since the arrival of the Spanish silver refiners. The total emissions within Mexico from silver refining from the last fifty years of the nineteenth century would have been chemically identical and quantitatively close to the total emissions from the sixteenth to the eighteenth century. Regla is thus a microcosm that faithfully embodies the practices of historical silver refining in the New World. This whole work is predicated on ignoring the distracting multiplicity of what is undoubtedly a complex historical scenario, while trying to use basic chemistry and physics to focus on the metaphorical forest instead of so many trees. Wind patterns are site-specific, the architecture will change the relative height differential between chimneys and the perimeter walls, the efficiency of the processes will improve with experience and design, refining haciendas will be isolated units in the countryside or embedded within city limits, silver ores come in one hundred different flavours, but the picture that emerges from Regla transcends these inevitable accidents of time and location. The lead content in the lead fume was the only heavy metal that was discharged to the air in great quantities, as a result of historical silver refining activities in New Spain / Mexico, right up to the end of the nineteenth century. Calomel, the solid and insoluble chlorine salt of mercury (i), trapped nearly all the consumption of mercury and was washed downstream from all the haciendas de patio. The immediate danger of mercury to the people in the hacienda lay in its constant contact with the skin, and in the accumulation of liquid mercury in the soil within the perimeter walls, or present in the waterways downstream from each hacienda. This exercise in industrial reconstruction, based on account books and extant physical structures, has opened a window on the operational challenges faced by colonial silver production in New Spain. It provides clues as to how

254

chapter 7

they may have been addressed, not only on an industrial scale in Mexico towards the end of the nineteenth century, but also over the previous colonial period. This quantitative profile of the industrial architecture and operations required by the patio process and smelting, represents the final stage of an evolution in spaces, equipment and processes that began in the sixteenth century in New Spain. It has also aimed at focusing attention on the urgent need to interpret and conserve the remaining industrial heritage of Mexico related to the historic refining of silver ores. Many of its refining haciendas constitute unique examples of a novel industrial architecture within their historical period, developed locally in the New World, yet time, neglect and reconversion works threaten to obliterate the last physical traces of the history of the silver refining industry in New Spain and Mexico. The value of the account books kept for Regla and the other refining haciendas of the Compañía de Minas del Real del Monte y Pachuca is not limited to these insights. By providing information on the economies of each refining process they also become the key to understanding the choices open to the refiners of silver in the New World, whether it was the chemistry of the ores or the production cost that tilted the balance of refining options. This will be the subject of the next chapter.

chapter 8

The Economies of Refining Silver It is at this point, of course, that lack of knowledge of the costs of mining becomes a true hindrance to explanation. In particular, it would be desirable to know at what yield, all other things being equal, ores became profitable. p.j. bakewell, Silver Mining and Society in Colonial Mexico, Zacatecas 1546–1700 (1971)

∵ Roads to Riches When the first Conde de Regla died on the 27th November 1781, he left what has been termed as ‘probably the largest estate of any noble in the colony [New Spain]’, a sum that may have reached up to 5 million pesos.1 When Josiah Wedgwood died in England fourteen years later, ‘one of the two dozen wealthiest men in Great Britain’, an icon of success in the England of the Industrial Revolution, he left an estate valued at 300,000 pounds, around one third of the wealth left by the Conde de Regla.2 The rise of a private individual well above his social origins to levels of great wealth, thanks to his business acumen and industrial empire, did not only take place within the newly industrialising European core. The opportunity arose because the private sector in New Spain was given a margin of around 80 percent of the value of all the silver refined, to cover its operating costs and gain its profit.3 For nearly 300 years Spain placed

1 Couturier, The Silver King, 172. 2 Finlay, The Pilgrim Art, 291. I have used an approximate exchange rate of 4s 6d for 1 silver peso, as reported in Marichal, Bankruptcy of Empire. 3 Humboldt estimated that 13 to 19% of the value of silver was retained by the Crown. Humboldt, Essai politique, Tome iv, 144–146; 78.8% remained to the private silver refiner, according to Kendall W. Brown, A History of Mining in Latin America: From the Colonial Era to the Present (Albuquerque: University of New Mexico Press, 2012), 23; ‘silver taxes accounted for at least 12 percent and mintage fees for another 6 percent of silver’s total value … transportation and miscellaneous expenses … another 3 to 4 percent’ in Garner and Stefanou, Economic Growth

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_010

256

chapter 8

the whole financial risk of refining the silver ores of New Spain (and elsewhere in the New World) on the pockets of venture capital. The Conde de Regla was part of the very small group of hugely successful investors and entrepreneurs who profited from this scheme. A greater anonymous mass of Spanish mineros and hacendados participated in the silver lottery of the New World, a segment where failure and bankruptcy was more common.4 As to the greater number of indigenous paid workers and supporting local communities, the historical opportunity cost of silver mining and refining remains to be examined. Nevertheless, the fact remains that mining and refining of silver ores in the Hispanic New World had to be a profitable industry at different scales and levels, for sufficient private individuals to have maintained continuous production for more than three centuries. The profits generated from silver rippled across the local economy, and made the tax payers of New Spain rich enough to convert the province into a ‘fiscal submetropolis’ within the Spanish Empire, capable of financing many of the region’s needs.5 Yet even though we know how much silver was registered, how much mercury was sold and at what price, and how much was raised in taxes, it is not clear how and why these profits were even possible in the first place. There was no guarantee that refining by the patio process, a process never applied to silver ores in Europe, could provide profits to its practitioners in the New World. In addition, who could foretell that smelting silver ores, without the benefit of concurrent sales of lead or copper, would make the refiners very rich? This chapter focuses on finding an answer to these questions, even if unfortunately, there is little colonial data to guide us, since the historiography is quite miserly on the detailed refining costs of silver from ores in the New World.6 The isolBourbon Mexico, 119. ‘The pro-business institutional framework in the mining industry … instead of a predator state … the institutional framework of the mining industry was … even more liberal than, those that co-existed in western Europe’, according to Rafael Dobado and Gustavo A. Marrero, “The Role of the Spanish Imperial State in the Mining-led Growth of Bourbon Mexico’s Economy,” The Economic History Review 64, no. 3 (2011): 862. 4 In the Zacatecas of 1626, after 15 years of ‘unprecedented amounts of silver’, only 4 out of 95 owners of haciendas and mines were truly rich, according to Bakewell, Silver Mining in Zacatecas, 207. See also “Colonial Mining,” 131–133. According to Brading, 8 out of 10 miners lost their money. Brading, Miners Bourbon Mexico, 169–170. 5 Marichal, Bankruptcy of Empire, 19. For events on the other side of the Atlantic, Stein and Stein present an overview of the role of silver in the evolution of the economy of Spain, vis a vis the rest of Europe, throughout the colonial period. S.J. Stein and B.H. Stein, Silver, Trade, and War: Spain and America in the Making of Early Modern Europe (Johns Hopkins Univ Pr, 2000), 19–31. 6 See comments on the dearth in production economic data in Bakewell, Silver Mining in

the economies of refining silver

257

ated clues available relate mainly to the patio process, not smelting. I will begin by reviewing some of the main sources on refining costs, that date mainly from the eighteenth century onwards. Once I have presented the available data, I will present the numbers provided by the accounting books of Regla, since they provide a wealth of information on refining costs. The period of research covers monthly accounts from 1875 to 1888, which have allowed me to arrive at a complete profile of refining costs for both the patio process and smelting, as practised at an industrial scale refining hacienda in the latter part of the nineteenth century. That information is then used to project back in time the deemed costs of both processes relative to one another. I have chosen two periods for this exercise, the initial phase of refining in the second half of the sixteenth century, and the period between mid seventeenth and mid eighteenth century that is known to coincide with issues related to the pricing and availability of mercury in New Spain. I then take a critical look at some conclusions reached in the historiography that merit a closer analysis, in the light of the new data I present in this chapter.

Refining Costs in New Spain, as Reported I begin with one of the most perceptive analyses of refining costs of the patio process practised in New Spain, that dates from the mid-eighteenth century. It is a discussion in print between the Accountant General of the Royal Mercury (Contador General de los Reales Azogues), José Antonio de Villaseñor y Sánchez, and the Overseer of the Royal Mint (Guardavista de la Casa de la Moneda), José Antonio Fabry.7 It takes place between 1741 and 1743, and contains as its central theme the two main arguments used at this time in favour and against Zacatecas, 187, 207; Brading, Miners Bourbon Mexico, 154, 158; Garner and Stefanou, Economic Growth Bourbon Mexico, 118; Jaime J. Lacueva Muñoz, La plata del rey y de sus vasallos: minería y metalurgia en México (siglos xvi y xvii) (Sevilla: Consejo Superior de Investigaciones Científicas, Escuela Superior de Estudios Hispano-Americanos; Universidad de Sevilla; Diputación de Sevilla, 2010), 58; Lara Meza, Haciendas de beneficio de Guanajuato, 102, 105. 7 José Antonio Fabry, Compendiosa demostracion de los crecidos adelantamientos, que pudiera lograr la real haciencda de su Magestad mediante la rebaja en el precio del azogue que se consume para el laborio de las minas de este reyno … con una previa impugnacion à las reflexiones del contador Joseph de Villa-señor y Sanchez (México: Impressa por la viuda de J.B. de Hogal, 1743). José Antonio de Villaseñor y Sánchez (1703–1759) was the author of a detailed description of the main cities and provinces of New Spain, published in two volumes as the Theatro Americano. For a biography see Alejandro Espinosa Pitman, José Antonio de Villaseñor y Sánchez, 1703–1759 (San Luis Potosí, s.l.p., México: Universidad Autónoma de

258

chapter 8

a decrease in the price of mercury, which had been held at 82 pesos per quintal for nearly 150 years. Villaseñor is defending the revenues to the Crown from the sale of mercury. He argues that most of the costs incurred in the patio process can be considered as fixed, irrespective of the silver content of the ore. The only cost that varies significantly with silver content is that of mercury consumed, which by his estimates is a minor fraction of the total cost. He then states that even with mercury at 82 pesos per quintal, the refiner could still make a profit with ores of 2oz of silver per quintal (approximately 0.13% silver content by weight). Lowering the price of mercury would not help the refining of ores with low silver content. Since the expense for mercury is proportional to the silver content, ores with little silver would only incur a small cost for mercury, and yet still have to meet the much greater relative cost of all other refining expenses, out of an ever-decreasing monetary value of silver content. He concludes it would not make sense to refine these ores even if mercury was given away for free. He also considers the case of tailings, whose cost of production can be considered nil since it is now a sunk cost. He estimates that even at 82 pesos for mercury the refiner could make a profit working with tailings that had 0.06% silver content. By highlighting the influence of the extraction cost of the ore on the final profit level of the refiner, he makes an extremely pertinent observation: what makes mines unprofitable is the labour-intensive drainage, the tough nature of the ore wall, the timbering of the supports, the cutting and removal of the ores, the expense on the workers, the need of food, forage for the animals powering the drainage … even if they gave away lakes of mercury, these costs would never be met for lower silver contents … what restricts silver mining is the lack of ingredient [mercury], not its price.8 What is impressive about Villaseñor’s line of argument is not so much its mathematical conclusions, which as always depend on the validity of the starting values, but the method for analysing the sensitivity of refining costs to the silver content of the ore. I have translated his written line of argument into tabular form in Table xi. He freezes as a virtual fixed cost all the costs of refining, irrespective of the silver content of the ore, except for mercury. He includes the extraction cost of the ore to calculate the level of profit for the refiner. He San Luis Potosí, 2003). He was the senior official in New Spain in charge of accounting for the distribution of mercury under the Crown monopoly. 8 Fabry, Impugnacion a reflexiones de Villaseñor, 5.

259

the economies of refining silver table xi

Interpretation of Villaseñor’s working examples and method that sustained his argument against decreasing the price of mercury. Source data from footnote 7. Ore source Mine

Tailings

% silver

by weight

0.5

0.25

0.125

0.063

0.063

silver content

oz / quintal

8

4

2

1

1

marks / 100 quintales

100

50

25

12.5

12.5

value of silver in 100 quintales @ 8 pesos 6 tomines per mark of silver

873

436.5

218.25

109.125

109.125

cost of mercury consumed

82

41

20.5

10.25

10.25

other refining costs, including the extraction cost of the ore @ 1 peso/quintal

150

150

150

150

pesos

other refining costs, excluding the extraction cost of the ore

50

total variable costs of refining

232

191

170.5

160.25

60.25

profit or loss

641

245.5

47.75

-51.125

48.875

differentiates the ores from tailings by assuming the extraction cost of the latter is by now a sunk cost, thus nil. He recognises that the higher expense of mercury for ores with higher silver content is compensated by the increase in value of the total silver refined. All that is missing is the fixed capital cost to the refiner and its impact on the break-even pricing for mercury.9

9 He also provides the historian with one of the first indications as to an order of magnitude of ore extraction costs in mid eighteenth century (1 peso per quintal), as well as that of the costs of the patio process net of mercury (0.5 pesos per quintal). Ibid., 1–5.

260

chapter 8

Fabry countered by arguing that a large fraction of ores being mined lay below the breakeven point calculated by Villaseñor for a mercury price of 82 pesos per quintal. By questioning the values adopted by Villaseñor, he concludes that at 41 pesos per quintal it would be possible for refiners to make a profit using the patio process on the abundant ores with a lower silver content. He then calculates that the increase in tax revenues from the additional production of silver would more than compensate the Crown for the decrease in the revenues from mercury. Fabry’s line of reasoning highlights that the aim in the reduction of the price of mercury was to increase the production of silver, not to lower its price, and thus compensate through increased royalties for the loss of mercury revenues to the Crown monopoly.10 A very similar argument to that of Fabry is used in a 1774 report addressed to the Spanish King, Charles iv, by the Guanajuato medical doctor and miner Manuel José Domínguez de la Fuente. The reduction in the pricing of mercury from 82 to 62 pesos per quintal had not pacified the miners and refiners, who sensed that the Bourbon King was willing to decrease the price even further, since he was asking for their official view on what should be the ultimate lowest price of mercury. It is in this light that any economic breakdown provided by Domínguez de la Fuente should be read: it is an exercise in lobbying the Crown in favour of lower mercury prices, not an accounting book.11 He argues for a lower price for mercury since this would allow miners to market the majority of the ore extracted, that at the time was being rejected as unprofitable by the refiners.12 He argues that the Crown would compensate the drop in mercury

10

11

12

Fabry questions, amongst other issues, both the monetary value of one mark of silver (8 pesos 6 tomines instead of 7 pesos 5 tomines) and the cost of ore extraction employed by Villaseñor. Ibid., 6–36. As a miner of modest means he faced the mining/refining model of the eighteenth century, whereby those who only refined the ore by tolling (maquila) in haciendas took most of the profit, and left the risk to those who only extracted the ore: ‘the Miner is he who works; and the Refiner he who benefits’, a pun on the refining (beneficio) of ores. He claimed that the owners of refining haciendas purchased the ores from miners ‘under very questionable assumptions [as to deemed silver content, in the absence of assaying] … the Miner leaves with his doubt and mistrust and the Buyer keeps his reserve [the margin between the deemed and the real silver content in the ore] and his caution’. The value of silver extracted was paid two and a half months after tolling in Guanajuato, and the cost of the maquila was considered ‘very burdensome and because of that very disheartening’. Domínguez de la Fuente, Leal Informe Politico-Legal, 102, 104, 199. According to Domínguez de la Fuente the most common ores had a silver content of 3 to 5 marks per montón of 32 quintales I have inferred he refers to marks per montón, since

the economies of refining silver

261

price by selling more mercury to refine by the patio process the new quantities of ore previously discarded by refiners.13 Once the price of mercury had been halved after 1776, as part of the Bourbon reforms on mining, the issue of mercury revenues versus price no longer crops up. Refining costs are reported but with no details on how they are calculated. Humboldt, an otherwise good source of silver refining data, is uncharacteristically silent on refining costs.14 Sonneschmidt’s claim that compared to refining with mercury, as practised in New Spain, ‘there is no cheaper refining cost even in Europe’, is not backed up by any data, and in any case is comparing apples and pears, as I will argue towards the end of this chapter.15 From the mid-1820s to the end of the century, much more information on refining comes out of the new independent Mexico, concurrent with the opening up of the silver mines to foreign investment. The most detailed economic data on the patio process and smelting in Mexico during the nineteenth century come from three sources, St Clair Duport (1842), Buchan (1856) and Laur (1871). These authors provide an oasis of detailed information on refining costs and practices based on a wide selection of Mexican refining operations, within an otherwise barren landscape on either side of the nineteenth century.

13

14

15

marks per carga would indicate too rich an ore not to be purchased. At the time refiners in Guanajuato were not buying ores for the patio process with less than 5 marks of silver (0.083%) per montón. He proposes that refining centres (Haziendas Refaccionarias) be set up to work expressly with the large amounts of ore available with just 4 marks per Guanajuato montón (0.06% silver), to be sold at a price from 3 to 8 reales per carga. His quantitative argument is not easy to follow. Ibid., 90–91. The compensation in direct revenues to the Crown comes from the opportunity cost of both additional silver production and mercury sales that would not otherwise have been possible at the higher price for mercury, if Villaseñor’s arguments are discounted for now. Thus assuming a correspondencia of 100 marks per quintal, a deemed value of 8 pesos per mark of silver and direct government revenues of 20 % of the silver produced, a drop from 62 pesos to 41 pesos per quintal required a 10% increase in mercury sales just for the Crown to breakeven in direct revenues. In turn this would have meant a 25 % increase in ore purchased by the refining haciendas, an obvious windfall for miners. Humboldt did compare the extraction costs for the mines of Valenciana (Guanajuato) and Himmelfürst (near Freiberg, Germany) in an average year at the end of the eighteenth century. At Valenciana the extraction cost was 6.9 livres tournois (1.3 pesos) per quintal, or 60.4 livres tournois (11.5 pesos) per kg of silver refined. For the European mine, it comes to an extraction cost of 17.1 livres tournois (3.3 pesos) per quintal of ore, 104.3 livres tournois (19.9 pesos) per kg of silver refined. Humboldt, Essai politique, Tome iii, 413. Sonneschmidt and de Fagoaga, Tratado de la amalgamación de Nueva España, 92.

262

chapter 8

Duport’s monograph on Mexican silver production is a work that well merits the recognition given by his peers at the time.16 Based on his extensive experience in Mexico, he was another staunch European supporter of the patio process. His breakdown of its refining costs, excluding the cost of ore, is very perceptive. He captures in a single set of numbers its essence: pure animal and human energy (34%) mixed with chemical reagents (60 %), with no complicated equipment in-between. He believed the major challenge for the patio process was the search for lower costs of milling the ore, since this stage absorbed more than half the total refining cost, and around 19 % of the final value of silver.17 Thus the importance of having access to water power, which was virtually free, instead of having to rely on animal power, which required feeding the animals, or steam engines, which required firewood in great quantities. Duport includes the variant of roasting the ores prior to the patio process, as practised in Taxco, where the threshold for profits was a silver content of 0.125 %.18 He considered smelting as practised in Mexico up to the early 1840s to be inefficient as to fuel consumption, with no effort made to improve furnaces or the fluxes used, losing 15% of silver in the process. For Duport there was not sufficient incentive to improve current practices, since he considered it was not an option to refine the most common ores with a silver content between 0.15 and 0.2% of silver.19 In 1855 John Buchan made public his report to the Directors of the Compañía de Minas del Real del Monte y Pachuca, on the workings of the mines

16

17 18

19

The treatise by St. Clair Duport was judged by Percy, a major author in the nineteenth century on metallurgy, to be ‘one of the best on the subject’ Percy also mentions that he could not add the details on operations at Pachuca because a Mr. Buchan of the Real del Monte Company died suddenly a few days after promising to add to the text to be published on the patio process. Percy, Metallurgy, 576. Randall has pointed out that Duport did not include the major operations at Real del Monte and Regla, apparently due to conflicting commercial interests. Randall, Real del Monte, 236. Duport, Métaux précieux au Mexique, 370–375, 400. The extraction cost of ore accounts for 52%, at 1.08 pesos per quintal, 19% to mercury, 6% to salt and 4% to magistral. The cost translates to 30.1 pesos per kg of refined silver, which is getting close to the deemed silver value within the ore, but it already includes a 3.5% margin for the operator of the refining hacienda. According to Duport, under these conditions (mercury at 130 pesos per quintal, salt at 3 pesos per quintal and magistral at 3.5 pesos per quintal) the operator will not be able to refine at a profit ores with less than 0.125% silver content, since already at this level the cost of production rises to 8 pesos per mark of silver (35 pesos per kg of silver). Ibid., 340–341. The cut-off value corresponds to Villaseñor’s limit for the patio process at 2oz of silver per quintal of ore. Ibid., 398–399.

the economies of refining silver

263

and the refining of silver in their haciendas, including Regla. It is a source of generous information on the cost structure of a working commercial concern, rare for the degree of detail provided to the public.20 Due to the relevance of its information to this chapter, I will address its content within the context of my own analysis of the accounting data from Regla in the following sections. The third author of note is Laur. He estimated that the range of silver content of most ores in Mexico lay between 0.1 and 0.27%, which in his view was sufficient to cover the variable mining and refining costs of the ore. Based on his data from average haciendas de patio, the variable refining cost (without including the extraction cost of the ore or fixed capital costs) for ores with a silver content between 0.09 and 0.2%, was equivalent to 30 to 42 g of silver per 100kg of ore, thus at least equivalent to 20% of the silver content of the ore.21 In the case of smelting, he concluded that in general its costs were very high due to a lack of fuel, unless the ore contained 20–25 % of lead, in which case with a silver content of 0.1% smelting could be profitable. For lead-poor silver ores the threshold silver content for smelting to be profitable in Mexico was 0.5%. He mentions a total cost for smelting in Mexico between 165 francs (15 pesos) per ton of ore to 800 francs per ton (72 pesos), but does not explain what determines the range. The partial variable costs were location-specific: excluding the extraction cost of the ore, labour could make up from 19 to 55 %, fuel 25 to 41%, and litharge when required up to 33 %.22 By the end of the nineteenth century all major metallurgical texts published in English included technical details on the patio process. Collins provides an interesting table of comparative costs for different haciendas in Mexico, which includes the sourcing of power, though it is the cost of the inputs, and not energy, that determines the major differences in his range of refining costs.23 I have set out a representative set of reported refining costs in Table xii.24

20 21

22 23 24

Buchan, Report Real del Monte. In addition to the economic data on refining costs, Laur discusses the issue of the sourcing of power, since steam and animal sources incur costs, while water was nominally free. In the case of animal power it was subject to oscillations in the pricing of animal feed, and he cites the example of a major refining hacienda in Fresnillo that switched to steam engines around 1850 after the yearly expenditure on animal feed rose five-fold. He argues that silver production in Guanajuato decreased as the price of maize increased from 1862 to 1864. Laur, “De la metallurgie de l’argent au Mexique,” 51,62, 198–201. Ibid., 106, 243–254. Collins, Metallurgy of Lead & Silver, 60–61. a) Fabry, Impugnacion a reflexiones de Villaseñor b) Domínguez de la Fuente, Leal Informe

264

chapter 8

Because the haciendas refining ores by tolling (maquila) were required to post separately the cost of mercury, the data in the table for the patio process excludes it. The range in values reflect the changes in ore quality and the variation in local costs of fuel or animal power, manpower and materials. A quick glance through Table xii shows an evident lower range of the values reported for refining costs of the patio process compared to smelting. At first sight, this seems to confirm the consensus in the modern historiography that the patio process is the lowest cost option to refine ores with ‘low’ silver content.25 Smelting is held to be more expensive than the patio process, except for ores with ‘high’ silver content. It has also been argued that the higher the silver content, the higher the absolute consumption of mercury, thus making the patio process too expensive for ores rich in silver. The threshold of silver content that forces a refiner to switch from the patio process to smelting is proposed to lie between 0.26 to 0.52%.26 In addition, mercury is considered ‘the single most expensive item in the refining operations’.27 All these assumptions will be questioned in the following paragraphs. To begin with, the results in Table xii for the patio process and smelting are not comparable on a straightforward basis, even though I have listed them side by side on purpose, to highlight the pitfalls involved. First, the chemical constraints are quite clear: regardless of silver content, the patio process is not a refining option for argentiferous lead ores. A refiner working with lead ores had no choice. He had to make a profit using smelting, or leave the ore in the ground. Thus, one cannot explain the choice of refining method by simply comparing the smelting cost of an argentiferous lead ore with the cost of refining a silver sulphide ore with the patio process, because the former could never be processed under the same conditions as the latter. On the other hand, smelting could refine in principle any silver ore, so costs are comparable for

25 26

27

Político-Legal c) de Sarria, Ensayo de metalurgia d) Garcés y Eguía, Nueva teórica del beneficio de plata e) Sonneschmidt and de Fagoaga, Tratado de la amalgamación de Nueva España f) Duport, Métaux précieux au Mexique g) Laur, “De la metallurgie de l’ argent au Mexique” h) Collins, Metallurgy of Lead & Silver, Vol. 2. For example, Bakewell, Silver Mining in Zacatecas, 138; Brading, Miners Bourbon Mexico, 137; Garner, “Long-Term Silver Mining,” 909. 4oz in Manuel Castillo Martos and M.F. Lang, Metales preciosos—unión de dos mundos: tecnología, comercio y política de la minería y metalurgia iberoamericana (Sevilla: Muñoz Moya y Montraveta Editores, 1995), 140; 8oz in M.F. Lang, El monopolio estatal del mercurio en el México colonial (1550–1710) trans. R.C. Gómez (México: Fondo de Cultura Económica, 1977), 50–51. Lacueva Muñoz, La plata del Rey, 72.

265

the economies of refining silver table xii

A sampling of costs reported for the patio process and smelting in New Spain/Mexico. Sources are indicated in footnote 24.

Location

generic

Period

Smelting costs (excluding cost of extraction of ore, capital cost)

Silver content

Pesos / quintal ore

Pesos / quintal ore

%

Source

0.5

0.125 to 0.5

a, 1–4

Guanajuato

0.5

0.08–0.13

b, 90–91

generic

0.5–0.6

?

c, 131

0.25

d, 144–146

low to medium

e, 92 f, 83–84

generic

mid 18c

Patio process costs (excluding cost of mercury, cost of extraction of ore, capital cost)

end 18c

3–5

0.625

generic

4

generic

0.5–0.75

Zacatecas

5.3–6.7

‘rich minerals’

Sombrerete

2.7–3.3

‘rich minerals’

Nieves

1.3–1.7

?

Zacualpan

5

?

f, 85

Guanajuato

0.6–0.8

0.08–0.23

f, 232–235

Zacatecas

0.6 0.7

0.46 0.14

f, 252 f, 275

1

0.15

f, 340–341 g, 246–248

mid 19c Taxco Cerro San Pedro, San Luis Potosí

1

0.3 (35 % lead)

Sombrerete

2.4

up to 1.75, medium g, 248–250 to low lead

Catorce

7

0.15 to 0.9; 10 % galena

g, 251–252

Regla

6.1

1.7; poor in lead

g, 253

266

chapter 8

table xii Location

A sampling of costs reported for the patio process and smelting (cont.)

Period

Patio process costs (excluding cost of mercury, cost of extraction of ore, capital cost)

Smelting costs (excluding cost of extraction of ore, capital cost)

Silver content

Pesos / quintal ore

Pesos / quintal ore

%

Durango

0.90

0.18

Zacatecas

0.32

0.07

0.33

0.05

Fresnillo

0.46

0.06

Taxco

0.41

0.32

Pachuca

0.41

0.30

Source

h, 60

Zacatecas end 19c

h, 61

all silver ores processed by smelting. As to the argument that a refiner would switch from the patio process to smelting at a higher silver content, because this type of ore consumes more mercury, it is not correct. What matters is not the total cost of mercury consumed, but the refining cost of the patio process per kg of silver refined, as a function of the silver content in the ore. For any refining process, the production cost of each kg of silver refined depends on the silver content of the ore. It is therefore impossible to make any meaningful comparison of refining costs between the patio process and smelting unless the ore in question can be refined by either method, and the function of refining cost versus silver content must be established. This will be one of the principal aims of this chapter. It is also necessary to understand better the historical role of mercury pricing, an issue that had already been questioned by Villaseñor in the eighteenth century. The patio process has overshadowed much of the discussion, and it is common to find sweeping but unsupported statements such as ‘mercury was essential to colonial silver mining because without it most of the silver ore could not be profitably refined’.28 Dissenting voices have started to question the monothematic insistence on mercury:

28

Garner and Stefanou, Economic Growth Bourbon Mexico, 132.

the economies of refining silver

267

the frequent correlation that is proposed between the lack of mercury with the mining crises [in New Spain], underestimates how the miners could compensate using smelting according to the silver content of the ores being extracted … it is necessary to investigate the regional differences to establish the vulnerability of the economy to any restriction in the supply of mercury.29 Back in 1943, the Mexican historian Mendizábal had already proposed factors totally unrelated to mercury, to explain the unexpected increase in silver production registered in the district of Rosario at the end of the eighteenth century.30 A necessary starting point, though not as novel as repressed sexual drives, is the quantitative analysis of the cost of refining silver. Bakewell’s common sense dictum that ‘mining would obviously not have survived for long unless someone were making a profit from it’ was correct in stating the obvious: if both the patio process and smelting had prevailed over centuries, and made some people very rich, it obviously made economic sense to have used them.31 What has been missing from the discussion is more data on the costs of each refining process. Fortunately for the historian, the account books of Regla have survived to provide detailed answers to fill some of these gaps in our knowledge.

29

30

31

Rosa Alicia Pérez Luque and Rafael Tovar Rangel, La contabilidad de la Caja Real de Guanajuato. Una aproximación a su historia económica 1665–1816 (Guanajuato: Universidad de Guanajuato, 2006), 88–89. The many factors that account for variations in silver output in each mining region is well illustrated by the case of silver, Rosario and sex. In order to explain why the mining district of Rosario, in the period 1785 to 1789, contributed greater silver revenues than the more renowned Cajas of Guadalajara, Pachuca, Bolaños and Zimapán, Mendizábal proposed that it was due to the proletarianisation of the indigenous workforce, following the expulsion of the Jesuits from New Spain. As soon as the indigenous males of Rosario were freed from the Jesuit restrictions of working in mining, an occupation that was held to expose them to alcohol consumption, gambling and sexual relations, they flocked to the mines and haciendas of Rosario to make up for lost time, and thus created the peak in silver production. Mendizábal, La minería mexicana, 61–62. Bakewell, Silver Mining in Zacatecas, 187.

268

chapter 8

The Refining Costs at Regla The accounting books of Regla provide a historical data base on refining costs that does not incorporate more modern elements such as steam or electricity.32 They represent the detailed inner workings of a commercial industrial patio process and smelting operation, from which the cost structure of refining can be calculated.33 There is no substitute for long time series of prices of consumables for each process, an operational accounting of their consumption, a detailed breakdown of labour costs and structure, and the amount and price of the product. The historical value of the accounting records of refining costs kept by the Compañía de Minas del Real del Monte y Pachuca lies not only in their detail that cover decades, but also because this company was one of the very few mining and refining conglomerates in Mexico that ran three distinct refining methods concurrently on an industrial scale (patio process, barrel process and smelting). In the period 1853 to 1888, it spread its refining activity over five different refining haciendas, each with a specific remit as to the refining process it used to produce silver. Thus Regla applied the patio process and was the only refining unit to smelt ores; Sánchez, Velasco and San Miguel only processed ores using the barrel (toneles) process, along the lines of the Alonso Barba/Born/Freiberg process; Loreto used mainly the patio process though in at least one year it is listed as also using the barrel process.34 The costs incurred by each refining unit were also compared to each other, in the way modern business compares the profitability of discrete refining units within an overall corporate cost structure.35

32 33

34

35

See Appendix a. Initially the new Mexican owners of the company decided that its accounting should be carried out by the Treasury of the Casa de la Moneda, and then by the Compañía de Tabaco, both enterprises also managed by the owners during this period. It is thanks to the insistence of one of the owners, Nicanor Beistegui, that the company was allowed to keep its own accounts as of 1852, a decision that has served so well this chapter. Ruiz de la Barrera, “La Empresa de Minas del Real del Monte,” 76, 81, 83. The fame of Regla as a smelting centre transcended Real del Monte: ‘the most important factory in Mexico, for the refining of mineral ores by smelting, is Regla’, according to Laur, “De la metallurgie de l’argent au Mexique,” 252. Laur argues that accounting records from Mexican refining haciendas may be subject to an under-accounting of received ores, to hide losses or pilfering or to present a more efficient picture of its refining activity. In the case of Regla I am accepting its data at face value. Ibid., 91.

the economies of refining silver

269

figure 78 The evolution of the silver to gold ratio from 1690 to 1900. Data from footnote 36.

The Macroeconomic Context in the Nineteenth Century When the Mexican investors stepped into the shoes of the Compañía de Minas del Real del Monte y Pachuca, they would have been forgiven for thinking that past performance of silver pricing was a very good guarantee of future results. For nearly two hundred years the silver to gold ratio had been stable (Figure 78).36 For any capital investor, the degree of confidence on future revenue streams is a major element in determining the expected economic viability of an industrial project. In the late 1840s the future price of silver was not a factor of concern to the silver refiners of Mexico, judging by the stability of its international price at around 60 pence per oz of fine silver in the London market. And yet Duport ends his extensive monograph on silver in Mexico, published in 1846, with a phrase that captured both the hubris of three centuries and an intimation of its mortality: ‘a time will come, give or take a century, when the only limit to the production of silver will be imposed by the accelerated decrease of its value’.37 As of the 1870s the floor shifted in a major way from under the silver market, at the time the u.s.a. became the world’s major producer of silver. By 1902

36

37

The same cannot be said of the millesimal fineness of silver bars during the colonial period. The initial official value was 0,930, decreasing to 0,9027 by 1826, according to Joaquin D. Casasus, La Question de l’argent au Mexique (Paris: impr. de Chaix, 1892), 28– 30. The data in Figure 78 is sourced from ‘Average commercial ratio of silver to gold each calendar year since 1687’ in the Report of the Director of the Mint contained within the Annual Report of the Secretary of the Treasury on the State of the Finances for the Fiscal Year ended June 30 1921, p. 654. http://fraser.stlouisfed.org/docs/publications/treasar/AR _TREASURY_1921.pdf. Duport, Métaux précieux au Mexique, 426.

270

chapter 8

the price of silver in London, the benchmark for silver sales in the world and the destination for much of the silver exported from Mexico in the nineteenth century, had plummeted to less than half its 1870 value, a near mirror-image of the devaluation of silver with respect to gold.38 In 1873 H.R. Linderman, the Director of the u.s. Mint, included in his annual report an analysis of the developing weakness in the value of silver and its impact on the main Western economies. His commentary on the world-wide concatenation of events leading to the drop in the value of silver merits an extended quote: the steady value of the money-unit … can only be maintained by making one of the precious metals the standard or measure of value, and assigning a subordinate position as to coinage for the other … gold, being less variable than silver, and of superior value, has been adopted [by] … Japan, Germany, the United States of America, Denmark, Sweden and Norway [Great Britain in 1816] … this system [single standard] … enhances the value of the one, and depreciates that of the other … large quantities of silver hitherto in circulation as standard money … will … be thrown on the market as bullion, and aid in its further depreciation … India has for many years past been the principal market for silver … the decline [in demand is] due principally to the fall in the price of cotton, soon after the close of the late Civil War in this country.39

38

39

‘Highest, lowest and average price of bar silver in London, per ounce British standard (0.925), since 1833; and the equivalent in United States gold coin, of an ounce 1.000 fine, taken at the average price and par of exchange’ in the Report of the Director if the Mint contained within the Annual Report of the Secretary of the Treasury on the State of the Finances for the Fiscal Year ended June 30 1921, p. 653. http://fraser.stlouisfed .org/docs/publications/treasar/AR_TREASURY_1921.pdf. Part of this Mexican silver was re-exported to Asia. For data on the amounts of Mexican silver sent from London to China and the Federated Malay States Silver from 1864 to 1902 see Eduardo Flores Clair, Cuauhtémoc Velasco Avila, and Elia Ramírez Bautista, Estadísticas mineras de México en el siglo xix, vol. ii (México: Instituto Nacional de Antropología e Historia, 1985), 140– 141. H.R. Linderman, ‘Report of the Director of the Mint, November 1st 1873’ in the Annual Report of the State of the Finances to the Forty-Third Congress, First Session, December 1, 1873, Washington, Government Printing Office, 1873, 476–477. http://fraser.stlouisfed.org/ docs/publications/treasar/AR_TREASURY_1873.pdf. For current research on the global impact of the fall in the value of silver in the nineteenth century, see the work of the damin group (Dépréciation de l’Argent Monétaire et relations INternationales) at http:// www.anr-damin.net.

the economies of refining silver

271

Despite these major changes in the international valuation of silver, Mexican silver production continued to increase. The reasons that have been given are that most refining costs were paid in local currency, and local inflation did not react in proportion to the international decrease in the value of silver. In addition, the Mexican State took specific actions to protect its most valued industry, such as keeping the dual gold and silver standard until 1905, so refiners could exchange silver for gold at an attractive rate, and allowing the export of silver in bullion, foregoing the previous restriction of only allowing the export of coin after payment for coinage. Finally, the Mexican industry begun the sale of other metals produced during refining, such as lead.40 In the case of Regla, we saw in the previous chapter that total silver production, during the period when silver prices dropped by one third in the London market, was relatively stable. How did the cost of the reagents and fuel required to produce silver vary for Regla during this period? As observed in Figure 79, they showed in general a surprising independence to the events overtaking the international silver market. The background to each profile varied. In the case of salt, as Buchan explained in 1855, Amongst the materials required for the reduction of the ores, Salt is one of the most costly and difficult to obtain … to secure the certain supply of this most necessary ingredient, and also with the hope of reducing its cost, we have commenced … the formation of large salt-works on the Lake of Tezcoco … and we hope in time to render them adequate to all our wants. In 1855 freight cost 42 dollars per ton, 0.48 pesos per arroba, to bring it ‘from the State of San Luis, being a distance of some 300 miles’.41 Freight was therefore a major fraction of the cost of salt at the plant gate of Regla. The expenditure on salt shows a marked downward trend from 1852 to 1888 over all the haciendas of the Company, with the average at Regla from 1872 to 1888 of 0.72 pesos per arroba (Figure 79a). The decrease in expenditure on salt over time was due to the successful efforts of the Company to find a closer supply source, and possibly a lowering of quality, since the consumption of salt per kg of silver refined at Regla rose at the end of the period (as presented in Figure 68a, Chapter Seven). 40 41

Velasco Avila et al., Estado y minería en México, 271–275. Buchan, Report Real del Monte, 18–19. Lake Tezcoco was situated less than 100 km from Regla, so freight costs would have decreased substantially. It is not known if the project succeeded. Salt was also brought in from salt pans in Campeche, using the ports of Tampico and Tuxpan, at the same cost.

272

chapter 8

figure 79a

Yearly average expense on salt. Values calculated from data in Informe Mensual correspond to the expenditure on salt at Regla. Values calculated from data in Estados Comparativos correspond to the average expenditure on salt registered for all the haciendas of the Compañía de Minas del Real del Monte y Pachuca in any given year.

figure 79b

Yearly average expense on copper sulphate. All values calculated from data in Informe Mensual.

the economies of refining silver

273

figure 79c

Yearly average expense on charcoal. All values calculated from data in Informe Mensual.

figure 79d

Yearly average expense on litharge. All values calculated from data in Informe Mensual.

274

chapter 8

figure 80 Yearly average expenditure on mercury. Values calculated from data in Informe Mensual correspond to the expenditure on mercury at Regla. Values calculated from data in Estados Comparativos correspond to the expenditure on mercury registered for all the haciendas of the Compañía de Minas del Real del Monte y Pachuca.

Copper sulphate is reported as having been one of the items imported by the Company.42 Its cost shows a remarkable resiliency to macroeconomic events in the 1873 to 1888 timeline, as evidenced in the nearly flat set of data points in Figure 79b. The average expenditure over the period was 0.13 pesos per lb. In the case of charcoal, the profile shows a very stable expenditure until 1881, after which there is an approximate 40% increase (Figure 79c). This could be explained by fuel demands finally outstripping supply in this period of continuous smelting, coupled with the high demand for fuel for the barrel process, or by a change in sourcing.43 The average over the whole period is 1.3 pesos per carga of charcoal. In contrast, the historic expenditure on litharge shows step decreases punctuating periods of stable pricing, with an average of 0.08 pesos per kg (Figure 79d). Mercury had to be imported in its totality and thus would be expected to be most sensitive to the impact of international macroeconomic factors that begin around 1873. However, other market factors specific to mercury would come into play. Figure 80 shows the vicissitudes the Compañía de Minas del Real del 42 43

Ruiz de la Barrera, “La Empresa de Minas del Real del Monte,” 293. There is one reference to charcoal being imported from England and Germany in the nineteenth century, and brought to the Compañía de Minas del Real del Monte y Pachuca by rail from the port of Veracruz, in Saavedra Silva and Sánchez Salazar, “Espacio PachucaReal del Monte,” 93. It does not specify either period, quantities or pricing, and I have no other source to confirm this. A switch to imported charcoal could explain the increase in price.

the economies of refining silver

275

figure 81 Monthly production costs of silver refined by the patio process at Regla (1872–1888). Values calculated from data in Informe Mensual.

Monte y Pachuca went through in the nineteenth century with regards to the cost of mercury. Not since the late sixteenth century had prices of mercury reached over 1.6 pesos per pound (160 pesos per quintal) in New Spain / Mexico. However, after the peak in pricing subsided, mercury prices are unexpectedly stable and in fact slowly decrease, from 1879 onwards, as the depreciation of the international price of silver was gathering steam, only picking up again after 1887. The average cost from 1878 to 1888 was 0.6 pesos per lb, 60 pesos per quintal. The attempt by the Rothschilds to profit from their dominant market position in mercury was the cause of the peak in pricing.44 However, the appearance in the market of new supplies of mercury from production in the United States of America starting in 1874 and lasting to 1884, led to mercury prices from New Almaden (u.s.a.) dropping ‘from $ 126.22 per flask [of 76 lbs] in 1874 to $49.75 in 1875, and thereafter until 1883 the average price per flask was about $30.00’.45 The new Mexican owners of Regla would reap the benefit from this drop in the cost of a principal component of the patio process recipe.

44

45

An analysis of the involvement of the Rothschilds in the mercury market of the nineteenth century is given in Miguel A. López Morell and Jose M. O’Kean, “Seeking out and building monopolies, Rothschild strategies in non ferrous metals international markets (1830– 1940),” in 14th Conference of the European Business History Association (Glasgow 2010). Henry Winfred Splitter, “Quicksilver at New Almaden,” Pacific Historical Review 26, no. 1 (1957): 36.

276

chapter 8

table xiii Mining and other costs for Real del Monte mines in the period 1849 to 1854, raw data adapted from footnote 46

General expense [overheads] Drainage of mines Extraction cost Freight to refining haciendas total cargas of ore production cost per carga

May 1849 to December 1852

1853

1854

92,456 215,541 758,906 105,283 1,172,186 311,765 3.76

30,152 69,334 294,874 63,768 458,128 181,151 2.53

36,410 83,707 322,812 82,640 525,569 192,982 2.72

all costs in dollars (equivalent to pesos)

The Variable Costs of the patio Process at Regla, 1872–1888 In Figure 81, I plot the monthly variable patio process costs registered in the account books of Regla, net of the cost of extracting the ore. From mid 1872 to early 1875 the profile reflects the sudden surge of mercury pricing around 1874. Because of the monthly interruptions of production by the patio process observed in the monthly accounts from 1874 to mid 1875, I have calculated the average cost from mid 1875 to mid 1888. Excluding the cost of ore, it averages at 7.8 pesos per kg of refined silver. The accounting books do not include the monthly cost of the ore delivered to Regla. I have found two earlier sources of extraction costs for silver ore by the Compañía de Minas del Real del Monte y Pachuca. The first is derived from data published by Buchan on the costs for the ore produced in the period May 1849 to December 1852, and the years 1853 and 1854 (Table xiii).46 The unit cost per carga of mined ore, including freight to the refining haciendas, shows a marked decrease over this short period. The average of the years 1853 and 1854 is 2.6 pesos per carga of ore at the plant gate. The second source is the weighted average of extraction costs reported between the 8th March and 20th December 1862. The data covers a variety of mines and ore qualities, for a total of 161,173 cargas, at a weighted average of 3.1 pesos per carga.47 I have opted to

46 47

Buchan, Report Real del Monte, 26–31. AHCRMyP, Sección: Ymporte de la Memoria de la Mina, Serie Informes de Minas i, Vol. 13.

the economies of refining silver

277

figure 82 Percentage breakdown of the total variable cost of the patio process at Regla (1872–1888, excluding 1874). Extraction cost (1853–1854) from Table xiii, other data from Table xiv. pie chart reproduced courtesy of taylor & francis from footnote 48

use as a working figure the former, since it covers a longer period. The average percentage of silver extracted from the ores processed by the patio process at Regla was 0.17% by weight, so I arrive at an average extraction cost for the ore at 11.1 pesos per kg of silver refined using mercury. The total average variable cost of the patio process is now 18.9 pesos per 1 kg of refined silver. The pie chart of variable refining costs, including the extraction cost of the ore, is shown in Figure 82, based on the data from Table xiv. It confirms that the cost of extracting ores was the major contributor to the total variable cost of the patio process (nearly 60%), and not mercury (10 %), as was correctly pointed out by Villaseñor in 1741. Salt at Regla represented a nearly equal share in the costs as mercury. As discussed in the previous chapter, fuel is a negligible factor. Labour costs at Regla represent 7 % of the total cost.48 However, it has been estimated that labour costs made up 85 % of the total extraction cost of mining ores in Mexico in the nineteenth century.49 If this 48 49

Saul Guerrero, “The History of Silver Refining in New Spain, 16c to 19c: back to the basics” History and Technology 32, no. 1 (2016): 11. Eugène Viollet, “Le problème de l’Argent et l’Etalon d’ Or au Mexique” (PhD diss., Université de Paris, 1907), 121.

278

chapter 8

table xiv The percentage breakdown of the main variable costs of the patio process at Regla, excluding the cost of ore at the plant gate. The percentage values were calculated from the individual headings within the monthly account data, and then averaged for the year. A total of 153 data sets are represented in the table. Source data from Informe Mensual.

Patio process Year mid 1872 1873 1874 mid 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 mid 1888 whole period 1875–1888

Labour Mercury

Salt

Copper sulphate Other

Total

24% 23%

24% 27%

24% 20%

7% 7%

21 % 25 %

100 % 100 %

15% 15% 17% 16% 18% 19% 16% 16% 14% 15% 16% 18% 18% 19% 17% 17%

43% 36% 28% 23% 19% 19% 21% 22% 22% 20% 22% 23% 24% 21% 25% 24%

14% 23% 25% 24% 24% 22% 23% 24% 23% 27% 24% 23% 20% 22% 23% 23%

7% 7% 11 % 13 % 16 % 13 % 11 % 10 % 11 % 9% 7% 5% 5% 4% 9% 9%

21 % 19 % 20 % 23 % 23 % 27 % 29 % 28 % 29 % 29 % 31 % 32 % 33 % 34 % 26 % 27 %

100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 %

estimate is correct, then 55% of the cost of the patio process at Regla was due to the cost of labour in mining and refining, with up to an additional 10 % in cost of local labour hidden under the headings of salt, fuel and others. The total cost of the patio process at Regla would need to incorporate as well the fixed capital cost due to the investment in infrastructure. Since in 1846 most of the infrastructure at Regla was bought at a pittance from its previous English owners, from an accounting point of view the servicing of this fixed capital cost is not representative of an hacienda de patio built from scratch. The report by John Buchan includes a listing for ‘reforming and enlarging reduction works at

the economies of refining silver

279

Regla’ of $20,000, compared to the capital cost of ‘erecting the new reduction work of Velasco’ at $209,750, based on the barrel process.50 The original cost of construction of Regla in the late eighteenth century has been reported at levels over $500,000, but there are strong reasons not to take even the original investment as a guide to the capital cost of a major refining hacienda.51 Visitors to Regla wondered ‘why his walls were built so thick, or why so many massive arches should have been constructed, is an enigma to the present generation, as they could by no means have been intended for a fortress down in a barranca’.52 As Humboldt pointed out, one of the advantages of the patio process was precisely its low capital expenditure in plant infrastructure.53 A major part of the construction cost of Regla was unrelated to the needs of both processes, so its original capital cost is not a guide to an average cost in the silver refining business. Duport reports that the rental in Guanajuato of an hacienda de patio would amount to 50 pesos per year per arrastre.54 If this benchmark were applied to Regla, with 24 arrastres, it would have induced a rent of 1,200 pesos per year, a minor amount compared to its average production of nearly 20,000 kg of silver by the patio process during this period. In the light of all these considerations, I will therefore ignore the fixed cost of capital at Regla and focus only on the total variable cost. Since I will require an estimate on a generic fixed capital cost of the patio process for my sensitivity runs later in this chapter, I will use Duport’s information that the rental cost of an hacienda de patio in the State of Guanajuato, capable of processing 2,565 montones of ore a year, represented 5 % of the variable refining cost of silver.55

50

51

52 53 54 55

Buchan, Report Real del Monte, 26. Another reference point is the cost of construction of the major hacienda de patio of Proaño in Fresnillo, Zacatecas, with a larger refining capacity than Regla. It is reported as having cost 300,000 pesos in the first half of the nineteenth century, before steam engines were installed. The fixed annual cost of servicing that investment is set at 5%. Duport, Métaux précieux au Mexique, 263, 278. According to Terrero, 2 million pesos were spent to construct the haciendas of Regla, San Francisco Javier, San Miguel and San Antonio. Manuel Romero de Terreros, Antiguas haciendas de México (Editorial Patria, 1956), 300. A figure of 425,708 pesos for Regla is given in Doris M. Ladd, The Making of a Strike: Mexican Silver Workers’ Struggles in Real del Monte, 1766–1775 (Lincoln: University of Nebraska Press, 1988), 144. Wilson, Mexico and Its Religion, 366. Humboldt, Essai politique, Tome iv, 84. Duport, Métaux précieux au Mexique, 237. Ibid., 232.

280

chapter 8

figure 83 Monthly production costs of silver refined by smelting at Regla (1875–1886). Values calculated from data in Informe Mensual.

The Variable Costs of Smelting at Regla, 1875–1886 The variable cost per kg of refined silver obtained via smelting at Regla averaged 5.2 pesos in the period June 1875 to January 1886, excluding the extraction cost of the ore (Figure 83). To better illustrate the grouping of values, despite outlying data points, I have opted to represent visually the scatter of the data. The account books provide data on variable refining costs, but not on the variable cost of ore (Table xv). I have therefore taken as a working figure a production cost of ore for smelting the same as that for the patio process, 2.6 pesos per carga or 1.04 pesos per kg of silver.56 Combining all the data, Figure 84 provides the total breakdown of the main variable costs of smelting as carried out at Regla, from 1875 to 1888. Fuel (31%), then labour (21%), are the main cost components of the process. The influence of the cost of ore and litharge on the final refining cost of smelting are equal (17%). The cost of labour in mining and refining would amount to around 35 % of the final refining cost by smelting, but the local labour fraction of fuel 56

At 2.6 pesos per carga production cost (1853–1854), for an ore with 1.9 % silver this is equivalent to a cost of ore of 1 peso per kg of silver refined by smelting. From another context Duport states: ‘one would need to be able to separate the costs of extracting the ore for smelting from those of the ore destined for amalgamation, and this is impossible; since, in all mines the selection is made on the mass of mineral that comes out of the mine, from where the richest ore is selected for smelting’. Ibid., 369.

281

the economies of refining silver table xv

The percentage breakdown of the main variable costs of the smelting process, excluding the cost of ore at the plant gate. The percentage values were calculated monthly, and then averaged for the year. A total of 103 data sets are represented in the table. Source data from Informe Mensual.

Smelting Year mid 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 average

Labour

Litharge

Charcoal

Others

Total

27% 33% 26% 27% 21% 21% 26% 25% 26% 29% 23% 26%

26% 20% 24% 26% 21% 21% 14% 19% 12% 20% 21% 20%

35% 34% 34% 32% 32% 34% 37% 34% 50% 45% 46% 38%

12 % 14 % 16 % 15 % 25 % 25 % 23 % 22 % 12 % 6% 10 % 16 %

100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 % 100 %

costs and others would probably add another 15%. Though less than the 65 % contribution of labour costs to the total estimated for the patio process, it still represents half the total variable cost for smelting at Regla in this period. Refining Cost as a Function of Silver Content Regla used two refining processes to extract silver from ores of two different chemical natures and two different levels of silver content. To be able to compare these costs, only for the case of ores that could be refined by either process, it is necessary to establish the function of refining cost versus the silver content of the ore being refined. This function can be calculated from the data previously derived from the account books for each refining process at Regla, as I show in Tables xvi and xvii.57

57

My calculations and the logic followed by Villaseñor are analogous, though arrived at by different paths.

282

chapter 8

figure 84 Percentage breakdown of the total variable smelting cost at Regla (1875–1886), extraction cost from Table xiii and other data from Table xv pie chart reproduced courtesy of taylor & francis from footnote 48

For the patio process (Table xvi), the first column provides the average variable operating costs per kg of silver registered in Regla, for an ore with an average silver content of 0.19%. The column highlighted with a grey background includes, from top to bottom, the amount of silver in a montón (assuming a total extraction rate of silver of 90%), and the value of the refined silver in pesos (adopting an equivalence of 38 pesos per kg of silver).58 The lower tranche of values corresponds to the average variable costs expressed in pesos per montón, for a silver content of 0.19%. A matrix of projected values for other silver content in the ore is then assembled from this column, by choosing different silver percentages and adjusting the cost of mercury (at a mercury to silver weight ratio of 1.3 to reflect the operation at Regla) and fuel, maintaining constant all the other values. The fuel costs vary as they reflect the amount of amalgam fired in a capellina and the bars cast. The cost of salt and copper sulphate are assumed not to vary significantly with silver content. Labour and other costs are deemed constant throughout the range.59 Finally, the bottom row expresses the value of variable refining cost, in pesos per kg extracted silver, as a function of silver content. 58 59

The value of 38 pesos per kg of silver is calculated from the data from 1849 to 1854 in Buchan, Report Real del Monte, 26–31. I draw attention to the fact that the cost of mercury consumed during the patio process at Regla increases with the silver content of the ore, but only when calculated based on a montón. On the basis of the total variable refining costs, it decreases at higher silver content, when calculated per kg of silver refined.

Fuel Mercury Salt Copper Sulphate Labour others ore total

Patio process

variable production cost pesos per kg silver

0.19 1.91 1.78 0.72 1.30 1.93 11.1 18.93

variable production cost pesos per kg silver

% silver in ore kg of silver in montón value silver pesos

0.00 0.00 4.21 1.69 3.07 4.54 26.2 39.71

0.00 0.00 0

161.96

0.05 0.48 4.21 1.69 3.07 4.54 26.2 40.23

0.02 0.25 9

82.03

0.09 0.95 4.21 1.69 3.07 4.54 26.2 40.75

0.04 0.50 19

0.08 0.99 38

0.12 1.49 57

0.19 2.36 90

0.60 7.45 283

55.39

0.14 1.43 4.21 1.69 3.07 4.54 26.2 41.27 42.07

0.19 1.90 4.21 1.69 3.07 4.54 26.2 41.80 28.74

0.28 2.85 4.21 1.69 3.07 4.54 26.2 42.84 18.93

0.44 4.52 4.21 1.69 3.07 4.54 26.2 44.67

7.43

1.40 14.27 4.21 1.69 3.07 4.54 26.2 55.37

variable production cost in pesos per montón

0.06 0.75 28

5.30

2.33 23.78 4.21 1.69 3.07 4.54 26.2 65.82

1.00 12.42 472

3.78

4.43 45.18 4.21 1.69 3.07 4.54 26.2 89.32

1.90 23.60 897

table xvi Matrix to calculate the variable cost of refining by the patio process at Regla as a function of silver content in the ore. Values derived from Table xiii and Informe Mensual.

3.17

7.00 71.33 4.21 1.69 3.07 4.54 26.2 118.04

3.00 37.26 1416

the economies of refining silver

283

Fuel Litharge Labour others ore total

Smelting

table xvii

variable production cost, pesos per kg silver

1.94 1.05 1.32 0.85 1.04 6.21

variable production cost pesos per kg silver

% silver in ore kg of silver in carga value silver pesos

5.09 0.00 3.47 2.22 2.73 13.51

0.00 0.00 0

245.78

5.09 0.06 3.47 2.22 2.73 13.57

0.04 0.06 2

164.20

5.09 0.09 3.47 2.22 2.73 13.60

0.06 0.08 3

123.42

5.09 0.12 3.47 2.22 2.73 13.63

0.08 0.11 4

0.12 0.17 6

0.16 0.22 8

0.19 0.26 10

0.40 0.55 21

98.94

5.09 0.15 3.47 2.22 2.73 13.65 82.63

5.09 0.17 3.47 2.22 2.73 13.68 62.24

5.09 0.23 3.47 2.22 2.73 13.74 52.58

5.09 0.28 3.47 2.22 2.73 13.79

25.53

5.09 0.58 3.47 2.22 2.73 14.09

variable production cost in pesos per carga

0.10 0.14 5

17.37

5.09 0.87 3.47 2.22 2.73 14.38

0.60 0.83 31

10.84

5.09 1.46 3.47 2.22 2.73 14.96

1.00 1.38 52

6.21

5.09 2.76 3.47 2.22 2.73 16.27

1.90 2.62 100

4.32

5.09 4.37 3.47 2.22 2.73 17.87

3.00 4.14 157

Matrix to calculate the variable cost of smelting at Regla as a function of silver content in the ore. Values derived from Table xiii and Informe Mensual.

284 chapter 8

the economies of refining silver

285

A similar method is applied in Table xvii for the case of smelting.60 It is assumed that the only cost that varies in a manner directly proportional to the silver content of the ore is the cost of litharge, and that all the other costs remain at the level that corresponds to the smelting of the 1.9 % ore. Since the whole mass of ore is heated, the cost of charcoal will not vary with silver content; if it had, then results would have been more favourable to smelting. Figure 85 provides one answer, specific to time and place, to the question that I quoted in the epigraph to this chapter. The margin of profit for each process is a function of the silver content of the ore being refined, regardless of the process used. For the average silver content of the ores sent to be refined at Regla, 0.19% for the patio process and 1.9% for smelting, the margin ranged from 50% (patio process, Figure 85a) to 85% (smelting, Figure 85b), of the value of the extracted silver. If the Crown retained approximately a further 20 % of the value of silver extracted, Regla could extract its net profit from the remaining margin of 30% to 65%. The total volume of ore being refined by each process would also play an important part in the final revenues of the company, since the lower margin of the patio process would be compensated by a greater volume of ore being processed. The same graphs indicate that the lowest break-even silver content for each type of ore, required to cover each of the refining costs at Regla and the tax to the Crown, was approximately 0.12% silver content for the patio process, and 0.4 % for smelting.61 For both processes, the refining cost decreases as the silver content of the ore increases. This reflects what Villaseñor had pointed out, that most of the refining cost is basically a fixed cost, so that as silver content increases, the cost expressed per kg of refined silver is being divided by an ever-increasing quantity. The next question is whether the sulphidic ores refined by the patio process could have been smelted at a profit at Regla. The answer is given in Figure 86, that compares side by side the variable refining cost of each process as a function of the silver content in the ore.62 The curve for smelting corresponds to the structure of costs for argentiferous lead ores that require additional litharge, so for the sake of this argument it is an acceptable approximation for dry silver sulphide ores. 60 61

62

The account books list side by side costs in pesos per montón for the patio process and pesos per carga for smelting, so care must be taken in comparing values. The theoretical projection also correlates well with the operational records at Regla. The lowest tranche registered in the accounts for the patio process corresponds to ores with 0.7 to 0.10% silver content. For smelting the lowest tranche is 1 %. Guerrero, “The History of Silver Refining in New Spain, 16c to 19c: back to the basics,” 13–14.

286

chapter 8

figure 85a

Projected margins at Regla, as a function of silver content: the patio process. Data from Tables xvi and xvii, respectively.

figure 85b

Projected margins at Regla, as a function of silver content: smelting. Data from Tables xvi and xvii, respectively.

The result is specific to Regla and to the period 1875 to 1888: the patio process would have been a more profitable option than smelting to refine ores with a silver content up to 3%, in the absence of lead in the ores. The accounting records of Regla confirm this theoretical prediction. They show the use of ores with a silver content of 61 marks per montón (1 %) in the preparation of tortas for the patio process.63 Figures 85b and 86 indicate that an increase in silver

63

Record of tortas prepared in the period of five weeks ending on November 28th, 1885, as registered in the Informe Mensual.

the economies of refining silver

287

figure 86 Variable refining cost as a function of silver content. Data from Tables xvi and xvii. A comparison between the two curves is only valid for silver ores without lead. based on footnote 62, courtesy of taylor & francis

content did not force a refiner to switch from the patio process to smelting, in order to lower his cost of refining by saving on mercury. Silver content, by itself, would not have precluded the use of the patio process at Regla in the period under study. A Window to the Past: Extrapolating the Results from Regla The refining costs of Regla are simply one combination of values within a historical range of costs, as set out in Table xviii, with the caveat that reported values do not necessarily include freight, and the quality of each ingredient is not the same over time and place.64 Despite these limitations, the comparison tells us that Regla between 1875 and 1888 had specific advantages, such as a low fixed capital cost, free and constant hydraulic power, very efficient smelting furnaces, and medium to low cost mercury. It also had other costs well within the middle range of the historical spread of values. What it also underlines is the circumstantial nature of the conclusions drawn so far for the case of 64

a) García Mendoza, “Minas de plata en Taxco” b) Menegus Borneman, “Las comunidades productoras de sal y los mercados mineros: los casos de Taxco y Temascaltepec” c) Sentir de Don Manuel de Aldaco, no page number, in Fabry, Impugnacion a reflexiones de Villaseñor d) Domínguez de la Fuente, Leal Informe Político-Legal e) Brading, Miners Bourbon Mexico f) Duport, Métaux précieux au Mexique g) Laur, “De la metallurgie de l’ argent au Mexique” h) West, The Parral Mining District.

288

chapter 8

table xviii

Selection of historical costs of refining in New Spain / Mexico, from the sixteenth to nineteenth century, and those for Regla from 1875 to 1888. Sources in footnote 64.

Period

Cost

Unit

Source, page

Regla, 1875–1888

16c

190

pesos/quintal

a, 54

60

17c to 18c

41 to 82

16c

0.2

17c

0.1

b, 26

18c

0.2 to 1.8

b, 26; c; d, 208; e, 153

19c

0.1 to 1

f, 87; g, 63,66–67

Copper sulphate

19c

0.14

pesos/pound

f, 98

0.13

Litharge (greta)

16c

0.06

pesos/kg

a, 54

0.08

18c

0.01

d, 208

19c

0.09

f, 75

16c

0.3

19c

1.5 to 2.5

17c

0.7–1

18c

1–2.5

e, 155

19c

3.4–8.3

f, 223, 246, 259, 308

Furnace efficiency

16c to 18c

1,000

kg charcoal/ kg silver

see Chapter 7

200

Fixed capital cost

19c

approx. 5% of variable cost

pesos

f, 278

very low

Source of power

16c to 19c

nil

water power

19c

12% of variable cost

animal power

Mercury

Salt

Charcoal

Ores with silver content 0.06% to 0.3%

see Figure 92 pesos/arroba

pesos/carga

a, 54

a, 54

0.7

1.3

f, 278; g, 83 pesos/carga

h, 95–96

2.6

water, nil f, 278

289

the economies of refining silver

table xix Values of projected refining costs of silver in the second half of the sixteenth century in New Spain. Data on Regla from Informe Mensual and Tables xvi and xvii. Data on projected historical ranges derived from Table xviii.

Regla 1875–1888

Projected value for 16c context

Basis for projected value

Pesos per kg of silver patio process mercury labour ore fuel salt power fixed capital

1.9 1.3 11.1 0.2 1.8 0 0

9.8 0.1 0 0 1.8 1.2 0.5

a factor of 5.12 = (180/60)*(2.1/1.3) 10 % of 19c value tailings (0.3/1.3) no change approx. 12 % variable cost approx. 5 % variable cost

1.9 1.1 1.3 1.0 0 0

2.2 0 0.1 0 0 0

a factor of 1.15 = (1,000/200)*(0.3/1.3) lead rich ores 10 % of 19c value rich surface deposits low cost inexpensive infrastructure

smelting fuel litharge labour ore power fixed capital

Regla, and that under other sets of historical values, the comparative advantage of one refining process over the other for sulphide ores may change. If Regla is one of the bookends to the history of silver refining with mercury in New Spain, the other corresponds to the scenario during the second half of the sixteenth century, when refiners were still grappling with smelting, and the use of mercury to refine silver ores was beginning to make inroads in the New World. For smelting it was characterised by inexpensive infrastructure, negligible extraction costs for the initial rich surface deposits, abundant and nearby fuel sources, inefficient furnaces and ores rich in lead in some locations. As to the use of mercury-based refining, it was initially applied to ores obtained from tailings, thus with zero extraction cost, mercury was sold at the highest

290

chapter 8

figure 87 Refining costs of the patio process and smelting in the context of the second half of the sixteenth century as a function of the silver content of the ore. Calculation based on data in Appendix b. graph based on footnote 62, courtesy francis & taylor

levels of historical prices, the novelty of the process resulted in some very high mercury to silver ratios, investment in new stamp mills was the major capital cost involved, and water was not always available to drive them. This new context can be translated into numbers that can be used to calculate a new set of comparative refining costs for both processes. The aim is not to arrive at absolute values but to project the relative profile of their refining cost as a function of silver content. The method is analogous to that applied in Tables xvi and xvii, except that now I will change certain key values as set out in Table xix. For the purpose of this approximation, I will ignore any difference between the refining carried out in canoas and that of the latter patio process, since a more inefficient canoa-based process would only accentuate the conclusions I reach. Once the new values are substituted in a calculation matrix, a new plot of comparative refining costs as a function of silver content can be generated.65 Again I must caution the reader that the curve for the patio process does not apply to argentiferous lead ores, and that only silver ores without lead would have offered the refiner the choice of using mercury. With this caveat firmly in mind, Figure 87 indicates that according to my projection: a) contrary to Regla, in this period smelting would have been the more profitable option

65

For calculation matrix see Annex b.

the economies of refining silver

291

figure 88 Refining costs of the patio process and smelting in the context of the period from the mid-seventeenth to mid-eighteenth century as a function of the silver content of the ore. Calculation based on data in Appendix b. graph based on footnote 62, courtesy of taylor & francis

for any ore with a silver content above 0.4% b) the patio process had a higher refining cost than in Regla, but could still show a margin of profit for ores with a silver content above 0.1% silver. As to the period between the bookends, the decades from mid seventeenth century to mid eighteenth century, comments in the historiography point to a fine balancing act during this period between the patio process and smelting. The appraisal by Humboldt that ‘usually it is only the abundance of mercury and the ease of procuring it that determine the choice of the miner on the method he will employ’, by its silence on refining costs is pregnant with the implicit assumption that smelting was only displaced by an opportunistic supply of mercury, not by an inherent economic shackling of its feet.66 Garner comments that in the years between 1801 and 1802 in Zacatecas, cutting the diezmo to one half triggered an increase of 2.5 times the previous amount of silver smelted, which decreases in 1803 by one-third when the full diezmo is reestablished.67 The decrease in the diezmo is equivalent to approximately two pesos in the value of one kg of silver, just 5% of the value of the silver being refined. It was a very fine line that at times kept smelting and refiners apart.

66 67

Humboldt, Essai politique, Tome iv, 56. Garner and Stefanou, Economic Growth Bourbon Mexico, 135.

292

chapter 8

table xx

Values of projected refining costs of silver from mid 17c to mid 18c. Data on Regla from Informe Mensual and Tables xvi and xvii. Data on projected historical ranges derived from Table xviii.

Original Regla data

Projected value for 17–18c context

Explanation

Pesos per kg of silver patio process mercury labour ore fuel salt power fixed capital

1.9 1.3 11.1 0.2 1.8 0 0

3.6 0.7 16.7 0 3.7 3.0 1.4

a factor of 1.89 = (82/60)*(1.8/1.3) 50% of 19c value a factor of 1.5, less efficient deep mining (0.3/1.3) a factor of 2, salt sources not optimised 12% variable cost 5% variable cost

1.9 1.1 1.3 1.0 0 0

2.2 0 0.7 0 0 0

a factor of 1.15 = (1,000/200)*(0.3/1.3) lead rich ores 50% of 19c value cost borne by ore for patio process low cost inexpensive infrastructure

smelting fuel litharge labour ore power fixed capital

This is a period where infrastructure costs become important as major industrial concerns arise that cover mining and refining of their own ores as well as the ores from smaller miners via the lucrative maquila, the toll charged for refining ores. In contrast, smelting continued to require a much lower capital fixed cost, and could always be carried out at a much smaller scale. Mercury prices would not be at their lowest level yet, and mercury to silver ratios would remain high except where iron was used as an additive or the nature of the ore dictated otherwise. Furnace efficiency remained low. Table xx does not attempt to represent specific locations or years within this very complex period, it only represents a context that is feasible in the light of what is now known. The plots in Figure 88 prove it is possible to find

the economies of refining silver

293

a set of values within this historical context, where the curves of refining cost versus silver content in the ore, for the patio process and smelting, are nearly superimposed on each other. For some refiners faced with a shortage of mercury to refine silver sulphide ores, Figure 88 demonstrates how in theory they could switch to smelting without a major sacrifice in profit, even for ores with 0.3% silver content (this would have been impossible in the context of Regla, 1872 to 1888). Charcoal and lead fluxes would have been needed at low prices to offset the inefficient furnaces. Low margins of profit would have to be offset by the opportunity cost of being able to smelt rather than close down operations for lack of mercury. Under the scenario of Table xx and Figure 88, a smelter could have been facing similar margins for the same ore quality as haciendas de patio where salt was expensive, water power absent, and the capital cost of infrastructure and deep mining costs high. None of these conditions can be described as unrealistic for the period. However, there are still too many gaps on economic data of refining for this period to take this analysis any further.

The False Positives of the patio Process At the end of a report on smelting in Utah in the late nineteenth century, the author concludes that ‘the proper treatment [is that] which, however wasteful, costly, or even unscientific, enables the owner to make the most money out of his ore’.68 This common-sense dictum is a timely reminder of what mattered most to the refiner in the field when he held a lump of silver ore in his hand. If it was very heavy he would think lead and smelting, and not mercury and the patio process, regardless of the silver content of the ore. If it was red-orange in colour, or greyish to dark in tone, he could then think of mercury to refine his ore. At any time, he could have switched from the patio process to smelting if the costs favoured the change, but never the other way around if he worked with argentiferous lead ores. There was never a competition between the patio process and smelting, though there would be a tilting of the playing field to make life easier for the refiners using the former. In this last section of my analysis of refining costs, I turn my attention to other false positives on the patio process, characteristic factors said to be present in the history of colonial silver refining, though the facts argue strongly otherwise.69

68 69

As quoted in Eissler, The Metallurgy of Argentiferous Lead, 349. One area of contention that the records of Regla cannot resolve is the issue whether the

294

chapter 8

Apples and Pears The following statement is a very good example of the pitfalls to be avoided in comparing the production costs of both refining processes: ‘John Phillips found that for the year 1840 the cost of smelting was only 34 % of the value of silver produced, while that of patio amalgamation was 46.25 percent’.70 The first impression is that smelting offers a larger margin than the patio process, but this initial impression is false. To begin with, if an ore destined for smelting contained lead, then Phillips was simply stating two different facts that have no common ground, and a comparison of values becomes irrelevant. However, even if the percentage margins suggested by Phillips applied to the refining of a lead-free silver ore, by the patio process and by smelting, then the silver content of the ore had to be stated for the comparison to be valid. Figure 85b illustrates why this is necessary: at Regla, a lead-free ore with 0.2 % silver (the type of ore refined by the patio process) could not have been refined at a profit using smelting, since the value of its silver was less than the cost of refining it. The problem also lies in attempting to compare refining costs on both sides of the Atlantic. There is an assumption in the historiography that silver from the New World was less expensive to produce, compared to silver from European mines. Frank argues that ‘the supply price of silver was relatively low where it was abundant at the mine head, especially in the Americas’, but abundance of ore does not translate necessarily into better profit margins for the refiner in the New World, compared to European refiners using a completely different

70

patio process required less or more manpower than smelting. Proponents of the former have been Medina, as quoted in Castillo Martos, Bartolomé de Medina, 112, and Berthe, “Le mercure et l’industrie mexicaine au xviéme siècle,” 145–146. One proponent of the latter has been Lara Meza, Haciendas de beneficio de Guanajuato, 101. The smelting overcapacity installed at Regla in the nineteenth century and the irregular supply of ore fit for smelting rule out any meaningful comparison between the labour cost and manpower dedicated to each process. The most conservative interpretation of the accounting data for now is that the main difference was the level of skill required (much greater for smelting, as seen in the level of wages), and that the patio process also required ample manpower in certain areas, such as the planilleros. Quoted from a report to the Directors dated 29 June 1841, Real del Monte Proceedings, in Randall, Real del Monte, 114. Phillips, as many of the English managers of the company of the period, was strongly in favour of replacing the Mexican patio process with the English tradition of smelting, and offered his data as support to his argument. Whether he was being disingenuous at the time or believed he was comparing costs on the same basis is open to question. According to my calculations, the corresponding values for the third quarter of the nineteenth century are 15% for smelting ores with 1.9 % silver, and 50 % for the refining by the patio process of ores with 0.19% silver.

the economies of refining silver

295

technical and business model.71 With regards to refining costs on both sides of the Atlantic, there is no common ground for a straightforward comparison. The majority of silver ores in Europe at the time were smelted, and silver was the minor product, while copper and lead constituted the main source of revenues to the refiner. On both sides of the Atlantic, the sources of revenue to the refiner remain totally different. The following three cases of European smelting costs from the nineteenth century are plucked from a non-exhaustive search in the literature, and represent Germany, France and England. Under Napoleon, the French oversaw silver production in the German mines of the Harz region, and as a result Héron de Villefosse drew up a report that contains partial refining costs both for mining of lead-silver ores and for their smelting for the years around 1805. Based on his data it is possible to calculate that the argentiferous lead ores with a silver content between 0.07 and 0.18 % could be smelted at a profit. This would not have been possible at Regla using smelting, not for technical reasons, but simply because the value of the silver content alone could not have covered the cost of smelting. However, what allowed the German smelters to obtain a profit was the amount of revenues from the sale of litharge, lead and copper, which in total amounted to 75% (Lautenthal), 94 % (Frankenscharner), 104% (Altenau) and 39% (Andreasberg) of the smelting costs.72 In the case of smelting silver ores at Pontgibaud (France), from 1838 to 1849, reported by Rivot, the average cost of refining one metric quintal of ore with 0.1% silver was 23.45 francs. Approximately 160 times more lead and litharge was produced than silver, but because of the price differential between the products at Pontgibaud, silver ended up contributing 75% of the total revenues, which amounted to approximately 24 francs per metric quintal, thus barely exceeding the refining costs. The only reason the French smelters could operate at a profit was the average additional revenue of 8 francs per metric quintal from the sale of lead products.73 Rivot also refers to English smelters

71

72

73

Frank, ReOrient, 134. Previously, Flynn had assumed that production costs for silver in the New World were consistently lower than in Europe at the time. Dennis O. Flynn, “Comparing the Tokugawa Shogunate with Hapsburg Spain: two silver-based emprires in a global setting” in The Political Economy of Merchant Empires: State Power and World Trade, 1350–1750, ed. James D. Tracy (Cambridge: Cambridge University Press, 1991), 337–340. See also Brown, History of Mining, 42. Antoine-Marie Héron de Villefosse, De la richesse minérale considérations sur les mines, usines et salines des différens états, et particulièrement du Royaume de Westphalie, pris pour terme de comparaison (Paris: Levrault, 1810). Raw data for my calculations taken from tables facing p. 102 and p. 116. Louis Edouard Rivot, Description des gites métallifères, de la préparation mécanique et du

296

chapter 8

in Flintshire in the nineteenth century working with raw ores that contain mainly lead (ca. 75%) and little silver (less than 0.03 %) in the ore. He could not include actual refining data from these works in his very extensive book on metal smelting since ‘the directors do not like to provide foreigners with exact information on their commercial affairs’, a sentiment that has dogged most archival research for this topic. His case study for a generic smelter of Flintshire is based on a process using crystallisation (Patterson’s process) to enrich the lead prior to cupellation. For this location Rivot assumes that silver only provides 10% of the revenues, thus leaving lead products to bear the brunt of making the operation profitable.74 All three cases illustrate the symbiotic relationship that characterised the silver smelting business in Europe. Lead, litharge and copper had a local market, and thus would provide the revenues required to make the smelting of these silver ores profitable to the refiner. It does not come as a surprise that Percy reports that in England ‘foreign silver ores, chiefly from South America [with a silver content upwards of 0.8%], have been largely imported and smelted for the past 40 years … the business appears to have been highly profitable’.75 A comparison between silver refining costs in Europe and in the Hispanic New World is thus not possible, not for the patio process, which was never applied in Europe, and not even for smelting of the same type of ore on each side of the Atlantic. The only factor shared by all refiners was the value of silver, set historically in Europe at a ratio around 15 to 1 to gold (see above), in a context completely divorced from the production realities of its major sources in the New World. The real competition for silver refiners in the New World was not between the patio process and smelting, or with refiners in Europe, but between their local costs of refining and the rigid ceiling on the price of their product. ‘Ganando indulgencias con escapulario ajeno’76 Jakob Fugger, the founder of the banking dynasty, had managed to translate the very risky business of lending money to powerful, unpredictable and profligate royal debtors into a huge personal fortune, by claiming as collateral two valuable royal assets: land and minerals. Well before the first Conde de Regla

74 75 76

traitement métallurgique des minerais de plomb argentifères de Pontgibaud (Paris: CarilianGoeury et Vor Dalmont, 1851), 193–197. Principes généraux du traitement des minerais métalliques (Paris: Dalmont et Dunod, 1859), Vol. ii, 317–319, 386–388. Percy, Metallurgy, 524–525. ‘earning indulgences with the scapular of another’. A Spanish saying with strong Catholic

the economies of refining silver

297

bought his entry to Spanish nobility thanks to the refining of silver, Jakob Fugger had become wealthy from the mining and refining of silver and copper in Mitteleuropa. Columbus had not yet sailed to the New World when it was already an accepted practice of the Hapsburgs to raise loans by allowing the lender to take over the mining and refining of whatever commercial minerals happened to lie within their lands. Unpaid loans to the Crown came to be considered by the Fuggers as potential strategic losses, to be compensated by net overall gains, whether as interest payments, minting rights or royal patronage in turbulent times of defaults by the Crown. As Charles v, and then his son Philip ii, managed to extract even greater sums from the heirs of Jakob Fugger, so was the Fugger family drawn into the sphere of royal assets in Spain, claiming as collateral the revenues from the Maestrazgos and ultimately the running of the silver mine at Guadalcanal. One of the prime assets of the Maestrazgos was the mercury mine of Almadén. The Fugger banking house would manage the mine in a series of periods that started in 1525 and would continue, with some lapses, until 1645, by which time the financially depleted banking family had ceded the production contract of Almadén back to the Crown.77 Matilla Tascón comments that the first contract to operate Almadén was offered as early as 1525 ‘in order to compensate him [Anton Fugger], according to what is said, for all that was owed to him by Charles v’.78 The trail of debts owed by the Spanish sovereigns to the Fuggers up to mid seventeenth century does not differentiate what was owed due to mercury supplied from Almadén and what corresponded to unpaid capital and accrued interest. The following snapshots however provide enough information to suggest a financial crash in slow motion. By 1560 the total debt owed to them by the Spanish Crown reached the sum of 2,975,797 ducats of 375 maravedies (over 4 million pesos). In 1582 the Fuggers failed to produce mercury, and Matilla Tascón conjectures that this may have been linked to the fact that the Crown could not be trusted to pay them on time.79 The representative of the Fuggers at Almadén complained to the Crown officials that no payments had been received for shipments of mercury from 1631 to 1633, and only partial ones had been made for 1630. In at least 1639 and 1645 remittances from New Spain derived from the sale of

77 78 79

overtones, it refers to getting the credit (religious indulgence) thanks to the sacrifices made by someone else (the symbolic commitment of having to wear a scapular under one’s clothing). Antonio Matilla Tascón, Historia de las minas de Almadén, 1 (desde la época romana hasta el año 1645) (Madrid: Gráficas Osca, 1958), 29ff. Ibid., 37. Ibid., 87, 108.

298

chapter 8

mercury were diverted to pay pressing obligations of the Crown, not to pay outstanding obligations on mercury to the Fuggers.80 In the end, the Fuggers could not meet the delivery targets of the contract that run from 1636 to 1645.81 The leasing of Almadén to the Fuggers took place well before silver changed the order of magnitude of the market for mercury, yet I suggest it would lead to a singular solution for the fiscal needs of Spain. On the one hand Almadén would serve as a collateral whose value grew apace both with the need for mercury in the New World and with the increasing amount of debt owed to the Fuggers. At the same time, part or all of the debt outstanding to the Fuggers could be considered by the Crown as accounts receivable of the Fuggers, to be borne in increasing amounts by Almadén, repayable by the Crown at some point in the indeterminate future. From 1560 to 1639 a total of approximately 8 million kg of silver were refined in New Spain.82 Assuming that at least one quarter of this total was smelted in this period, this means that anywhere from 5 to 20 % of the total silver produced using mercury in New Spain by 1640, would have been due to mercury seemingly gifted by the Crown but ultimately paid for (as irrecoverable debt) by the Fuggers.83 80 81 82 83

Lang, Monopolio Estatal, 79. Bakewell, Silver Mining in Zacatecas, 166–167. Based on a total of approximately 8 million kg of silver produced in New Spain from 1560 to 1639, according to Table 3–1 in TePaske and Brown, Gold and Silver, 113. Almadén was the major, and at times the sole, source of mercury in New Spain. The mercury debt seems to have peaked at around four million pesos, while a persistent debt at around one million pesos is reported up to the end of the eighteenth century. Thus, according to Lang, the debt run up on mercury by silver refiners in New Spain had already reached 1,1 million pesos as early as 1580. Lang, Monopolio Estatal, 361. According to Bakewell the debt in New Spain by 1590 had increased to 1,8 million pesos, and would still remain at 1.1 million pesos by 1597. To place these sums in perspective, new sales of mercury to the provinces of New Spain and Pachuca-Pánuco from 1590 to 1597 totalled some 2,5 million pesos. Peter J. Bakewell, “Notes on the Mexican Silver Mining Industry in the 1590’s,” in Mines of Silver and Gold in the Americas, ed. Peter J. Bakewell (Aldershot (gb); Brookfield (Vt.): Variorum, 1996), 176–177. By the end of the sixteenth century it was acknowledged that ‘the [mercury] debt of the miners was irreversible’, in G. Cubillo Moreno, “Los dominios de la plata: el precio del auge, el peso del poder. Empresarios y trabajadores en las minas de Pachuca y Zimapán, 1552–1620,” (México: Instituto Nacional de Antropología e Historia/Consejo Nacional para la Cultura y las Artes, 1991), 166. Bakewell and Lang report a persistent debt on mercury throughout New Spain during all the seventeenth century. Bakewell, Silver Mining in Zacatecas, 207.; Lang, Monopolio Estatal, 361. Castillo Martos states that the debt continued to be rolled over, never repaid, and a century later, in 1763, was approximately 1.1 million pesos for New Spain. Castillo Martos and Lang, Metales preciosos—unión de dos mundos, 54, 145.

the economies of refining silver

299

The role of the Fuggers throughout this period raises at least two questions: why did they persevere until they went bankrupt, and why did they not reap the benefit from a textbook example of a new technology creating a completely new market, for a product only produced by them under licence from the Crown? To the first there is an article by the German historian of the Fuggers, Hermann Kellenbenz, which asks through its title whether the rent of the Maestrazgos was good business for the Fuggers in Spain after 1562. His short answer is yes, due to payments of interest rates that offered a way of slowly amortising the outstanding debt and from collateral business opportunities, such as profits from arbitrage with Spanish silver minted at Hall (Tyrol) and Venice. Events not analysed by Hellenbenz however indicate that ultimately by the mid-1640s the inclusion of Almadén had only aggravated their problem instead of solving it. The Fuggers had tried to extract themselves from the Spanish Monarchy’s drain on their financial resources as early as the 1550s. However, the lure of fresh future cash flows offered as bait on each new demand for loans, extended their line of credit to the breaking point.84 As to the brave new world for mercury created by its use to refine silver ores, there is a certain irony in the fact that by inadvertently gifting part of the mercury required for the patio process, the Fuggers became too weak to benefit from it. The Succour to Refiners The mercury debt owed by refiners to the Crown from the introduction of the patio process in New Spain had been continually rolled over by successive Viceroys, until Madrid issued instructions in 1634 to demand prompt payment from refiners for any new supplies of mercury. This in turn has been interpreted as a watershed in the history of refining in New Spain, since the new and more inflexible attitude of the Crown with respect to mercury sales is claimed to have triggered a major switch from the patio process to smelting after the 1650s. Lacueva places the context of this decision in 1634 within the new financial strategy of the Count-Duke de Olivares, who since the ascension to the throne of the young King Philip iv in 1621 had been implementing new ways to strengthen the collection of royal revenues.85 However, the evidence presented in the previous section could also point to the realisation by the Crown that the loans by the Fuggers, that had kept the mercury debt of New Spain rolling over for nearly one hundred years, were coming to an end. 84

85

Hermann Kellenbenz, “Los Fugger en España en la época de Felipe ii. ¿Fue un buen negocio el arrendamiento de los Maestrazgos después de 1562?,” in Dinero y crédito (Siglos xvi al xix), ed. Alfonso Otazu (Madrid-Villalba-Segovia: Banco Urquijo 1977), 19–36. For example, Lacueva Muñoz, La plata del Rey, 79–87.

300

chapter 8

The history of the pricing of mercury to the refiners in New Spain was the result of the opposing tensions in the designs of the Crown on its two major sources of mineral wealth, silver and mercury. These were now tied by the patio process into a potentially conflicting trajectory, if it were not carefully handled. On the one hand silver was produced by private investors who clamoured for help, rightly or wrongly, in the shape of lower mercury prices. If no such assistance was provided, or supply of mercury faltered, these refiners could shift in theory to smelting, and reduce substantially their consumption of mercury. Should this occur, tax revenues on silver might remain at a similar level, but the revenues from the monopoly on mercury would suffer.86 Direct revenues from the sale of mercury were very important for the Spanish Treasury. According to Herbert Klein ‘there is little doubt that mercury sales were the single largest generator of income [from monopoly taxes] in … Mexico’.87 The level of that net income, once production costs at Almadén or payments for mercury from Idria have been accounted for, has not been calculated, but the gross figures are impressive. From 1680 to 1809, the gross revenue from the mercury monopoly in New Spain was approximately 30% of the total revenues obtained by the Crown from mining activities.88 A breakdown reported by Mendizábal for Pachuca in the early seventeenth century indicates 50,000 pesos in revenues from silver taxes, and 30,000 pesos from the sale of mercury.89 Even if adjusted by half to take into account production costs and freight to New Spain (inland freight was charged and accounted for separately within New Spain), the data

86

87

88

89

In Chapter 4, I pointed to the zeal in prosecuting innovators in Potosí, who in the late sixteenth century were searching for a more efficient use of mercury, which was feared to decrease the mercury revenues to the Crown. Herbert S. Klein, The American Finances of the Spanish Empire: Royal Income and Expenditures in Colonial Mexico, Peru, and Bolivia, 1680–1809 (Albuquerque: University of New Mexico Press, 1998), 18. Based on the data in Table 5.2 (Estimated Average Annual Income from Mining Taxes, Vice-Royalty of New Spain, 1680–1809) and Table 5.3 (Estimated Average Annual Monopoly Tax Incomes, Vice-Royalty of New Spain, 1680–1809), assuming 30 % of the average values in table 5.3 correspond to the sale of mercury (the percentage of total monopoly revenues corresponding to mercury related revenues is provided by Klein). Ibid., 19, 80, 86. Tax revenues from silver of the New World contributed at its peak one third of the revenues of the Spanish Treasury, according to Mauricio Drelichman and Hans-Joachim Voth, “Institutions and the Resource Curse in Early Modern Spain,” in Institutions and Economic Performance ed. Elhanan Helpman (Cambridge, Mass.: Harvard University Press, 2008), 137. This would have made mercury sales the contributor of up to 10 % of total revenues for the Treasury in Spain. Mendizábal, “Minerales de Pachuca,” 274.

the economies of refining silver

301

confirm that mercury was an important source of revenue to the Crown in its own right. The question was, how far would the Crown sacrifice its mercury revenues in order to assist refiners? The texts of the eighteenth century explicitly evaluate the choice in mathematical terms, judging decisions by balancing the expected increase in silver taxes to the Crown against the expected loss of mercury revenues incurred by decreasing its price. The Spanish authorities managed to handle these tensions with extreme pragmatism. For the first hundred years, I have shown that the Fuggers shouldered the unpaid debt on mercury supply, whilst Spain reaped the benefit of silver produced by the patio process and the revenues of its mercury monopoly. At the same time the Spanish Crown had sought from the beginning to obtain a high price for mercury in New Spain. There is an instruction from Joanna of Austria, the Princess Governess of Spain during the temporal absence of Philip ii, who in 1559 issued a Royal Decree (Real Cédula) stating that ‘we would benefit greatly [on the sale of mercury in the New World], and with that mercury would earn double of what it costs over here [at Almadén]’.90 The initial pricing levels per quintal in New Spain were certainly more than a doubling of the prices in Spain, as can be observed in Figure 89. The highest values of all its colonial market history are registered precisely in the first decades when the use of mercury for refining is being implemented in New Spain. This is the stage one expects a subsidy to be applied, to assist the new technology in gaining a foothold against the much more traditional process of smelting. Once the patio process had taken hold, a major incentive to sacrifice revenues to the Crown from mercury sales disappeared. The empirical fact that even at the high end of mercury prices, New World silver flooded over world markets, overwhelming at the beginning the European silver industry, gave the pragmatic answer to the Spanish officials that the mercury price being applied was enough to insure continued production of silver, an argument that would hold well into the eighteenth century. Garner considers that the profit on mercury gained by the Spanish Crown ranged from 100 to 300 %.91 The visual evidence in Figure 89 indicates that only in the last 30 years of this 250-year period could it be argued that mercury was by State design sold

90

91

As quoted in Manuel Castillo Martos, “Primeros beneficios de la plata por amalgamación en la América colonial (1565–1600),” in Minería y metalurgia. Intercambio tecnológico y cultural entre América y Europa durante el período colonial español, ed. Manuel Castillo Martos (Sevilla, Bogotá: Muñoz Montoya y Montraveta Editores, 1994), 378. It is not clear if she is referring to a sale price or net profit equal to double the production cost. Garner and Stefanou, Economic Growth Bourbon Mexico, 138.

302

chapter 8

figure 89 Prices in Spain and price of mercury in New Spain, based on footnote 62 courtesy of francis & taylor

in New Spain at or below its price in Spain.92 In 1767 the first decrease in price from 82 to 61 pesos was implemented, with the final reduction to 41 pesos decreed in 1776.93 Underlying the question as to when, or even whether, a subsidy (in the modern meaning of the term) ever existed with regards to

92

93

Matilla Tascón, Minas de Almadén (to 1645), 37–183; Historia de las minas de Almadén, 2 (desde 1646 a 1709) (Madrid: Instituto de Estudios Fiscales, 1987), 96–117. Complimentary information from Bakewell, Silver Mining in Zacatecas, 172. The end of the eighteenth century also saw the import of mercury from Idria, the ‘German mercury’ that appears in the accounts of the various Cajas Reales of New Spain, which was priced in the range of 52 to 62 pesos per quintal towards the end of the eighteenth century. Even at the lower price Humboldt claimed that Idria mercury still provided a profit of 23% to the Spanish Crown. See Garner and Stefanou, Economic Growth Bourbon Mexico, 139; Humboldt, Essai politique, Tome iv, 87, 89. This change in Crown policy is one of the consequences of ‘early Bourbon reformism’ and is interpreted as signalling a decisive change from mercantilist policies on mercury to the proactive fostering of silver production, according to Dobado and Marrero, “The Role of the Spanish Imperial State in the Mining-led Growth of Bourbon Mexico’s Economy,” 869. Production costs at Huancavelica were much higher, in the 58 to 73 pesos per quintal range. Brown, History of Mining, 32. However, mercury at Potosí was priced only slightly higher than in New Spain, as shown by the following values per quintal: 105 pesos in the sixteenth century, 97 pesos in 1645, 79 pesos in 1779, 71 pesos in 1787 and 50 pesos in 1809. Bakewell, “Colonial Mining,” 122. This conclusion remains valid even taking into account transport to Seville (1.7 pesos per quintal in 1646), freight costs to the New World (2.4 to 4.8 pesos per quintal, from 1568 to 1598) and packing (under 3 pesos per quintal in 1619). Inland freight

the economies of refining silver

303

mercury pricing set by the Crown in New Spain, lie two more fundamental issues.94 First, the price of silver in the period of interest was independent of the price of mercury. Lowering the price of mercury did not make silver from New Spain more competitive with respect to its sole competitor, European production. The sole purpose of the decrease in mercury prices was to increase the production of silver, by allowing the patio process to refine at a profit a greater range of silver content in new extracted ores, as well as incorporating into the production process the masses of tailings accumulated over decades with a silver content below 0.08%. The sunk cost of zero of these tailings reduced by nearly 60% the total variable refining cost of the patio

94

in New Spain was 3 to 4 pesos per quintal in the 1620s. Matilla Tascón, Minas de Almadén (to 1645), 98, 221–222; Bakewell, Silver Mining in Zacatecas, 171. The Spanish Crown did offer its indirect support to miners and refiners in the New World, though the local elites would have been expected to reap the most from this assistance (preferential supply of mercury to major refining concerns, dispensations from the payment of royalties, and other major tweaks to the system). The fact that the English investors in the region around Pachuca as late as the nineteenth century had to invest major capital simply building roads, is an indication of the limited extent of the Crown’s involvement in the general infrastructure required by its silver industry. The extent to which the ordinary miner-refiner benefitted from the assistance of the Spanish Crown can be better judged in relation to what the German miners still received from their King in the region of the High Harz at the beginning of the nineteenth century. The King had the right to retain one tenth of the value of the minerals extracted (silver, lead, copper); to have a share in each mine, around 3%; to build all the common infrastructure required by mining and refining (water reservoirs, galleries to discharge water from mines, for crushing the ore, and smelters) and to operate the washing and smelting of ores exclusively, all for a modest fee payable by each mine. Finally, the Crown bought exclusively the lead and copper at a price lower than market prices, and profited from the difference (silver was bought at three quarters its coined value). On the other hand, the King had the obligation to provide all the wood required by mining and refining at no cost except that of cutting and freight; all other industrial consumables were sold at a regulated price, below market; the chief officials for mining, smelting and forests were paid by the Crown, all other wages by each mine; the Crown would provide a fixed amount of cereal (rye) at a low price to each worker; the Crown would partially compensate the mine for unexpected increases in certain materials if it could prove unable to meet the new prices; it would also compensate for certain increases in the price of oats required for the animals used in mining; injured or sick workers were cared for, and a fund established for widows. There is a major difference between the above and providing a legal framework for the claims of mines, selling mercury above its cost of production, and providing escorts for the shipments of silver. Héron de Villefosse, De la richesse minérale du Royaume de Westphalie, 89–94.

304

chapter 8

process, a much greater impact on refining costs than the reduction in the price of mercury, according to the results presented in the early part of this chapter. Thus, through the increase in silver taxes, the Crown compensated the decrease in mercury revenues, without having to sacrifice them altogether, which would have been the case had smelting taken the place of the patio process.95 This brings us to the second fundamental issue. It is not evident that the pricing of mercury, or even the implementation of the patio process, determined the existence and survival of the silver refining industry in the New World. The core refining by smelting of argentiferous lead ores in New Spain remained impervious to any strategy on supply or pricing of mercury. The Crown could have implemented a more pro-active strategy with respect to smelting, such as offering incentives in the search for lead deposits in New Spain, or in the supply of cheaper charcoal, or by promoting improvements in furnace efficiency. From a chemical point of view, smelting could have processed the totality of the silver ores in New Spain, whilst the patio process would have always been limited to approximately two thirds of the total amount of silver historically refined in New Spain, regardless of the price of mercury. This is an argument along the lines of counterfactual history that I will complete with the data I present in Chapter 9. Barrel Process The barrel (toneles), or Freiberg process, was not a novel refining technology, but a variation on Alonso Barba’s cazo process, now promoted by Baron Born in Europe at the end of the eighteenth century. At the time of its introduction it led to strong but very contrasting opinions on the utility of the process in New Spain. Humboldt was one of the earliest supporters of the Freiberg process, but pointed out that the sheer scale of the ores waiting to be refined in New Spain dwarfed the extent of its application in Europe.96 Sonneschmidt, an initial supporter of the method, ended by concluding that after ten years of trials in New Spain it had failed to convince its users since it was more costly, extracted less silver than the patio process, consumed more mercury than expected and 95

96

Dobado and Marrero conclude that ‘it was in the interest of the Crown that amalgamation should be the preferred technique for silver refining’, in Dobado and Marrero, “The Role of the Spanish Imperial State in the Mining-led Growth of Bourbon Mexico’s Economy,” 866. According to Humboldt 60 thousand quintales were refined yearly using roasting and the barrel process in Europe, while in New Spain 10 million quintales would have been the required quantity. Humboldt, Essai politique, Tome iv, 85.

the economies of refining silver

305

produced impure silver.97 Humboldt’s approval may have been a factor in the enthusiastic backing it received from the English managers of the Adventurers Company of Real del Monte, from the very beginning of their project.98 At that stage, 1825, they had no experience in refining silver ores of the New World on which to base their decision.99 Mexican opinion was dismissive of the level of imported English expertise: ‘the recently arrived English miners … in general do not know any other minerals than copper, tin, iron and coal, and they are completely ignorant of the patio refining process which is so important [in Mexico]’.100 John Buchan later proceeded to convince the new Mexican owners to implement it on a much greater scale than the English had attempted, even though in the initial years his own report indicated that the barrel process only extracted on average 80% of silver in the ores.101 Though it never threatened to displace the patio process, its initial appearance coincided with the Bourbon initiatives at the end of the eighteenth century, that sought to increase the production of silver. Its subsequent adoption during the phase of foreign capital investment in Mexico, has made it an 97 98

99

100

101

Sonneschmidt and de Fagoaga, Tratado de la amalgamación de Nueva España, iii, vi, ix–x. Letter dated 23rd August 1825 to Roger Morgan, Esq., at Regla from James Vetch at Mineral del Monte (Real del Monte): AHCRMyP, Fondo Siglo xix, Sección: Correspondencia, Serie: Compañía a Varios, Subserie: Correspondencia General, 8–1: 20 Abril 1825–1 Noviembre 1825. The experience of silver refining by the English investors was very limited. John Taylor never set foot in Mexico, see for example Roger Burt, John Taylor. Mining Entrepeneur and Engineer 1779–1863 (Hartington: Moorland Publishing Company, 1977). England by 1854 was producing just 70,000 pounds troy of smelted silver compared to Mexico’s output of 1,750,000 pounds troy coming mainly from the patio process. Production data from J.D. Whitney, The metallic wealth of the United States, described and compared with that of other countries (Philadelphia: Lippincott, Grambo & Co., 1854), 506. In Viajero, “Las Minas de México,” 182. An anonymous Englishman returning from Mexico published in England in 1856, under the pseudonym ‘Traveller’ (Viajero), a scathing criticism of the way English capital had been extravagantly wasted on Mexican mining ventures. In a footnote, the Mexican translator appended his opinion on the skills of English miners. The opinion of modern Mexican historians is equally scathing: ‘the arrogant English investors and administrators … believed they possessed a vast knowledge … much greater than the aggregate practical experience of the Spanish and Criollo owners [of refining haciendas] in Mexico. The economic failure of the English company can be explained to a large degree on the blind faith placed on those principles’. Inés Herrera Canales, Cuahtemoc Velasco Avila, and Eduardo Flores Clair, Etnia y clase, los trabajadores ingleses de la Compañía del Monte y Pachuca, 1824–1906 (México: inah, 1981), 7. Ruiz de la Barrera, “La Empresa de Minas del Real del Monte,” 113; Buchan, Report Real del Monte, 18.

306

chapter 8

figure 90 Annual average of cargas of silver ore processed at the refining haciendas of the Compañía de Minas del Real del Monte y Pachuca in the period 1853 to 1873. b: barrel process p: patio process. Source data from Estados Comparativos.

obligatory reference in discussions on the technical relevance of both historical events.102 It has been claimed as being ‘the most important innovation bequeathed by the English who exploited the mines of Pachuca and Real del Monte’.103 According to another modern historian ‘it definitely improved on Mexico’s traditional ore-reduction method … to supplant altogether the traditional Mexican method of treating silver ore … from the patio … to barrels, where it was quick and less destructive of that costly commodity [mercury]’.104 There has also been more restrained praise: ‘to them [the English investors in Real del Monte] can be attributed as innovation … [the] many attempts to substitute the patio amalgamation process that resulted in minor modifications in

102

103 104

The barrel process had been tried out in Oaxaca and Bolaños prior to the 1840s, a test is reported with ore from Catorce, but the Real del Monte Company was the first to apply it in any major scale in Mexico. Duport, Métaux précieux au Mexique, 51; Laur, “De la metallurgie de l’argent au Mexique,” 224–225; John Phillips, Descriptive Notice of the Silver Mines and Amalgamation Process of Mexico. Extracted from the Railway Register (London: Pelham Richardson, 1846), 16. From 1877 to 1893 the barrel process contributed just 4 % of total silver refined in Mexico. Flores Clair, Velasco Avila, and Ramírez Bautista, Estadísticas mineras, ii, 161–162. Rina Ortiz Peralta, “El beneficio de minerales en el siglo xix: el caso de la Compañía Real del Monte y Pachuca,” Tzintzun: Revista de Estudios Históricos no. 14 (1991): 75. Randall, Real del Monte, 109, 118.

307

the economies of refining silver

table xxi Comparison of major refining parameters for the various haciendas of the Compañía de Minas del Real del Monte y Pachuca, in the period 1853 to 1873 (except for 1856–1858 and 1866–1868). The extraction cost of the ore is not included. Source data from Estados Comparativos.

average silver content of ore (%)

0.24

0.24

0.18

0.21

0.26

Sánchez (b)

Velasco (b)

San Miguel (b)

Regla (p)

Loreto (p)

average variable production cost, pesos per kg silver

10.2

10.1

13.1

9.7

11.5

average unextracted silver (%)

16.7

14.0

17.6

9.9

6.1

average mercury to silver ratio

0.8

0.9

0.8

1.6

1.6

the methods to reduce ores’.105 Since no economic analysis has been provided to sustain these conclusions, and the historical failure of the barrel method to displace the patio process is significant, it is interesting to find out what the accounting records of Regla have to say. There is no doubt that the Mexican Compañía de Minas del Real del Monte y Pachuca, in the period 1853 to 1873, believed in the barrel process. The Hacienda Velasco came to apply it to over twice the amount of ore that was processed by the patio process by its sister refining haciendas. However, by the early 1870s, Velasco undergoes a marked decrease in the amount of ore processed, to be overtaken by Regla in 1873 (Figure 90). Why the barrel process was used on such a large scale is not immediately evident, when some of the main production parameters from each refining hacienda are compared (Table xxi). Mercury consumption was lower than for the patio process, as expected from Alonso Barba’s original method, but total refining costs were higher. Furthermore, during the first three years of operation, on average up to one quarter of the silver could be left unextracted in the ores processed at Sánchez and San Miguel, and only slightly lower at Velasco. Velasco, the more efficient of the barrel process haciendas, still left 40% more unextracted silver than by the patio process at Regla. 105

Velasco Avila et al., Estado y minería en México, 106–107. For the view that the more effective Bourbon reforms were administrative rather than technical, see Castillo Martos and Lang, Metales preciosos—unión de dos mundos, 183.

308

chapter 8

table xxii Average costs required to refine 1kg of silver using the patio and barrel processes. The haciendas in italics used the barrel process, and the haciendas in normal script used the patio process. The data has been calculated for the period 1853 to 1873 using as source the Estados Comparativos. The extraction cost of the ore is not included.

Costs (pesos) / 1kg silver Mercury Salt Copper sulphate Fuel Others+labour Total Mercury+fuel Sánchez Velasco San Miguel Regla Loreto

1.03 1.43 1.18 2.45 2.84

2.03 2.15 3.20 2.21 2.34

0 0 0 0.95 0.81

2.67 2.42 2.68 0.20 0.20

4.52 4.10 6.00 3.84 5.27

10.25 10.10 13.06 9.65 11.46

3.70 3.86 3.86 2.65 3.04

A more detailed breakdown of the variable refining costs for each hacienda during this period shows what was happening. In the case of the barrel process, the savings on mercury were overshadowed by the greater spending on the fuel required to roast the silver ores with salt, and then to heat the barrels. Thus the column in Table xxii indicating the total pesos spent on both mercury and fuel to produce 1kg of silver, is higher for the barrel process than for the patio process. Since the cost of salt is roughly equivalent for both processes, it turns out that the barrel process offered no net advantages on cost even when it has saved at times on mercury consumption. Why then was the barrel process ever implemented by the Compañía de Minas del Real del Monte y Pachuca? Buchan had access to all this information, yet persisted, arguing that the patio process would have not been able to extract as much silver as the barrel process from recalcitrant ores.106 In addition, by cutting down on the reaction period it could partially offset its lower percentage of silver extraction with a higher output of silver per unit of time. It may be significant that Velasco retained its pre-eminence when it was able to process 2.5 times the amount of

106

Buchan, Report Real del Monte, 16–18. The table with the comparative costs of production for the year 1854 appears on page 17 of his report. According to Duport, the silver ores that did not respond well to the patio process in Zacatecas were the black or red sulphides of antimony and silver, which could retain up to 40% of their silver after refining. Duport, Métaux précieux au Mexique, 246.

the economies of refining silver

309

ore compared to Regla. Once their throughput became very similar, for example towards the end of the 1853–1873 period, Regla would have been a much better economic choice to process ores, under equal chemical conditions. It can also be argued that even its capacity to treat difficult ores owed more to the prior roasting with salt, that aided the conversion of silver compounds to silver chloride, than to the barrel process itself.107 It never displaced the patio process as the most cost-effective option to refine the majority of Mexico’s silver ores.108 The irony remains that even some who have recognised its failure still appeal to the ‘timelessness’ of the Mexican scenario. ‘In Europe where speed was important and labour expensive, the much faster … more wasteful Born process was preferable’ while in Mexico ‘time was not important and labour cheap … [so that] it was more profitable to use the slower, less wasteful patio process’.109 The barrel process did not fail because ‘time was not important’ in Mexico. It failed for the simple reason the patio process was never a ‘Mexican trick’, but an efficient industrial process, whose refining costs per kg of silver extracted from the typical ores of Mexico were lower than those of the European barrel process.110

107

108

109

110

Alonso Barba had recognised nearly three centuries before that the cazo process (on which the barrel process is based) worked better with ores high in native silver and/or silver halides, as was the case in Catorce at the end of the eighteenth century. Alonso Barba, Arte de los metales, 111–112. Roasting with salt converts silver compounds in the ore to silver chloride (a silver halide), as indicated in Chapter 4. Thus it is a valid question whether prior roasting with salt before the patio process would have been another option to refine these recalcitrant ores. It never achieved a major penetration either of the refining market in Europe. In Spain in the nineteenth century 60 barrels were employed to refine ore from Hiendelaencina at La Bella Raquel, the largest facility employing the Freiberg process according to Kerl, Crookes, and Röhrig, Prof. Kerl’s Metallurgy, 331. Clement G. Motten, Mexican silver and the enlightenment (Philadelphia: University of Pennsylvania Press, 1950), 53. The stereotype on time and Mexico is repeated by other authors: ‘[refining by the patio process] costs little … cheapness of plant compensates for the time … in Mexico … time has no value’. Egleston, The Metallurgy of Silver, 311. ‘in a sense, the English expended huge amounts of time, energy and money in a fruitless effort to learn a Mexican trick they did not even like [emphasis added]’, in Randall, Real del Monte, 115.

310

chapter 8

Silver in the Context of Other Commodity Trades At the risk of over-simplifying quite complex scenarios, it is nevertheless possible to single out differences in the economic context that allowed each of the major commodities of the Early Modern period to occupy its distinctive historical role. The spice trade was a typical case along pre-capitalist lines of a market opportunity of buying cheap to sell dear, with no technology required to add further value in the chain to the final consumer. For cloves the markup in Lisbon in times of scarcity could reach 240 times the price at source in the Moluccas; for nutmeg, up to 840 times between London and the Bandas. From late seventeenth to late eighteenth century, the mark-up on cloves in Amsterdam ranged from 15 to 25 times the price at source, and for pepper approximately five times.111 In the case of cotton, technology developed by and applied exclusively at the European imperial core played a decisive historical role. A combination of very low prices and production based on massive slave labour at the colonial periphery and later in the u.s.a., together with the introduction of technological advances in its industrial conversion into cloth, led to its critical role in shaping the Industrial Revolution in England. The mechanisation of cotton picking was only developed and introduced in the twentieth century.112 Sugar, despite its market cycles, required over time a very significant lowering of its retail price to create a larger body of consumers, in the case of England in synergy with the increasing consumption of tea. The final refining stage of sugar was for the most part limited to the European imperial core, with texts of the nineteenth century still debating whether it made more economic sense to allow sugar to be refined at source. From a technological point of view, sugar production and refining was a straightforward process of liquid-solid separation, and the main innovation in the nineteenth century was the use of a partial vacuum to decrease the temperature range at which water was removed prior to the crystallisation stage.113 Silver also presents its own specific profile. Compared to the spice trade, even at the height of the arbitrage trade in silver to China, the level of markup in its final prime market was no more than twice its value after coining 111 112 113

Ronald Findlay and Kevin H. O’Rourke, Power and Plenty. Trade, War and the World Economy in the Second Millenium (Princeton: Princeton University Press, 2007), 179, 208. Joel Mokyr, The Enlightened Economy. An Economic History of Britain 1700–1850 (New Haven: Yale University Press, 2009), 81–83, 129, 162. For an overview of the state of sugar production, technology involved and market data in England in the nineteenth century, see William Reed, The History of Sugar and Sugar Yielding Plants (London: Longmans, Green and Co., 1866).

the economies of refining silver

311

at a Mint in the New World. Even this modest mark-up lasted only a few decades: ‘by 1640, bimetallic ratios around the world converged’.114 Between the sixteenth and the final decades of the nineteenth century the value of silver in the European market, as set by the bimetallic ratio to gold, had shown very little change. The only route to profit in the production of silver lay in the control of operating costs. The margins calculated for Regla are more similar to a modern industrial operation than to the heady heights of the spice trade. In comparison to cotton, there was no need, or even opportunity, to apply exclusive European technology to add substantial value to the commodity. I have argued in the previous chapters that the technological premium for silver had been developed and applied immediately at source, in the New World. Silver was refined, minted and exported from the Spanish colonial periphery as bullion, or in specie as pesos, pieces of eight. The difference in the value of silver between the colonial periphery and an equivalent operation in Spain, would have been due solely to normal marketing costs, not technology: the various freight and import costs, and where necessary the payments to third parties who acted as middlemen in the chain.115 The silver refiners in New Spain did have a major advantage that a producer of cotton, sugar, and porcelain would not have. It was never enough to efficiently produce a piece of cloth, a pound of sugar, or a bowl of porcelain, at a cost below the expected market price. The manufactured product still faced the most difficult hurdle, its transformation into specie through an actual sale to a fickle consumer. The market could refuse to buy it because of changes in taste or style, or there was not enough surplus specie in the households to cover sumptuary consumption, and the producer would be left with an overflowing inventory of unsold produce, which could ultimately bankrupt his operation. For a private individual mining and refining silver in the Hispanic New World, the Spanish Crown never placed a limit to his production, and for the appropriate fee (señoreage, see next chapter), allowed him to convert his output directly into coins of the realm. This meant that whatever he produced could be converted into the most sought-after specie of the period if he so chose, without any limit other than his capacity to operate at a profit, and free from consumer whims. This was a guaranteed revenue that would have been welcomed by any other industry, both then and now. 114 115

Giraldez, The Age of Trade, 91. One example of middlemen are the aviadores of New Spain, who functioned both to provide specie to refiners who did not want to travel to the mint in Ciudad de México, and also as providers of credit. According to Bakewell, a refiner at the end of the sixteenth century could exchange with an aviador nine reales of registered silver for eight reales in coin. Bakewell, Silver Mining in Zacatecas, 211.

312

chapter 8

The Bottom Line Whatever their size or success, the mining and refining operations created a workplace with novel occupational hazards, the human cost of which, for the indigenous workers, their communities and their environment, is still unknown. The extent of this outstanding social and cultural debt, on which the ultimate profitability of the Spanish silver extraction industry was built, remains invisible in the accounts analysed in this chapter. Bearing this limitation in mind, I have used the account books from Regla to explain why refiners in the New World could have reaped profits from their operations. I have argued that in its operating essence, the nineteenth century hacienda at Regla is a sister to any other hacienda of the previous 300 years, and not a distant and unrecognizable relative. Thus, the diagnostic carried out on its profile of refining costs in the nineteenth century could be used to recreate two other historical stages, in the evolution of the relative costs of the patio process and smelting. Thanks to the initial stability over two centuries of the silver to gold ratio, and the immutability up to 1888 of refining costs amidst the collapse of silver prices, I would argue that the errors in my projections affect both processes in the same manner, and do not jeopardize the comparisons. For some 350 years neither technology, the newcomer patio process and the traditional smelting, changed in a substantive manner, and neither managed to displace the other. The former because it could not refine argentiferous lead ores, the latter because production costs in New Spain favoured, for most of the period, the refining of ores with a silver content around 0.2 % by way of the patio process. Even then, for nearly a century (mid 17c to mid 18c), when mercury supplies or restricted credit became a problem, these hard-headed, self-made but very pragmatic individuals had found little to choose between the two refining processes, at least on economic terms. The answer that the accounts from Regla have provided will bolster the argument, to be rounded off in the next chapter, that the relation between the patio process and smelting was not a one-dimensional, zero-sum game. At the very least, increases in production by the patio process could not come at the expense of smelting of argentiferous lead ores. Tailings have played an important role in this history, at the beginning and at the end of the colonial period, and perhaps beyond. Their sunk cost of zero decreased the refining cost of the patio process by around 60%. Villaseñor was right in pointing out that extraction costs, and not the price of mercury but its availability, were the critical economic drivers behind the patio process. The only period mercury was the major influence on the costs of the patio process in New Spain was in the second half of the sixteenth century, when tailings were already being

the economies of refining silver

313

raided and the Crown sold mercury for the highest price it could set, without even having to pay for part of it. For the next three hundred years, it was the extraction cost of the ore that determined the profit margin of a refiner, as had been pointed out by Villaseñor. Reducing the price of mercury increased the impact of salt prices on the total cost structure until they became equal in importance, around 10% of total cost each at Regla, and yet salt was never the intense focus of the lobby by refiners as was mercury. The cost of power for haciendas de patio completes the trilogy of refining costs that most influenced the margin of profit, yet it is even more absent than the cost of salt from many a modern analysis. Copper sulphate, or its less refined variety by the name of magistral, was the most cost effective of all ingredients for the process. Fuel was a very minor contributor to the cost of the patio process, thus isolating it from the vicissitudes of searching for sufficient woodland resources, as well as making its impact on this renewable resource negligible. As to the fixed cost of capital, once the major hurdle of raising capital was cleared, its service was not a major burden on refining costs. As always, regardless of the breakdown of production costs, the need to continuously provide working capital can make or break a business, thus the importance given in the historiography to the allocation and sources of capital in the history of silver refining in the New World. More data on other haciendas are necessary, to establish whether Regla was an isolated case of a very efficient patio process, or whether this was a more potent refining process than has been credited. Smelting remained throughout a viable option, first of all for the obvious reason that for some ores rich in lead it was the only effective way to refine them. The minimum silver content that made smelting profitable in New Spain lay in the range of 0.3 to 0.5%, though in practice at Regla no ores below 1 % silver were smelted. Taylor had identified back in the 1820s the two major challenges the English faced at Regla in order to implant smelting as the preferred option: dressing the ore to increase its silver content, and finding enough fuel to feed the blast furnaces. Smelting of silver ores made sense in the England of the Cornishmen and the Erzgebirge of the Germans, because the sale of lead and copper made profitable their economies of refining. In New Spain / Mexico the whole weight of meeting the costs of smelting fell only on silver and whatever gold was present in the ore. This was the challenge faced by smelting, not the scale of available ore that loomed so large an obstacle in the imagination of Humboldt. Argentiferous copper was never a major component of Mexico’s silver ore deposits, in contrast to the European scenario, so the production of copper was not an option to defray the costs of refining silver in New Spain or Mexico. The production of lead for the local market would only become important in the twentieth century.

314

chapter 8

I have stressed that the historical continuity of silver production was both the consequence of an efficient technical operation, and of the fact that individual refiners, the large concerns and the small operators, all found the necessary level of profit that satisfied each of their individual expectations. This was the business model implemented by the Spanish Crown in the New World, and it worked, if we judge by the flow of silver that emerged from it. However, there was a shift in the profile of the refining industry in New Spain, from the sixteenth to the end of the nineteenth century, that will require a more detailed analysis in future research. Initially the Crown did not allow anyone who was not extracting silver ore to refine it, thus combining the higher risk of the former with the profit potential of the latter. Then by the eighteenth century the Crown allowed the larger refining haciendas to refine silver ores from third parties, under a specific set of accounting procedures for the costs involved. In the words of Domínguez de la Fuente, the split in the business model was now evident: ‘the Miner is he who works; and the Refiner he who benefits’. This decision would have been a game-changer for the mining and refining community, and would have favoured the emergence of major refining centres, such as Regla. How much of the silver produced came from integrated mining operations, how much by tolling, and what was the evolution in refining capacity of haciendas throughout the colonial period and beyond, has not yet been established. The economics of refining silver did not follow the traditional lines that apply to the other commodity icons of the Early Modern period, such as cotton or sugar. Profit margins were sufficient if the operation was run with an industrial efficiency, but never reached the magnitudes that applied to the spice trade. There was no scope for European technology to increase the value of the peso minted in Mexico City, after silver had been refined using a patio process brought to maturity in the colonial periphery of the New World. Its producers had the enviable guarantee that they could convert to the world’s foremost specie every single unit of silver produced, if they so wished. Furthermore, I cannot find any other example of a commodity, other than gold, that from the sixteenth century was produced both in Europe and the New World, under an industrial context where the market price was fixed at basically the same level for all producers on both sides of the Atlantic during three hundred years. During this span of time, refiners had to fit their local variations in wages, fuel, reagents, infrastructure costs and government duties into a box of just one size, the valuation of silver that remained nearly unchanged for over twelve generations of refiners. The fact that this valuation was completely divorced from the costs of a refining process that was virtually unused in Europe, but that produced a major share of all the new silver that spanned the

the economies of refining silver

315

globe from the seventeenth century onwards, makes the situation more unique still. It would take the appearance of a huge wave of silver suddenly coming to the market from the refining mills of Nevada in the 1870s to shake the whole silver price structure to its roots. Hobsbawm has argued that in pre-industrial societies, the drive for innovation in manufacturing processes was constrained by the absence of a mass consumer market, and it made more sense to concentrate production on small volume but high profit items only for the elite that possessed the greater purchasing power.116 The high-volume production of silver during the Early Modern period did produce innovation in the pre-industrial Hispanic New World, driven solely by the drive for profit from the private sector charged with its production. Even more, the source of profit lay in factors common to modern mass consumer industries: controlling refining costs and increasing the volume of production, since the price of silver was fixed by factors totally exogenous to refining conditions in the Hispanic New World. It would be logical to argue that a lower wage scale in the New World compared to Europe would have been a significant help to keep competitive the variable costs of the patio process, of which around two thirds are total labour costs, or one half in the case of smelting. This fact should not overshadow recognition of the efficiency of the mining workforce in the New World. It turns out that time was just as important in Mexico as in Europe. To their efficiency must be added their attrition, and those of their communities, in human lives, occupational diseases and environmental damage from mining and refining, a hidden but very real contribution to maintaining a competitive refining cost of silver. By the nineteenth century European silver production had recovered, and now even offered better smelting economies for ores imported from the Mexican mines. In the tradition of the hare and the tortoise, smelting would outlive the patio process in New Spain. Looking back, it is tempting to conclude that the environmental history of the New World veered in a totally new direction as soon as some of the first miners of New Spain found they were left with more coin at the end of the day, if they could use cold mercury instead of a hot and fussy furnace. In this case, however, chemistry trumps economics. Lead would continue to be the main environmental hazard for silver refiners, a heavy metal whose compounds issued to the air in New Spain, in greater quantities than any volatile emissions of mercury. How much, when and where is the subject of the next and final chapter. 116

Hobsbawm, Industry and Empire, 18.

chapter 9

The Environmental Impact of Silver Refining: A Shift of Paradigm They suffer infinitely, the miserable smelters during this hour of immense fatigue, because the furnace is the mouth of Hell … the smoke … poisonous … this evil being so necessary, yet so terrible for those who toil at this work, so important for the Republic. francisco xavier de gamboa, Comentarios a las Ordenanzas de Minas (1761)

∵ The Base Line The silver coins that poured out of the Hispanic New World spread globally throughout the Early Modern period, a specie that became the world’s major trading currency well before the u.s. dollar took that role in the modern market. However, the global trust placed on the Spanish pieces of eight was not based on a promise to redeem a paper instrument. It depended on something much more tangible, a combination of a global standard of silver purity and weight, plus the technical ability to have sustained production and refining of silver ores at a level never seen before, during some three hundred years. The previous chapters have highlighted that the technical feat was not accidental, but the result of a conscious effort to apply best metallurgical practices adapted and developed to the realities in the field. This history of technology, or for that matter, global financial impact, cannot be studied in isolation from its collateral impact on the local environment, at every location associated with the mining and refining of silver ores. As Richards has pointed out, ‘as yet, the true environmental costs of silver have not been fully explored or acknowledged by scholars’.1 The first step in addressing this absence is to identify the main chemical compounds, mass balances and energy requirements characteristic of each refining process. The next step is to quantify the environmental

1 Richards, The Unending Frontier: An Environmental History of the Early Modern World, 366.

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_011

the environmental impact of silver refining

317

impact. In the previous chapters I have established the generic mass balances of raw materials, chemical by-products, waste and fuel consumption that apply to every kilogram of silver refined by either smelting or the patio process. In this final chapter I will apply these ratios to estimate the quantitative environmental impact of silver refining by Caja in New Spain, and on Mexico in the nineteenth century. The first challenge to any such analysis is that the whole framework within which silver production and mercury sales were registered was tinged with corruption: The Royal Treasury in New Spain was an efficient organization, carefully controlled and with very precise working guidelines, but at the same time it was a centre of corruption and traffic of influences. Then as now, these two aspects coexisted without interfering with each other, so that as long as the accounting figures matched precisely, other parallel practices were accepted so as to privilege certain miners in the distribution of mercury or in the evasion of taxes.2 Rampant corruption and contraband have been pointed out in the historiography as distorting the historical production data, so that up to an estimated two thirds of the total silver registered has been deemed to have been excluded from the official record of New Spain.3 The only way to circumvent such an elusive figure as contraband, is to accept the official values on silver and mercury simply as a base line, with real levels of silver refined and mercury consumed to have been higher. This implies that the absolute magnitude of the environmental impact vectors could have been substantially greater than any estimates to be calculated in this chapter. The relative magnitudes of these vectors to one another however remain a valid guide to their relative impact on the environment. The starting point in my estimate is the official production data for silver in New Spain and nineteenth century Mexico. Many historians have dedicated a

2 Pérez Luque and Tovar Rangel, Caja Real Guanajuato, 13. 3 See footnote 27 in Dennis O. Flynn and Arturo Giraldez, “Cycles of Silver: Global Economic Unity through the Mid-Eighteenth Century,” Journal of World History 13 (2002): 404; Bernd Hausberger, La Nueva España y sus metales preciosos (Frankfurt am Main: Vervuert, 1997), 41– 44; Peter J. Bakewell, “Registered Silver Production in the Potosí District 1550–1735,” Jahrbuch für Geschichte Lateinamerikas 12 (1975): 80; see footnote 3 in Saul Guerrero, “Chemistry as a Tool for Historical Research: Estimating the Contraband of Silver from Potosí and Oruro, 1576–1650,” Bulletin for the History of Chemistry 37, no. 2 (2012): 79.

318

chapter 9

major effort to establishing these data sets, based on official records. TePaske authored one of the most complete reviews of published data sets and provided a compilation of the total production of silver in New Spain by Caja (regional Treasury, not always a mining district), from the sixteenth century up to independence. Refining haciendas sent their silver to be taxed and stamped at the Cajas. Between 1521 and 1810 the Cajas of interest are, in descending order, Zacatecas (21% of total colonial silver production), México (18 %), Guanajuato (17%), Durango (12%), San Luis Potosí (9%), Guadalajara (8 %), Pachuca (5 %), Sombrerete (4%), Bolaños, Rosario and Zimapán (2 % each) and Chihuahua (< 1%).4 The Cajas however were not fiscal districts set in stone. Their relative ranking and the geographical reach of each Caja changed over time, as did the Reales de Minas which sent them the refined silver, the chemical nature of the ores being mined, and the refining process adopted. Their history can only be interpreted as a fluid mosaic in constant motion, not as a still life stagnant over time. In addition to this time variable, there was also a space variable, since the environmental impact calculated from the registered silver input to each Caja varied as to the locus of its effects. Thus, the environmental impact calculated for the Cajas of Guanajuato and Pachuca was concentrated over a much more reduced geographical area than in the case of the Cajas for Durango and Guadalajara. However, while the haciendas reporting to the Caja of Guanajuato were within or close to the city of Guanajuato, in the case of Pachuca only the Hacienda de Loreto was within the city limits, and all the other refining activity was well away from the main urban centres. To this geographical diversity must be added a technical diversity, since as I will show some Cajas will be associated more with either the patio process or smelting, but could also alternate over time between the two. The fluidity of this matrix will become evident in the following sections. The data on silver production in the nineteenth century are not as detailed as the data for the colonial period. Its records suffered on par with the political situation in Mexico, and the most detailed statistics are found only for the latter part of the century. I have set out in Table xxiii the main sources in the historiography on the production of silver in this period. The figures calculated by Soetbeer up to 1875 are the common source quoted for many published estimates.5 For 1876 up to 1899 I include the data from González Reyna and

4 TePaske and Brown, Gold and Silver, 82–83, 115–116. 5 Adolf Soetbeer, Edelmetall-produktion und werthverhältniss zwischen gold und silber seit der entdeckung Amerika’s bis zur gegenwart (Gotha J. Perthes, 1879), 60. Other works that rely on Soetbeer’s data are Merrill, Summarized Data of Silver Production; Jenaro González Reyna,

319

the environmental impact of silver refining

from Flores Clair, Velasco Avila and Ramírez Bautista.6 I have not found an equivalent breakdown of data by region as is available for New Spain, so only a gross national approximation can be attempted based on the total silver production from 1820 to 1899. table xxiii

Silver production in Mexico, nineteenth century. Sources in footnotes 5 and 6. The data after 1875 corresponds to fiscal years beginning in the year indicated.

Period

Silver production (kg) Soetbeer

1801–1810 1811–1820 1821–1830 1831–1840 1841–1850 1851–1855 1856–1860 1861–1865 1866–1870 1871–1875 1821–1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887

González Reyna

Flores Clair et al.

5,538,000 3,120,000 2,648,400 3,309,900 4,203,100 2,330,500 2,239,000 2,365,000 2,604,500 3,009,000 22,709,400 546,410 588,518 610,683 643,907 694,000 714,573 718,658 748,681 793,377 824,079 876,724 939,779

570,000 635,572 641,502 704,783 756,505 743,372 756,345 810,448 849,579 873,996 959,215 1,005,080

Minería y Riqueza de México, Monografías Industriales (México: Banco de México, 1944); Flores Clair, Velasco Avila, and Ramírez Bautista, Estadísticas mineras, ii. 6 González Reyna, Minería y Riqueza de México, 22–23; Flores Clair, Velasco Avila, and Ramírez Bautista, Estadísticas mineras, ii, 17–18.

320 table xxiii

chapter 9 Silver production in Mexico, nineteenth century (cont.)

Period

Silver production (kg) Soetbeer

1888 1889 1890 1891–1899 1876–1899 1821–1899

González Reyna

Flores Clair et al.

986,382 983,799 990,237 13,250,485 24,910,292

1,051,995 998,742 1,068,088 13,796,861 26,222,083 48,931,483

No primary source has been reported that breaks down the total production of silver, from the sixteenth to the nineteenth century on a yearly basis, according to whether smelting or any refining process based on mercury was used. Only partial breakdowns per Caja are for now available, starting from the late seventeenth century onwards. Since this is the critical information required to apply the mass balance ratios that quantify the environmental impact of silver refining, I have had no other recourse but to extract from the available data an estimate of this breakdown, using assumptions based mainly on the chemistry of the processes. My primary source of information is the raw primary tax and mercury sale data of each Caja as has been gathered and transcribed by Tepaske, Klein and other collaborators in Mexico and Spain.7 Their data set (to which I will refer henceforth as the tk set) includes the tax and señoreage (duty on coinage). Both are reported either for a total amount of silver registered at each Caja, or correspond to plata de fuego (silver refined by smelting) or plata de azogue (silver refined using mercury, mostly by the patio process).8 When these tax revenues are reported under ‘1 % y Diezmos de Plata de Azogue’ and ‘1% y Diezmos de Plata de Fuego’, the fraction of silver refined

7 Prof. Herbert S. Klein of Columbia University very kindly provided me with his Excel files containing the raw information collated from primary sources on tax revenues and mercury revenues, during the colonial period for the provincial Cajas. Any error in the sorting and calculations based on the raw data transcribed in his files is solely this author’s responsibility. 8 To avoid overloading with unnecessary precision an exercise that is solely a first approximation, ‘patio process’ will also encompass the early period when canoas were used to carry out the reaction (Chapter 5). I will point out where pertinent, instances where either the cazo process or the barrel process were later involved.

the environmental impact of silver refining

321

by each type of process can be calculated directly from the peso amounts registered under each heading. I plot these fractions for each Caja under the heading ‘duties’, in the following section. In addition, in certain periods a tax on the minting of coins (señoreage or señoreaje) appears, as ‘1 % Diezmos Señoreage Plata de Azogue’ or ‘1% Diezmos Señoreage Plata de Fuego’.9 On the assumption that the silver allotted to the minting of coins reflects the original split of production between the two refining processes, I have also plotted these data points under ‘duties & mintage’. The smooth merging of the data between tax levied on silver and duty on coinage confirms my assumption that the production split between refining processes is reflected even when silver was converted to specie. The detailed mathematical projections for each Caja, based on the tk set, are set out in Appendix c.

An Estimate of the Breakdown of Silver Production by Refining Process by Caja Caja of Zacatecas The profile of Figure 91 indicates a gradual transition in the main refining process, from smelting to the patio process, from the second half of the seventeenth century to end of the eighteenth century.10 The reason is primarily a change in the chemical nature of the ore from the mines reporting to this Caja over time. As of the 1680s, the silver from the lead rich ores of Sombrerete were no longer registered at the Zacatecas Caja. In Figure 92, I plot the estimated profile of silver production by refining process, based on projections set out in Table c-i (Appendix c), and the effect of this reassignment after 1680 of ores that could only be smelted is evident. The patio process is seen to peak in the 1730s (exact reason unknown, an increase in the extraction of silver sulphide ores, or a better supply of mercury), suffers a significant downturn and then responds to the price decrease of mercury in the 1770s. The nature of the ore prevailed: ores that could be smelted continued to be smelted, and a lower mercury price made it profitable to process more of the ore that could be refined 9 10

‘señoreage: a tax on the minting of coins by private individuals’ in Pérez Luque and Tovar Rangel, Caja Real Guanajuato, 61. I have also included the data reported by Lacueva, who has published annual data sets of silver production in Zacatecas, since it covers a period prior to 1700 when the time periods for the tk set do not always correspond exactly to twelve month intervals. Lacueva Muñoz, La plata del Rey, 391. Bakewell provides a figure of 30 % by smelting in Zacatecas in the 1720s and 1760s. Bakewell, “Colonial Mining,” 145.

322

chapter 9

figure 91 The fraction of plata de azogue registered at the Caja of Zacatecas in the period 1670 to 1820. Prior to 1700 the time intervals are exactly one year only for the Lacueva data. Fractions calculated based on the raw data from tk set and footnote 10.

figure 92 Estimate of silver registered at the Caja of Zacatecas according to refining process. Data from Appendix c, Table c-i.

by the patio process.11 On average over this period, the patio process is estimated to have accounted for two thirds and smelting for one third of total silver produced. During the period the ores from Sombrerete were included in this Caja, the fraction of plata de fuego registered in the Caja represented over half the total of silver produced.

11

Lacueva has analysed the role played by smelting in Zacatecas irrespective of the supply of mercury. See Lacueva Muñoz, La plata del Rey, 187–210.

the environmental impact of silver refining

323

figure 93 The fraction of plata de azogue registered at the Caja of Guanajuato in the period 1679 to 1816. Prior to 1720 the time intervals of the raw data in the tk set have been approximated to the calendar years.

Caja of Guanajuato Prior to 1650 refiners of the Guanajuato area registered their silver in the Caja of México (Ciudad de México). After that date the level of silver ore extracted in the mines around the city of Guanajuato made it necessary to set up a separate Caja. Most the silver registered at the Caja of Guanajuato came from refining haciendas close to Guanajuato.12 Figure 93 shows how the impact of lower mercury pricing as of the 1780s, is reflected in the steady increase of the fraction of plata de azogue that already dominated production of silver in this area, starting from a market share of approximately 65 %.13 Figure 94 shows how the baseline of smelting had been reached before the first of the decreases in mercury pricing. Guanajuato is a Caja where the patio process always dominated, with an estimated average of 71 %, and the profile shows a strong correlation with the decreases in the price of mercury. However, from mid seventeenth to mid eighteenth century the two refining processes were more evenly balanced. Caja of México The Caja at México was the first established by Spain to channel the silver product refined in New Spain. As production grew, other regional Cajas were

12 13

TePaske and Brown, Gold and Silver, 95. Bakewell reported 35% for smelting in the 1730s and 27 % in the 1770s. Bakewell, “Colonial Mining,” 145.

324

chapter 9

figure 94 Estimate of silver registered at the Caja of Guanajuato according to refining process. Data from Appendix c, Table c-ii.

figure 95 The fraction of plata de azogue registered at the Caja of México in the period 1786 to 1816. The raw data are from the tk set.

created, which means that the tax and revenue records of this Caja not only reflect the refining activity in the vicinity of Ciudad de México (58 Reales de Minas by mid 1760s, including the major production centre at Taxco), but at different times have also included the silver and mercury that were later reported by the new Cajas of Guanajuato, San Luis Potosí, Zimapán and Pachuca.14 This creates major problems for the analysis of the tk set for this Caja. The distinction between plata de azogue and plata de fuego only appears very late in the records, as can be seen in the very limited results reported in Figure 95. It is also impossible to work with the figures on mercury sales, since they aggregate without distinction mercury used for various other Cajas. Table c-iii

14

TePaske and Brown, Gold and Silver, 84–87, 115–116.

the environmental impact of silver refining

325

figure 96 The fraction of plata de azogue registered at the Caja of Durango in the period 1696 to 1813. Between 1737 and 1765 no distinction was made between refining processes in the tax register. Prior to 1713 I have approximated the irregular time series of the raw data in the tk set to their nearest calendar years.

(Appendix c) is based on data from secondary sources, and my estimates are biased in favour of the patio process. Caja of Durango Established in 1599, the Caja of Durango reflects the production of mines (42 Reales de Minas in 1761–1767) around the capital of the colonial province of Nueva Vizcaya, some two hundred miles to the northwest of Zacatecas. Among the contributors were Parral and Chihuahua, the latter becoming a Caja in its own right as of 1785.15 Figure 96 shows how the fraction of plata de azogue increased from a very low level until becoming the major type of silver produced, over the period 1679 to 1816. The registries of tax for the two decades between 1740 and 1760 do not discriminate revenues according to refining process. but the data on mercury sales in the tk set for this hidden period confirm an increasing use of the patio process. Figure 97 traces the two estimated production profiles. A marked decrease in smelting is observed at the same time that refining by the patio process increases in the second half of the eighteenth century. There is an obvious exhaustion of lead-rich ores towards the end of the eighteenth century, and increases in supply and the decrease in the price of mercury favoured an increase in the patio process to refine the available ores. On average smelting accounted for an estimated 61% of the total silver registered at the Caja of Durango. 15

Ibid., 91–93.

326

chapter 9

figure 97 Estimate of silver registered at the Caja of Durango according to refining process. Data from Appendix c, Table c-iv.

figure 98 The fraction of plata de azogue registered at the Caja of San Luis Potosí in the period 1713 to 1806. Source of raw data is the tk set.

Caja of San Luis Potosí The split between plata de fuego and plata de azogue reflected in the tax records of the Caja of San Luis Potosí is the best example why these curves must first be interpreted through the nature of the ore being processed. At a first reading of Figure 98, the change in refining process coincides so well with the decrease in the price of mercury in the 1770s that a causal link seems the explanation. However, it is the change of the type of ore, and not the price of mercury, that determines the profile in Figure 98. The initial period corresponds predominantly to the smelting of lead rich ores, first found in the

the environmental impact of silver refining

327

figure 99 Estimate of silver registered at the Caja of San Luis Potosí according to refining process. Data from Appendix c, Table c-v.

mines of the Cerro San Pedro on the hills that surround the town, and then from other locations such as Charcas and Guadalcazar.16 In the 1770s silver deposits were discovered at Catorce, 250 km to the north of the town of San Luis Potosí.17 The ore, composed of native silver and silver halides, was refined using the cazo process, complemented by a secondary extraction using the patio process.18 It is the sudden surge in production from Catorce that explains the predominance of plata de azogue in the production of silver registered at the Caja of San Luis Potosí as of the 1780s, observed in Figure 98. While the decrease in the price of mercury would have benefitted these refiners, the lower consumption of mercury in the cazo process and the scale of the new deposits, would have made the surge in silver production inevitable, irrespective of the new prices of mercury. 16

17 18

During this period Bakewell reports 86% of silver produced by smelting in the 1730s dropping to 54% by the 1760s. Bakewell, “Colonial Mining,” 145. Smelting is reported as accounting for 92% of production in 1718, and then decreasing to 48.6 % in 1761–1767 and then virtually disappearing at 1.6% by 1785–1789 and later years, in Inés Herrera Canales, “El método de refinación con azogue en la minería potosina colonial: del fuego al cazo” in La plata en Iberoamérica: Siglos xvi al xix, ed. Jesús Paniagua Pérez and Nuria Salazar Simarro (León: Universidad de León, 2008), 68. For a general history on the discovery and development of the mining activity at Catorce, see Montejano y Aguiñaga, Real de Minas Catorce. The average mercury to silver weight ratio calculated from the tk set for San Luis Potosí averages 1.7 between 1710 and 1780, and 0.9 from 1781 to 1806 (Appendix c, Table c-v). The use of the cazo process cut average mercury consumption by half, though this decrease is very dependent on the nature of the ore.

328

chapter 9

figure 100

The fraction of plata de azogue registered at the Caja of Guadalajara in the period 1691 to 1804. Prior to 1699 I have approximated the irregular time series of the raw data in the tk set to their nearest calendar years.

The plot in Figure 99 shows two clear refining periods. The first, from the early 1600s to 1780, where an estimated average of 82 % of all silver was produced by smelting. The second, as of 1780, produced 96 % of the silver by the cazo and patio process, due to the entry of the mines of Catorce. The baseline of smelting remained fairly constant over the whole period, only decreasing substantially as of the 1790s. The environmental impact of silver refining for this Caja is also divided along geographical lines. Smelting would impact the region around and within the town of San Luis Potosí for some 160 years. In contrast, the more isolated area around the mines of Catorce would bear the completely different brunt of the environmental impact of both cazo and patio processes, and was spared the consequences of smelting. Caja of Guadalajara The silver registered at the Caja of Guadalajara came from multiple small and medium mines, with 46 Reales de Minas operating in the mid-1760s.19 Figure 100 shows that the patio process was the main refining process used, and the increase of its share from approximately 60 %, until it became the predominant route to silver, had begun even before the decrease in price of mercury, so that an increase in available supply of mercury made have had a role.20

19 20

TePaske and Brown, Gold and Silver, 90–91. Bakewell reported 26% smelting in the 1730s dropping to 8 % by the 1770s. Bakewell, “Colonial Mining,” 145.

the environmental impact of silver refining

329

figure 101

Estimate of silver registered at the Caja of Guadalajara according to refining process. Data from Appendix c, Table c-vi.

figure 102

The fraction of plata de azogue registered at the Caja of Pachuca in the period 1667 to 1820. Prior to 1706 I have approximated the irregular time series of the raw data in the tk set to their nearest calendar years.

Figure 101 shows how the patio process starts to pull away in the 1740s, while after the 1760s the balance between the processes had shifted from a close pairing at the beginning of the century to an evident dominance of the patio process by the end. The Caja of Guadalajara registered an estimated average of 73 % of plata de azogue and 27% of plata de fuego. Caja of Pachuca From the start of mining in the region of Pachuca in 1552, until 1667, all its silver production was registered at the nearby Caja of México. Its registry corresponds to refining activities concentrated around two main sites, Pachuca and Real

330

chapter 9

figure 103

Estimate of silver registered at the Caja of Pachuca according to refining process. Data from Appendix c, Table c-vii.

del Monte.21 Figure 102 shows the evolution in the fraction of plata de azogue over the period 1679 to 1816. At first sight it indicates an unexpected increase in smelting after the price of mercury had decreased.22 However, as the fraction of plata de azogue decreases towards the end of the eighteenth century, so did the level of silver production decrease as well (Figure 103). It was therefore not a case of a sudden increase in plata de fuego, but of a significant decrease in the production of plata de azogue, which impacted the whole output registered to this Caja. The pricing of mercury could not compensate the drop in ore available for refining by the patio process. Despite of the impression given by Figure 102, the silver registered at the Caja of Pachuca was refined mainly in haciendas de patio, which supplied an estimated 73% of the total silver produced. Smelting would contribute with 27% of the production. Caja of Sombrerete The Caja at Sombrerete was established in 1683, and the silver from its ores had been registered until then at the Caja of Zacatecas. The bell-shaped profile of the plot in Figure 104, of the fraction of plata de azogue, shows an impressive alternation between smelting and the patio process as the main refining route. 21 22

TePaske and Brown, Gold and Silver, 96–98. Bakewell reports 27% by of silver produced by smelting in the 1720s, decreasing slightly to 23% by the 1760s. Bakewell, “Colonial Mining,” 145. The migration of lead rich ores to the new Caja of Zimapán as of the 1730s (see below) may have contributed to the low spike in the fraction of plata de azogue observed around the 1760s.

the environmental impact of silver refining

figure 104

331

The fraction of plata de azogue registered at the Caja of Sombrerete in the period 1680 to 1820. Prior to 1760 I have approximated the irregular time series of the raw data in the tk set to their nearest calendar years.

It is reported that from the 1760s more than 90% of the registered silver came from local mines of Sombrerete, and most from ‘the rich vein of El Pabellón … [which had] a high lead content’.23 A high lead content in ores rules out the patio process, so it explains the resurgence of smelting after the 1760s irrespective of the price of mercury. In Figure 105, the peak in the fraction of plata de azogue may correspond to an increase in the volume of mercury supply, tied to more sulphide ores becoming available, even before the price of mercury had been decreased. The level of smelting on either side is fairly constant, only rising steeply at the end of the period. As in other examples, the nature of the ore was more important than the price of mercury. The Caja of Sombrerete was in smelting territory. An estimated 68% of the total silver registered at the Caja would have been produced by smelting haciendas, 32% from the patio process.

23

TePaske and Brown, Gold and Silver, 98–99. Bakewell reported 68 % for silver by smelting in the 1720s, decreasing to 33% by the 1760s, but he did not extend his data to the end of the century and the new increase in smelting lies outside his data. Bakewell, “Colonial Mining,” 145. Lacueva indicates that at the end of the seventeenth century smelting was used in the new mines of Sombrerete, where in the period from 1688 to 1699 up to 88 % of silver would be produced by smelting of ores. Lacueva Muñoz, La plata del Rey, 401.

332

chapter 9

figure 105

Estimate of silver registered at the Caja of Sombrerete according to refining process. Data from Appendix c, Table c-viii.

figure 106

The fraction of plata de azogue registered at the Caja of Bolaños in the period 1753 to 1804. Raw data from tk set.

Caja of Bolaños Major mining in the region around Bolaños started in 1747, and the Caja was established some six years later. Records are affected by a fifteen-year tax exemption on the diezmo granted to one of the principal miners of this region as of 1789.24 The data from the tk sets as plotted in Figure 106 show that Bolaños was dominated by the patio process, even before the price decrease of mercury in the 1760s, with over 94% of all the silver registered (see Appendix c, Table cix).

24

TePaske and Brown, Gold and Silver, 101.

the environmental impact of silver refining

figure 107

333

The fraction of plata de azogue registered at the Caja of Rosario in the period 1770 to 1813. Raw data from tk set.

Caja of Rosario The Caja of Rosario was set up in 1770, moved to Alamos in 1783 and then to Cosalá around 1807.25 The records of the tk set are simply identified as Rosario. Figure 107 shows smelting maintaining a relatively constant fraction under one third, even after the decrease in mercury pricing. Of the total silver registered in this Caja, an estimated 71% corresponds to the patio process and 29 % to smelting (see Appendix c, Table c-x). Caja of Zimapán The silver from Zimapán was initially registered as of the sixteenth century, first in the Caja of México and then after 1667 in the Caja of Pachuca, but in 1729 it was awarded its own Caja. By the 1760s, the mines of Zimapán contributed 86 % of the total argentiferous lead ores.26 It is the presence of lead that determines the profile seen in Figure 108, to all practical purposes indicating a total absence of plata de azogue in the silver registered at Zimapán. Caja of Chihuahua According to TePaske this was the ‘last mining Caja created in New Spain’.27 The observed split between the two refining processes is impervious to the decrease in mercury prices, most probably due to the lead content of the ores. Santa Eulalia and Santa Bárbara, the principal mines feeding the refining 25 26

27

Ibid., 103. Ibid., 100. There were more than 100 smelting furnaces in Zimapán in 1795, ‘with no trace of refining by amalgamation’, according to Sonneschmidt and de Fagoaga, Tratado de la amalgamación de Nueva España, 62. TePaske and Brown, Gold and Silver, 104.

334

chapter 9

figure 108

The fraction of plata de azogue registered at the Caja of Zimapán in the period 1729 to 1806. Raw data from tk set.

figure 109

The fraction of plata de azogue registered at the Caja of Chihuahua in the period 1788 to 1813. Raw data from tk set.

haciendas that reported to the Caja in Chihuahua, are linked to the few known argentiferous lead deposits of New Spain.28 It was only in the nineteenth century that the fraction of plata de azogue increased (Figure 109). Overall smelting provides 60% of the silver registered at this Caja, and the patio process the remaining 40% (Appendix c, Table c-xii).

28

Claude T. Rice, “The Silver-Lead Mines of Santa Barbara, Mexico,” The Engineering and Mining Journal lxxxvi, no. 5 (1908): 208–209.

335

the environmental impact of silver refining table xxiv

Silver production by Caja, according to refining processes, over the whole colonial period in New Spain. Adapted from footnote 29, courtesy Taylor & Francis.

Caja

Plata de azogue Plata de fuego Patio process Smelting t t

Zacatecas Guanajuato México Durango San Luis Potosí Guadalajara Pachuca Sombrerete Bolaños Rosario Zimapán Chihuahua total New Spain

6,522 5,895 6,346 2,382 2,249 2,656 1,781 536 1,112 777 0 97 30,353

3,262 2,322 2,357 3,677 1,697 1,003 668 1,115 67 314 799 149 17,429

67 % 72 % 73 % 39 % 57 % 73 % 73 % 32 % 94 % 71 % 0% 39 % 64 %

33 % 28 % 27 % 61 % 43 % 27 % 27 % 68 % 6% 29 % 100 % 61 % 36 %

Aggregate Totals for New Spain Based on my estimated breakdown by Caja, smelting predominated historically in San Luis Potosí (except for Catorce), Durango, Sombrerete, Zimapán, and Chihuahua. The patio process predominated in Zacatecas, Guanajuato, México, Pachuca, Guadalajara, Bolaños and the cazo process for the silver from the mines of Catorce. On average, the patio process accounted for approximately 64 % of the silver produced in New Spain, and smelting 36 % (Table xxiv).29 My aggregate projection coincides with the statement in 1777 by the Administrador General de Minas, who reported that 40% of all production was by smelting.30 It also shows a much more balanced distribution between the patio process and smelting output than what at times has been reported in the historiography.31 29 30

31

Guerrero, “The History of Silver Refining in New Spain, 16c to 19c: back to the basics,” 8. As stated in Humboldt, Essai politique, Tome iv, 51. Up to the eighteenth century at least half of the silver ores mined in Honduras were smelted, according to Linda A. Newson, “Silver Mining in Colonial Honduras,” Revista de Historia de América, no. 97 (1984): 52. An extreme case is the claim that 95% of all silver was produced by refining with mercury. Castillo Martos, “Alquimia en la metalurgia de plata,” xxiv. Humboldt estimated that the

336

chapter 9

figure 110

Approximate areas in black correspond to Cajas where on average, over the colonial period, silver was refined mainly by smelting, those in grey by the patio process. Note the enclave of the cazo process at Catorce (dashed grey line) reporting to the Caja of San Luis Potosí. adapted from the map by hausberger in footnote 32

A rough approximation of what the average regional distribution of the two refining processes would have been over the whole colonial period is given in Figure 110.32 When the data are plotted by decade, three stages can be discerned in Figure 111. The initial period up to the 1640s is the one subject to the greatest uncertainty of all. Data are at times scarce, and the divergence observed between the two curves is a function of how efficient one deems the patio process, or its pre-

32

overall ratio of silver refined by the patio process to smelted silver was 3.5 to 1, possibly influenced by the production after 1777 of Catorce. Humboldt, Essai politique, Tome iv, 49– 50. Santiago Ramírez (1884), estimated the split was 75% patio process and 25 % smelting, as quoted in Mendizábal, La minería mexicana, 72–73. Map of the distribution of mining areas reporting to the Cajas in the 1760s, in Hausberger, Metales preciosos, 64.

the environmental impact of silver refining

figure 111

337

Registry of silver by refining process, as projected for New Spain adapted from footnote 29, courtesy taylor & francis

decessor using canoas, to have been in the first century of its implementation. It is a period marked by a very lenient policy on the payment of mercury supplied to the refiners, together with a source of tailings with a sunk cost of zero that could be refined using mercury. Both factors can explain the peak in the fraction of silver refined by the patio process, since once these factors recede, the production from both refining processes becomes more balanced. From mid seventeenth to mid eighteenth century, the patio process and smelting share equally the production of silver in New Spain. It reflects the more natural balance between the two refining processes based on the chemistry of the ores available, with any bias on the part of the Crown towards promoting the use of mercury somewhat muted. During this period, ores with a silver content below 0.08% continue to build up over decades in the mounds of tailings. Starting in the 1740s, according to Dobado and Marrero, the supply of mercury from Almadén increases, which would allow refiners to process a larger amount of suitable ore by the patio process.33 Until mid-eighteenth century, the production from both processes follow parallel lines in their positive slope, with the patio process showing a slight advantage in total output. It is in the 1750s that the patio process now begins to pull ahead in its contribution to total silver output. The divergence of curves becomes more significant once the step decreases in the price of mercury are implemented in 1767 and 1776. The steep rise in silver production using mercury in the latter part of the eighteenth century brings to mind the same drastic increase seen in Potosí in the 1570s. Among the causes were a sudden increase in silver sulphide 33

Dobado and Marrero, “The Role of the Spanish Imperial State in the Mining-led Growth of Bourbon Mexico’s Economy,” 867.

338

chapter 9

figure 112

The fraction of total silver produced in Mexico, 1876 to 1892, according to refining process. Raw data from footnote 34.

ore extraction, the contribution from the cazo process due to the bonanza at Catorce (which contributed 20% of the major peak in production observed in Figure 111), and not least the refining of tailings accumulated over centuries. These could now be a source of profits, due to the low price of mercury and because their sunk cost of extraction was effectively nil, cutting their refining cost by the patio process approximately by 60 %. Under these conditions it would have been impossible for smelting to compete for the refining of these ores, unless the Crown had offered major incentives on lead and charcoal.

Aggregate Totals for Mexico, 1820 to 1900 In this period that covers republican Mexico, unfortunately the reported data that quantify silver production according to refining process is limited to the fiscal years 1877 to 1896. Figure 112 plots the contribution of smelting and the patio process to the total production of silver, in a period when the barrel process (4%) and leaching (lixiviación) (6%) were also being applied.34 For lack of more detailed data at present, I have assumed as a working figure that on average during the nineteenth century in Mexico, the patio process accounted for 80% of silver refined, and smelting 20%. 34

Flores Clair, Velasco Avila, and Ramírez Bautista, Estadísticas mineras, ii, 161–162. The yearly totals are between 20% to 40% less than the yearly data in Table xxiii. Due to the years in question, I interpret the leaching process mentioned in the statistics as the hyposulphite process, such as described in Collins, Metallurgy of Lead & Silver, 186–251.

the environmental impact of silver refining

339

Environmental Impact Vectors, Sixteenth to Nineteenth Century The estimated breakdown in production by refining process is the required starting point to project the base-line magnitude of the main environmental impact vectors for each Caja in New Spain, and for the total territory in republican Mexico during the nineteenth century. Appendix c has the details of the estimations used to arrive at the projected chemical consumption of mercury as calomel, and its physical loss as liquid mercury and as emissions to the air. The other magnitudes correspond to salt and copper sulphate consumed and washed away from the treated ore, of solid waste washed away in waterways and woodland required for firewood. For smelting, the main vectors correspond to lead lost as lead fume, woodland required to produce charcoal and solid waste as slag. I have chosen to report for lead lost to the air in lead fume, a range of 5 to 10kg per kg of smelted silver. The ratios used to project these magnitudes from the amount of silver produced in New Spain, according to each refining process, are set out in Table xxv, and correspond to values derived in the previous chapters.35 The values for woodland consumed are overestimated, since I have not factored in the natural cycle of regeneration, which could reduce the projections by at least 50%. By applying each of these ratios to the estimates of silver refined by each process for each Caja, an order of magnitude is generated that is important as an indication of the relative impact of each vector by region (Table xxvi). The total environmental impact due to smelting could be greater, in aggregate, in regions nominally dominated by the patio process, than in regions where smelting was the preferred refining option. This is clear in the series shown in Figure 113 a) and b), where I plot the magnitude of the environmental impact vectors corresponding to smelting in New Spain. While Durango remains the silver catchment area that would have been most affected by the environmental impact of smelting, it is closely followed by areas where the patio process predominated, such as Zacatecas, Mexico and Guanajuato. In the case of the major impact vectors associated with the patio process, they reflect more closely the ranking of Cajas where it dominated, as illustrated in Figures 113 c), d) and e). For nineteenth century Mexico, it is not possible yet to reach a similar breakdown of estimate by refining region. To reach the gross magnitudes reported in

35

Laur and Duport estimated that smelting produced 10 % of silver in nineteenth century Mexico, though Laur cautioned that ‘it is however not possible to establish with certitude their relative importance’. Laur, “De la metallurgie de l’ argent au Mexique,” 106; Duport, Métaux précieux au Mexique, 86. Guerrero, “The History of Silver Refining in New Spain, 16c to 19c: back to the basics,” 17–18.

Smelting

Patio process

refining process

85

Mercury, in soil and water

14 n/a

1

Volatilised mercury

% of total mercury consumed

Mercury, as calomel

29.3

Salt

2.6

Copper sulphate

Charcoal

49

609 1,000

7

kg per kg silver refined

Mineral waste

5 to 10

n/a

Lead in lead fumes

0.4

0.003

ha per kg silver refined, no regeneration

Woodland

table xxv Weight ratios of consumables to silver refined by the patio process (ores with 0.2 % silver content) and smelting (ores with 1.9% silver content), for New Spain. Adapted from footnote 35, courtesy of Taylor & Francis.

340 chapter 9

341

the environmental impact of silver refining table xxvi

Caja

Zacatecas Guanajuato México Durango San Luis Potosí Guadalajara Pachuca Sombrerete Bolaños Rosario Zimapán Chihuahua total New Spain

figure 113a

Estimated magnitudes of main environmental impact vectors by Caja in New Spain. All units in thousand tons, except for woodland depletion expressed in thousand ha. Reproduced from footnote 35, courtesy of Taylor & Francis.

Total Total Total Total Total Total Total mercury in liquid salt copper mineral lead in woodland calomel mercury sulphate waste lead fume 11 9 10 4 2 5 3 1 2 1 0 0.2 47

2 1 2 1 0.3 1 1 0.4 1 0.4 0 0.2 10

200 170 190 70 70 80 50 16 30 25 0 3 904

17 15 16 6 0 7 5 1 3 2 0 0 72

4,000 3,600 3,900 1,500 1,400 1,600 1,100 330 700 500 0 60 18,690

15 to 30 12 to 24 12 to 24 18 to 36 8 to 16 5 to 10 3 to 6 6 to 12 0.5 to 1 2 to 4 4 to 8 1 to 2 90 to 180

Ranking of Cajas by the magnitude of the environmental impact vector corresponding to loss of lead in lead fume. Data from Table xxvi.

1,320 917 969 1,507 707 408 275 452 33 132 320 60 7,100

342

chapter 9

figure 113b

Ranking of Cajas by the magnitude of the environmental impact vector corresponding to woodland consumed. Data from Table xxvi.

figure 113c

Ranking of Cajas by the magnitude of the environmental impact vector corresponding to mineral waste voided into waterways. Data from Table xxvi.

the environmental impact of silver refining

figure 113d

Ranking of Cajas by the magnitude of the environmental impact vector corresponding to salt voided into waterways. Data from Table xxvi.

figure 113e

Ranking of Cajas by the magnitude of the environmental impact vector corresponding to mercury in calomel voided into waterways. Data from Table xxvi.

343

344

chapter 9

Table xxvii, I have assumed an average mercury to silver ratio of 1.8 and for charcoal, a more efficient ratio of 300 to one. All the other factors remain as in Table xxvi. The projected breakdown of the magnitudes of the environmental impact vectors can be adjusted in the future as better figures come to light on a regional basis. I have set out side by side my projections for nineteenth century Mexico and New Spain in Table xxvii, to highlight the fact that in the case of the former, approximately two-thirds of this impact took place in just fifty years, during the second half of the nineteenth century. Even though each Caja also had periods of very intense activity, so that the environmental impact was also concentrated in periods shorter than the whole colonial span, the contrast in the intensity of the environmental impact between New Spain and republican Mexico remains striking. It is in this context that the picture that emerges in Table xxvii for New Spain and Mexico should be considered. In both cases, it confirms that smelting caused a greater environmental impact than the use of mercury in the patio process. In the air it is the lead content in the lead fumes that dominates, up to two orders of magnitude greater any expected loss of volatile mercury. The impact on woodland is also greatest for smelting, with a factor of 40 to 80 times greater than for the patio process. Since smelting produced at the most just over half the silver produced by the patio process, this is a significant difference in the emission of heavy metals to the air, and in the rate of woodland depletion for both processes. In terms of solid waste voided into waterways, the patio process dominates with nearly 19 million t during approximately 250 years for New Spain, and around 22 million t in just 80 years in Mexico.36 This highlights the more concentrated environmental stress from silver refining during the latter period. Since the order of magnitude of silver refined in nineteenth century Mexico is the same as for the whole colonial period, it is not surprising that the magnitudes of the rest of the environmental impact vectors from the patio process (calomel, salt, copper sulphate) are quite similar for both New Spain and Mexico.

36

These magnitudes are dwarfed by the environmental impact of hydraulic mining during the California gold rush in the latter part of the 19c. It has been reported that just at the North Bloomfield mine, in the 1880s, up to 40 million cubic yards of material had been gouged out of the earth. The amount of soil and mineral debris that clogged up the lower reaches of the Yuba river has been estimated between 360 to 600 million cubic yards (at least 360 to 600 million tons), in a period of around 60 years. Deposited debris from hydraulic mining in the Sierra Nevada in the late nineteenth century has been estimated as high as 1,555 million cubic yards. See Randall Rohe, “Man and the Land: Mining’s Impact in the Far West,” Arizona and the West 28, no. 4 (1986): 316, 328, 330.

New Spain Mexico

table xxvii

31 36

Total silver patio process

47 54

Total mercury in calomel 10 9

Total liquid mercury 1 1

Total volatile mercury 900 1,000

Total salt

70 90

Total copper sulphate 90 100

Total woodland

19,000 22,000

Total mineral waste 20 11

Total silver smelted 90 to 180 50 to 100

Total lead in lead fume 1,000 700

Total slag waste

7,000 4,300

Total woodland

Comparative total magnitudes of main environmental impact vectors for New Spain and nineteenth century Mexico. All units are thousand tons, except for woodland depletion, in thousand ha.

the environmental impact of silver refining

345

346

chapter 9

To transcend the gross estimates in Table xxvii will require a deeper analysis for every region reporting its silver to a Caja, and their evolution from the sixteenth to the nineteenth century. It will be necessary to map, as a function of time, waterways and water basins, the location of historical landfills from refining waste dumped in waterways, wind roses and geographical contours, relative distribution of refining haciendas by location, size and refining process, population centres, agricultural centres and cattle rearing, woodlands used to source fuel, regeneration rates for the local woodland, and transit routes to and within each region. In addition, a detailed knowledge on the changes in the architectural details of the refining haciendas of each region over time is required, to be able to model the deposition of heavy metals and their byproducts around each establishment.37 The Environmental Impact on New Spain and 19c Mexico: The Modern Legacy To define the environmental legacy of historic silver refining in Mexico will require data on the traces of lead and lead compounds, calomel and mercury in the soil that still remain at the present time. The Cajas whose refiners would have exhibited the highest peaks in yearly averages of lead emissions are Durango, Zacatecas, Guanajuato, Sombrerete and San Luis Potosí. For nineteenthcentury Mexico it is not possible with the data available to calculate regional peaks. As I already mentioned in Chapter 3, modern studies of children living close to historical mining sites in Mexico show very high levels of lead in their blood, a finding that highlights the importance of reviewing the historical role of smelting and its environmental impact in New Spain and Mexico.38 Mercury can still be found in the subsoil of the patio reactors in Mexico, as was shown in Figure 26. The lifecycle of calomel in soil and waterways needs to be studied, to

37

38

In Chapter 2, I referred to the regional studies identifying primary economic functions carried out by Salazar González in the region around the town of San Luis Potosí. Another relevant study has centred on determining the role of ‘mining as a creator of economic spaces due to its great organising power’. The authors draw up maps of geographical networks that grew around the mining and refining activities in Pachuca and Real del Monte in the nineteenth century, as set out in Saavedra Silva and Sánchez Salazar, “Espacio Pachuca-Real del Monte,” 82–101. Each cluster of refining haciendas creates by their sole presence a source of an economic force field, attracting by its needs the other economic activities that sustain or are generated by refining. Overlaid on this economic ‘gravitational’ contour map would lie the vectors of environmental impact, due initially to the original refining of silver and then to the growth of its collateral economic activity. See footnote 50, Chapter 3.

the environmental impact of silver refining

347

determine how much of the original by-product of the patio process remains in river beds, in areas where the waste dredged from riverbeds was landfilled, and even in the subsoil of historic patios. Calomel is a non-soluble mercury compound, which has not received attention regarding its life-cycle in aquatic environments, or its impact on organisms at different levels of concentration. What is reported in the historiography on calomel is very limited: it has been used as a diuretic, a laxative, a means to increase the rosiness of the cheeks of babies, a skin-whitening agent and as a topical disinfectant. Until the whole life cycle of calomel in waterways, river beds and landfills is established, its effect as mercury source to the environment in the long term remains unknown. The shipment of hundreds of tons per year across the Atlantic of mercury for a purely industrial use, from the latter part of the sixteenth century onwards, is a very early harbinger of the modern traffic in toxic chemicals across oceans. The resulting loss of liquid mercury to the rivers and soil, at an average of 40 t/y continuously over 250 years of the colonial period (over 100 t/y in Mexico during the nineteenth century), is again an early example of anthropogenic chemical stresses on indigenous communities and their environment, due solely to an imported industrial-scale production activity. The legacy on woodland cover in Mexico is difficult to isolate from the encroachment in modern times imposed by the growth in population and industry. A total of 7 million ha of woodland in New Spain are projected to have been required for refining of silver, ignoring natural cycles of regeneration. The forest cover of Mexico in the year 2010 was 65 million ha. The forest depletion rate between 2005 and 2010 was measured at 0.24 % per year.39 A loss of 7 million ha represents just over 10% of the modern forest cover over 250 years, without the attenuation of natural recovery or forest husbandry. Modern population pressures on these natural resources have been greater than the pressure exerted historically from the colonial silver refining processes. A Change of Paradigm The only explicit paradigm that has guided most of the recent research in various fields on the environmental legacy of historic silver refining, is based on the assumption that all the mercury consumed during its use for the refining of silver ores was due to physical losses, of which the majority (65 to 85 %) is posited as having taken place during the heating stage of the amalgam.40 This paradigm was initially adopted in order to estimate projections of global

39 40

Global Forest Resources Assessment 2010, 228, 233. Nriagu, “Legacy of Mercury Pollution”; Robins, Mercury, Mining and Empire.

348

chapter 9

deposition of mercury over time, but it ignores the chemical role of mercury during the refining of silver ores. The only modern laboratory research on the chemistry of the patio process, was the paper published by Johnson and Whittle which confirmed the formation of calomel during the patio process, but it has found very limited recognition in the narrative by modern historians.41 The current paradigm also ignored the important body of observations made in the historical texts up to the nineteenth century. The manner in which calomel disappeared from most modern models on global deposition of mercury, despite the wealth of data in the historiography of the nineteenth century, is worthy of its own history. According to Kuhn, a paradigm begins to be questioned when it is unable to explain anomalies in its field.42 The current paradigm not only has ignored the chemical role of mercury during the refining of silver ores. It also presents four major anomalies, both on historical and scientific grounds, that cannot be explained by assuming a 100% physical loss of mercury during the process: 1.

2. 3.

4.

41 42

43

The relatively constant value of the correspondencia, for some three centuries, within a range very close to the value dictated by the chemistry of the process. This cannot be explained by contingent losses of mercury, each loss subject to multiple aleatory degrees of carelessness on the part of operators. No first-hand observers of the process in the field ever reported constant and high levels of mercury loss to the atmosphere. The historical silence on mercurialism, incompatible with the amounts of mercury claimed to have been issued to the air, in the close vicinity of major urban centres, such as Potosí in Upper Peru, or Guanajuato in New Spain. The absence of evidence for high levels of mercury deposited from the air over the past centuries, close to historic areas where mercury was used intensively for the refining of silver ores.43

Johnson and Whittle, “The Chemistry of the Hispanic-American Amalgamation Process.” Kuhn describes the moment when ‘an anomaly … cannot, despite repeated effort, be aligned with professional expectation’, in Thomas S. Kuhn, The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1965), 6. Colin A. Cooke et al., “Over three millennia of mercury pollution in the Peruvian Andes,” Proceedings of the National Academy of Sciences 106, no. 22 (2009); Engstrom et al., “Atmospheric Hg Emissions from Preindustrial Gold and Silver Extraction in the Americas: A Reevaluation from Lake-Sediment Archives.”

the environmental impact of silver refining

349

There is an alternative context within which it is possible to account for these anomalies, but it requires discarding the previous paradigm and building upon what was already known in the nineteenth century, complemented by the contributions from modern science. Its main tenets are: 1.

2.

3.

4.

The correspondencia is not an empirical number but the reflection of the stoichiometric relation between mercury and silver, imposed by the chemistry of the reactions that take place during the refining of silver ores using mercury. Standard operational practice during the recycling stage of mercury from the silver amalgam, required heating under controlled conditions in special vessels. The production of most of the silver refined using mercury in New Spain and Mexico, took place between the seventeenth and nineteenth centuries, when metal desazogaderas or capellinas were mostly used. As indicated by first-hand observers, physical losses of volatile mercury during the heating cycle were minimal under normal conditions. The main consumption of mercury was through the formation of calomel. This, together with the operational care mentioned in the previous point, kept cases of mercurialism due to inhalation of volatilised mercury mainly limited to workers involved in accidents, the azogados.44 The main legacies of silver refining around historic mining centres are not mercury from air emissions, but lead compounds, calomel and liquid mercury in the soil. The fact that in New Spain the patio process and smelting were used in parallel, and at times evenly balanced, means that lead in lead fume played a greater role as an environmental impact vector than mercury emissions to the air.

A change of paradigm is long overdue, to allow different fields of knowledge and research to better integrate their results, based on a new set of common guidelines and premises. The previous chapters have shown the benefits of merging historical archives with chemical laws and field studies by environmental scientists, since none of these disciplines alone can provide an answer. The synergy between these studies is evident, and a future collaborative effort can only serve to clear up the pending issues. 44

There is no doubt that every day the workers at a refining hacienda were exposed to the handling of liquid mercury with bare hands, or the contact of mercury in the ore slurry with bare feet, and communities exposed to mercury in waste water voided into streams. The concern should therefore be on these cases of exposure, and not on a non-existent persistent loss of major amounts of volatile mercury.

350

chapter 9

What Did They Know and When Did They Know It? The previous sections have shown that both lead and mercury were an indispensable part of the history of silver refining in the New World. In modern times, they have been identified as heavy metals whose anthropogenic emissions to the environment or in the workplace need to be closely monitored and controlled, due to their toxicity.45 The question arises as to how much was known of the occupational hazard they represented during the historical period I have been covering, and if this recognition in any way influenced the decisions taken by refiners or authorities. In the case of lead, it is acknowledged that even though it was a metal used in major quantities in Roman times, the textual record on its toxicity is very sparse until the nineteenth century.46 Bakewell interpreted Viceroy Toledo’s mining ordinance of 1574 which requires chimneys from smelters to be higher than chimneys from ingenios in Upper Peru, “so that the Indians shall not receive the smoke in any fashion”, as recognising the toxic nature of lead fumes.47 The greater concern shown over lead may have been due more to the much greater amount of smoke coming out of a smelter (Chapter 3), than to a pondered judgment based on the relative risks to health posed by both metals. In fact, in some of the main metallurgical texts of the period it is only mercury that is clearly identified as a poison, while no similar warning is made as to lead. Biringuccio had written: ‘[Mercury] is numbered among the poisons. It has the property of contracting the nerves of those workers who extract it from ore if they are not very careful, and it makes the limbs of those who continually handle it weak and paralyzed’.48 Ercker concurred: ‘be careful lest the smoke or vapour [of mercury] enter your body, since it is a poisonous, cold vapour that paralyzes and kills’.49 When Agricola states that ‘lead is a pestilential and noxious metal’ he is not discussing lead fumes or plumbism, but only repeating the claims made by critics of mining, who link lead to its use in musket balls to wound and kill, or the threat of molten lead in the hands of ‘Cruel Necessity’ mentioned by Horace.50 In contrast, texts highlighting the toxicity of mercury were published throughout the period when

45 46 47 48 49 50

Lead and mercury figure among the top ten chemicals of ‘major public health concern’ by the World Health Organization. For a recent review see Sven Hernberg, “Lead Poisoning in a Historical Perspective,” American Journal of Industrial Medicine 38, no. 3 (2000). Bakewell, Miners of the Red Mountain, 150. Biringuccio, The Pirotechnia, 81. Ercker, Treatise on Ores and Assaying, 112. Agricola, De re metallica, 11.

the environmental impact of silver refining

351

mercury was being used to refine silver.51 Even more important, on both sides of the Atlantic there was prior experience of the dangers of handling and using this liquid metal. In the New World, Garcilaso de la Vega’s (1539–1616) views on the responsibility of a monarch to any of his subjects handling mercury indicates an understanding of the risks involved: The Inca Kings … felt that it [mercury] was harmful to the welfare of those who extract it and refine it, for they saw it caused them tremors and the loss of consciousness. In view of which, as Kings that care so much for the welfare of their subjects, according to their name “Lover of the Poor”, they prohibited by law its extraction, or any memory of it, so much did the Indians abhor it, that they erased its name from their memory and language.52 In the case of Spain and its authorities, who implemented a policy of exporting massive amounts of mercury to the New World for its industrial use, their knowledge of the occupational hazards involved was first-hand and without equal in Europe. Spanish workers had been producing mercury from Almadén before the New World was conquered. By the 1550s these workers were awarded tax exemptions due to the risk in their work, which ‘damaged their mouths’. A doctor and apothecary was also assigned, since ‘there are no people on Earth … so subject to illness as the azogados [the afflicted by mercurialism]’. Such were the perceived dangers of that workplace that in the absence of free

51

52

Rosen has listed the following authors on mercury poisoning: Gabriele Fallopius (1523– 1562), in his treatise De Meteallis et Fossilibus, noted the poisoning by mercury of workers at the mines and mentioned they could not work more than three years as a result; Andrea Mattioli, a contemporary of Fallopius, comments on the chronic mercurialism of the workers at the mercury mines at Idria (modern day Slovenia); Pieter van Foreest (1522– 1597) of Delft; Paracelsus dedicates the third book of his monograph on miners’ diseases to mercury poisoning (1567); in 1665 Walter Pope comments with detail on the symptoms of mercury poisoning in workers; Bernard de Jusein in 1719 presented a memoir on the situation of workers at the Almadén mercury mine in Spain to the Academy of Sciences; some fifty years later Giovanni Scopoli described the effects of mercury poisoning on miners of Alto Isonzo. George Rosen, The History of Miners’ Diseases. A Medical and Social Interpretation (New York: Schuman’s, 1943), 39–133. I have not come across any studies during this period on the effects of mercury or lead on workers at silver refining haciendas of the Hispanic New World. Garcilaso de la Vega, Comentarios reales de los Incas, Primera Parte (Lisboa: Pedro Crasbeeck, 1609), 224v.

352

chapter 9

workers, slaves were required to work at Almadén.53 In the 1560s the Fuggers had requested additional manpower for the mercury mines at Almadén, to meet the increasing demand from the New World. Some 40 convicts destined to serve their sentence as galley-slaves were now sent to work at Almadén, a punishment that must have been deemed not much better than the death sentence of rowing on a Mediterranean galley. In 1593 a special judge, Mateo Alemán, drew up a report for the Spanish King on the conditions under which these slaves and other workers were exposed to the effects of mercury while working in the mine managed by the Fuggers. Among his detailed report of interviews with workers at Almadén, he writes down the words of Miguel del Aldea, a convict from Taracona, sent some four years back by a tribunal of Pamplona to work in the mines: it is true that to be present during the heating of the ores when said mercury is being made is very dangerous to health because the smoke from the ore … causes great harm and many lose their wits and others are azogados and this work is of the greatest danger for the health of the men and this man who is declaring has seen that many have died due to that [work].54 Alemán would then emigrate and die in New Spain ca. 1615, though little has been published on his activity in the New World.55 He represents at least one direct channel for the transmission of knowledge on the toxicity of mercury, to the authorities and mining community in New Spain, during the period that saw the introduction of major quantities of mercury to refine silver. As far as I can ascertain, no similar study was commissioned by the Crown on the effect of mercury on the workers in the New World that refined silver with the aid of mercury.

53 54

55

Matilla Tascón, Minas de Almadén (to 1645), 79. Bleiberg, “Informe Secreto,” 373–374. It is worth reading the whole set of interviews, which leave no doubt that at the end of the sixteenth century, it was reported to the highest level in Spain that mercury vapours were very toxic to workers exposed to it. I have not come across a similar study of the period commissioned for the effect of lead fumes from smelters. Alemán is better known as the author of a novel, Guzmán de Alfarache, that together with the now more famous Don Quixote by Cervantes became a widely-quoted example of the Spanish picaresque style. Irving A. Leonard, “Mateo Alemán in Mexico: A Document,” Hispanic Review 17, no. 4 (1949).

the environmental impact of silver refining

353

In the light of this state of knowledge, the textual silence on any sign of widespread mercurialism in the New World as a consequence of silver refining is both striking and very significant to anyone approaching the subject for the first time. There is no mention in the selections of letters that have been published from Spanish immigrants to New Spain to their families in Spain.56 The more militant and vocal workers of the eighteenth century, when the yearly average of mercury consumption would have reached its peak in New Spain, did not include mercurialism among their list of workplace grievances, where mining risks dominated their concerns.57 I have not come across any reports on mercurialism from the English expatriates supervising the refining of silver in the haciendas of the Real del Monte Company.58 Nor have I seen reports of complaints against mercurialism by neighbours of haciendas de patio. The only reports of mercurialism come from accidents during the heating stage of the amalgam.59 Further afield, there is no report of mercurialism from the refining

56 57 58

59

Otte and Albi Romero, Cartas privadas de emigrantes a Indias, 1540–1616; Fernández Alcaide, Cartas de particulares en Indias del siglo xvi: edición y estudio discursivo. Ladd, The Making of a Strike, 19–28. In 1825 a very graphic portrayal of the deemed toxicity of the patio process was published in England: ‘the men employed barefooted [in mixing the tortas], soon became salivated and paralytic, by the absorption of the mercury, and ultimately died the most painful deaths’, from William Rawson, “The Present Operations and Future Prospects of the Mexican Mine Associations Analysed, by the Evidence of Official Documents, English and Mexican, and the National Advantages Expected from Joint Stock Companies, Considered, in a Letter to the Right Hon. George Canning,” (London: J. Hatchard & Son, 1825), 19. The text is very relevant, not for the improbable scenario it depicts, but for being an early example of a strand of thought that has judged the indigenous workforce as being incapable of running an efficient operation with regards to the recovery of mercury, coupled to a supine docility in the face of such claimed immediate mortal dangers. Indigenous workers suffered great hardships in the mining and refining industries of New Spain and Mexico to be able to feed their families, but there is no evidence they were a completely inefficient and passive workforce. Not even an industry based on a slave workforce could bear the economic impact of workers sickening and dropping like flies soon after trampling the tortas. The gist of the first part of the letter is that English knowhow will allow the wealth of the silver mines in Mexico to be properly extracted. It was written before operations at Regla under its new English investors were underway. There is no evidence that Sir William Rawson ever travelled to Mexico, his source is not identified, and by providing a lurid caricature of events he diverts attention from the real problems of historical refining of silver ores. This extract is a late version of the negative European reports mentioned by Sonneschmidt in footnote 67, this chapter. For example, Sonneschmidt and de Fagoaga, Tratado de la amalgamación de Nueva España, 51.

354

chapter 9

mills using mercury in Nevada in the latter part of the nineteenth century, a world-leader in the production of silver by way of mercury at the time.60 Was historicity at work, with the Spanish and English sources hiding the truth on the ravages of mercurialism behind the nested stories within a story? The initial answer is that there was no mention of widespread mercurialism simply because the aggregate amounts of volatile mercury issued to the air over time were the results of isolated accidents, not part of the standard operational practice of the patio process, and were confined to the immediate surroundings of the heating area of the capellinas. When volatile mercury did escape, there is no doubt its effects on the workers was immediate and toxic, but it was never a case of major losses of volatile mercury affecting both the hacienda and the surrounding communities.61 It was solid, water-insoluble calomel and liquid mercury that were handled and spread in major amounts during the patio process. The only problem with this answer is that while mercury is the usual suspect in any environmental issue related to historical silver refining, so too should be lead. Lead fume was issued to the atmosphere in major amounts within and around each smelting hacienda, under workplace concentrations that exceeded any modern guideline. The greatest harm would fall on the smelting teams closest to the furnaces and then on the rest of the workforce within the smelting hacienda, but as reported in Chapter 3, even the communities around smelters were aware of the toxic nature of the smoke from the smelting furnaces. The visual evidence from historical photographs and drawings leave no doubt as to the degree of air pollution from these smelting haciendas. However, we again find that specific references in the historiography to the effects of lead fume as a consequence of the smelting of silver ores are few and far between, in part due to the ignorance in colonial times on its specific toxicity.62 The fact seems to be that the general silence on occupational hazards began at home, with the way Spain ignored its own workers at Almadén: ‘During the firing of the ores, the vapours from the furnaces … extend over the town [of

60 61

62

De Quille, The Big Bonanza; Smith, The History of the Comstock Lode. This conclusion only applies to the silver refining haciendas. Mercurialism was very much present in the mercury mines of Huancavelica and Almadén. For a historical analysis of the hazards of mercury at Huancavelica, see Brown, “Workers Health and Colonial Mercury Mining at Huancavelica, Peru.” Robins provides historical texts on the dangers of processing cinnabar at Huancavelica and from the mills crushing ores, but no quantity of historical quotes on wide-spread mercurialism either in the refining workforce or among the inhabitants of Potosí. Robins, Mercury, Mining and Empire. One of the few exceptions is the following observation: ‘poisonous are the Smelters and amalgamation [haciendas]’ in de Gamboa, Comentarios ordenanzas de minas, 462.

the environmental impact of silver refining

355

Almadén] so that its inhabitants are under the pernicious effect of a mercurial atmosphere’.63 It would be hard to argue that the Spanish Crown would take greater care of its new subjects in the New World than it took of its old ones in Spain. Thus, any negative consequences from the use of mercury or lead on the health of the local communities was no greater than the risks already borne in Europe. In another context, Rosen has argued that while mining was the activity of slaves and criminals it impeded any interest in documenting systematically their diseases, and that only the appearance of free men as miners, with an intrinsic value now placed on their welfare and the need to safeguard their increased technical knowledge, transformed them into capital assets that justified the creation of a corpus of detailed medical literature related to their occupational diseases.64 To complement Rosen it can also be argued that in the context of the perils of mining and in the light of the immediate harm caused by silicosis due to the fine milling of ores or the incidence of hernias on the workforce, or mortal accidents within the mines, long-term lead poisoning or mercurialism was the least of the worries in the minds of the miners and refiners.65 It is thus much more common to find references to the hazards of mining or milling, than to the perceived hazards of lead and mercury within the labour-force. This normal response to the clearest and most present dangers, lowered the visibility of the potent but longer term threat of lead emissions in those areas where smelting was carried out, though it did not quite erase it from the historiography. Was there a conscious decision to downplay the occupational, safety and health aspects of the refining haciendas up to the end of the nineteenth century? Smelting would have created more problems among the workers and surrounding communities than the patio process, since mercury in calomel would 63 64

65

José María Pontes y Fernández, Historia de la antigua ciudad de Sisapón, hoy Almadén del Azogue (Madrid: Enrique Rojas, 1900), 12. Rosen, History of Miners’ Diseases, 8–38. This disdain of the local workforce is not a phenomenon restrained to the Spanish Crown, authorities and miners of the sixteenth century. The silence of the English managers, investors and technical staff on mercurialism can be explained by calomel, but the absence of measures to control the lead in the smoke from the smelting furnaces signals that the workforce in the Mexico of the nineteenth century may not have yet been considered an asset in the eyes of the English investors or their Mexican successors. During a panel discussion on modern artisanal gold mining at the 11th International conference on Mercury as a Global Pollutant (Edinburgh, 2013), members of the audience pointed out that the concerns of researchers did not necessarily match the concerns of the artisanal refiners of gold on the subject of workplace safety, for whom the dangers of inhaling mercury fumes were the least of their daily problems.

356

chapter 9

have posed a negligible short-term health hazard compared to lead fume. The fact lead fume did not receive any major attention in the historical texts, points to a more complex answer to this question. Mining and refining drove the economy of New Spain and Mexico during the period of interest. The common interest of the private entrepreneurs and the Crown to maximize the extraction of silver, the merging of refining and farming interests in the hands of strong owners of capital within New Spain, the absence of the social concern and knowledge base that would lead to the modern discipline of occupational medicine, and finally the entry of overseas investment allied to newly emerging capital and industrial Mexican groups, would have created overall a very strong reason to silence, or ignore, the negative effects of refining on the local communities. In the case of private letters or journals written by Europeans, I would argue that the paradox of Father Brown’s postman was also at work.66 The travails of the indigenous population of refining workers of Mexico barely figure in the commentaries of European observers in the nineteenth century. Whatever their illnesses may have been, they were not part of the mental landscape reflected in these texts, except in the more lurid accounts already mentioned. The history of silver refining through the eyes and words of the indigenous workers of the New World remains to be written. The answer to the question that heads this section is therefore straightforward. The impact on worker’s health in the presence of volatile mercury was well known in the sixteenth century, and onward, at the highest levels of policy makers in Spain. In the case of the toxicity of lead fumes from smelters, the dangers seem to have become documented only later, towards the nineteenth century. There is no evidence that this knowledge, at any point in time, influenced any major decision related to the practice of refining silver ores in New Spain / Mexico from the Spanish authorities. Even in the nineteenth century, I have not found any evidence of concern expressed by the English managers on the possible hazards to health from the refining activities. Overall the production of silver in this time remained unfettered by strict controls on the use of lead and mercury. To judge the mentalité of the period on occupational health, it should be borne in mind that in eighteenth century Germany, one of the important reasons to protect workers from the toxic smoke of the smelting furnaces, was because it tended ‘to kill the best and most useful of the smelters’.67 66 67

From G.K. Chesterton’s short story “The Invisible Man” in G.K. Chesterton, Father Brown, Selected Stories (London: Oxford University Press, 1961), 74–94. Schlutter, De la fonte des mines, Tome Second, 2. In the case of the patio process, Sonneschmidt addresses the issue of worker’s health in a separate section of his work. He

the environmental impact of silver refining

357

The collateral damage on the health of the smelters or workers in an hacienda de patio, caused by both lead fume and mercury, was judged by the norms of the period to be an acceptable cost of each process. However, there was always a limit to what workers and refiners could accept. In the case of mercury, I have argued that its transformation into calomel attenuated to a high degree the impact on the human environment, though a loss of 15 % as elemental mercury is still a major historical health concern to all the communities exposed to its effects. What would have been the response in the workplace to a more immediate toxic impact from the patio process recipe? On the 7th February 1561, a petition was forwarded to the Viceroy of New Spain, Luis de Velasco, requesting a merced for a new refining recipe that included the addition of solimán, mercuric chloride. To prove that its use was not injurious to health, the promoter of the method, Pedro Martín from Taxco, swilled his mouth with the water used to wash the ore after treatment, though the only ones forced to actually swallow the liquid were a cockerel from Castile and a horse, to no apparent ill effect. On the 10th March 1561, the authorities in Seville were quickly apprised of the claimed benefits of this new recipe and requests made for new shipments of solimán at an attractive price. By July of that year the same authorities in New Spain would start to point out the hazard posed by the use of solimán: ‘it is dangerous for the blacks [slaves], we give notice to Your Majesty of this fact’ (24 July 1561). By the time a year had gone by it was clear that ‘there are great dangers to using solimán, for which reason the miners do not make use of it’ (2 April 1562).68 Such was the risk involved, and its lack of technical success, that it soon disappeared from the repertoire of refining methods in New Spain. The toxic effects of mercuric chloride on the human body are much more immediate and evident than the longer term and more insidious effects of either mercury or lead poisoning, leading to retention of urine, vomiting, bloody diarrhea and ultimately death unless treated.69 Even if solimán had

68

69

concludes that in spite of negative reports in Europe on the effects of this process on the workforce, he had met in his visits to Reales de Minas workers with up to 40 years working in haciendas de patio who did not show any health problems as a result. Sonneschmidt and de Fagoaga, Tratado de la amalgamación de Nueva España, 94–95. Castillo Martos, Bartolomé de Medina, 210–213. What a difference a state of oxidation makes in life. Solimán and calomel are both chloride salts of mercury, the former a mercuric salt (Hg+2), the latter a mercurous salt (Hg+1). Solimán is an extremely toxic mercury compound, while calomel is not. Laszlo Magos and Thomas W. Clarkson, “Overview of the Clinical Toxicity of Mercury,” Annals of Clinical Biochemistry 43(2006): 259–260.

358

chapter 9

been an effective refining reagent (for which there is no evidence), the very high level of toxicity would have precluded its widespread use in the New World. The obvious and natural reticence of humans to work with evident poisons was the ultimate threshold. Had the short-term toxicity of mercury been equivalent to that of solimán, the patio process would not have been used to refine silver ores.

Was Mercury the Indispensable Key to Silver in the New World? As I approach the end of my arguments, which have been guided by the basic chemical fundamentals of each refining process, to better interpret the human choices that have determined the direction and environmental impact of the history of silver refining in New Spain, there remains one last issue I wish to address. For many observers, both past and present, mercury was the only key that could unlock the wealth of the silver mines in the New World. I have a reasonable doubt as to the veracity of such a strong claim. To begin with, for chemical reasons, any process based on mercury could not have refined the totality of the silver ores of New Spain, and according to my projections, smelting during certain historical periods accounted for half the total silver production. The level of silver content in the ores was never the real obstacle to smelting all the ores, the real hurdle would have been sourcing sufficient charcoal. In contrast to the case of England, where by the year 1400 already only 17% of England and Wales’ usable land was covered by forests, Spain found in the New World a far vaster virgin resource of wood for making charcoal.70 Even nearly five hundred years later, the modern states of Mexico, Bolivia and Peru still have from 30 to over 50% of their land covered in forests.71 Even so, dressing of the ore would have been a necessary pre-requisite to smelting, otherwise the need to heat the totality of an ore containing on average 0.2 % silver would have required an estimated additional 105 million ha of woodland for charcoal, more than twice the area of woodland existing in Mexico in the year 2010.72 Dressing the ore to a final silver content greater than 0.6 % would

70

71 72

Jed O. Kaplan, Kristen M. Krumhardt, and Niklaus Zimmermann, “The prehistoric and preindustrial deforestation of Europe,” Quaternary Science Reviews 28, no. 27–28 (2009): 3023. Global Forest Resources Assessment 2010, 228. A split of 60:40 in silver produced by the patio process and smelting, based on an average silver content of 0.2% for the former and 2% for smelting, corresponds to a 15 to 1 ratio in the weight of ore processed by each route. Thus if 7 million ha were required to heat all the ore used for smelting in New Spain, fifteen times that amount would be needed to

the environmental impact of silver refining

359

have brought down the additional requirement of woodland to at least one third of the previous estimate, and at the same time offered a degree of profit to the smelters, net of dressing costs.73 Lead would not have been a material bottleneck, had Spain chosen to smelt all the ore. Up to an additional 300,000 t over 250 years would have had to be sourced locally or imported from Europe (and serving as needed ballast to the outgoing convoys from Seville), to make up for losses during the smelting of the additional ore. Mexico was producing yearly well over 60,000 t of lead by the end of the nineteenth century, so if the Spanish Crown had promoted the mining and production of lead, the local deposits would have been sufficient to sustain the requirements of smelting.74 When the initial wave of complaints rose from the ranks of amateur refiners in New Spain, that they were running into problems refining silver, the Crown had the option of improving the smelting skills of this population, or promoting the search for lead within New Spain. Smelting was the universal refining process, not limited like the patio process to certain types of ore. The rapid response of the Crown to the potential use of mercury to refine silver ores, commented upon in Chapter 4, leads to the unsurprising conclusion that the ownership of the Almadén mine played a critical role in the refining policy adopted by the Crown. There is no doubt that the patio process was a technique better suited to most amateur refiners that sought wealth in New Spain after its conquest. However, as the history of San Luis Potosí shows, smelting was implemented when they were faced with silver-rich lead ores. With sufficient lead and after dressing ores, the experience at San Luis Potosí could have been the model for the expansion of silver production by smelting in New Spain as of the seventeenth century. The case of Japan in the late sixteenth and early seventeenth century, refining major amounts of silver from silver sulphide ores by smelting, constitutes a timely reminder that mercury was never the only metallurgical key. In any case, for approximately 100 years smelting refined silver ores on par with the patio process in New Spain. The use of smelting would however have posed an opportunity cost to the Crown. If mercury was

73

74

heat the ore that would have otherwise been refined using mercury. The data on the forest cover in Mexico in 2010 (65 million ha) are from ibid. I am not proposing that an option that would have decimated the forests of Mexico would have been preferable, but simply pointing out that a policy consonant with the historical period could have been adopted, even more so if forest regeneration cycles operated over a 250-year span, as would be expected. The future use of coal, local or imported, to substitute charcoal in smelters cannot be ruled out in this counterfactual scenario. Flores Clair, Velasco Avila, and Ramírez Bautista, Estadísticas mineras, ii, 155.

360

chapter 9

used, it opened an additional important revenue stream to the Treasury, in parallel to the revenues from the royalty and other taxes on silver, as discussed in Chapter 8. This opportunity cost could have reached some 10 % of the total revenues to the Crown in the early seventeenth century, a major sacrifice in earnings. Thus, to the operational advantages of the patio process must be added the fiscal advantage over smelting it represented to the Royal Treasury, made even more attractive during the first century of the patio process in New Spain by the role of the Fuggers. So the answer to my final question is no, mercury was not the only technical and financially viable key to unlock the silver from the silver sulphide ores in the New World, and was certainly not a technique justified solely by the deemed low silver content of any ore. The patio process offered some important operational advantages over smelting, and it processed vast amounts of silver ores without placing great stress on local woodlands. What made a critical difference between the patio process and smelting was the ownership by Spain of two of the three major mercury sources available in the world during the Early Modern period, first Almadén and then Huancavelica. Fiscal reasons thus provided the final argument to favour the use of mercury over smelting, a decision evident from the actions of the Crown and its colonial authorities. The unintended consequence of the promotion of the patio process was that the population and the environment of New Spain was spared a worst-case scenario, of large amounts of lead fume issued to the air around each of the smelting haciendas, and a major consumption of woodland. Instead they would be subject to millions of tons of silt suspended in water and settling along river beds, entombing mercury transformed into calomel, together with liquid mercury entrained with dissolved salt and copper compounds. Within the haciendas de patio the environmental legacy would be a soil soaked in mercury. Finally, but significantly in terms of quantity and impact, air emissions of mercury would be two orders of magnitude lower than lead and its compounds exiting the chimney stacks.

Epilogue There’s the story, then there’s the real story, then there’s the story of how the story came to be told. Then there’s what you leave out of the story. Which is part of the story too. margaret atwood, MaddAddam (2013)

∵ For most of the Early Modern period, Spain would control the world’s largest reserves of silver and of mercury, a historically unique geological and geopolitical triangulation that would make it the master of the silver market. By conquering the western chain of mountains from the Andes to the Cordillera of North America, Spain conjoined its own vast reserves of mercury at Almadén, with the deposits of silver that had been generated by a process of subduction of tectonic plates all along the eastern Pacific Rim. Nowhere else in the world is subduction as active, nor such a chain of major primary silver deposits to be found, much younger and of a different geological origin than their more modest counterparts in MittelEuropa. The only common denominator was the presence on both sides of the Atlantic of argentiferous lead ores. Most of the silver ore deposits of Europe were mined at a profit thanks to their copper or lead content, rather than silver, while for the ores of the New World, silver was the only economic reason that justified their refining. The absence of major preConquest extractive industries kept intact the composition of their weathered surface, native silver and silver chlorides, which made possible their refining even by the inexpert hands of the first generations of refiners. The three hundred and fifty years of uninterrupted silver production in New Spain, and then Mexico, regardless of major political upheavals, were based on the development of efficient mining and refining practices by private entrepreneurs, whose only motivation was personal profit. The critical part of the technology that allowed the refining of silver sulphide ores using mercury, including innovations in milling equipment and industrial architecture, was developed in the Hispanic New World. Experiential knowledge had been seeded in the sixteenth century in New Spain by European artisans, but from the 1570s onward it was local ingenuity in the Andes that made possible the refining of vast amounts of silver ore. How this came about is a fascinating puzzle of trial and error, a mixture of alchemy and perseverance, of being right

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_012

362

epilogue

for all the wrong reasons, of operational discipline at every level of the workforce. The historical trail begins with the scavenging instincts of the initial swarm of Spanish colonists, avid for wealth and lacking skills, who left behind generous mounds of discarded weathered ores, all the mineral that did not immediately promise silver to their untrained eye or yield it to their primitive smelting efforts. As they dug into the mineral veins, they complained to the authorities of decreasing silver content, when in reality what was changing was the nature of the mineral, from weathered silver ore to deeper and more intractable silver sulphides. The initial technical challenge was solved thanks to imported German know-how in smelting. The environment of New Spain would feel the first wave of lead fume from the smoke of smelting furnaces, and woodlands the appetite of charcoal burners smoldering their way through their midst. As the mountains of discarded ore grew, so did the frustration of the first generation of self-taught refiners. In New Spain the metallurgical use of mercury did not arrive on the back of a sudden decrease in silver production. It arrived from Europe as a process that was known to work with gold, a method better suited for untrained refiners than smelting. Spain saw a quadruple opportunity open up: more silver could be produced by its untrained colonists foiled by smelting and intractable ores, it could gain much more revenues from its mercury mine at Almadén, it could use the Fuggers to provide the mercury on credit as well as collateral for future loans, and by controlling mercury supply as a State monopoly it could gain a measure of control over the contraband of silver. The hazard posed by mercury was well known, but in practice was not seen to be worse than the burden imposed upon its own people at Almadén. Mercury, aided by its aura as the alchemical precursor to silver and gold, was thus applied to the mounds of cheap discarded ores and to existing superficial deposits. These responded well to the primitive refining method that was known to work with gold. Had the first generation of refiners been thorough in their triage and smelting, only lead-rich slags would have been left behind. In such a case the primitive recipe used for gold would not have worked either on these or on the deeper silver sulphides or negrillos. As the mounds of discarded ore that were easy to refine with mercury were run down, so did silver production suffer. Then events in the Andes would radically change the industrial potential of the basic gold amalgamation recipe. In a short burst of impressive technical creativity within the Spanish refining ingenios of the altiplano, the chemical roots of the patio process were generated, a powerful refining method that was able to reduce the silver sulphide present in negrillos into amalgamated metallic silver. Though an earlier incarnation had

epilogue

363

been applied previously in the Schio mines by Venetians in the early sixteenth century, it was recreated independently through the stubborn efforts of the selfmade refiners around Potosí. As the patio process gained momentum, mercury was consumed in quantities never observed before in any human industry. The workers of the haciendas de patio and the population of New Spain were spared the ravages of mercurialism on a major scale both by the safeguards adopted during the heating cycle of the amalgam, and by the chemistry of the patio process that consumed mercury by converting it to solid, insoluble calomel. Nevertheless, tons of liquid mercury were squeezed through the fingers of workers onto the slurries of ore, and liquid mercury was still washed away in water or lost by seepage to the soil. Very little mercury escaped to the air during the regular heating cycle of the amalgams. Streams and water basins around the clusters of haciendas de patio became their waste disposal units. Salt seeped into the ground of the patios and was washed away together with copper compounds, and the millions of tons of fine mineral silt that were useless to the refiner once all possible silver was extracted. All these materials would contaminate the water downstream from each refining unit, and compromised its use for consumption and irrigation. The patio process represented a technology best adapted to the medium where it was implemented, competitive with the more traditional route of smelting. The longevity of its recipe was never a sign of backwardness or technical stagnation, in the same manner the chemical immutability of the smelting process over thousands of years has never been regarded a drawback. It embodied all the elements of a modern industrial process: planning of inventories, carefully concatenated stages of physical treatment and chemical reactions, avoidance of operational bottlenecks to achieve a smooth production output, clear separation of technical functions within its labour force. The patio reactor that evolved in New Spain was an efficient answer to the amounts and nature of the ore that had to be treated, and to the materials at hand. The industrial architecture it gave rise to was born in the New World, an autochthonous spatial answer to the efficient concatenation of a chain of operational stages. The history of silver refining in New Spain was determined as much by smelting as by the patio process. Smelting contributed with just under 40 % of all the silver produced during the colonial period, and possibly one quarter to one fifth during most of the nineteenth century. Lead and its compounds, within the lead fume that issued from the chimneys of smelters, represented the main source of heavy metals issued to the air from the historical silver refining in New Spain, on average two orders of magnitude greater than the total air emissions of mercury. The historical impact of emissions of lead fume from the smelting and cupellating furnaces was the aggregate of multiple but

364

epilogue

singular depositional footprints determined by the local wind rose, furnace efficiency, skill of the smelter, lead content of each ore, and by the architectural trace of the hacienda and the location of its mounds of grasas. Some mining regions in New Spain would never be exposed to much lead, others would have known no other airborne heavy metal, but those who did live near smelters had no doubts on the toxicity of its smoke on animals and humans. The other major environmental impact of smelting was on woodlands, consuming them at a rate over 50 times higher, per kg of silver refined, than the patio process. This depredation of woodlands was only attenuated by the increased efficiency of blast furnaces in the nineteenth century, which decreased by an order of magnitude the rate of consumption of charcoal per kg of silver smelted. This history is interwoven with many silences. The environmental cost from silver refining as imposed on the indigenous communities and the new settlers was never addressed in a significant manner in the texts of the period. This silence reflects many realities: the greater, more immediate dangers of mining silver ore; the acceptance of occupational health risks from a workforce whose pressing issue was to earn a living for their families; the potential for profit at every level of the mining and refining workforce, overriding concerns for personal safety; the cloak of social invisibility that covered most of the indigenous population in the eyes of others. The silence encompassed both the Spanish and English contingents who came to extract silver from New Spain and then Mexico, as well as the new local owners and operators of the later republican period. Another is the silence in the modern historiography on lead and calomel, a very puzzling nested history within this history. It is not clear why the current narrative on colonial silver refining opted to equate its well documented and careful operational handling of the heating stage of the amalgam, to the level of the careless and primitive methods of artisanal gold amalgamation carried out in modern times. The technological and chemical gulf between the two could not be greater, yet both have been placed at the same level of mercury losses to the air, irrespective of the voices of first-hand observers, or of the tenets of the chemistry that defines the action of mercury during the refining of silver ores. The most original operational practices of silver refining developed in the Hispanic New World, have been judged to be no better than the worst of modern Third World gold refining techniques. Consequently, calomel disappeared from most modern research on the topic, and deemed major losses of volatile mercury dominated the study of its environmental impact. This has led to a dead-end mirage of copious mercury fumes emanating from haciendas de patio, which has eclipsed the more serious reality of dense smoke containing lead fumes spewing forth from smelting haciendas.

epilogue

365

Lead and its compounds, not volatile mercury, represented the main environmental hazard from heavy metals issued to the air, as a result of silver refining in the period I have covered. The silver of New Spain and Mexico could have been extracted only with smelting, had this been necessary. Enough lead and woodland existed in its vast territory to have covered the needs of this refining process. Without recourse to mercury, refiners would have sought a greater efficiency from the furnaces, recovered lead from the fumes, spent more manpower in dressing the ores and would have developed a secondary market for its smelted non-precious metals, much as republican Mexico finally did as of the late nineteenth century. Had smelting prevailed, the total environmental impact of its lead emissions and destruction of woodland would have been more severe than from the historical mix of patio process and smelting. Mercury as the lesser evil: this was the ultimate paradox of the history of silver refining in New Spain and Mexico.

appendix a

The Accounting Books of Regla The account books that were consulted in the Archivo Histórico de la Compañía de Minas del Real del Monte y Pachuca (AHCRMyP), as the source of the inventories, mass balances and production costs, correspond to the Fondo Siglo xix. They comprise the following sections, series and sub-series:

1

Sección: Explotación y Beneficio, Serie: Informes de Haciendas de Beneficio, Subserie: Informes Mensuales Hacienda de Regla Vol. 225, Exp. 3: 29 Jun 1872–27 Oct 1888

This is the single tome accounting ledger with production data for Regla, referred to in the main text as Informe Mensual. The ledger registers the accounts in four or five week intervals, dated according to the final day of each period. It tracks the quantities of incoming ore according to silver content and ultimate destination (patio process or smelting), together with the final amount of silver extracted. The data on the patio process cover a fifteen year interval, though the period from 1874 to mid 1875 is completely atypical in that silver refining plummeted, the patio process was suspended in many months and smelting carried out mainly on slags. Data from this period will be excluded from the general analysis. Smelting data from ores was only reported for the period June 1875 to January 1886, with some monthly interruptions. Each monthly account sheet contains the following information on the costs incurred: 1.

2.

A report on the monthly consumption by weight and total cost (in pesos) incurred of the following major consumables, under the heading Almacén (warehouse): salt (sal), mercury (azogue), copper sulphate (sulfato de cobre), litharge (greta), charcoal (carbón), barley (cebada), straw (paja), corn (maíz), animals in stock and losses by death.1 The monthly production costs (Costo de Beneficio) are reported within a separate boxed-in area of each monthly account sheet. They are presented under some fifteen different headings, some of which change during the 1872 to 1888 period.

1 Firewood is only reported in the period 1872 to 1873, and is incomplete. A more complete and detailed breakdown of both charcoal and firewood is provided both in the Memorias de Gastos and Estados Comparativos.

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_013

368

appendix a However it is fairly straightforward to group these costs under the following sub-sets: Labour costs; Mercury, Salt, Copper Sulphate, Litharge and Charcoal; Other costs. In the case of smelting, I use the accountants’ figure on total monthly smelting costs, subtract the costs for litharge, charcoal and labour (reported as fundición), and the net amount I register as ‘other costs’ for smelting.

2

Sección: Negociaciones, Serie: Haciendas de Beneficio, Subserie: Hacienda de Regla i, Vol. 22: 1875–1878

I refer to this source in the main text as the Memorias de Gastos. Each Memoria is numbered and corresponds to a weekly account of certain expenditures (gastos) at Regla. It provides a wealth of detail on the labour component of the two refining processes used at Regla: the name of the worker in some cases, hours worked per week, wage per hour and approximate description of work carried out or skills. It also contains total firewood and charcoal consumption that appears in the weekly accounts, as detailed under the heading Varios Efectos (Miscellaneous). For the earlier years, 1853–1873, the Estados Comparativos provide more information on fuel, as set out below.

3

Sección: Contabilidad de la Dirección, Serie: Producción y Gastos, Subserie: Estados Comparativos, 1853–1855, 1859–1865, 1869–1873

This source is referred to as Estados Comparativos, and is a monthly and yearly comparative summary of selected expenditures and costs per montón or carga of refined ore, for all the refining haciendas active in any given period. The comparative tables of production costs incurred at each hacienda contain the following information reported both on a monthly, quarterly and yearly basis: Variable production costs; Information on the ore processed at each hacienda: number of cargas; the average total silver content (ley, expressed in marks per montón) of the ore being processed, and the silver actually extracted, reported also as a ley; the percentage loss of silver and the loss of mercury expressed as ounces of mercury per mark of silver. Some of this information also appears in the monthly accounts of Regla within the Informe Mensual; Information on the consumption of main reagents: total weight consumed of salt, mercury, copper sulphate, magistral (only in some years), litharge, firewood, charcoal; Information on the consumption of sundry items: the same yearly average (total weight and total expense) is reported for all the sundry materials used at the haciendas.2 2 Efectos Diversos is mostly made up of: tools (herramientas), machinery (maquinarias), bricks

the accounting books of regla

4

369

Sección: Ymporte de la Memoria de la Mina, Serie Informes de Minas i, Vol. 13

This was the source of production costs of the ore. (ladrillos), lime (cal), planks (tablón), wood (madera), leather (cueros), lead (plomo), iron (hierro) and steel (acero), nails (clavos), refractory stones (piedra refractaria), limestone (piedra de cal), bone ash (ceniza de hueso), stones for arrastres (piedras voladoras), capellinas, and dead animals (animales muertos). The expense on fodder (maize, barley, straw) is absent.

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_014

Fuel Mercury Salt Copper Sulphate Labour others ore cost power capital cost total

0.04 9.80 1.78 0.72 0.13 1.93 0 1.20 0.50 16.10

variable production cost pesos per kg silver

silver in ore kg of silver in montón value silver in pesos

0.02 4.26 4.21 1.69 0.31 4.54 0.0 2.82 1.18 19.04

43.79

n/a

0.04 % 0.43 17

0.00 0.00 4.21 1.69 0.31 4.54 0.0 2.82 1.18 14.76

0.00 % 0.00 0

Mercury-based refining, sixteenth century context

Mercury-based refining process

table b-i

Sensitivity Matrix for Refining Costs

appendix b

33.61

0.03 6.09 4.21 1.69 0.31 4.54 0.0 2.82 1.18 20.87

0.05 % 0.62 24

0.08 % 0.99 38

0.12% 1.49 57

0.19% 2.36 90

0.60% 7.45 283

0.04 9.74 4.21 1.69 0.31 4.54 0.0 2.82 1.18 24.54

0.06 14.61 4.21 1.69 0.31 4.54 0.0 2.82 1.18 29.43

0.10 23.13 4.21 1.69 0.31 4.54 0.0 2.82 1.18 37.99

0.32 73.04 4.21 1.69 0.31 4.54 0.0 2.82 1.18 88.12

29.65

24.70

19.75

16.10

11.83

variable production cost in pesos per kg silver

0.03 7.30 4.21 1.69 0.31 4.54 0.0 2.82 1.18 22.09

variable production cost in pesos per montón

0.06 % 0.75 28

11.03

0.54 121.74 4.21 1.69 0.31 4.54 0.0 2.82 1.18 137.03

1.00% 12.42 472

10.47

1.02 231.30 4.21 1.69 0.31 4.54 0.0 2.82 1.18 247.08

1.90% 23.60 897

10.24

1.61 365.22 4.21 1.69 0.31 4.54 0.0 2.82 1.18 381.58

3.00% 37.26 1416

Fuel Litharge Labour others ore total

2.23 0.00 0.13 0.85 0.00 3.21

variable production cost pesos per kg silver

Smelting silver in ore kg of silver in a carga value of silver in pesos

5.27 0.00 0.31 2.00 0.00 7.58

5.27 0.00 0.31 2.00 0.00 7.58

5.27 0.00 0.31 2.00 0.00 7.58

5.27 0.00 0.31 2.00 0.00 7.58

50.85

31.29

10.17

7.63

variable production cost in pesos per kg silver

5.27 0.00 0.31 2.00 0.00 7.58

174.33 122.03 101.69 76.27

5.27 0.00 0.31 2.00 0.00 7.58

n/a

5.27 0.00 0.31 2.00 0.00 7.58

5.27 0.00 0.31 2.00 0.00 7.58

5.27 0.00 0.31 2.00 0.00 7.58

variable production cost in pesos per carga

6.10

5.27 0.00 0.31 2.00 0.00 7.58

5.08

5.27 0.00 0.31 2.00 0.00 7.58

3.21

5.27 0.00 0.31 2.00 0.00 7.58

2.03

5.27 0.00 0.31 2.00 0.00 7.58

0.00 % 0.04 % 0.05 % 0.06 % 0.08 % 0.12 % 0.20 % 0.60% 0.80% 1.00% 1.20% 1.90% 3.00% 0 0.0435 0.0621 0.0745 0.0994 0.149 0.2422 0.7452 0.9936 1.242 1.4904 2.3598 3.726 0 2 2 3 4 6 9.2 28 38 47 57 90 142

table b-ii Smelting, sixteenth century context

sensitivity matrix for refining costs

371

0.04 3.62 3.66 0.72 0.67 1.93 16.7 2.95 1.38 31.61

variable production cost pesos per kg silver

silver in ore kg of silver in montón value silver in pesos

0.02 1.57 8.64 1.69 1.58 4.54 39.3 6.97 3.25 67.55

155.40

n/a

0.04 % 0.43 17

0.00 0.00 8.64 1.69 1.58 4.54 39.3 6.97 3.25 65.96

0.00 % 0.00 0

109.88

0.03 2.25 8.64 1.69 1.58 4.54 39.3 6.97 3.25 68.23

0.05 % 0.62 24

Patio process, seventeenth and eighteenth century context

Fuel Mercury Salt Copper Sulphate Labour others ore cost power capital cost total

Patio process

table b-iii 0.08 % 0.99 38

0.12% 1.49 57

0.19% 2.36 90

0.60% 7.45 283

0.04 3.60 8.64 1.69 1.58 4.54 39.3 6.97 3.25 69.60

0.06 5.39 8.64 1.69 1.58 4.54 39.3 6.97 3.25 71.42

0.10 8.54 8.64 1.69 1.58 4.54 39.3 6.97 3.25 74.60

0.32 26.96 8.64 1.69 1.58 4.54 39.3 6.97 3.25 93.25

92.18

70.05

47.92

31.61

12.51

variable production cost in pesos per kg silver

0.03 2.70 8.64 1.69 1.58 4.54 39.3 6.97 3.25 68.69

variable production cost in pesos per montón

0.06 % 0.75 28

8.97

0.54 44.94 8.64 1.69 1.58 4.54 39.3 6.97 3.25 111.44

1.00% 12.42 472

6.46

1.02 85.38 8.64 1.69 1.58 4.54 39.3 6.97 3.25 152.37

1.90% 23.60 897

5.43

1.62 134.82 8.64 1.69 1.58 4.54 39.3 6.97 3.25 202.39

3.00% 37.26 1416

372 appendix b

Fuel Litharge Labour others ore total

2.23 0.00 0.66 0.85 0.00 3.74

variable production cost pesos per kg silver 5.27 0.00 1.56 2.00 0.00 8.83

5.27 0.00 1.56 2.00 0.00 8.83

5.27 0.00 1.56 2.00 0.00 8.83

5.27 0.00 1.56 2.00 0.00 8.83

59.24

36.45

11.85

8.89

variable production cost in pesos per kg silver

5.27 0.00 1.56 2.00 0.00 8.83

203.09 142.16 118.47 88.85

5.27 0.00 1.56 2.00 0.00 8.83

n/a

5.27 0.00 1.56 2.00 0.00 8.83

5.27 0.00 1.56 2.00 0.00 8.83

5.27 0.00 1.56 2.00 0.00 8.83

variable production cost in pesos per carga

7.11

5.27 0.00 1.56 2.00 0.00 8.83

5.92

5.27 0.00 1.56 2.00 0.00 8.83

3.74

5.27 0.00 1.56 2.00 0.00 8.83

2.37

5.27 0.00 1.56 2.00 0.00 8.83

0.00 % 0.04 % 0.05 % 0.06 % 0.08 % 0.12 % 0.20 % 0.60% 0.80% 1.00% 1.20% 1.90% 3.00% 0 0.0435 0.0621 0.0745 0.0994 0.149 0.2422 0.7452 0.9936 1.242 1.4904 2.3598 3.726 0 2 2 3 4 6 9.2 28 38 47 57 90 142

Smelting, seventeenth and eighteenth century context

Smelting silver in ore kg of silver in a carga value of silver in pesos

table b-iv

sensitivity matrix for refining costs

373

appendix c

Estimates of Silver Production by Caja and Refining Process, Including Balance of Mercury Consumption and Physical Losses In the following tables two sets of data are calculated in a straightforward manner. Using the tax receipts, I have repeated with the tk set one of the classical calculations carried out in the historiography of silver refining, the conversion of silver tax revenues into a weight of silver production.1 In parallel, I have converted the revenues from the sale of mercury by each Caja into an approximate weight of mercury, using the generic price ranges for mercury for each period in question.2 Freight does not have to be subtracted from these values since this was a separate cost and included as such in the accounts of the Caja.3 Silver taxes and mercury sales registered in the same period are not necessarily correlated, due to credits extended, late payments by refiners, inventory build-ups, or even contraband of mercury. By aggregating the mercury data into a decade whenever possible, the potential for mismatches will tend to be minimised but not eliminated. With this caveat in mind, I proceeded to correlate mercury consumed with silver refined by the patio process, using a deemed value of correspondencia, based on the values of this ratio available for later periods in each Caja. Since the correspondencia could vary substantially with the recipe (iron or copper will lower it), this is the assumption that introduces the widest margin of error in my estimate, but at least it attempts to maintain the regional idiosyncrasy of this factor. Knowing the total silver registered, and estimating the amount of silver refined by the patio process that can be assumed to correspond to the sale of mercury in any given period, it is possible to arrive at an estimate of the breakdown between smelting and the patio process. For all the Cajas, I also include the projection of mercury consumed as calomel, and physical losses of liquid mercury in the soil and water, and volatilised mercury, using the breakdown established in Chapter 4 (85%, 14% and 1 %, respectively, of the total

1 I have followed Bakewell’s path in applying factors of 10.9 and 20.8 to reverse calculate from the tax data under 1% and diezmos and quintos the value of silver produced in pesos (of 272 maravedies). I have used his value of 8 pesos 1 real for a mark of silver up to the year 1700, and 8 pesos 6 reales after that date. Bakewell, Silver Mining in Zacatecas, 245. 2 For the calculation of the total weight of mercury I have used the Almadén price even for the mercury brought from Idria and Peru. 3 Pérez Luque and Tovar Rangel, Caja Real Guanajuato, 37.

© koninklijke brill nv, leiden, 2017 | doi: 10.1163/9789004343832_015

estimates of silver production by caja and refining process

375

amount of mercury consumed). When the mercury to silver ratio is above 2.3 for any period, a limit is imposed on the value of mercury consumed as calomel, and the excess is credited to losses as liquid mercury.

Caja of Zacatecas (Table c-i) To arrive at an estimate of production by each refining process, I first generate average values of the mercury to silver weight ratio per time periods as close to a decade as the data allows. The average of this ratio for the period 1670 to 1810 is 2.05. Between 1670 and 1650 the tk set provides the total silver produced, and the amount of mercury sales. Using the average historical ratio of 2.05, I can estimate the amount of plata de azogue, and consequently plata de fuego. Prior to 1651, the tk set does not provide sufficient data so I turn to other sources. For the period 1611 to 1650, I use the total silver produced as reported by TePaske, and the amount of mercury distributed to Zacatecas from late 1608 to 1649, as reported by Bakewell (figures in bold).4 I then proceed as above. For the period from 1590 to 1610, for which I have no information on mercury sales, I work in reverse. I assume the same patio process fraction of 0.7 as in the period 1611–1650, and apply it to TePaske’s figure on silver production. I then estimate the amount of mercury consumed based on the calculated amount of silver produced from the patio process and a deemed ratio of 2.05.

Caja of Guanajuato (Table c-ii) The projections are carried out along analogous lines as explained for Zacatecas.

Caja of México (Table c-iii) Though the silver tax records on silver from other regions are identified, which helps to avoid double accounting of their totals, mercury revenues are reported as an aggregate, with no such distinction. I have therefore opted to use Tepaske’s data on silver production for the Caja during the period 1521 to 1810, and have assumed that the patio process accounted for a fraction of 0.8, based on Figure 99 (Chapter 9), and a mercury to silver ratio of 1.8 to apply over the whole period.5 4 TePaske and Brown, Gold and Silver, 115–116. For the data on mercury see Bakewell, Silver Mining in Zacatecas, 251. 5 TePaske and Brown, Gold and Silver, 84–87, 115–116.

376

appendix c

Caja of Durango (Table c-iv) The average mercury to silver ratio, from 1689 to 1810, was 2.1. This ratio is then projected for the earlier timeframe of 1578 to 1688, and is used in combination with mercury sales, total silver refined (complemented by Lacueva’s data for the years 1578 to 1598), and the plata de azogue fraction projected for 1622 (0.97) applied to the previous period, to generate the estimated production of silver according to refining process.6

Caja of San Luis Potosí (Table c-v) The data on silver production by refining process begin to be recorded in the tk set as of 1712. Prior to that date, the data provides only total taxes paid on silver. Revenues from mercury sales only appear from 1672 onwards, and I am assuming smelting completely dominated production in the early years. From 1672 to 1710 I divide the weight of mercury sold by 1.7 (the average mercury to silver ratio from 1710 to 1770 according to the tk data that does not involve the cazo process) so as to arrive at a deemed weight of silver obtained by the patio process, and from there I project its corresponding silver fraction for each period.

Caja of Guadalajara (Table c-vi) The fraction of plata de azogue is calculated directly from the tk set from 1690 onwards. The average ratio of mercury to silver calculated for this period is 2.1. To estimate the fraction of plata de azogue for the earlier periods I apply this ratio to the data from the tk set on sales of mercury from 1611 to 1690. The average projected fraction of plata de azogue for this period is 0.7. I then apply this fraction to the period where I have no data on mercury sales, 1568 to 1611, to estimate a total amount of silver refined by the patio process, from where I obtain the deemed quantities of mercury consumed using the average mercury to silver ratio of 2.1.

Caja of Sombrerete (Table c-viii) The mercury to silver ratio during this period shows an interesting behaviour. On average it has a value of 2.1, which falls within the expected historical range. However,

6 Lacueva Muñoz, La plata del Rey, 397.

estimates of silver production by caja and refining process

377

when calculated by decade it shows an abnormal range after 1780, reaching the value of 9.2 from 1791 to 1800. The average from 1683 to 1780 is 1.6, but from 1781 to 1816 it increases to 4.8, at the same time the amount of silver produced by the patio process reaches a minimum. One explanation is that inexpensive mercury was being used in a very inefficient manner to attempt to refine by the patio process any lead-rich ores, thus leading to its waste from 1780 to 1810.7 During this period more liquid mercury would have been lost while calomel amounts reach their chemical ceiling value (figures in bold).

Caja of Bolaños (Table c-ix) The mercury to silver ratio again shows a step increase that coincides with the decrease in the price of mercury, from 1.9 prior to 1760 to an average of 3.2 from 1761 to 1804, reaching a value of 5.2 in the following decade. A non-efficient use of inexpensive mercury again would implicate a greater loss of liquid mercury instead of the production of calomel.

Caja of Rosario (Table c-x) Again the mercury to silver ratio is higher than the historical range for New Spain in the first two decades just after the price decrease of mercury, but not to the extent of potential waste observed in the previous two Cajas. The calomel projection is fixed at its ceiling value (figures in bold) during these decades.

Caja of Chihuahua (Table c-xii) The abnormal range of mercury to silver ratios up to the end of the eighteenth century places a ceiling on the possible value of calomel produced (numbers in bold), and

7 It could also be argued that an abnormally high mercury to silver ratio indicates a combination of bad practice and contraband of mercury. Why refiners would suddenly become extremely bad operators after the experience shown in the 1760s with mercury to silver ratios well below 2, or why it would make sense to contraband the least expensive mercury in the history of New Spain, weaken these alternative explanations. A lack of short-term correlation in the accounting figures between mercury sales and silver produced could also create these spikes in the mercury to silver ratio, but even adopting longer periods does not eliminate a higher range of ratios.

378

appendix c

increases the deemed loss of liquid mercury. The data in the tk set have an entry for the calendar year 1790 that indicates the purchase of 114,000 pesos of mercury, approximately 128,000kg of mercury. This amount of mercury does not correlate with the production level of silver in the previous years, so I have placed it as mercury consumed in the following period. There is a gap in the tk set on silver production from 1791 to 1796 which I fill in with TePaske’s figure of 55,029kg (figure in italic bold).8 table c-i

Period

Total silver kg

1591– 1,500,000 1610 1611– 1,650,000 1650 2/1651– 315,388 4/1661 5/1661– 242,674 4/1663, 5/1664– 6/1670 7/1670 to 670,575 6/1681 1/1686 to 200,432 2/1692 3/1692 to 250,211 12/1700 1/1701 to 316,089 12/1710

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Zacatecas. In this and following tables p: Patio process, s: Smelting.

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.7 0.7 0.5 0.3

0.4 0.6 0.5 0.5

p s p s p s p s

p s p s p s p s

Breakdown (kg) As calomel

1,050,000 2,152,500 1,829,625 450,000 1,185,402 2,430,075 2,065,564 464,598 144,558 296,345 251,893 170,829 68,658 140,750 119,637 174,015

268,230 402,345 120,259 80,173 125,105 125,105 158,044 158,044

8 TePaske and Brown, Gold and Silver, 104.

As As liquid volatilised mercury mercury

Mercury to silver ratio

301,350

21,525

2.05

340,211

24,301

2.05

41,488

2,963

2.05

19,705

1,407

2.05

468,508

398,232

65,591

4,685

1.7

218,731

185,921

30,622

2,187

1.8

281,986

239,688

39,478

2,820

2.3

310,918

264,280

43,529

3,109

2.0

estimates of silver production by caja and refining process

Period

Total silver kg

1/1711 to 536,709 12/1720 1/1721 to 544,886 12/1730 1/1731 to 415,326 12/1740 1/1741 to 349,798 12/1750, exc 1749 1/1751 to 313,580 12/1760 1/1761 to 241,316 12/1770 1/1771 to 494,484 12/1780 1/1781 to 559,775 12/1790 1/1791 to 606,623 12/1800 1/1801 to 575,722 12/1810 ex 1802 total 9,783,589

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.6 0.8 0.7 0.8

0.6 0.6 0.7 0.8 0.8 0.8

p s p s p s p s

322,025 214,683 435,909 108,977 290,728 124,598 279,839 69,960

p s p s p s p s p s p s

379

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

649,013

551,661

90,862

6,490

2.0

742,401

631,041

103,936

7,424

1.7

616,362

523,908

86,291

6,164

2.1

620,106

527,090

86,815

6,201

2.2

188,148 490,145 125,432 144,790 290,293 96,526 346,139 587,454 148,345 447,820 1,185,986 111,955 485,298 790,299 121,325 460,578 1,287,572 115,144

349,955

135,288

4,901

2.6

246,749

40,641

2,903

2.0

499,336

82,244

5,875

1.7

832,945

341,182

11,860

2.6

671,754

110,642

7,903

1.6

856,675

418,021

12,876

2.8

13,559,445 11,045,956 2,377,895 135,594

380

appendix c

table c-ii Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Guanajuato

Period

Total silver kg

5/1665 to 60,870 2/1671 3/1671 to 168,370 3/1681 4/1681 to 49,969 7/1684 6/1690 to 212,323 1/1701 2/1701 to 217,069 2/1710 3/1711 to 263,708 12/1720 1/1721 to 422,272 12/1730 1/1731 to 565,316 12/1740 1/1741 to 792,108 12/1750 1/1751 to 650,639 12/1760 1/1761 to 637,455 12/1770 1/1771 to 1,058,474 12/1780 1/1781 to 931,462 12/1790 1/1791 to 1,371,119 12/1800

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.70 0.70 0.73 0.65 0.65 0.59 0.64 0.64 0.58 0.63 0.69 0.73 0.78 0.83

p s p s p s p s p s p s p s p s p s p s p s p s p s p s

Breakdown (kg) As calomel

42,609 96,080 81,668 18,261 117,859 191,991 163,192 50,511 36,477 44,231 37,596 13,492 138,010 171,413 145,701 74,313 141,095 177,727 151,068 75,974 155,588 270,894 230,260 108,120 270,254 448,296 381,051 152,018 361,802 530,439 450,873 203,514 459,422 830,679 706,077 332,685 409,902 647,787 550,619 240,736 439,844 828,293 704,049 197,611 772,686 1,414,740 1,202,529 285,788 726,541 1,610,538 1,368,957 204,922 1,138,029 1,588,932 1,350,592 233,090

As As liquid volatilised mercury mercury

Mercury to silver ratio

13,451

961

2.3

26,879

1,920

1.6

6,192

442

1.2

23,998

1,714

1.2

24,882

1,777

1.3

37,925

2,709

1.7

62,761

4,483

1.7

74,261

5,304

1.5

116,295

8,307

1.8

90,690

6,478

1.6

115,961

8,283

1.9

198,064

14,147

1.8

225,475

16,105

2.2

222,450

15,889

1.4

381

estimates of silver production by caja and refining process

Period

Total silver kg

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

1/1801 to 815,637 12/1806 total 8,216,791

table c-iii

Period

0.84

p s

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

685,135 1,493,699 1,269,644 209,118 14,937 130,502 10,345,738 8,793,877 1,448,403 103,457

Mercury to silver ratio

2.2

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of México. Values in bold from footnote 5.

Total silver kg

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

1521 to 1560

770,410

0

1561 to 1600

2,251,290

0.8

1601 to 1700

2,783,840

0.8

1701 to 1810

2,897,520

0.8

total

8,703,060

p s p s p s p s

0 770,410 1,801,032 450,258 2,227,072 556,768 2,318,016 579,504

0

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

0

0

0

3,241,858 2,755,579

453,860

32,419

4,008,730 3,407,420

561,222

40,087

4,172,429 3,546,564

584,140

41,724

11,423,016 9,709,564 1,599,222

114,230

382

appendix c

table c-iv

Period

1578– 1598 1/1599 to 6/1611 7/1611 to 4/1615 5/1622 to 4/1625 6/1632 to 6/1641 7/1641 to 12/1650 1/1651 to 12/1659 exc 1654 1/1664 to 5/1673 6/1673 to 7/1677 1/1685 to 12/1688 1/1689 to 6/1700 7/1700 to 6/1711 7/1711 to 12/1720

Total silver kg

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Durango. Silver production figure in bold from footnote 6, ceiling for calomel estimates indicated in bold italic figures.

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

109,516

0.97

105,916

0.97

37,422

0.97

29,828

0.97

223,591

0.36

234,813

0.30

171,920

0.41

204,639

0.36

100,278

0.18

75,456

0.44

168,564

0.16

203,352

0.21

355,211

0.23

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

p s p s p s p s p s p s p s

106,231 3,285 102,739 3,177 36,299 1,123 28,845 984 80,938 142,653 69,274 165,539 70,368 101,552

223,084

189,621

31,232

2,231

2.1

215,752

183,389

30,205

2,158

2.1

76,228

64,794

10,672

762

2.1

60,574

51,488

8,480

606

2.1

169,969

144,474

23,796

1,700

2.1

145,476

123,655

20,367

1,455

2.1

147,773

125,607

20,688

1,478

2.1

p s p s p s p s p s p s

73,541 131,097 18,221 82,056 33,010 42,445 26,970 141,593 42,704 160,648 81,699 273,513

154,437

131,272

21,621

1,544

2.1

38,265

32,525

5,357

383

2.1

69,321

58,923

9,705

693

2.1

91,629

50,165

40,548

916

3.4

97,567

82,932

13,659

976

2.3

107,857

91,679

15,100

1,079

1.3

383

estimates of silver production by caja and refining process

Period

Total silver kg

1/1721 to 404,195 12/1730 1/1731 to 415,980 12/1740 exc. 1734 1/1741 to 532,680 12/1750 1/1752 to 428,258 12/1760 1/1761 to 518,147 12/1770 1/1771 to 478,270 12/1780 1/1781 to 342,124 12/1790 exc 1787 1/1791 to 474,878 12/1800 1/1801 to 444,626 12/1810 exc 1806 total 6,059,663

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.11 0.1

0.2 0.3 0.41 0.53 0.61

0.66 0.68

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

p s p s

44,461 359,734 41,598 374,382

51,731

43,972

7,242

517

1.2

44,639

37,943

6,249

446

1.1

p s p s p s p s p s

106,536 426,144 128,477 299,780 212,440 305,707 253,483 224,787 208,696 133,428

85,549

72,717

11,977

855

0.8

159,861

135,882

22,381

1,599

1.2

410,587

348,999

57,482

4,106

1.9

677,749

471,478 199,493

6,777

2.7

509,953

388,174 116,680

5,100

2.4

p s p s

313,419 161,458 302,346 142,280

735,234

624,949 102,933

7,352

2.3

705,805

599,934

98,813

7,058

2.3

4,979,041 4,054,570 874,680

49,790

384

appendix c

table c-v Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of San Luis Potosí

Period

up to 3/1629 2/1630 to 4/1640 12/1640 to 2/1651 12/1653 to 6/1661 7/1661 to 10/1671 12/1672 to 2/1675, 11/1677 to 4/1681 5/1681 to 7/1684, 3/1686 to 3/1688 1/1690 to 4/1701 1/1706 to 12/1710 1/1712 to 12/1720 1/1721 to 12/1730 1/1731 to 12/1740 1/1741 to 12/1748

Total silver kg

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

27,519

0

172,664

0

155,489

0

82,141

0

112,388

0

p s p s p s p s p s p s

0 27,519 0 172,664 0 155,489 0 82,141 0 112,388 3,396 72,129

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

5,774

4,908

808

58

1.7

75,525

0.04

70,374

0.10

p s

6,844 63,529

11,635

9,890

1,629

116

1.7

128,682

0.04

6,935

1,142

82

1.7

0.03

2,099

1,784

294

21

1.7

68,138

0.08

8,165

6,940

1,143

82

1.5

91,046

0.07

7,250

6,163

1,015

73

1.1

110,512

0.14

34,540

29,359

4,836

345

2.2

71,885

0.08

4,800 123,882 1,235 40,597 5,451 62,687 6,373 84,673 15,472 95,040 5,751 66,134

8,159

41,832

p s p s p s p s p s p s

11,788

10,020

1,650

118

2.0

385

estimates of silver production by caja and refining process

Period

Total silver kg

1/1752 to 252,907 12/1760 1/1761 to 186,820 12/1770 exc 1765 1/1771 to 318,643 12/1780 1/1781 to 738,851 12/1790 1/1791 to 815,894 12/1800 1/1801 to 424,969 12/1806 total 3,946,278

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.37 0.46

0.39 0.92 0.97 1

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

p s p s

93,576 159,332 85,937 100,883

188,859

160,530

26,440

1,889

2.0

133,103

113,138

18,634

1,331

1.5

p s c+p s c+p s c+p s

124,271 194,372 679,743 59,108 791,417 24,477 424,969 0

138,863

118,033

19,441

1,389

1.1

659,313

560,416

92,304

6,593

1.0

571,467

485,747

80,005

5,715

0.7

406,692

345,688

56,937

4,067

1.0

2,187,709 1,859,553 306,279

21,877

386

appendix c

table c-vi

Period

1568, 1578, 1579 1/1581 to 12/1590, exc. 1585, 1588, 1589 1/1591 to 3/1601 4/1601 to 3/1611, exc. 1605 4/1611 to 3/1621 4/1621 to 4/1631 5/1631 to 4/1641 5/1641 to 5/1651 6/1651 to 12/1660 1/1661 to 2/1671 3/1671 to 3/1681 4/1681 to 6/1690

Total silver kg

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Guadalajara

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

57,647

0.7

p s

40,353 17,294

84,390

71,731

11,815

844

2.1

37,100

0.7

p s

25,970 11,130

54,311

46,165

7,604

543

2.1

70,457

0.7

87,671

14,440

1,031

2.1

0.7

49,320 21,137 45,508 19,504

103,143

65,012

p s p s

95,171

80,895

13,324

952

2.1

80,698

0.7

98,599

16,240

1,160

2.1

0.6

151,261

128,572

21,177

1,513

2.1

88,119

0.9

162,430

138,066

22,740

1,624

2.1

85,969

0.6

104,486

88,813

14,628

1,045

2.1

136,504

0.7

213,927

181,838

29,950

2,139

2.1

175,092

0.4

163,555

139,022

22,898

1,636

2.1

187,842

0.6

223,607

190,066

31,305

2,236

2.1

181,195

0.7

55,468 25,231 72,329 42,532 77,670 10,450 49,962 36,006 102,294 34,209 78,208 96,885 106,923 80,919 120,816 60,379

115,999

114,861

p s p s p s p s p s p s p s p s

252,662

214,762

35,373

2,527

2.1

387

estimates of silver production by caja and refining process

Period

Total silver kg

7/1690 to 146,283 6/1701 ex mid 1693 to mid 1696 1/1701 to 154,299 12/1710 1/1711 to 170,740 12/1720 1/1721 to 172,303 12/1730 1/1731 to 199,184 12/1740 1/1741 to 204,671 12/1750 1/1751 to 184,246 12/1760 1/1761 to 296,484 12/1770 1/1771 to 304,919 12/1780 1/1781 to 252,230 12/1790 1/1791 to 215,008 12/1800 1801 to 78,527 1804 total 3,659,392

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

0.53

p s

76,919 69,364

144,062

122,453

20,169

1,441

1.9

0.66

p s p s p s p s p s p s p s p s p s p s p s

101,471 52,828 99,955 70,785 111,278 61,025 150,680 48,504 159,427 45,244 147,564 36,682 279,318 17,166 268,537 36,382 186,589 65,641 183,414 31,594 66,009 12,518

177,326

150,727

24,826

1,773

1.7

195,663

166,313

27,393

1,957

2.0

249,515

212,087

34,932

2,495

2.2

298,825

254,001

41,835

2,988

2.0

307,543

261,412

43,056

3,075

1.9

356,895

274,469

78,857

3,569

2.4

572,694

486,790

80,177

5,727

2.1

527,232

448,147

73,812

5,272

2.0

483,742

347,055 131,850

4,837

2.6

330,124

280,606

46,217

3,301

1.8

167,555

122,777

43,103

1,676

2.5

5,536,120 4,593,040 887,719

55,361

0.59 0.65 0.76 0.78 0.80 0.94 0.88 0.74 0.85 0.84

388

appendix c

table c-vii

Period

9/1667 to 2/1671 3/1671 to 1/1680 11/1680 to 5/1693 1/1706 to 12/1710 1/1711 to 12/1720 1/1721 to 12/1730 1/1731 to 12/1740 1/1741 to 12/1749 1/1751 to 12/1760 1/1761 to 12/1770 1/1771 to 12/1780 1/1781 to 12/1790 1/1791 to 12/1800

Total silver kg

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Pachuca

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

34,341

0.89

87,016

0.90

184,106

0.90

60,077

0.88

189,289

0.68

327,528

0.72

191,368

0.80

134,187

0.76

253,687

0.83

299,372

0.75

228,999

0.69

153,629

0.53

193,246

0.56

p s p s p s p s p s p s p s p s p s p s p s p s p s

30,553 3,788 77,996 9,020 165,695 18,411 52,859 7,218 128,283 61,007 236,640 90,888 152,517 38,851 101,799 32,388 210,418 43,269 223,331 76,040 158,277 70,722 82,008 71,621 108,221 85,025

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

52,305

44,460

7,323

523

1.7

154,706

131,500

21,659

1,547

2.0

376,266

319,826

52,677

3,763

2.3

85,314

72,517

11,944

853

1.6

255,624

217,281

35,787

2,556

2.0

479,629

407,685

67,148

4,796

2.0

295,395

251,086

41,355

2,954

1.9

209,256

177,868

29,296

2,093

2.1

341,768

290,503

47,847

3,418

1.6

341,426

290,212

47,800

3,414

1.5

299,692

254,738

41,957

2,997

1.9

279,349

0

0

2,793

3.4

246,137

209,217

34,459

2,461

2.3

389

estimates of silver production by caja and refining process

Period

1801 to 1804, 1806 total

Total silver kg

111,808

2,448,652

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.47

p s

52,269 59,539

171,709

Breakdown (kg) As calomel

0

As As liquid volatilised mercury mercury 0

1,717

3,588,578 2,666,892 439,253

35,886

Mercury to silver ratio

3.3

390

appendix c

table c-viii

Period

5/1683 to 5/1690, exc mid 1684– mid 1688 5/1690 to 5/1701 6/1701 to 3/1711 4/1711 to 12/1720 1/1721 to 12/1730 1/1731 to 12/1740 1/1741 to 12/1750 exc 1747, 1748 1/1753 to 12/1760 1/1761 to 12/1770 1/1771 to 12/1780 1/1781 to 12/1790 1/1791 to 12/1800

Total silver kg

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Sombrerete

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

Breakdown (kg) As As As calomel liquid volatilised mercury mercury

Mercury to silver ratio

119,115

0.2

p s

23,823 95,292

32,071

27,261

4,490

321

1.3

189,032

0.19

63,151

10,401

743

2.1

0.31

26,411

22,450

3,698

264

1.3

40,832

0.29

23,069

19,609

3,230

231

1.9

32,122

0.37

21,349

18,147

2,989

213

1.8

128,583

0.65

130,547 110,965

18,277

1,305

1.6

152,455

0.78

35,916 153,116 20,641 45,942 11,841 28,991 11,885 20,237 83,579 45,004 118,915 33,540

74,296

66,583

p s p s p s p s p s p s

184,837 157,111

25,877

1,848

1.6

40,616

0.76

45,723

0.71

118,356

0.58

107,826

0.39

183,605

0.09

p s p s p s p s p s

30,868 9,748 32,464 13,260 68,647 49,710 42,052 65,774 16,524 167,081

65,535

55,705

9,175

655

2.1

33,103

28,137

4,634

331

1.0

61,405

52,194

8,597

614

0.9

158,534

78,217

78,731

1,585

3.8

151,716

30,735 119,463

1,517

9.2

391

estimates of silver production by caja and refining process

Period

1/1801 to 12/1809 exc 1807 1/1811 to 12/1816 total

Total silver kg

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

Breakdown (kg) As As As calomel liquid volatilised mercury mercury

Mercury to silver ratio

339,262

0.10

p s

33,926 305,336

133,794

63,103

69,354

1,338

3.9

87,100

0.06

p s

5,226 81,874

12,567

10,682

1,759

126

2.4

1,109,234 737,467 360,675

11,092

1,651,212

392

appendix c

table c-ix

Period

Total silver kg

1/1753 to 460,551 12/1760 1/1761 to 166,019 12/1770 1/1771 to 214,718 12/1780 1/1781 to 218,996 12/1790 1/1791 to 108,061 12/1800 1/1801 to 10,479 12/1804 total 1,178,824

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Bolaños

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.99 0.93 0.89 0.92 0.93 0.82

p s p s p s p s p s p s

455,770 4,781 155,080 10,939 190,762 23,957 201,123 17,873 100,453 7,607 8,598 1,880

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

864,622

734,929 121,047

8,646

1.9

376,216

288,449

84,005

3,762

2.4

478,827

354,817 119,222

4,788

2.5

637,264

374,089 256,802

6,373

3.2

519,384

186,843 327,347

5,194

5.2

6,879

231

2.7

2,899,416 1,955,120 915,302

28,994

23,103

15,993

393

estimates of silver production by caja and refining process

table c-x Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Rosario

Period

Total silver kg

1/1770 to 87,354 12/1780 1/1781 to 268,810 12/1790 1/1791 to 373,230 12/1800 exc 1794 1/1801 to 360,896 12/1809 exc 1806 total 1,090,290

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

0.66 0.73 0.78

0.64

p s p s p s

57,654 29,700 196,231 72,579 291,119 82,111

p s

231,756 129,141

Mercury losses As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

149,415

107,236

40,685

1,494

2.6

559,577

364,990 188,991

5,596

2.9

585,950

498,057

82,033

5,859

2.0

490,704

417,099

68,699

4,907

2.1

1,785,646 1,387,382 380,408

17,856

394 table c-xi

appendix c Estimate of the breakdown of silver production by refining process for the Caja of Zimapán

Period

Total silver kg

Fraction Process Silver produced kg plata de azogue

1/1729 to 12/1740

67,661

0

1/1741 to 12/1748

60,545

0

1/1752 to 12/1759

71,005

0

1/1761 to 12/1770

106,210

0

1/1771 to 12/1780

150,355

0

1/1781 to 12/1790

125,489

0

1/1791 to 12/1800

143,017

0

1/1801 to 12/1806

74,912

total

799,194

0.02

p s p s p s p s p s p s p s p s

0 67,661 0 60,545 0 71,005 0 106,210 0 150,355 0 125,489 0 143,017 1,498 73,413

395

estimates of silver production by caja and refining process table c-xii

Period

Estimate of the breakdown of silver production by refining process, and magnitude of the environmental impact vectors for mercury in calomel, liquid mercury and volatile mercury, for the Caja of Chihuahua

Total silver kg

1/1785 to 32,882 12/1790 exc 1787 1/1791 to 55,029 12/1796 1/1797 to 39,496 12/1800 1/1801 to 93,014 12/1810 1/1811 to 25,023 12/1814 total 245,443

Fraction Process Silver Mercury plata de produced consumed azogue kg kg

Breakdown (kg) As calomel

As As liquid volatilised mercury mercury

Mercury to silver ratio

0.36

p s

11,837 21,044

76,709

22,018

53,925

767

6.5

0.34

p s p s p s p s

18,710 36,319 12,639 26,857 40,926 52,088 12,762 12,261

114,000

34,800

78,060

1,140

6.1

34,084

23,508

10,235

341

2.7

129,563

76,122

52,145

1,296

3.2

20,599

17,509

2,884

206

1.6

173,957 197,248

3,750

0.32 0.44 0.51

374,955

Glossary of Technical Terms in Spanish Afinación: second stage of smelting process where silver is separated from lead and litharge (greta) is produced (cupellation) Alcribís: see Tuyère Arrastre: circular grinding equipment using horizontal stones, powered by water or animal power Aviador: person who supplied materials on credit to miners and refiners Azogado: person intoxicated with mercury Azogar: to intoxicate with mercury Azogue: mercury Azoguería: room where mercury was handled (stored, weighed, extracted by squeezing through a cloth containing the amalgam) Azoguero: in New Spain, applied to the master in charge of the patio process Barra: bar of silver-enriched lead from first stage of smelting (see refinación) Beneficio: term applied to refining, ie beneficio de plata por azogue meant refining of silver ores using mercury, beneficio de plata por fuego, refining by smelting Bonanza: a very rich zone of precious metal in a deposit Caja: regional Treasury, usually but not always a mining district Canoas: rectangular vats used in New Spain in the sixteenth century, where refining by mercury was carried out Capellina: 1. Equipment to recover mercury from amalgam by heating, consisting of a metallic top cover and a base placed on the water channel that condensed mercury 2. In Guanajuato, also applied to the building that housed the capellina ensemble Caperuza: upper part of early version (16c) of equipment to recover mercury from amalgam, made from clay or metal Carbón: charcoal Cárcamo: channel to drain waste water laden with mineral silt, that ran through haciendas de patio Cazo: a copper pot or vessel, used by Alonso Barba for his cocimiento (cooking) process using mercury Cendrada: bone ash impregnated with litharge, material used in the cupel (vaso) that held the barras for the afinación Correspondencia: amount of silver, in marks, which was produced for every 100 quintales of mercury consumed Desazogadera: equipment to recover mercury from amalgam by heating under controlled conditions, see capellina Desmontes: tailings from a mine Diezmo: tax of a tenth applied to silver registered at each local Treasury (Caja)

glossary of technical terms in spanish

397

Fundición: smelting of silver compounds in the presence of lead to obtain elemental silver, which is then absorbed by the molten lead. The silver-enriched lead is cast into barras (pigs) Galena: lead sulphide (PbS), which can contain silver that can be extracted by smelting Grasas: slag from smelting furnace Greta: litharge, lead oxide (PbO) Hacienda: original term referred to the creation of wealth, was then applied initially to silver refining units in New Spain (Hacienda de beneficio, Hacienda de patio, called ingenios in Peru) and to agricultural and livestock economic units Horno Castellano: initially very simple smelting furnaces, in the form of a pillar with a square or circular cross-section, built from mortar and stones and with a low chimney outlet Ingenio: originally refers to a machine, and in Peru was used to denote a silver refining facility (an hacienda in New Spain). Lavado de metales: concentrating the silver content of an ore by washing away the less dense fractions with water. In English the term is dressing. Maestrazgo: land and mining rents to the Spanish Crown from territories that historically were under the control of Spanish military orders Manga: vertical cloth filter used to squeeze excess mercury from amalgam Maquila: business model whereby an hacienda refines silver ores that belong to third parties, for a fee that covers its operational costs plus a profit margin Merced: a royal grant awarded to technical innovations by the Spanish authorities. Metal: ore Minero: applied both to miners and refiners of silver ores Molino: circular stone set on its edge and driven by water or animal power, used to crush ore Montón: literally mound, was a unit of measure in the patio reactor, thus a torta at Regla was composed of 20 montones, and each montón represented 30 cargas (see Guide to the text). These are not universal values and can vary according to mining location. Mortero: mill that uses stamp-heads made of stone or metal to crush ore, driven by human, animal or water power Negrillos: term used to denote the darker and deeper silver ores in a deposit, mainly silver sulphide compounds Patio: the courtyard that functioned as a chemical reactor, where tortas were spread out until the silver refining process was deemed completed Planillas: inclined planes to separate entrained amalgam, mercury or silver ore from the washings of the torta Planilleros: workers stationed at the planillas Plata de azogue: silver refined using mercury Plata de fuego: silver refined by smelting

398

glossary of technical terms in spanish

Real de Minas: legally established mining interests and community recognised by the Spanish Crown and subject to its legislation Señoreage: duty paid for coining silver Solimán: mercuric chloride (HgCl2) Tahona or taona: see arrastre Torta: ore slurry in the shape of a flat cake Tuyère: element in back furnace wall with orifice to hold the nozzle (cañón) of the bellows Vaso: term used for the ensemble of reverberatory oven and cupel used during the afinación Zangarro: patio process refining unit smaller than the hacienda

Archival Sources Archivo General de la Nación, Ciudad de México (agn) Instituciones Coloniales / Minería

Archivo Histórico de la Compañía de Minas del Real del Monte y Pachuca (AHCRMyP) Fondo Siglo xix Sección: Explotación y Beneficio Serie: Informes de Haciendas de Beneficio Subserie: Informes Mensuales Hacienda de Regla Vol. 225, Exp. 3: 29 Jun 1872–27 Oct 1888 Sección: Negociaciones Serie: Haciendas de Beneficio Subserie: Hacienda de Regla i, Vol. 22: 1875–1878 Sección: Correspondencia, Serie: Compañía a Varios, Subserie: Correspondencia General Sección: Contabilidad de la Dirección Serie: Producción y Gastos Subserie: Gastos de Haciendas de Beneficio, Marzo 1871–Nov 1877 Gastos de Haciendas de Beneficio, Sept. 1896–Sept 1902 Estados Comparativos, 1853–1855, 1859–1865, 1869–1873. Sección: Administración Interna Serie: Departamento de Ingenieros Subserie: Croquis y Planos Sección: Ymporte de la Memoria de la Mina Serie: Informes de Minas i, Vol. 13

400

archival sources

Archivo Histórico del Estado de Zacatecas (ahez) Notarías/Colonia Poder Judicial-Civil Real Hacienda Serie Civil Fondo Mapas e Ilustraciones, Serie v: Planos Siglos xviii al xx

Archivo Histórico de San Luis Potosí (ahslp) Fondo Alcaldía Mayor Colección Miguel Iwadare Colección Powell

Archivo Histórico de la Universidad de Guanajuato (ahug) Protocolo de Minas Minería Actas de Cabildo Bienes Difuntos Mapoteca

Mapoteca Manuel Orozco y Berra, Ciudad de México (mmob) Colección General

British Library, London, u.k. (bl) General Reference Collection

Bibliography Agricola, Georgius. De re metallica. Translated by Herbert Clark Hoover and Lou Henry Hoover. New York: Dover Publications, 1950. Agricola, Georgius. Bermannus. Translated by Robert Halleux and Albert Yans. Paris: Belles Lettres, 1990. Aitchison, Leslie. History of Metals. London: Macdonald & Evans, 1960. Alonso Barba, Alvaro. Arte de los metales. Barcelona: Editorial Labor, 1977. Alonso, Ricardo N. Diccionario Minero. Glosario de voces utilizadas por los mineros de Iberoamérica. Madrid: Consejo Superior de Investigaciones Científicas, 1995. Amador, Manuel. Tratado práctico y completo de trabajos de minas y haciendas de beneficio. México: E. Sánchez, Editor, 1901. Anonymous. “Journal Descriptive of the Route from New York to Real del Monte by Way of Tampico.” The London Magazine (1826): 145–171. Antúnez Echegaray, Francisco. Monografía histórica y minera sobre el Distrito de Guanajuato. México: Consejo de Recursos Naturales No-Renovables, 1964. Arzáns de Orsúa y Vela, Bartolomé. Historia de la villa imperial de Potosí. Edited by Lewis Hanke and Gunnar Mendoza L. Providence: Brown University Press, 1965. Bachmann, Hans-Gert. “Archäometallurgische Untersuchungen zur antiken Silbergewinnung in Laurion. ii. Charakterisierung von Bleiverhüttungsschlacken aus Laurion= Archaeometallurgical Investigation on Ancient Silver Smelting at Laurion. ii. Characterisation of Lead Smelting Slags from Laurion.” Erzmetall 35, no. 5 (1982): 246–251. Bachmann, Hans-Gert. The Identification of Slags from Archaeological Sites. London: Institute of Archaeology, 1982. Bailly-Maître, Marie-Christine. L’argent: du minerai au pouvoir dans la France médiévale. Paris: Picard, 2002. Bakewell, Peter J. Silver Mining and Society in Colonial Mexico: Zacatecas, 1546–1700. Cambridge, u.k.: University Press, 1971. Bakewell, Peter J. “Registered Silver Production in the Potosí District 1550–1735.” Jahrbuch für Geschichte Lateinamerikas 12 (1975): 67–103. Bakewell, Peter J. Miners of the Red Mountain. Indian Labor in Potosí, 1545–1650. Albuquerque: University of New Mexico Press, 1984. Bakewell, Peter J. “Mining in Colonial Spanish America.” In The Cambridge History of Latin America, edited by Leslie Bethell, 105–151. Cambridge: Cambridge University Press, 1984. Bakewell, Peter J. “The First Refining Mills in Potosí: Design and Construction.” In In Quest of Mineral Wealth: Aboriginal and Colonial Mining and Metallurgy in Spanish America., edited by Alan K. Craig and Robert C. West, 287–302. Baton Rouge: Geoscience Publishers, 1994.

402

bibliography

Bakewell, Peter J. “Notes on the Mexican Silver Mining Industry in the 1590’s.” In Mines of Silver and Gold in the Americas, edited by Peter J. Bakewell, 383–409. Aldershot (gb); Brookfield (Vt.): Variorum, 1996. Bakewell, Peter J. “La transferencia de la tecnología y la minería hispanoamericana, siglos xvi y xvii: algunas observaciones.” In Hombres, técnica, plata: minería y sociedad en Europa y América, siglos xvi–xix, edited by Julio Sánchez Gómez, Guillermo Mira Delli-Zotti and Francisco A. Rubio Durán, 363–375. Sevilla: Aconcagua Libros, 2000. Bakewell, Peter J. A History of Latin America. 2nd ed. Oxford: Blackwell Publishing, 2004. Bargalló, Modesto. La minería y la metalurgia en la América española durante la época colonial. México: Fondo de Cultura Económica, 1955. Bargalló, Modesto. La química inorgánica y el beneficio de los metales en el México prehispánico y colonial. México: Universidad Nacional Autónoma de México, 1966. Bartels, Christoph. “The Production of Silver, Copper and Lead in the Harz Mountains from Late Medieval Times to the Onset of Industrialization.” In Materials and Expertise in Early Modern Europe: Between Market and Laboratory, edited by Ursula Klein and E.C. Spary, 71–100. Chicago: University of Chicago Press, 2010. Berthe, Jean Pierre. “Le mercure et l’industrie mexicaine au xviéme siècle.” In Ciencia, vida y espacio en Iberoamérica: trabajos del programa movilizador del c.s.i.c., edited by José Luis Peset Reig, 141–149. Madrid: Consejo Superior de Investigaciones Científicas, 1989. Biringuccio, Vannoccio. The Pirotechnia. Translated by Cyril Stanley Smith and Martha Teach Gnudi. Cambridge, Mass.: The m.i.t. Press, 1966. Blanchard, Ian. “England and the International Bullion Crisis of the 1550s.” In Precious Metals in the Age of Expansion: Papers of the xivth International Congress of the Historical Sciences, edited by Hermann Kellenbenz, 87–118. Stuttgart: Klett-Cotta, 1981. Blanchard, Ian. Russia’s “Age of Silver”. Precious-metal Production and Economic Growth in the Eighteenth Century. London; New York: Routledge, 1989. Blanchard, Ian. “Technical Implications of the Transition from Silver to Lead Smelting in Twelfth Century Britain.” In Boles and Smeltmills Seminar, edited by Lynn Willies and David Cranstone, 9–11. Reeth, Yorkshire: Historical Society, Ltd., 1992. Blanchard, Ian. Mining, Metallurgy and Minting in the Middle Ages. Continuing AfroAmerican Supremacy 1250–1450. Vol. 3. Munich: Franz Steiner Verlag Stuttgart, 2005. Bleiberg, German. “El “informe secreto” de Mateo Alemán sobre el trabajo forzoso en la Minas de Almadén.” Estudios de Historia Social, no. 2–3 (1977): 357–443. Bordeaux, Albert. Le Mexique et ses mines d’argent. Paris: Plon-Nourrit et cie., 1910. Born, Inigo. Baron Inigo Born’s New Process of Amalgamation of Gold and Silver Ores, and Other Metallic Mixtures: As by His Late Imperial Majesty’s Commands Introduced in Hungary and Bohemia. Translated by R.E. Raspe. London: T. Cadell, 1791.

bibliography

403

Bowden, Mark. Furness Iron: the Physical Remains of the Iron Industry and Related Woodland Industries of Furness and Southern Lakeland. Swindon: English Heritage, 2000. Brading, D.A., and Harry E. Cross. “Colonial Silver Mining: Mexico and Peru.” The Hispanic American Historical Review 52, no. 4 (1972): 545–579. Brading, D.A. Miners and Merchants in Bourbon Mexico 1763–1810. London: Cambridge University Press, 1971. Braudel, Fernand. The Wheels of Commerce. Translated by Sian Reynolds. London: Collins/Fontana Press, 1988. Braunstein, P. “Les entreprises minières en Vénétie au xve siècle.” Melanges d’archéologie et d’histoire 77, no. 2 (1965): 529–607. Brocklehurst, Thomas Unett. Mexico To-day: a Country with a Great Future, and a Glance at the Prehistoric Remains and Antiquities of the Montezumas. London: J. Murray, 1883. Brown, Kendall W. “Workers Health and Colonial Mercury Mining at Huancavelica, Peru.” The Americas 57 (2001): 467–496. Brown, Kendall W. A History of Mining in Latin America: From the Colonial Era to the Present. Albuquerque: University of New Mexico Press, 2012. Buchan, John H. Report of the Real del Monte Company, Mexico. London: Taylor, Printer, 1855. Burkart, J. “Du filon et des mines de Veta-Grande, près de la ville de Zacatecas, dans l’état du même nom, au Mexique.” Annales des Mines, 3eme série 8 (1835): 55– 87. Burkart, Johann. “Memoria sobre la explotación de minas de los distritos de Pachuca y Real del Monte de México.” Anales de la Minería Mexicana (Revista de Minas) i (1861). Burt, Roger. John Taylor. Mining Entrepeneur and Engineer 1779–1863. Hartington: Moorland Publishing Company, 1977. Calderón de la Barca, Fanny. Life in Mexico; the Letters of Fanny Calderón de la Barca, with New Material from the Author’s Private Journals. Edited by Howard T. Fisher and Marion Hall Fisher. Garden City, n.y.: Doubleday, 1966. Canterla y Martín de Tovar, Francisco. Vida y obra del primer Conde de Regla. Sevilla: Escuela de estudios hispano-americanos, 1975. Capoche, Luis. “Relación General de la Villa Imperial de Potosí.” In Biblioteca de Autores Españoles, edited by Lewis Hanke 115–189. Madrid: Atlas, 1959. Cárdenas, Juan de. Primera parte de los problemas y secretos maravillosos de las Indias. México: Academia Nacional de Medicina, 1980. Cargill, S.M., D.H. Root, and E.H. Bailey. “Resource Estimation from Historical Data: Mercury, A Test Case.” Mathematical Geology 12, no. 5 (1980): 489–522. Casasus, Joaquin D. La Question de l’argent au Mexique. Paris: impr. de Chaix, 1892. Castillo Martos, Manuel. “Primeros beneficios de la plata por amalgamación en la

404

bibliography

América colonial (1565–1600).” In Minería y metalurgia. Intercambio tecnológico y cultural entre América y Europa durante el período colonial español, edited by Manuel Castillo Martos, 375–406. Sevilla, Bogotá: Muñoz Montoya y Montraveta Editores, 1994. Castillo Martos, Manuel. Bartolomé de Medina y el siglo xvi. Santander: Servicio de Publicaciones de la Universidad de Cantabria, 2006. Castillo Martos, Manuel. “Alquimia en la metalurgia de plata y oro en Europa y América.” In Informes para obtener plata y azoque en el mundo hispánico, xiii–lxix. Granada: Universidad de Granada, 2008. Castillo Martos, Manuel, and M.F. Lang. Metales preciosos-unión de dos mundos: tecnología, comercio y política de la minería y metalurgia iberoamericana. Sevilla: Muñoz Moya y Montraveta Editores, 1995. Chesterton, G.K. Father Brown, Selected Stories. London: Oxford University Press, 1961. Cobo (s.j.), Bernabé. Historia del Nuevo Mundo. Biblioteca de Autores Españoles. edited by Francisco Mateos (s.j.). Madrid: Atlas, 1964. Coll-Hurtado, Atlántida, María Teresa Sánchez-Salazar, and Josefina Morales. La minería en México: geografía, historia, economía y medio ambiente. México: unam, 2002. Collins, Henry F. The Metallurgy of Lead & Silver. London: Griffin & Co., 1899. Constable, Olivia Remie. Trade and Traders in Muslim Spain: the Commercial Realignment of the Iberian Peninsula, 900–1500. Cambridge; New York: Cambridge University Press, 1994. Contreras, Manuel Maria. “Empleo de los ensayes de pella y de residuos para determinar los adelantos y fin de la amalgamación de la plata en el beneficio de patio.” In Historia de la ciencia en México: Estudios y textos., edited by Elías Trabulse 184–196. México: Conacyt y Fondo de Cultura Economica, 2003. Cooke, C.A., P.H. Balcom, C. Kerfoot, M.B. Abbott, and A.P. Wolfe. “Pre-Colombian Mercury Pollution Associated with the Smelting of Argentiferous Ores in the Bolivian Andes.” ambio: A Journal of the Human Environment 40, no. 1 (2011): 18–25. Cooke, Colin A., Prentiss H. Balcom, Harald Biester, and Alexander P. Wolfe. “Over three millennia of mercury pollution in the Peruvian Andes.” Proceedings of the National Academy of Sciences 106, no. 22 (June 2, 2009 2009): 8830–8834. Cotton, S.A. Chemistry of Precious Metals. Bristol: Blackie Academic & Professional, 1997. Couturier, Edith Boorstein. The Silver King: the Remarkable Life of the Count of Regla in Colonial Mexico. Albuquerque: University of New Mexico Press, 2003. Craddock, P.T. Early Metal Mining and Production. Washington, d.c.: Smithsonian Institution Press, 1995. Cranstone, David. “Excavations at Old Gang Smeltmills: an Interim Report.” In Boles and Smeltmills: Report of a seminar on the History and Archaeology of Lead Smelting,

bibliography

405

edited by Lynn Willies and David Cranstone, 28–31. Reeth, Yorkshire: Historical Metallurgical Society, 1992. Cubillo Moreno, G. “Los dominios de la plata: el precio del auge, el peso del poder. Empresarios y trabajadores en las minas de Pachuca y Zimapán, 1552–1620.” México: Instituto Nacional de Antropología e Historia/Consejo Nacional para la Cultura y las Artes, 1991. Cunill Grau, Pedro. “El paisaje andino: Punas, Salares y Cerros.” In Potosí: plata para Europa edited by José Villa Rodríguez, 75–103. Sevilla: Universidad de Sevilla, 2000. Cunningham, C.G., H. Aparicio, F. Murillo, N. Jimenez, E.H. McKee, G.E. Ericksen, and F. Tavera. “Relationship between the Porco, Bolivia, Ag-Zn-Pb-Sn deposit and the Porco caldera.” Economic Geology 89, no. 8 (1994): 1833–1841. Dahlgren, Charles Bunker. Historic Mines of Mexico. A Review of the Mines of that Republic for the Past Three Centuries. New York: Printed for the author, 1883. Day, John. “The Great Bullion Famine of the Fifteenth Century.” Past & Present, no. 79 (1978): 3–54. de Acosta, José. Historia natural y moral de las Indias. Edited by Fermín del Pino. Madrid: Consejo Superior de Investigaciones Científicas, 2008. de Gamboa, Francisco Xavier. Comentarios a las Ordenanzas de Minas. Madrid, Oficina de Joaquín Ibarra. 1761. de Mesa, José, and Teresa Gisbert. “Oruro. Origen de una villa minera”. In La minería hispana e iberoamericana. Ponencias del i coloquio internacional sobre historia de la minería., 559–589. León: Cátedra de San Isidro, 1970. d’Oliveira, M.V.N., E.C. Alvarado, J.C. Santos, and J.A. Carvalho Jr. “Forest Natural Regeneration and Biomass Production after Slash and Burn in a Seasonally Dry Forest in the Southern Brazilian Amazon.” Forest Ecology and Management (2011): 1490–1498. de Oñate, Juan. Nuevas leyes de las minas de España: 1625 edición de Juan de Oñate: con tratado de re Metalica de Juan de Oñate. Santa Fe, New Mexico: Sunstone Press, 1998. d’ Orbigny, Alcide Dessalines. Malerische Reise in Süd- und Nordamerika. Leipzig: Baumgärtners Buchhandlung, 1839. de Sarría, Francisco Javier. Ensayo de metalurgia. México: D. Felipe de Zúñiga y Ontiveros, 1784. De Quille, Dan. The Big Bonanza. London: Eyre & Spottiswoode, 1947. De Quille, Dan. A History of the Comstock Silver Lode & Mines, Nevada and the Great Basin Region: Lake Tahoe and the high Sierras … The Mineral and Agricultural Resources of “Silverland”. Virginia, Nevada: F. Boegle, c. 1889. Díaz del Castillo, Bernal. Historia verdadera de la conquista de la Nueva España: Manuscrito “Guatemala”. México: Colegio de México: Universidad Nacional Autónoma de México, 2005. Dickinson, William R. “Anatomy and Global Context of the North American Cordillera.” In Backbone of the Americas: Shallow Subduction, Plateau Uplift, and Ridge and

406

bibliography

Terrane Collision, edited by Suzanne Mahlburg Kay, Víctor A. Ramos and William R. Dickinson, 1–29. Boulder, Colo.: Geological Society of America, 2009. Diringer, Sarah, William Pan, Beth Feingold, Ernesto Ortiz, and Heileen Hsu-Kim. “Mercury Biogeochemistry and Artisanal and Small-Scale Gold Mining in Madre de Dios, Peru.” Poster, 11th International Conference on Mercury as a Global Pollutant, July 28-August 2, Edinburgh 2013. Dobado, Rafael, and Gustavo A. Marrero. “The Role of the Spanish Imperial State in the Mining-led Growth of Bourbon Mexico’s Economy.” The Economic History Review 64, no. 3 (2011): 855–884. Domínguez de la Fuente, Manuel José. Leal Informe Político-Legal (Observaciones sobre la maniobra de las minas, hechas en el Real de Guanajuato en 1774). Guanajuato: Ediciones La Rana, 1999. Drelichman, Mauricio, and Hans-Joachim Voth. “Institutions and the Resource Curse in Early Modern Spain.” In Institutions and Economic Performance, edited by Elhanan Helpman, 120–147. Cambridge, Mass.: Harvard University Press, 2008. Driscoll, Charles T., Robert P. Mason, Hing Man Chan, Daniel J. Jacob, and Nicola Pirrone. “Mercury as a Global Pollutant: Sources, Pathways, and Effects.”Environmental Science & Technology 47, no. 10 (2013): 4967–4983. Duport, Saint Clair. De la production des métaux précieux au Mexique, considérée dans ses rapports avec la géologie, la métallurgie et l’ économie politique. Paris: Firmin Didot Frères, 1843. Egleston, Thomas. The Metallurgy of Silver, Gold, and Mercury in the United States. London; New York: J. Wiley & Sons, 1887. Eissler, Manuel. The Metallurgy of Argentiferous Lead. London: Crosby Lockwood & Son, 1891. Eltis, Davis, and David Richardson. Atlas of the Transatlantic Slave Trade. New Haven, London: Yale University Press, 2010. Emmons, William Harvey. The Enrichment of Ore Deposits. Washington: United States Geological Survey, 1917. English, Henry. A General Guide to the Companies Formed for Working Foreign Mines with their Prospectuses, Amount of Capital, Number of Shares, Names of Directors, &c. London: Boosey & Sons, 1825. Engstrom, Daniel R., William F. Fitzgerald, Colin A. Cooke, Carl H. Lamborg, Paul E. Drevnick, Edward B. Swain, Steven J. Balogh, and Prentiss H. Balcom. “Atmospheric Hg Emissions from Preindustrial Gold and Silver Extraction in the Americas: A Reevaluation from Lake-Sediment Archives.” Environmental Science & Technology 48, no. 12 (2014): 6533–6543. Epstein, Stephan R. “Transferring Technical Knowledge and Innovating in Europe, c.1200–c.1800.” Paper presented at the Conference on Endogeneous Institutional Change, Stanford Institute for Economic Policy Research, March 4–5, 2005.

bibliography

407

Ercker, L. Treatise on Ores and Assaying. Translated by Anneliese Grünhaldt Sisco and Cyril Stanley Smith. Chicago: University of Chicago Press, 1951. Eslava Galán, Juan. Cinco tratados españoles de alquimia. Madrid: Tecnos, 1987. Espinosa Pitman, Alejandro. José Antonio de Villaseñor y Sánchez, 1703–1759. San Luis Potosí, s.l.p., México: Universidad Autónoma de San Luis Potosí, 2003. Evans, Anthony M. Ore Geology and Industrial Minerals: An Introduction. Oxford; Boston: Blackwell Scientific Publications, 1992. Ewald, Ursula. The Mexican Salt Industry, 1560–1980: A Study in Change. Stuttgart; New York: G. Fischer, 1985. Fabbrizzi, L. “Communicating about Matter with Symbols: Evolving from Alchemy to Chemistry.” Journal of Chemical Education 85, no. 11 (2008): 1501–1511. Fabry, José Antonio. Compendiosa demostracion de los crecidos adelantamientos, que pudiera lograr la real haciencda de su Magestad mediante la rebaja en el precio del azogue que se consume para el laborio de las minas de este reyno … con una previa impugnacion a las reflexiones del contador Joseph de Villa-señor y Sanchez. México: Impressa por la viuda de J.B. de Hogal, 1743. Fernández Alcaide, Marta. Cartas de particulares en Indias del siglo xvi: edición y estudio discursivo. Madrid; Frankfurt am Main: Iberoamericana; Vervuert, 2009. Fernández de Oviedo y Valdés, Gonzalo. Historia general y natural de las Indias. Madrid: Ediciones Atlas, 1959. Fernández de Pinedo y Fernández, E. “Influencias recíprocas de las técnicas extractivas entre la minería vasca y la americana en la Edad Moderna.” Areas, no. 16 (1994): 35– 46. Fernández del Castillo, Francisco. “Algunos documentos nuevos sobre Bartolomé de Medina.” In Memorias de la Sociedad Alzate, 207–251. México: Talleres Gráficos de la Nación, 1927. Fernández Pérez, Joaquín. “La amalgamación de los minerales de plata.” In El oro y la plata de las Indias en la época de los Austrias, edited by Concepción Lopezosa Aparicio, 141–155. Madrid: Fundación ico, 1999. Findlay, Ronald, and Kevin H. O’Rourke. Power and Plenty. Trade, War and the World Economy in the Second Millenium. Princeton: Princeton University Press, 2007. Finlay, Robert. The Pilgrim Art. Cultures of Porcelain in World History. Berkeley, Los Angeles: University of California Press, 2010. Flores Clair, Eduardo, Cuauhtémoc Velasco Avila, and Elia Ramírez Bautista. Estadísticas mineras de México en el siglo xix, Vol. ii. México: Instituto Nacional de Antropología e Historia, 1985. Flores, Teodoro. Yacimientos minerales de la República Mexicana, con algunos datos relativos a su producción. México: Instituto Geológico de México, 1933. Flynn, Dennis O., and Arturo Giraldez. “Cycles of Silver: Global Economic Unity through the Mid-Eighteenth Century.” Journal of World History 13 (2002): 391–427.

408

bibliography

Flynn, Dennis O. “Comparing the Tokugawa Shogunate with Hapsburg Spain: two silver-based emprires in a global setting”. In The Political Economy of Merchant Empires: State Power and World Trade, 1350–1750 edited by James D. Tracy, 332–359. Cambridge: Cambridge University Press, 1991. Food and Agriculture Organization of the United Nations, Global Forest Resources Assessment 2010. Rome, 2010. Fortey, Richard A. Earth: An Intimate History. London: Harper Perennial, 2005. Frank, Andre Gunder. ReOrient: Global Economy in the Asian Age. Berkeley: University of California Press, 1998. Freestone, I.C., A.P. Middleton, P.T. Craddock, L.K. Gurjar, and D.R. Hook. “Role of Materials Analysis in the Reconstruction of Early Metal Extraction Technology: Zinc and Silver-Lead Smelting at Zawar, Rajasthan.” Materials Research Society Symposia Proceedings 185 (1990): 617–626. Friede, Juan. “La introducción de mineros alemanes en América por la compañía Welser de Augsburgo.” Revista de Historia de América, no. 51 (1961): 99–104. Gale, N.H., and Z.A. Stos-Gale. “Cycladic Lead and Silver Metallurgy.” The Annual of the British School at Athens 76 (1981): 169–224. Galindo y R, José J. El distrito minero Pachuca-Real del Monte. [Pachuca?]: Cia. de Real del Monte y Pachuca, 1957. Galván Arellano, Alejandro. Arquitectura y urbanismo de la ciudad de San Luis Potosí en el siglo xvii. San Luis Potosí: Editorial Universitaria Potosina, 1999. Garcés y Eguía, Joseph. Nueva teórica y práctica del beneficio de los metales de oro y plata por fundición y amalgamación. México: D. Mariano de Zúñiga y Ontiveros, 1802. García de Llanos. Diccionario y maneras de hablar que se usan en las minas y sus labores en los Ingenios y beneficios de los metales (1609). La Paz, Bolivia: musef, 1983. García Mendoza, Jaime. “La administración de las minas de plata y haciendas de beneficio de la familia Sandoval en Taxco (1562–1564).” In La plata en Iberoamérica: Siglos xvi al xix, edited by Jesús Paniagua Pérez and Nuria Salazar Simarro, 39–59. León: Universidad de León, 2008. Garner, Richard L. “Long-Term Silver Mining Trends in Spanish America: A Comparative Analysis of Peru and Mexico.” The American Historical Review 93, no. 4 (1988): 898–935. Garner, Richard L., and Spiro E. Stefanou. Economic Growth and Change in Bourbon Mexico. Gainesville: University Press of Florida, 1993. Gasparrini, Claudia. “The Mineralogy of Silver and its Significance in Metal Extraction.” cim Bulletin 77, no. 866 (1984): 99–110. Gill, Michael C. “Analysis of Lead Slags.” In Boles and Smeltmills Seminar, edited by Lynn Willies and David Cranstone, 51–53. Reeth, Yorkshire: Historical Society, Ltd., 1992. Giraldez, Arturo. The Age of Trade. The Manila Galleons and the Dawn of the Global Economy. Lanham: Rowman & Littlefield, 2015.

bibliography

409

Gleeson, Janet. The Arcanum. London: Bantam Press, 1998. Goldschmidt, V.M. Geochemistry. Oxford: Clarendon Press, 1954. Gómez de Cervantes, Gonzalo. La vida económica y social de Nueva España al finalizar el siglo xvi. México: Antigua Librería Robredo de José Porrúa e hijos, 1944. González Reyna, Jenaro. Minería y Riqueza de México. México: Banco de México, 1944. González, Tomás. Noticia histórica documentada de las célebres minas de Guadalcanal, desde su descubrimiento en el año 1555, hasta que dejaron de labrarse por cuenta de la Real Hacienda. Tomo Segundo. Madrid: Don Miguel de Burgos, 1831. González Valdez, Eduardo, Ezequiel González Reyes, Carlos Bedolla Cedeño, Edith Lorena Arrollo Ordaz, and Eduardo Manzanares Acuña. “Niveles de plomo en sangre y factores de riesgo por envenenamiento de plomo en niños mexicanos.” Revista de la Facultad de Ingeniería de la Universidad de Antioquia, no. 43 (2008): 114–119. Goodman, David. “Philip ii’s Patronage of Science and Engineering.” The British Journal for the History of Science 16, no. 1 (1983): 49–66. Graybeal, F.T., and D.M. Smith Jr. “Regional Distribution of Silver Deposits on the Pacific Rim.” In Silver—Exploration, Mining and Treatment, 3–10. London: Institution of Mining and Metallurgy, 1988. Graybeal, Frederick T., Douglas M. Smith, and Peter G. Vikre. “The Geology of Silver Deposits.” In Handbook of Strata-Bound and Stratiform Ore Deposits. Part iv, edited by K.H. Wolf, 1–184. Amsterdam: Elsevier 1986. Guerrero, Saul. “Chemistry as a Tool for Historical Research: Identifying Paths of Historical Mercury Pollution in the Hispanic New World.” Bulletin for the History of Chemistry 37, no. 2 (2012): 61–71. Guerrero, Saul. “Chemistry as a Tool for Historical Research: Estimating the Contraband of Silver from Potosí and Oruro, 1576–1650.” Bulletin for the History of Chemistry 37, no. 2 (2012): 72–81. Guerrero, Saul. “The Environmental History of Silver Refining in New Spain and Mexico, 16c to 19c: A Shift of Paradigm.” PhD diss., McGill University, 2015. Guerrero, Saul. “The History of Silver Refining in New Spain, 16c to 19c: back to the basics”. History and Technology 32, no. 1 (2016): 2–32. Hagan, Nicole, Nicholas Robins, Heileen Hsu-Kim, Susan Halabi, Mark Morris, George Woodall, Tong Zhang, et al. “Estimating historical atmospheric mercury concentrations from silver mining and their legacies in present-day surface soil in Potosí, Bolivia.” Atmospheric Environment 45, no. 40 (2011): 7619–7626. Hall, Cris M., Pablo L. Higueras, Stephen E. Kesler, Rosario Lunar, Hailing Dong, and Alex N. Halliday. “Dating of Alteration Episodes Related to Mercury Mineralization in the Almadén district, Spain.”Earth and Planetary Science Letters 148 no. 1–2 (1997): 287–298. Hausberger, Bernd. “Una iniciativa ecológica contra la industria minera en Chihuahua (1732).” Estudios de Historia Novohispana 13, no. 13 (1992): 115–134.

410

bibliography

Hausberger, Bernd. La Nueva España y sus metales preciosos. Frankfurt am Main: Vervuert, 1997. Hausberger, Bernd. “El universalismo científico del Barón Ignaz von Born y la transferencia de tecnología minera entre Hispanoamérica y Alemania a finales del siglo xviii.” Historia Mexicana 59, no. 2 (2009): 605–668 Haven, Gilbert. Our Next-Door Neighbor: a Winter in Mexico. New York: Harper & Brothers, 1875. Heers, Jacques. Gênes au xve siècle, activité économique et problèmes sociaux. Paris: s.e.v.p.e.n., 1961. Hermosa, Francisco de Paula. Manual de Laboreo de Minas. Besanzon: Librería de Rosa Bouret y Cia, 1857. Héron de Villefosse, Antoine-Marie. De la richesse minérale. Paris: Levrault, 1810. Herrera Canales, Inés. “El método de refinación con azogue en la minería potosina colonial: del fuego al cazo.” In La plata en Iberoamérica: Siglos xvi al xix, edited by Jesús Paniagua Pérez and Nuria Salazar Simarro, 61–68. León: Universidad de León, 2008. Herrera Canales, Inés, Cuahtémoc Velasco Avila, and Eduardo Flores Clair. Etnia y clase, los trabajadores ingleses de la Compañía del Monte y Pachuca, 1824–1906. México: inah, 1981. Hillman, Aubrey L., Mark B. Abbott, JunQing Yu, Daniel J. Bain, and TzeHuey ChiouPeng. “Environmental Legacy of Copper Metallurgy and Mongol Silver Smelting Recorded in Yunnan Lake Sediments.”Environmental Science & Technology 49 (2015): 3349–3357. Hobsbawm, Eric. Industry and Empire. London: Penguin Books, 1999. Humboldt, Alexander von. Essai politique sur le royaume de la Nouvelle-Espagne. Paris: Chez F. Schoell, 1811. Hylander, Lars D., and Markus Meili. “500 Years of Mercury Production: Global Annual Inventory by Region until 2000 and Associated Emissions.” Science of The Total Environment 304, no. 1–3 (2003): 13–27. Ibarra Gueztaraen, Juan de, “Señor, Cinco cosas son las que la villa imperial de Potosi suplica a V.M. en este Memorial.” Madrid, 1617? Jasso-Pineda, Yolanda, Guillermo Espinosa-Reyes, Donají González-Mille, Israel RazoSoto, Leticia Carrizales, Arturo Torres-Dosal, Jesús Mejía-Saavedra, Marcos Monroy, Ana Irina Ize, Mario Yarto and Fernando Díaz-Barriga. “An Integrated Health Risk Assessment Approach to the Study of Mining Sites Contaminated With Arsenic and Lead.” Integrated Environmental Assessment and Management 3, no. 3 (2007): 344– 350. Jiménez de la Espada, Marcos. Relaciones geográficas de Indias-Perú. Vol. i and Vol. ii, edited by José Urbano Martínez Carrera. Madrid: Atlas, 1965. Johnson, D.A., and K. Whittle. “The Chemistry of the Hispanic-American Amalgamation Process.” J. Chem. Soc., Dalton Trans., no. 23 (1999): 4239–4243.

bibliography

411

Jones, G.D.B. “The Roman Mines at Riotinto.” The Journal of Roman Studies 70 (1980): 146–165. Kaplan, Jed O., Kristen M. Krumhardt, and Niklaus Zimmermann. “The prehistoric and preindustrial deforestation of Europe.” Quaternary Science Reviews 28, no. 27–28 (2009): 3016–3034. Karpenko, V.Í. “The Oldest Alchemical Manuscript in the Czech Language.” Ambix 37, no. 2 (1990): 61–73. Kellenbenz, Hermann. “Los Fugger en España en la época de Felipe ii. ¿Fue un buen negocio el arrendamiento de los Maestrazgos después de 1562?” In Dinero y crédito (Siglos xvi al xix), edited by Alfonso Otazu, 19–36. Madrid-Villalba-Segovia: Banco Urquijo, 1977. Kellenbenz, Hermann. “Final Remarks: Production and Trade of Gold, Silver, Copper and Lead from 1450 to 1750.” In Precious Metals in the Age of Expansion: Papers of the xivth International Congress of the Historical Sciences, edited by Hermann Kellenbenz, 307–361. Stuttgart: Klett-Cotta, 1981. Kerl, Bruno, William Crookes, and Ernst Otto Röhrig. A Practical Treatise on Metallurgy Adapted from the Last German Edition of Prof. Kerl’s Metallurgy. London: Longmans, Green, and Co., 1868. Klein, Herbert S. The American Finances of the Spanish Empire: Royal Income and Expenditures in Colonial Mexico, Peru, and Bolivia, 1680–1809. Albuquerque: University of New Mexico Press, 1998. Kobata, Atsushi. “The Production and Uses of Gold and Silver in Sixteenth- and Seventeenth-Century Japan.” The Economic History Review 18, no. 2 (1965): 245– 266. Körner, Christian. “A Re-Assessment of High Elevation Treeline Positions and Their Explanation.” Oecologia 115, no. 4 (1998): 445–459. Kuhn, Thomas S. The Structure of Scientific Revolutions. Chicago: University of Chicago Press, 1965. Lacueva Muñoz, Jaime J. La plata del rey y de sus vasallos: minería y metalurgia en México (siglos xvi y xvii). Sevilla: Consejo Superior de Investigaciones Científicas, Escuela Superior de Estudios Hispano-Americanos; Universidad de Sevilla; Diputación de Sevilla, 2010. Lacueva Muñoz, Jaime J. “Nueva Vizacaya y sus yacimientos minerales hasta el descubrimiento de San José del Parral.” In Ophir en las Indias. Estudios sobre la plata americana. Siglos xvi–xix, edited by Jesús Paniagua Pérez and Nuria Salazar Simarro, 89–108. León: Universidad de León, 2010. Ladd, Doris M. The Making of a Strike: Mexican Silver Workers’ Struggles in Real del Monte, 1766–1775. Lincoln: University of Nebraska Press, 1988. Lamborn, Robert H. The Metallurgy of Silver and Lead: A Description of the Ores; their Assay and Treatment, and Valuable Constituents. London: C. Lockwood, 1878.

412

bibliography

Lang, Mervyn F. El monopolio estatal del mercurio en el México colonial (1550–1710). Translated by R.C. Gómez. México: Fondo de Cultura Económica, 1977. Lang, Mervyn F. “Amalgamación y fundición en la minería americana.” In Estudios de historia de las técnicas, la arqueología industrial y las Ciencias, edited by Juan Luis García Hourcade, Juan M. Moreno Yuste and Gloria Ruiz Hernández, 671–677. Salamanca: Junta de Castilla y León, 1998. Lang, Mervyn F. “Silver Refining Technology in Spanish America (patio y fundición)” In 5th International Mining History Congress, edited by James E. Fell, P.D. Nicolaou and G.D. Xydous, 136–145. Milos Island: Milos Conference Center-George Eliopoulos, 2001. Lara Meza, Ada Marina. Haciendas de beneficio de Guanajuato. Tecnología y usos del suelo 1770–1780. Guanajuato: Presidencia Municipal de Guanajuato; Dirección de Cultura y Educación, 2001. Laur, M.P. “De la metallurgie de l’argent au Mexique.” Annales des Mines, 6th series, 20 (1871): 38–317. Laznicka, Peter. “Discovery of Giant Metal Deposits and Districts.” In Proceedings of the 30th International Geology Congress: Energy and Mineral Resources for 21th Century: Geology of Mineral Deposits. Mineral Economics, edited by Pei Rongfu, 355–366. Utrecht: vsp, 1997. Laznicka, Peter. Giant Metallic Deposits: Future Sources of Industrial Metals. Berlin: Springer, 2006. Lehmann, B., A. Dietrich, and A. Wallianos. “From Rocks to Ore.” International Journal of Earth Sciences 89, no. 2 (2000): 284–294. Leonard, Irving A. “Mateo Alemán in Mexico: A Document.” Hispanic Review 17, no. 4 (1949): 316–330. Linden, S.J. “Jonson and Sendivogius: Some New Light on Mercury Vindicated from the Alchemists at Court.” Ambix 24, no. 1 (1977): 39–54. Linden, S.J. “Alchemy and Eschatology in Seventeenth-Century Poetry.” Ambix 31, no. 3 (1984): 102–124. Lopez Morell, Miguel A., and Jose M. O’Kean. “Seeking out and building monopolies, Rothschild strategies in non ferrous metals international markets (1830–1940).” Paper presented at the 14th Conference of the European Business History Association, Glasgow, August 26–28, 2010. Lynch, Martin. Mining in World History. London: Reaktion Books, 2002. Lyon, G.F. Journal of a Residence and Tour in the Republic of Mexico in the Year 1826 with Some Account of the Mines of that Country. London: J. Murray, 1828. MacDonald, Bernard. “Old Mexican Methods.” Mining and Scientific Press (1907): 123– 126. Magnin, Lucile. “Les peintures de paysages de Johann Moritz Rugendas: un exemple de transferts artistiques entre Europe et Amérique Latine au xixe siècle.” Artl@s Bulletin 5, no. 1 (2016): 24–37.

bibliography

413

Magnus, Albertus. Book of minerals. Translated by Dorothy Wyckoff. Oxford: Clarendon Press, 1967. Magos, Laszlo, and Thomas W. Clarkson. “Overview of the Clinical Toxicity of Mercury.” Annals of Clinical Biochemistry 43 (2006): 257–268. Maier, Michael. Symbola aureae mensae duodecim nationum. Frankfurt: Antonius Hummius, 1617. Manzanares Acuña, Eduardo, Héctor René Vega Carrillo, Martha Celia Escobar León, Consuelo Letechipía de León, Laura J. Guzmán Enríquez, Víctor Martín Hernández Dávila, and Miguel Ángel Salas Luévano. “Evaluación de riesgos ambientales por plomo en la población de Vetagrande, Zacatecas.” Zacatecas: Universidad Autónoma de Zacatecas, Unidad de Estudios Nucleares, 2005. Marichal, Carlos. Bankruptcy of Empire: Mexican Silver and the Wars Between Spain, Britain, and France, 1760–1810. New York: Cambridge University Press, 2007. Martin, Percy F. Mexico’s Treasure House (Guanajuato). New York: The Cheltenham Press, 1906. Martín Torres, Eugenio. El beneficio de la plata en Guanajuato, 1686–1740. Guanajuato: Presidencia Municipal de Guanajuato, 2001. Maskall, J.E., and I. Thornton. “Chemical Partitioning of Heavy Metals in Soils, Clays and Rocks at Historical Lead Smelting Sites.” Water, Air, and Soil Pollution 108, no. 3– 4 (1998): 391–409. Matilla Tascón, Antonio. Historia de las minas de Almadén, 1 (desde la época romana hasta el año 1645). Madrid: Gráficas Osca, 1958. Matilla Tascón, Antonio. Historia de las minas de Almadén, 2 (desde 1646 a 1709). Madrid: Instituto de Estudios Fiscales, 1987. Mendizábal, Miguel Othón. La minería y la metalurgia mexicanas (1520–1943). México: Centro de Estudios Históricos del Movimiento Obrero, 1980. Mendizábal, Miguel Othón. “Los minerales de Pachuca y Real del Monte en la época colonial.” El Trimestre Económico, no. 2 (1941): 253–309. Menegus Borneman, Margarita. “Las comunidades productoras de sal y los mercados mineros: los casos de Taxco y Temascaltepec.” In Minería regional mexicana. Primera reunión de historiadores de la minería latinoamericana, edited by Dolores Avila Herrera and Rina Ortiz, 21–31. México: Instituto Nacional de Antropología e Historia, 1994. Menéndez Navarro, Alfredo. Un mundo sin sol. Biblioteca Chronica Nova de Estudios Históricos Granada: Universidad de Granada / Universidad de Castilla—La Mancha, 1996. Merrill, Charles White. Summarized Data of Silver Production. us Government Printing Office, 1930. Meyer, Michael C., and William L. Sherman. The Course of Mexican History. New York: Oxford University Press, 1979.

414

bibliography

Mintz, Sidney W. Sweetness and Power. New York: Penguin Books, 1986. Mitchell, A.H.G., and M.S. Garson. Mineral Deposits and Global Tectonic Settings. London; New York: Academic Press, 1981. Mokyr, Joel. The Enlightened Economy. An Economic History of Britain 1700–1850. New Haven: Yale University Press, 2009. Molenda, Danuta. “La métallurgie du plomb en Pologne au moyen age et aux xvie– xviiie siècles”. In Mines et métallurgie, edited by Paul Benoit, 45–57. Villeurbanne: Programme Rhône-Alpes recherches en sciences humaines, 1994. Montejano y Aguiñaga, Rafael. El Real de Minas de la Purísima Concepción de los Catorce, slp. San Luis Potosí: Editorial Universitaria Potosina, 1993. Moran, Bruce T. Distilling Knowledge: Alchemy, Chemistry, and the Scientific Revolution. Cambridge, Mass.: Harvard University Press, 2005. Moran, Bruce T. Andreas Libavius and the Transformation of Alchemy: Separating Chemical Cultures with Polemical Fire. Sagamore Beach, ma: Science History Publications/Watson Pub. International, 2007. Moreno y Castro, Juan. Arte o nuevo modo de beneficiar los metales de oro y plata. México: Imprenta Biblioteca Mexicana, 1758. Motten, Clement G. Mexican silver and the Enlightenment. Philadelphia: University of Pennsylvania Press, 1950. Nakanishi, Tetsuya, and Eiji Izawa. “Evolution of Silver-Smelting Technology of Japan in the Middle of the Sixteenth Century.” isij International 54, no. 5 (2014): 1093– 1097. Nef, John U. “Silver Production in Central Europe, 1450–1618.” The Journal of Political Economy (1941): 575–591. Nef, John U. The Conquest of the Material World. Chicago: University of Chicago Press, 1964. Newman, William R. The ‘Summa perfectionis’ of Pseudo-Geber: A Critical Edition, Translation and Study. Leiden: Brill, 1991. Newman, William R. Atoms and Alchemy: Chymistry and the Experimental Origins of the Scientific Revolution. Chicago: University of Chicago Press, 2006. Newson, Linda A. “Silver Mining in Colonial Honduras.” Revista de Historia de América, no. 97 (1984): 45–76. Nriagu, J. “Legacy of Mercury Pollution.” Nature 363 (1993): 589. Nummedal, Tara E. Alchemy and Authority in the Holy Roman Empire. Chicago: University of Chicago Press, 2007. Ortiz Peralta, Rina. “El beneficio de minerales en el siglo xix: el caso de la Compañía Real del Monte y Pachuca.” Tzintzun: Revista de Estudios Históricos no. 14 (1991): 72– 85. Otte, Enrique, and Guadalupe Albi Romero. Cartas privadas de emigrantes a Indias, 1540–1616. México: Fondo de Cultura Económica, 1993.

bibliography

415

Paracelsus. The Hermetic and Alchemical Writings of Aureolus Philippus Theophrastus Bombast, of Hohenheim, Called Paracelsus the Great: Now for the First Time Faithfully Translated into English. Translated by Arthur Waite 1976. Pardo Hernández, Berenice, and Oscar Sánchez Rangel. Mineral de la Luz. La obra fotográfica de John Horgan Jr. en México. Guanajuato: Editorial La Rana, 2010. Percy, John. The Metallurgy of Lead Including Desilverization and Cupellation. London: J. Murray, 1870. Percy, John. Metallurgy. The Art of Extracting Metals from their Ores. Silver and Gold. London: J. Murray, 1880. Pérez Luque, Rosa Alicia, and Rafael Tovar Rangel. La contabilidad de la Caja Real de Guanajuato. Una aproximación a su historia económica 1665–1816. Guanajuato: Universidad de Guanajuato, 2006. Pérez Sáenz de Urturi, J.E. “La minería colonial americana bajo la dominación española.” Boletín Millares Carlo, no. 7 (1985): 53–120. Phillips, John Arthur. Descriptive Notice of the Silver Mines and Amalgamation Process of Mexico. Extracted from the Railway Register. London: Pelham Richardson, 1846. Phillips, John Arthur. The Mining and Metallurgy of Gold and Silver. London: E. and F.N. Spon, 1867. Pique, Edward. A Practical Treatise on the Chemistry of Gold, Silver, Quicksilver and Lead, Tracing the Crude Ores from the Mines Through the Various Mechanical and Metallurgic Elaborations, Until the Pure Metal is Obtained. San Francisco: Towne & Bacon, printers, 1860. Pohl, Walter. Economic Geology: Principles and Practice: Metals, Minerals, Coal and Hydrocarbons—Introduction to Formation and Sustainable Exploitation of Mineral Deposits. Chichester, West Sussex; Hoboken, nj: Wiley-Blackwell, 2011. Pomeranz, Kenneth. The Great Divergence. China, Europe, and the Making of the Modern World Economy. Princeton, New Jersey: Princeton University Press, 2000. Pontes y Fernández, José María. Historia de la antigua ciudad de Sisapón, hoy Almadén del Azogue. Madrid: Enrique Rojas, 1900. Powell, Philip Wayne. Soldiers, Indians & Silver. Berkely and Los Angeles: University of California Press, 1952. Probert, Alan. “Bartolomé de Medina: the Patio Process and the Sixteenth Century Silver Crisis.” Journal of the West 8, no. 1 (1969): 90–124. Puerto Sarmiento, Francisco Javier. “La panacea aúrea. Alquimia y destilación en la corte de Felipe ii (1527–1598).” Dynamis (Granada, Spain) 17 (1997): 107–140. Ramos-Arroyo, Yann René, Rosa María Prol-Ledesma, and Christina Siebe-Grabach. “Características geológicas y mineralógicas e historia de extracción del Distrito de Guanajuato, México. Posibles escenarios geoquímicos para los residuos mineros.” Revista Mexicana de Ciencias Geológicas 21, no. 2 (2004): 268–284. Ramos, Demetrio. “Ordenación de la minería en Hispanoamérica durante la época pro-

416

bibliography

vincial (siglos xvi, xvii y xviii).” In La minería hispana e iberoamericana. Ponencias del i coloquio internacional sobre historia de la minería, 373–397. León: Cátedra de San Isidro, 1970. Randall, Robert W. Real del Monte: a British Mining Venture in Mexico. Austin: The University of Texas Press, 1972. Rankine, Margaret E. “The Mexican Mining Industry in the Nineteenth Century with Special Reference to Guanajuato.” Bulletin of Latin American Research (1992): 29– 48. Rawson, William. “The Present Operations and Future Prospects of the Mexican Mine Associations Analysed, by the Evidence of Official Documents, English and Mexican, and the National Advantages Expected from Joint Stock Companies, Considered, in a Letter to the Right Hon. George Canning.” London: J. Hatchard & Son, 1825. Read, John. Through Alchemy to Chemistry; A Procession of Ideas & Personalities. London: G. Bell, 1957. Rech, Georges. “La fonderie de Wisenbach (Vosges).” In Mines et métallurgie, edited by Paul Benoit, 173–188. Villeurbanne: Programme Rhône-Alpes recherches en sciences humaines, 1994. Reed, William. The History of Sugar and Sugar Yielding Plants. London: Longmans, Green and Co., 1866. Reyes, Aurelio de los. Los caminos de la plata. México: Universidad Iberoamericana, 1991. Reyes, Juan Carlos. “Introducción.” In La sal en México, edited by Juan Carlos Reyes, vii–ix. Colima: Universidad de Colima, Consejo Nacional para la Cultura y las Artes, 1995. Reza, Alma Linda. Guanajuato y sus miasmas. Higiene urbana y salud pública, 1792–1804. Guanajuato: Presidencia Municipal de Guanajuato, 2001. Rice, Claude T. “The Silver-Lead Mines of Santa Barbara, Mexico.” The Engineering and Mining Journal lxxxvi, no. 5 (1908): 207–211. Richards, John F. The Unending Frontier: An Environmental History of the Early Modern World. Berkeley: University of California Press, 2005. Rickard, T.A. Journeys of Observation. San Francisco: Dewey Pub. Co., 1907. Riello, Giorgio. Cotton. The Fabric that Made the Modern World. Cambridge: Cambridge University Press, 2015. Ripley, George. The compound of Alchemy. London: Thomas Orwin, 1591. Rippon, Stephen, Peter Claughton, and Chris Smart. Mining in a Medieval Landscape: The Royal Silver Mines of the Tamar Valley. Exeter: Univ of Exeter Press, 2009. Rivot, Louis Edouard. Description des gites métallifères, de la préparation mécanique et du traitement métallurgique des minerais de plomb argentifères de Pontgibaud. Paris: Carilian-Goeury et Vor Dalmont, 1851.

bibliography

417

Rivot, Louis Edouard. Principes généraux du traitement des minerais métalliques. Paris: Dalmont et Dunod, 1859. Robb, L.J. Introduction to Ore-Forming Processes. Malden, ma: Blackwell Pub., 2005. Robertson, William Parish. A Visit to Mexico, by the West India Islands, Yucatan and United States, with Observations and Adventures on the Way. London: Simpkin, Marshall & Co., 1853. Robins, N.A. Mercury, Mining, and Empire: The Human and Ecological Cost of Colonial Silver Mining in the Andes. Bloomington and Indianapolis: Indiana University Press, 2011. Rodríguez Marín, Francisco. Felipe y la Alquimia. Madrid: Tipografía de la Revista de Archivos, Bibliotecas y Museos, 1927. Rohe, Randall. “Man and the Land: Mining’s Impact in the Far West.” Arizona and the West 28, no. 4 (1986): 299–338. Rosen, George. The History of Miners’ Diseases. A Medical and Social Interpretation. New York: Schuman’s, 1943. Ruiz de la Barrera, Rocío. “La Empresa de Minas del Real del Monte (1849–1906).” PhD diss., Colegio de México, 1995. Saavedra Silva, Elvira Eva, and María Teresa Sánchez Salazar. “Minería y espacio en el distrito minero Pachuca-Real del Monte en el siglo xix.”Investigaciones Geográficas, Boletín del Instituto de Geografía, no. 65 (2008): 82–101. Saignes, Thierry. “Las técnicas mineras de Potosí según una relación inédita de 1600.” Arte y arqueología 8, no. 9 (1982/1983): 171–176. Salas, Guillermo P. Carta y Provincias Metalogenéticas de la República Mexicana. México: Consejo de Recursos Minerales, 1980. Salazar-Soler, C. “Innovaciones técnicas, mestizajes y formas de trabajo en Potosí de los siglos xvi y xvii.” In O trabahlo mestigo. Maneiras de pensar e formas de viver séculos xvi a xix, edited by Eduardo Franca Paiva and Carla Maria Junho Anastasia, 141–163. Sao Paulo: ufmg, 2002. Salazar González, Guadalupe. Las haciendas en el siglo xvii en la región minera de San Luis Potosí. Su espacio, forma, función, material, significado y la estructuración regional. San Luis Potosí: Universidad Autónoma de San Luis de Potosí, 2000. Salkield, Leonard U. “Ancient Slags in the South West of the Iberian Peninsula.” In La minería hispana e iberoamericana. Ponencias del i coloquio internacional sobre historia de la minería, 85–98. León: Cátedra de San Isidro, 1970. Sánchez-Crispín, Alvaro, Eurosia Carrascal, and Alejandrina de Sicilia Muñoz. “De la minería al turismo: Real de Catorce y Cerro de San Pedro, México. Una interpretación geográfico-económica.” Revista Geográfica, no. 119 (1994): 81–106. Sánchez Gómez, Julio. De minería, metalúrgica y comercio de metales: la minería no férrica en el Reino de Castilla, 1450–1610. Salamanca: Universidad de Salamanca: Instituto tecnológico geominero de España, 1989.

418

bibliography

Sasaki, Junnosuke. “Modes of Traditional Mining Techniques.” In Project on Technology Transfer, Transformation and Development: The Japanese Experience. Tokyo: United Nations University’s Human and Social Development Program, 1980. Schafer, Ernst. “Johann Tetzel, ein deutsche bergmann in westindien zur zeit Karls v.” Ibero-Amerikanisches Archiv 10 (1936): 162–170. Schlutter, Christophe-Andre. De la fonte des mines, des fonderies, &c. Tome Second. Translated by M. Konig. Paris: Jean-Thomas Herissant et Jacques-Noel Pissot, 1753. Scott, David A. Copper and Bronze in Art: Corrosion, Colorants, Conservation. Los Angeles: Getty Conservation Institute, 2002. Sempat Assadourian, Carlos. “Base técnica y relaciones de producción en la minería de Potosí.” In Ciencia colonial en América, edited by Antonio Lafuente and José Sala Catalá, 121–142. Madrid: Alianza, 1992. Sempat Assadourian, Carlos. Zacatecas, conquista y transformación de la frontera en el siglo xvi: minas de plata, guerra y evangelización. México, d.f.: Colegio de México, Centro de Estudios Históricos, 2008. Sheppard, H.J. “Alchemy: Origin or Origins?”. Ambix 17, no. 2 (1970): 69–84. Sillitoe, Richard H. “Relation of Metal Provinces in Western America to Subduction of Oceanic Lithosphere.” Geological Society of America Bulletin 83, no. 3 (1972): 813–818. Sillitoe, Richard H. “Supergene Silver Enrichment Reassessed.” In Supergene Environments, Processes and Products, edited by Spencer R. Titley, 15–32. Littleton, Colorado: Society of Economic Geologists, 2009. Singer, D.A. “World Class Base and Precious Metal Deposits; A Quantitative Analysis.” Economic Geology 90, no. 1 (1995): 88–104. Smith, Grant H. The History of the Comstock Lode. Reno: Nevada Bureau of Mines and Geology and University of Nevada Press, 1998. Smith Jr, D.M. “Geology of Silver Deposits along the Western Cordillera.” In Silver— Exploration, Mining and Treatment, 11–24. London: Institution of Mining and Metallurgy, 1988. Smith, Pamela H. “Vermilion, Mercury, Blood, and Lizards: Matter and Meaning in Metallurgy.” In Materials and Expertise in Early Modern Europe: Between Market and Laboratory, edited by Ursula Klein and E.C. Spary, 29–49. Chicago: University of Chicago Press, 2010. Soetbeer, Adolf. Edelmetall-produktion und werthverhältniss zwischen gold und silber seit der entdeckung Amerika’s bis zur gegenwart. Gotha: Justus Perthes, 1879. Sonneschmidt, F.T., and J.M. de Fagoaga. Tratado de la amalgamación de Nueva España. México: Galería de Bossange (padre), Antoran y Cia., 1825. Splitter, Henry Winfred. “Quicksilver at New Almaden.”Pacific Historical Review 26, no. 1 (1957): 33–49. Stein, Robert Louis. The French Sugar Business in the Eighteenth Century. Baton Rouge: Louisiana State University Press, 1988.

bibliography

419

Stein, Stanley J., and Barbara H. Stein. Silver, Trade, and War: Spain and America in the Making of Early Modern Europe. Baltimore and London: The Johns Hopkins University Press, 2000. Ströbele, F., T. Wenzel, A. Kronz, L.H. Hildebrandt, and G. Markl. “Mineralogical and Geochemical Characterization of High-Medieval Lead-Silver Smelting Slags from Wiesloch near Heidelberg (Germany)—An Approach to Process Reconstruction.” Archaeological and Anthropological Sciences 2, no. 3 (2010): 191–215. Studnicki-Gizbert, D., and D. Schecter. “The Environmental Dynamics of a Colonial Fuel-Rush: Silver Mining and Deforestation in New Spain, 1522 to 1810.” Environmental History 15, no. 1 (2010): 94–119. Suárez Argüello, Clara Elena, and Brígida Von Mentz. Epístolas y cuentas de la negociación minera de Vetagrande, Zacatecas, 1791–1794, 1806–1809. México: Centro de Investigaciones y Estudios Superiores en Antropología Social (ciesas), 2008. Tandeter, Enrique. Coercion and Market: Silver Mining in Colonial Potosí, 1692–1826. Albuquerque: University of New Mexico Press, 1993. Taylor, John. Selections from the Works of the Baron de Humboldt, Relating to the Climate, Inhabitants, Productions, and Mines of Mexico. With Notes by John Taylor. London: Longman, Hurst, Rees, Orme, Brown, and Green, 1824. TePaske, John Jay, and Kendall W. Brown. A New World of Gold and Silver. Leiden, Boston: Brill, 2010. Terreros, Manuel Romero de. Antiguas haciendas de México. Editorial Patria, 1956. Thibodeau, A.M., D.J. Killick, J. Ruiz, J.T. Chesley, K. Deagan, J.M. Cruxent, and W. Lyman. “The Strange Case of the Earliest Silver Extraction by European Colonists in the New World.” Proceedings of the National Academy of Sciences of the United States of America 104, no. 9 (2007): 3663–3666. Thomas, Hugh. Rivers of Gold: The Rise of the Spanish Empire. New York: Random House, 2005. Thomas, Hugh. The Golden Empire. Spain, Charles v, and the Creation of America. New York: Random House, 2010. Thornton, I., and P. Abrahams. “Historical Records of Metal Pollution in the Environment.” In Changing Metal Cycles and Human Health, edited by J.O. Nriagu, 7–25. Heidelberg, New York, Tokyo: Springer-Verlag, 1984. Todd, A.C. The Search for Silver. Cornish Miners in Mexico, 1824–1947. St. Austell, Cornwall: Cornish Hillside Publications, 2000. Tudor, Henry. Narrative of a Tour in North America Comprising Mexico, the Mines of Real del Monte, the United States, and the British Colonies: with an Excursion to the Island of Cuba: in a Series of Letters, Written in the Years 1831–2. London: J. Duncan, 1834. usepa. Air Quality Criteria for Lead epa/600/r-5/144aF. Vol. i, ii 2006. Vaněk, Václav, and Dalibor Velebil. “Staré hutnictví stříbra.” In Stříbrná Jihlava 2007.

420

bibliography

Studie k dějinám hornictví a důlních prací, 188–205. Jihlava: Archaia Brno / Muzeum Vysočiny Jihlava, 2007. Vega, Garcilaso de la. Comentarios reales de los Incas, Primera Parte. Lisboa: Pedro Crasbeeck, 1609. Velasco Avila, Cuahtémoc, Eduardo Flores Clair, Alma Laura Parra Campos, and Edgar Omar Gutiérrez López. Estado y minería en México (1767–1910). México: Fondo de Cultura Económica, 1988. Vergani, Raffaello. “La métallurgie des non-ferreux dans la république du Venise (xve– xviiie siècles).” In Mines et métallurgie, edited by Paul Benoit, 207–215. Villeurbanne: Programme Rhône-Alpes recherches en sciences humaines, 1994. Viajero. “Las Minas de México.” Revista de Minas Tomo i (1861): 176–183. Villaseñor y Sánchez, José Antonio de. Theatro americano: descripcion general de los reynos y provincias de la Nueva España y sus jurisdicciones. México: Imprenta de la Viuda de D. Joseph Bernardo de Hogal, 1746. Viollet, Eugène. “Le problème de l’Argent et l’Etalon d’ Or au Mexique.” PhD diss., Université de Paris, 1907. Wagner, Henry R. “Early Silver Mining in New Spain.” Revista de Historia de América, no. 14 (1942): 49–71. Wallace, T.C., M. Barton, and W.E. Wilson. “Silver & Silver-Bearing Minerals.” Rocks & Minerals 69, no. 1 (1994): 16–38. Ward, H.G. Mexico. London: H. Colburn, 1829. Waszkis, Helmut. Mining in the Americas: Stories and History. Cambridge: Woodhead Publishing Limited, 1993. West, Robert C. The Mining Community in Northern New Spain: the Parral Mining District. Berkeley and Los Angeles: University of California Press, 1949. West, Robert C. “Aboriginal metallurgy and metalworking in Spanish America: a brief overview.” In In Quest of Mineral Wealth. Aboriginal and Colonial Mining and Metallurgy in Spanish America, edited by Alan Craig and Robert West, 5–20. Baton Rouge: Geoscience Publications, 1994. West, Robert C. “Early Silver Mining in New Spain, 1531–1555.” In In Quest of Mineral Wealth. Aboriginal and Colonial Mining and Metallurgy in Spanish America, edited by Alan Craig and Robert C. West, 119–135. Baton Rouge: Geoscience Publications, 1994. Whitaker, Arthur P. “The Elhuyar Mining Missions and the Enlightenment.” The Hispanic American Historical Review 31, no. 4 (1951): 557–585. Whitney, J.D. The metallic wealth of the United States, described and compared with that of other countries. Philadelphia: Lippincott, Grambo & Co., 1854. Wild, Mike, and Ian Eastwood. “Soil Contamination and Smelting Sites.” In Boles and Smeltmills Seminar, edited by Lynn Willies and David Cranstone, 54–57. Reeth, Yorkshire: Historical Society, Ltd., 1992.

bibliography

421

Willies, Lynn. “Derbyshire Lead Smelting in the Eighteenth and Nineteenth Centuries.” Bulletin of the Peak District Mines Historical Society 11 (1990): 1–19. Wilson, Robert Anderson. Mexico and Its Religion; with Incidents of Travel in that Country During Parts of Years 1851–52–53–54. New York: Harper & Bros., 1855. Wilson, W.J. “An Alchemical Manuscript by Arnaldus de Bruxella.” Osiris 2 (1936): 220– 405. Wood, Ellen Meiksins. The Origin of Capitalism. A Longer View. New York: Verso, 2002. Young, Otis E. “The Spanish Tradition in Gold and Silver Mining.” Arizona and the West 7, no. 4 (1965): 299–314. Zavala, S. “La amalgama en la minería de Nueva España.” Historia Mexicana xi, no. 3 (1962): 416–421. Zhang, Yanxu, Lyatt Jaeglé, LuAnne Thompson, and David G. Streets. “Six centuries of changing oceanic mercury.” Global Biogeochemical Cycles 28, no. 11 (2014): 1251–1261.

Index acanthite 26n45 Acosta, José de 140 Adventurers and barrel process 305 initial impressions 202 losses of 196, 198 afinación 56, 219 See also cupellation Agricola, Georgius 21n33, 48n12, 58n46, 69, 100 on dressing ores 69 on lead 55n38, 78n, 90n27, 350 on refining with mercury 47, 112 alchemy 2, 103, 104n10, 117, 361 Arcanum 109 and metallurgy 108, 144, 151 and Philip ii 109, 142 See also Magisterio; magistral; mercury (alchemy) alcribís 69n75, 70, 71n77 Alemán, Mateo 352 Almadén 12, 102, 111, 147, 299, 361, 374n2 and choice of patio process 146, 359, 360 and convict labour 352 as fiscal solution for Crown 298, 359, 362 and Fuggers 298, 352 geology 37 mercurialism at 351, 352, 354, 355, 362 miners from 50, 112 supply from 112, 298n83, 337 total production 38 Alonso Barba, Alvaro 45, 71n80, 73n85 Arte de los Metales 44 and alchemy 110 and barrel process 138n106, 304 on dressing ores 69 on grasas 75n89 on hornos castellanos 69n75 on recycling mercury 165 on refining skills 49, 150 on smelting 44, 47, 98 on the use of mercury 46 See also cazo process altiplano 38, 148, 362 Alto Perú. See Upper Peru

amalgam 104 amalgam, silver 118, 120, 122 in lavaderos 161, 163, 210, 211 recovery of mercury from 163–169 See also piñas amalgamation of elemental silver 118, 119 of gold 6, 48, 102, 111–119, 135, 151, 362 misleading use of term 118n57, 131, 153 Andes innovation in 101, 133, 135, 139, 144, 148, 151, 362 major metallic deposits 12, 15, 18, 42, 361 orogen 18 antimony 29n57, 30n58, 55 interference in silver refining 28, 58, 123n67, 234n11, 308n106 aposento de azogue 163 See also azoguería aqueducts 187, 208 architecture of fear 170 industrial 3, 176, 183, 253, 254, 361, 363 plans of haciendas 176–183 argentite 26n45 arrastres 157, 188, 279 gallery of 179, 180 inventor of 158 See also Regla: milling; tahona arsenic 29n57, 30n58, 55, 79n, 85, 86 assaying gold ores 46, 112 limited use of 35, 149, 260 silver ores 46, 47, 239 August the Strong (Saxony) 109 Auld (General Manager), 209 aviadores 172n45, 311n115 azogados 349, 351, 352 azogue. See mercury azoguería 163, 164f, 212 azogueros 124, 149, 150, 161, 239 Azoth 106 barranca 196, 200, 209, 279 barras 73, 87 See also pigs

index barrel process 144, 153, 229, 268, 279, 304– 309 fuel consumption 138, 250, 274 mercury consumption 121, 307 in Mexico 306n102, 338 See also cazo process; Freiberg process Bassoco, Antonio de 150, 151n Beistegui, Nicanor 199n18, 268n33 Bell, R. (chief smelter, Regla, 1845) 201, 217 Bell, R. (Regla, 1865) 209 bellanding 89, 90n26 Bellange, Alejandro 199n18 bellows 63, 67n66, 69–71, 83 Belvio, Paolo 117 beneficio de azogue 119, 153 See also patio process beneficio de patio. See patio process Biringuccio, Vannoccio on lead losses 84 on mercury as poison 350 and refining with mercury 48, 107, 112, 115, 117, 141, 144, 148 and Venice 113, 114 ‘black smoker’ 13 blast furnace 73, 87n21, 90 efficiency of 94, 192, 364 See also under Regla bonanza 16, 18, 43, 198, 199n18, 338 Born, Inigo (Baron) 112, 127, 138n106, 268, 304, 309 Boteller, Mosén 143n121 Böttger, Johann Friedrich 109, 110 Bourbon mining reforms 138n106, 260, 261, 302n92, 305, 307n105 brine 115, 135, 140 Buchan, John H. 198, 261, 262 on accounting 199, 200 on barrel process 305, 308 conflict of interest 199n18 inventory Regla 205, 209, 213, 217, 219 on patio process 232, 237, 271, 276, 278 on woodlands 250, 251 buddling 91 See also silver ores: dressing buitrones 153 See also canoas Burkart, Johann 29, 34, 35

423 Cabildo de México 51 Cádiz 103, 197 Cajas 318 data as base line 317 and refining process 335 Caja of Bolaños 318 abnormal mercury to silver ratios 377 silver by refining process 332, 392 Caja of Chihuahua 318, 325 abnormal mercury to silver ratios 377 silver by refining process 333, 334, 395 Caja of Durango 318 silver by refining process 325, 326, 376, 382, 383 Caja of Guadalajara 318 silver by refining process 328, 329, 376, 386, 387 Caja of Guanajuato 318, 324, 375 silver by refining process 323, 380, 381 Caja of México 318, 323–325, 333, 381 Caja of Pachuca 318, 324, 329, 333 silver by refining process 330, 388, 389 Caja of Rosario 318, 377 silver by refining process 333, 393 Caja of San Luis Potosí 64, 318, 324, 376 silver by refining process 326–328, 384, 385 Caja of Sombrerete 318, 376 abnormal mercury to silver ratio 377 silver by refining process 330, 331, 390, 391 Caja of Zacatecas 318, 330 silver by refining process 321, 322, 375, 378, 379 Caja of Zimapán 318, 324, 330n22 silver by refining process 333, 394 calomel historical uses of 347 in historiography 131, 132, 348, 355n64, 364 lifecycle of 346, 347 loss, as solid waste 187, 189, 190 lower production 120, 121, 137, 139 from mercury 123, 128, 348, 349, 374 as mitigating factor 7, 152, 191, 253, 354, 357, 363

424 in New Spain and Mexico 339, 341 reports on (19c) 119, 124 See also under correspondencia; mercury; Regla (patio process) canoas 153, 176, 290, 320n8, 337 cañón 70 capellinas 166, 169 area requirements 182, 183, 190 as building (Guanajuato) 130n84, 168 heating cycle 168, 215 metals used for 127, 166, 168, 182, 192, 216n43 output in La Escalera 176 silver production using 128, 349 size and weight 168, 182, 190, 215n See also under Regla caperuzas 165, 166 Capoche, Luis 135, 147 carboneras 63, 96 cárcamo 161 Cárdenas, Juan de 110, 134, 142 Casa de la Moneda 124n70, 199n18, 257, 268n33 Catorce 77, 138n105, 138n106, 335 bonanza 327, 328, 338 silver ore from 29n54, 39, 64, 137, 309n107 cazo process 44, 110, 153, 338 advantages of 138 and Caja of San Luis Potosí 64, 327, 328 fuel consumption 138 mercury to silver ratio 121, 123, 141, 307, 327n18 ores, suitable 29n54, 64, 137, 308n107, 327 as two-step process 138n105 See also Freiberg, barrel process cendrada 57, 58n45, 79, 81 cerargyrite 29 Cerro San Pedro 24n41, 31, 45, 65, 99, 327 charcoal 63, 77, 99, 135 to displace patio process 358, 359 for metal ores 93 as reducing agent 54–56, 93, 100, 101 for silver ores 93, 94, 237, 364 See also under Regla (smelting); woodlands Charles iv (King of Spain) 260

index Charles v (King of Spain) 16, 51, 143n121, 297 chemical history 4 chemistry as gatekeeper to refining 7, 315 as historical tool 253 and history of refining 5, 192 Chichimeca 65, 170 Chilean mill 66, 67, 154, 156f, 204 China 40n85, 41, 104n10, 110n29, 132n88 lack of major silver deposits 9 silver imports 144, 270n38, 310 as source of mercury 37 chlorargyrite 27, 29 coagulation 110 See also transmutation coal 305, 359n73 Cobalt, Ontario 10 coinage, duty on. See señoreage Collins, Henry F. 263 colonial periphery 42, 144, 311 technology development 3, 5, 6, 148, 157, 314 terminology 40n85 colorados (New Spain) 28, 29, 30n58 Columbus, Christopher 19, 48, 112, 297 commodities displacement of sources 5, 40–42 trade 5, 6, 40, 310, 311 See also spice trade; cotton; sugar Compañía de Minas de Real del Monte y Pachuca accounting practice 268 barrel process 307, 308 mercury prices 274, 275 Mexican owners 199n18 ore distribution 224 ore extraction costs 276, 277 refining haciendas and processes 268 report to the Directors 262 Comstock Lode, Nevada 2n, 136n99, 144, 239 Mexican refining technology 5, 53n30 silver sulphide ores 29n57 Conde de Regla (First) 196–198, 200, 201, 235, 251, 296 estate 255, 256 contraband. See corruption; silver: contraband

425

index Contreras, Manuel María 124, 125 copapiri 137 copper argentiferous 21, 92, 111, 313 deposits, enrichment of 27 as revenue source 256, 295, 296, 303n94, 313, 361 copper pyrites 137, 141 copper sulphate 137, 141, 212, 222, 312 absence in recipe 142, 144, 146 chemical role in mercury refining 117, 137, 139, 229, 234 colour 137n102, 209 environmental impact 187, 345 soluble waste 190, 339, 360, 363 See also under Regla (patio process) Cordillera 10, 18, 31, 193, 361 Cornwall 20, 21, 195n6 Cornish miners 22, 196, 199n19, 313 Correa, Manuel 86–90 correspondencia 123, 126 and contraband 123 Duport on interpreting 125 historical variation of 123n68 mathematical modelling of 125–127 as stoichiometric ratio 125, 349 value of 123–125, 128, 348 ceiling on 375, 377 corruption, effect on data 151, 317 See also silver: contraband Cortés, Hernán 49 cotton 5, 270, 311, 314 displacement of sources 40–42 European technology 310 Crown. See Spanish Crown cupel 46, 56, 87 from animal bones 46n7, 57 from human bones 49 cupellation 56, 73, 219 and loss of lead 77, 84, 87 cupelling assembly. See vasos Cusano, Tommaso 113 cyanide process 6, 103, 144, 190, 192 De Quille, Dan 244 desazogaderas 166, 168, 190, 192, 349 See also caperuzas desmontes 135 See also tailings

Díaz, Bernal 49 Diezmos 320, 321, 332, 374n1 Domínguez de la Fuente, Manuel José 314 dressing. See under silver ores Duport, Saint Clair 261, 262, 269 on losses of mercury 127 ‘duties’ 52, 314, 321 ‘duties & mintage’ 321

260,

eastern Pacific rim 15, 361 Eden, Richard 12 Egleston, Thomas 127, 129 Eissler, Manuel 137 El Salto (waterfall) 196, 200, 204n Emmons, William Harvey 26 Enchel, Juan 52 England failed mining investments 195, 198n16, 305 Industrial Revolution 217, 255, 310 lead supply from 92 silver mines 21n35 silver production 305n99 silver refining expertise 39, 219, 313 entrepeneurs. See private enterprise environmental impact. See under refining methods and materials consumed environmental legacy 3, 190, 251, 346, 347, 360 epithermal deposits 14, 18, 20n31 Ercker, L. 47, 115, 164, 350 Erzgebirge 20, 21, 47, 51n24, 92, 313 Escandón, Manuel 199n18 Escuela Práctica del Colegio de Minería 181 Estados Comparativos 202, 368 Estados Unidos Mexicanos 192 Europe imperial trade 5, 6, 40–42, 310, 314 privileged access to silver 9, 42 experiential knowledge of silver refining 2, 19, 43 Europe and smelting 21, 111 global trails 6, 115, 144, 145 mercury-based, source of 115 planted in New World 2, 5, 117, 147, 148, 176, 361 shared by refining haciendas 3, 149 transmission ‘in the flesh’ 117, 147, 148

426 Fabry, José Antonio 257, 260 fao (Food and Agricultural Organization) 94 Ferdinand (King of Spain) 48 Fernández Montaño, Juan 137 Fernández de Oviedo y Valdés, Gonzalo 49 Fernández de Velasco 134 Flamel, Nicolas 106 fluxing agents 55 forest cover. See woodlands France. See New France Freiberg 21, 261 Freiberg process 121, 138, 144, 153, 268, 304, 309n108 See also cazo; barrel process fuelle. See bellows Fugger, Anton 297 Fugger, Jakob 296, 297 Fuggers 21, 22, 50–52, 352, 360, 362 Almadén, leasing of 297 debt owed to 297–299 and Guadalcanal 143n121 mercury gifted to New Spain 298, 299 and mining in Europe 296 fundición 51, 55 See also smelting furnaces. See under specific types galena, argentiferous 26, 361 assaying of 46, 47 and colonial refiners 32, 54 European deposits 20–22, 50 lead as key for smelting 21 Mexican deposits 30, 31, 39, 45 and patio process 122n67, 264 refining by mercury 111, 143n121 silver content 20 smelting 55, 99 weight as indicator of 98, 293 Gamboa, Francisco Xavier de 80 Garci Hernández (Venice) 114 García de Llanos 28 Garcilaso de la Vega 351 Geber (Jabir ibn Hayyan) 107, 110 German metallurgical artisans in 19c Mexico 201 in New World 51, 52, 54, 117, 137, 143n121, 157n6 smelting know-how 22, 53, 113

index Germany 270, 274n43, 295 assistance to refiners 303n94 gold assaying of, by mercury 112 placer 112, 117, 144 revenues from, in silver refining 24, 58– 60, 77 in silver ores of New Spain 99 gold amalgamation 48, 111, 114 in Africa 102, 103, 111n35, 135 confusion with silver refining 46, 118n57, 131 difference with silver refining 131, 152 in Europe 6, 112, 117, 118, 144, 147 modern artisanal practice 131, 164, 364 recipe 102, 108, 115, 117, 135, 151, 362 early use for silver ores 115, 117, 120, 135, 151, 362 modified for silver sulphides 118, 121, 135–137, 139 scale of use 112 stages 115 gold production Dominican Republic 19 Hispaniola 19, 43, 112 limited by colonial expertise 19 Gómez de Cervantes 45, 142, 153 gossan 28, 29n54 granzas 75 grasas 75 fields of 60, 61f, 62, 74, 76f, 97, 364 sink for chemical pollution 100 as source of lead and silver 64, 75, 81, 85 See also slag greta 56, 79, 97, 99 contraband of 81, 82 sale of 81 See also litharge Guadalcanal 22, 50, 114, 144, 297 failure of refining by mercury 50n19, 143n121 Guadalcázar 71n80, 73, 327 Gualdi, Pietro 181 Guanajuato, silver refining sector 171, 172 hacienda, definition 3n Hacienda Aranzazu, de 73, 74f Buen Suceso, del 150

index Casas Blancas 179, 180, 182 Gogorrón, de 97 hmc1 (Monte Caldera) 60 La Escalera 172, 173, 176, 220n58 La Luz 61 Las Mercedes 176–180, 182 La Purísima 173 Loreto, de 268, 318 Nuestra Señora de los Dolores 65 Nueva de Fresnillo 173, 181, 182, 184f, 220n58, 279n50 Rocha, de 180n52, 182, 220n58 Salgado, de 158f, 176n48, 180 San Antonio 173 San Antonio de Regla 200, 202, 248, 279n51 San Juan Nepomuceno 173 Santa María 60, 72f, 76f Santa María de Regla. See Regla San Miguel de Regla 200, 202, 248, 251, 268 Sánchez, de 268, 307 Velasco, de 268, 279, 307, 308 hacienda de beneficio de azogue. See hacienda de patio hacienda de patio 154 accounting records 191 architects 183 architectural plans 169, 176, 183, 191 azoguería 163, 164f capellinas 164–169 construction cost 279n50 as credit centre 172 environmental impact 187–189 environmental legacy 190, 191, 360 evolution of 176, 183 industrial nature of 170, 176, 183 mercury, control of 190 mills 154, 155, 157 patio area and output 234 patio reactor 158–161 perimeter walls 169, 170, 171f planillas 161, 162f range of sizes 170, 172, 173, 182, 183n rental cost 279 spatial requirements 183, 190 stables 178 See also under patio process; Regla (patio process)

427 hacienda, smelting 88f, 89f accounting records 59, 64, 65 architecture, reconstruction of 58 areas of occupational hazards 87, 89, 100 assets of 63, 64 bellows 70, 71 chimneys 71, 73 complaints against 86, 87 construction cost 63 dearth of plans 191 dressing of ores (see under silver ores) emissions of lead fumes 87–90, 100, 354, 360, 364 environmental impact 96–98 footprint 60 furnaces 59, 72f, 74f grasas, areas for disposal of 60, 61f, 62, 63, 74, 75, 76f hornos castellanos 69, 70 molinos 66, 67, 156f output per furnace 59 outsourcing of services 64, 77 rental costs 64 revenues from gold by-product 60 sale price 60, 63 silver production, estimated 59, 60 siting relative to water 99 vasos 59, 73, 74 water, requirements 69 wind rose and furnaces 90, 91 workers and families 97, 100 See also under Regla (smelting); smelting Harz 20, 21, 47, 113, 295, 303n94 Hayyan, Jabir ibn (Geber) 104 heavy metals 346, 350 lead as main source 7, 253, 315, 344, 363– 365 Hermosa, Francisco de Paula 128 Hernández, Pero 148 Herón de Villefosse, Antoine-Marie 295 Hiendelaencina 309n108 Hispanic New World, silver production 1 Hispaniola, copper 48 Honduras, smelting 335n30 horn silver. See silver chloride hornos castellanos 55, 69, 70, 88, 192 chimney height 71, 350 See also under Regla

428 horses 160n See also under Regla horseshoes, as source of iron 229 Huancavelica 12, 37, 43, 360 major supplier to Upper Peru 38 mercurialism 146, 354n61 production costs 302n93 Humboldt, Alexander von 25, 30n58, 255n3, 313 on barrel process 304 on chemical conversion of mercury 120n61 on lining patio reactor 158, 159, 190 on losses of mercury 127 on milling expertise 157 on patio process 159, 279 on profits from mercury 302n92 on refining choices 291 silence on refining costs 261, 291 on silver content 34 silver production by refining process 335n30, 336n31 hydraulic energy. See power, water hypogene 26, 27, 29n57 Idria 37, 300, 302n92, 351n51, 374n2 India, lack of major silver deposits 9 indigenous communities 170, 193 and charcoal production 96 hidden cost of silver production 6, 311, 347, 364 and silver mining 27 indigenous workforce 150n131, 256, 267n30 attrition of 315 as capital assets 63, 64 Europeans on 353n58, 356, 364 life of 147, 356 occupational hazards 100 punishments for 74n87, 81 skills of 30, 52, 315 wages, 315 (see also under Regla: labour force) Informe Mensual 202, 239, 367 ingenio (machine) 71, 157 ingenio (refining centre) 135, 147, 176n48, 350, 362 investors. See private enterprise iron 50, 55, 110, 115, 187, 220, 305

index capellinas made from 127, 165, 166, 168, 215n, 216 to coalesce lis 141 and correspondencia 125, 292, 374 horseshoes as source 229 lining patio reactor 159 and lower mercury consumption 119, 121, 136, 139 role in refining by mercury 141, 253 jales 97n, 161n17 James ii (King of England) 107 Japan 270 Jesuits and mercury recipe 144, 146 refining scenario 144, 146, 359 silver production 10, 144 subduction 16 jarositic ores 22, 57, 58 Joanna of Austria 143n121, 301 Jonson, Ben 107 Jung, Carl 106 Kutná Hora

21, 113

lamas 75, 141, 161n17 Landesio, Eugenio 209, 217 Laur, M.P. 168, 233, 261, 263 on mercury losses 128 Laurion 75n88, 85 lavaderos 161, 178 lavadores 69 lead airborne, historical deposition of 132 argentiferous (see galena) in assaying 46 barrier to mercury refining 21, 47, 122, 143n121, 234n11, 264 bars 55, 73, 87, 218 content in smelting recipe 82–84, 263 and copper liquation (see saigerprocess) as co-product of silver 36, 43, 256, 271, 294–296, 313, 361 endowment, modern Germany and Mexico 3, 16 environmental impact 341, 345, 346 (see also under lead fume) environmental legacy 344, 346, 349, 360 European deposits 21, 92 as fluxing agent 12, 21, 55, 99 fugitive, toxicity of 81

429

index and Japan 144, 146 leaching of, from slags 81 in lead fume 79, 82, 97, 100 from lead sulphide 55 from litharge 55n37, 57 losses, during smelting 77, 84, 85, 248– 250 mass balance of, in smelting 245, 248 melting point 55n37 Mexican deposits 24, 31, 39, 304, 359, 365 occupational hazard 80, 89, 100, 350, 355 in slags 75, 81, 85, 100 and smelting of silver ores 21, 24, 39, 47, 52, 55–58, 79, 99, 111 supply from England to New Spain 21, 92, 93 See also under Regla (smelting); smelting lead fume 79 arsenic in 85, 86 and chimney height 71, 350 composition of 79n, 82 depositional footprint 87, 88, 98, 364 emissions 79, 84, 248, 339 and flue traps 80, 89, 90 heavy metals, main source of 7, 253, 315, 344, 363–365 and Mexico 345 and New Spain 341, 345, 346 silence on 78n, 100, 352, 354, 356, 357 in smoke from smelters 86, 87, 350, 354, 363–365 toxicity of 7, 78n, 80, 89, 364 workplace concentrations 80, 87n21, 97, 100, 250, 354, 357 worst case scenario 360 See also under Regla (smelting) lead oxide. See litharge lead sulphide. See galena, argentiferous lead toxicity 80 and children 81, 97, 346 historical knowledge of 100, 350, 354, 356 influence on policy 7, 350–352, 356, 357 and livestock 89, 91 (see also bellanding) poisoning 355, 357 silence on 354–356, 364 and speciation 82 lime 69, 70, 141 Linderman, H.R. 270 Lira, José M. 179

lis 140, 141 litharge 56, 57, 93, 219, 263 See also greta lixiviación 338 Lobato, Laguna 132 Lohman, Gaspar 118, 134n90, 141 Loman, Gaspar 117, 118 López de la Madriz, Juan 59, 60 Lull, Raymond 105 machacado 28 Mackenzie, Roderick 216 Mackintosh, A. 216 Maestrazgos 297 Magisterio 141 magistral 137n102, 141, 142, 313 See also copper sulphate magma 13, 24 Magnus, Albertus, Book of Minerals 102 manga 163, 164, 190, 212 maquila. See tolling marqueta 166n32 Mauro, Giovanni Antonio 113 Maximillian, Joseph (Emperor) 193 mazamorras 75 Medina, Bartolomé de 100, 103, 118, 134, 141, 142, 143n121 Memorias de Gastos 202, 368 merced 118, 134n90, 143, 357 mercurialism 115, 349, 351, 355, 357n67, 363 accidents 128, 353, 354 in Huancavelica and Almadén 354n61 Rawson’s report on 353n58 silence on 2, 348, 353–355 See also azogados; mercury toxicity mercuric chloride 117, 134n90, 357 See also solimán mercurius vivus 108 mercurous chloride 7, 120, 357n68 See also calomel mercury airborne deposition, historical 132, 348 anthropogenic losses of 103, 347 and artisanal gold refining 131, 164, 364 assaying with 46, 47, 112 consumption breakdown 189, 374 contact with workers 188, 188n61, 253 conversion to calomel 7, 119, 121, 124, 191, 349, 363

430 counterfactual scenario 146, 304, 358, 360, 365 debt to Crown 298, 299 deceitful 46 density of 119, 120 deposits, and Spain 37, 43, 361 (see also Almadén, Idria, Huancavelica) domination of historical narrative 2, 4, 266, 267 dual role in refining 119, 123, 124, 152 effect of iron, copper 121, 136, 137, 139, 229 (see also correspondencia) and environment 187–189, 341, 343– 345 (see also under Regla (patio process)) environmental legacy 190, 253, 346, 347, 349, 360, 363 fiscal advantages of 299–304, 337, 359, 360, 362 and Fuggers 22, 296–299 global trail of refining with 115, 117, 143, 144 and Japan 359 as key to American silver 4, 266, 304, 358, 360 as lesser environmental evil 365 lifecycle, environmental 189n losses accidental 128, 188 casting of silver bars 127, 188 liquid 127–131, 161, 360, 363 minimal loss 127, 128, 191, 348, 349, 360, 363 observer’s reports on 3, 127, 128 physical 127, 128, 131, 191, 347 volatile, modern claims for 3, 131, 132, 152, 190, 364 See also under capellinas magnitude of use 176, 347, 363 mitigation of environmental impact 7, 8, 363 monopoly by Spain 362 occupational hazards 350, 351, 362 opportunity cost to Crown 301, 359, 360 and ores with high silver content 264, 266, 282n59 patents on refining with 113 production from cinnabar 8

index reactions during patio process 120–124 recovery from amalgam 128, 164–169, 349, 364 recycling from hacienda soil 129, 130 refining with, in Upper Peru 134, 135, 138, 139 revenues from Crown monopoly 136, 261n13, 300 and silver chloride 120 supply and bribery 150, 151, 317 surface tension 119 (see also lis) ‘sweet’ 120 use in refining silver ores. See patio process weight ratio to silver 124, 229, 375, 377 (see also correspondencia) mercury (alchemy) 103, 362 ontological role 104, 109 popular image of 107 as prima materia 104, 105 and refining lore 110, 124, 140 Sophic 105 and Spanish Crown 109 symbolism 106 transition to metallurgy 107, 108 and transmutation 106, 107, 110, 124 uses in 106 mercury amalgam with lead 47, 122 with silver 118–120 mercury prices decrease (18c) 258, 260, 261 fluctuations (19c) 274, 275, 301, 302 margin of profit by Spain 301 polemic on (18c) 258–260 and silver production and royalties 260, 304 subsidy, the question of 302, 303 and the switch to smelting 299 timeline 301–303 mercury toxicity 115, 350, 351n51, 358 influence on policy 7, 355, 356 knowledge of, in Spain 351, 352, 356, 362 prohibition by Inca kings 351 See also mercurialism; solimán metallogenesis Aristotle on 104 sulphur-mercury theory 104, 105

index methylmercury 189n Metztitlán, Laguna de 248 Metztitlán, Rio 248 Mexico (19c) English investment 198 environmental impact 344, 345 political turbulence 192–194 silver production 193, 194 ‘time not important’ 309 Mexico industrial heritage of 254 lead endowment 39 metallogenic provinces of 30, 31 woodland resources 94, 347 milling christening of mills 157 as cost factor 262, 290 dust 188 noise level 188, 205, 251 for patio process 115, 119, 152, 157n6, 355 for smelting 67n65, 355 See also under individual milling equipment mineral waste. See waste, mineral mineros 1, 45, 46, 256 attitude of 49 belief in transmutation 110, 124 as capital assets 355n64 early lack of skills 49, 50, 53, 54 mining colonial 8 immediate dangers of 355, 364 mining bubble, Anglo-Spanish American 195 mita 146 Mitteleuropa 1, 6, 22, 51, 297, 361 molinos 154, 156f, 157, 178 advantage of 155 circular stones 66, 68f introduction in New Spain 67 Monte Caldera 60, 65, 90 Montesinos, Fernando de 134 morteros 154, 155, 251 See also stamp mills Mota y Escobar, Alonso de la 52, 142 mules 71, 160n16, 172, 182, 188 See also under Regla

431 negrillos 29, 47, 121, 122, 135, 137, 139, 142, 362 Nevada, u.s.a. 2, 10, 137n103, 239, 244, 315, 354 See also Comstock Lode New Almaden 275 New France, and major silver source 10 New Spain environmental impact 339–343 ‘fiscal submetropolis’ 256 silver production 1, 194, 318, 335 Treasury 317 occupational hazards mentalité of the period 356, 357 silence on 355, 356, 364 Ojo de Agua 200 Olivares, de (Count-Duke) 299 operational practice 2, 7, 69, 128, 169, 349, 354, 364 ores, silver. See silver ores Ortega, José Antonio de 129 Ortiz de Zárate, Juan 136 Oruro 15 oxidised zone. See under silver ores: deposits Pachuca 7, 94, 103, 134, 195, 329 European visitors to 196 major epithermal deposit 18 revenues from mercury 300 Pacific Ocean, mid-ocean ridges 13 pacos 28, 30n58, 137 See also colorados pan process 138n106, 153 silver extraction rate 239 Pánuco, Real de 86 Paracelsus 108, 351n51 paradigms 14n11 anomalies of 348 new 349 and physical loss of mercury 347 Parkes’ process 219 patents, on silver refining with mercury 101, 113, 144 patio process 103n4, 153–155 advantages of 148, 149, 160, 312, 360, 362, 363 and alchemy 110, 124, 140 architecture, industrial 6, 152, 169, 170 and barrel process 304–309

432 chemistry, maturity of 124, 154, 190, 192, 195, 362 confusion with gold amalgamation 46, 131, 151, 152, 364 costs 263 capital 279 compared to smelting 264, 266, 293, 294 eighteenth century 257–260 and European silver refining 294–296 fixed 258 historical range 287, 288 mercury as highest cost factor 264 milling 262 nineteenth century 262, 263 projected (16c to 18c) 289–293, 370, 372 for ‘rich’ ores 264, 266 and silver content 258, 266, 294 sunk 258 and value of silver 296 and wages 315 counterfactual scenario 358–360 and cyanide process 103, 144, 192 environmental impact 170, 187, 189, 191, 245, 341–345, 363 environmental legacy 7, 190, 191, 346, 347, 360, 365 European expertise, limited role of 2, 53n30, 148 false positives 293–309 as finishing stage to cazo process 138n105 firewood, consumption 313, 364 fiscal advantages 300, 301, 303, 360 fringe process in Europe 2 historical stages 118, 133–143, 148, 192 Ibero-American singularity 103 industrial nature of 6, 152, 309, 361, 363 as key to American silver 45, 304 labour force 293n69 lead, interference from 46, 47, 58, 99, 122, 264, 290, 358 longevity of recipe 133, 139, 363 mature process 148, 194, 312, 314 merced for 118 mercury, consumption (see under correspondencia; mercury) as Mexican ‘trick’ 309 mixing, effect of 122n65, 234n10

index operational areas, relative size of 182, 186, 190 ore grain size 67, 157, 205 ore surface area 119 ores excluded from 47, 111, 123n67, 148, 290 as peerless process 3, 42, 148, 152, 154, 160 and political upheavals 194 ‘poor’ silver ores, narrative of 4, 33, 47, 264 production limit in New Spain 304 profits and efficient operation 2, 136, 222, 309, 311, 313, 314, 361 and innovation 148 and mercury price 257–260, 266 and ore extraction cost 258, 259 and threshold silver content 258, 261n12, 262, 263, 290–293, 312 ratio of consumables to silver refined 339–340 recipe and chemistry of ores 48, 108, 115, 117, 133–143 replication in laboratory 120, 348 right for all the wrong reasons 14, 139–141 silver extraction, time function 233 silver ores, loss by entrainment 161 and silver production 335, 338 and silver sulphide ores 47, 117, 121, 135– 137, 139, 361, 362 and skills 53, 149, 194, 359 and smelting 300, 312, 349, 365 and tailings 30, 135n96, 258, 303, 312 and Venice 22, 113, 114 as wet process 103, 104, 189 See also under Regla (patio process) patio reactor 153, 158f, 159f, 160f, 176, 183f, 184f, 363 area and production formula 182, 234 from canoas to 176 as chemical reactor 158, 159 flexibility of 160, 190 lining of 129n83, 158, 190, 209 and milling 178, 179, 234 roof 159 subsoil 161, 346 See also under hacienda de patio; Regla Pattinson process 219 pella 134, 211, 215

433

index Peñón Blanco 38 peones 239, 240 Percy, John 262n16, 296 Philip ii (King of Spain) 109, 114, 142, 297, 301 Philip iv (King of Spain) 299 Phillips, John 294 pieces of eight 96, 310, 316 pigs 55, 87 See also barras pilfering 190, 244, 268n35 piñas 165, 173 planillas 161, 162f, 210, 211, 222 planilleros 161, 162f, 240, 243, 294n69 plata de azogue 151, 320 plata de fuego 75, 77, 320 Plato 104 plomo rico 55 See also pigs plumbism 350 See also under lead toxicity poor lead 79, 81, 82, 219 porcelain 311 and alchemy 109, 110 displacement of production 41 Portugal gold (Africa, Brazil) 9, 42 sugar 40 Potosí 10, 15, 348, 354n61 early refining with mercury 135 historical deposition airborne mercury 132 innovations in recipe 122, 133–143, 300n86, 363 motto 16 nature of ores 135 peak in production 337 silver supergiant deposit 16 spike airborne lead and silver 132 power animal 67, 69n69, 157, 182, 262 cost of 262, 263n21, 264, 312 water 69n69, 71, 135, 157, 187, 262, 287, 293 priests, role in refining silver 1, 44, 46–48, 86n17, 149, 163n22, 267n30 Jesuits in Japan 144, 146 private enterprise and innovation 148, 314, 361

profit margin 255 risk of failure 2, 256 in San Luis Potosí 64, 75, 77, 100 and silver production 1, 256 support from Crown 5, 256n3, 311, 356 transmission of know-how 149 Proaño 181, 279n50 profits, of silver refining base metals, role of 2, 36, 58, 361 (see also under copper, lead) difference between Europe and New World 36, 294–296 difference with other commodities 310, 311 and efficiency of operation 2, 136, 222, 311, 313, 314, 361 and extraction cost of ore 258, 289, 312 guaranteed revenue at source 6, 311, 314 magnitude of 310, 311, 314 and production volume 315 role of indigenous labour 277, 278, 312, 315 and technology at source 314 See also under each refining process projection (alchemical) 106 See also transmutation proustite 29n57 pyrargyrite 28, 29n57 quemadero 178 See also capellinas Quevedo, Francisco de 107 quicksilver 93, 103, 107, 112, 113, 127, 129, 165n27, 209 See also mercury Rawson, William 353n58 Real de Minas 25, 170n41, 318, 324, 325, 328, 357n67 Real del Monte 31, 195n6, 198, 200, 216, 234n11, 262n16, 306, 330, 346n37 Real del Monte Company 201, 216, 262n16, 305, 353 reduction 24n39, 27, 39, 54n34, 55n37, 101, 123n68, 124, 125 refining costs. See under each refining process Regla 185f, 210f absence of plan 202, 213

434 accounting records 3, 7, 191, 200, 202, 222, 239, 268, 286, 367–369 afinación 219 architecture 198 artificer-architect 202, 222 azoguería 212 blast furnaces 218f, 243 doubts on new 216 efficiency of 237, 364 lead fume control 217, 250–252 overcapacity of 235 Satanic mills 217 capellinas 212–216, 214f, 215f, 222 as case study 195, 200, 253, 312 charcoal (see under Regla (smelting)) Church 209, 219 copper sulphate (see under Regla (patio process)) cost of construction 278, 279 dual refining methods 201 environmental impact, aggregate 248– 251 environmental legacy 251, 252, 360 footprint 203f guest quarters 204, 205, 219 hornos castellanos 204, 211, 216 chimney stacks 212f, 213 Christian inscriptions 212 smoke from chimneys 185f, 204, 251 horses 208, 209, 219, 220, 222, 251 horseshoes, as source of iron 229 interruption in production 367 inventory, management of 195, 224, 232 labour force daily life 244 efficiency of 195 maintenance 239, 240, 243 partitioning of 239–244 supervision of 240, 251 wage structure 239–243 work period 240 litharge (see under Regla (smelting)) location 200 materials control of 203, 211, 212 internal flow of 203, 222, 223 mass balance 220, 222, 249 mercury (see under Regla (patio process))

index milling arrastres 204, 205, 206f, 207f Chilean mill 204 and rainfall 227 stamp mills 204–206, 239, 251 synchronicity with patio operations 224 throughput 224n modern alterations to 220, 221f mules 200, 208, 220, 222, 229, 251 ores chemistry of 234n11, 237, 252 inventory, just-in-time 224 sampling errors 35, 237, 239 silver content 227, 237, 252 overdesign of 243, 279 overview 251–254 owners English 198 (see also Adventurers) Mexican 199n18 patio reactor 208 impact of rain on 209 perimeter walls 220 planillas 210, 211, 222 planilleros 243, 294n69 process areas 202, 204 and efficiency of layout 222 reverberatory furnaces 213, 216, 218, 219, 251 salt (see under Regla (patio process)) silver extraction rates, comparative 237– 239 silver production end 18c 198 1872–1888 220, 253 by refining process 198, 220 silver, storage of 212, 222 site 198 storage vaults 202, 203, 211, 212, 213f, 251 tortas 195, 208–210, 222, 232, 243, 245, 249, 251, 253, 286, 353n58 transit corridors, mass flows 222 visitors’ impression 196, 196n10, 196n9, 202, 205, 210, 213, 216, 244 washing tanks 210, 211, 222, 243 water, uses of 208, 222, 227, 243, 245, 251, 252 water channels network 205, 208 water-wheels 205

435

index Regla (patio process) 268, 276–279 calomel, in solid waste 245, 248, 251, 252 copper sulphate consumption of 229, 232 costs 276 inventory of 232 environmental impact (see under individual consumables) fuel use 234, 235, 250 iron from horseshoes 229 labour 239, 240, 243, 244 mass balance of materials 245–249 mercury consumption of 232, 245 costs 274, 275 entrained by workers 249 inventory of 232 losses of 245, 249 in soil 249 weight ratio to silver 229 milling output 224, 227 ores extraction costs 276, 277 silver content 227, 252 production costs 276–279, 285 profit margin 285–287 reaction period optimisation 232–234, 253 silver extraction as function of 233 recipe, interactive application of 229 salt consumption of 229, 232 costs, including freight 271 inventory of 232 silver production 227 value of 282n58 waste, solid losses of 245 voided into waterways 248, 251, 252 Regla (smelting) charcoal 222 consumption 237 costs 274 inventory 237 weight ratio to silver 237, 252 woodland, depletion and regeneration 95, 237, 250–252

environmental impact (see under individual consumables) initial strategy 216, 235 labour 241–244, 280, 281 lead, losses of as fugitive lead 249 as lead fume 90, 217, 248, 250–253, 355n64 weight ratio to silver 248 within operational areas 250 litharge consumption of 237 costs 274 weight ratio to silver 237 mass balance of materials 245, 247, 248 ores extraction cost 280 irregular supply 235 lead content 237 silver content 237, 252 overdesign of capacity 243 poor lead, as by-product 219, 245, 248 production costs 280, 281, 285–287 slag 220, 245, 248, 251 refiners alchemical lore 110, 124, 140 best operational practices 2, 7, 128, 316 challenge faced by 30 worst modern practices 364 reverberatory furnace 56, 73n85 See also under Regla. Rio Tinto 22, 57, 58, 95 Rivot, Louis Edouard 295 roasting 122, 139, 141, 149, 262, 304n96, 308 Rothschilds, and mercury monopoly 275 royalties 260, 303n94, 360 See also tax revenues Rugendas, Johann Moritz 204 Rule, John 216 saigerhütten 22 saigerprocess 22, 51n24, 92, 111, 149n130 salt (sodium chloride) 38 as critical ingredient 115, 122, 313 deposits 38 environmental impact 341, 343, 345 and mortification of ore 140

436 soluble waste 363 See also under Regla (patio process); roasting San Luis Potosí (State) local service industry 75, 77 major smelting region 58, 65, 99, 359 mines of argentiferous lead 3, 99 ruins of smelting haciendas 58, 99 See also Cerro San Pedro San Luis Potosí (city) and grasas 63 Sardaneta, Pedro de 158 scales, weighing 163, 212 Schio 101, 113, 114, 363 Schürren, Johann 143n121 Semería, Felipe 178 señoreage 311, 320, 321 Seville 50, 52, 92, 102, 103, 111n35, 357, 359 silences in historiography 2, 364 silicosis 355 silt 115, 141, 161, 187, 360, 363 See also waste, mineral silver amalgam 104, 118, 120, 122 arbitrage trade, China 310 bars, casting 127, 149, 188, 213n40, 234, 239, 245 commodity trade 5, 40–42, 310, 311 contraband 317, 362 elemental 26, 43, 54, 119–123, 126 endowment, Mexico, Peru and Bolivia 16 melting point 54 millesimal fineness 269n36 native 27–29, 51, 55n38, 126, 135, 137, 151 (see also silver: elemental) price, in London 269, 270 as primary metal in New World 20 impact on refining economy 295 refining costs. See under each refining process royalties and taxes 255n3, 320, 321, 360, 374 and mercury revenues 260, 300–304 secondary metal in European deposits 20 impact on refining economies 294– 296 value 285, 310, 314 at colonial source 311

index impact of single standard 270 set in Europe 296 See also silver to gold ratio volatilisation during smelting 54, 55n38 and weathering 27 silver chloride and beginner’s luck 57 in colorados 29, 30n58 ease of smelting 51, 54 melting point 54 in New World ores 27n51 reduction by charcoal 54, 57, 93, 100 copper or iron 121, 123n68, 124, 136, 137, 139, 229 mercury 120, 123n68, 124, 125, 139 and weathering 26, 27, 29, 51, 54 See also silver halides silver content and ease of smelting 35, 51 ‘low’, relative nature of 35, 36 narrative of ‘poor’ 4, 30 as primary metal 33, 34, 36 for patio process 182, 227, 262–264 for smelting 263 as red herring 32–36 as secondary metal 20, 36 for smelting 58 silver halides 20n31 and weathering 26, 29, 43, 54 See also silver chloride silver ores assaying of 35 chemical profile in New World 24–28 degree of milling for patio process 119, 157n6, 204 for smelting 67n65, 204, 205 deposits abundance in New World 9, 15n15 age of 20 altitude, as sign of age 32 and aridity 29 challenge posed to refiners 30, 43, 362 chemical profile of 25, 27, 28, 361 Europe and New World 6, 20, 32, 33, 361 geological genesis of 22, 24, 43 (see also subduction)

index global imbalance of 9, 12 and gold 24 as immoveable object 42 oxidised zone 26, 29 primary 16 supergene enrichment 27, 43 water table 27–29, 43 weathering of 26, 27, 29, 54, 120 dressing 67, 69, 75, 358 strategy of Adventurers 69, 313 See also buddling dry 33, 55, 65n62, 79, 92, 99, 114, 285 extraction cost 258, 259n9 See also under Regla (patio process) lead-poor. See dry ley. See silver content organoleptic inspection of 1 for patio process 258, 262, 263 sampling, problems of 35, 237 varieties of 25 silver oxide, absence of after heating in air 55, 56 in oxidised layer 26 silver production balance between refining processes 337 and geochemistry 12 Hispanic New World 1, 12 method of estimation 320, 321, 363, 374 Mexico (19c) 1, 318 aggregate, by refining process 338, 363 reasons for increase 271 by smelting 363 New Spain 1, 10n5 aggregate, by refining method 335, 363 by Caja and refining process 321–336 by decade and refining process 336– 338 effect of mercury pricing and supply 337 effect of silver sulphide ores 54 official tax records 318 sex as factor 267 silver refining in New World 1, 2 business model 5 cautionary tale, Hispaniola 18, 19 chemistry as gatekeeper 5, 7, 254

437 counterfactual history 146, 304, 359n73, 365 economic spaces 346n37 environmental impact 316 compared to California gold rush 344n36 and Crown policy 360, 365 economy of scale 187 geochemical context 7, 12, 315, 360 geographical mapping over time 98, 318, 346 on indigenous communities 6, 244 in Mexico 339–343 modern dating techniques 78 multidisciplinary approach 349 in New Spain 253, 318, 328, 339–346 See also under each refining process and European experiential knowledge 2, 6, 19, 43, 144, 147, 176 industrial nature of 3, 144, 152, 253, 347, 361 peerless technical innovation 3, 42, 314, 364 profits (see under each refining process) technology at colonial source 148 silver sulphide 26n45, 239, 252, 361 conversion to silver chloride 27, 54, 122, 123, 139 and Japan 144, 146, 359 limitations of mercury in assaying of 47 in reduction of 47, 121 as sole refining option 146, 360 melting point 54 in negrillos 29 reduction during smelting 362 refining challenge posed by 43, 50, 51, 57, 135–137, 139 role of cupric ions 117, 122, 137, 234n10 silver to gold ratio 269, 296, 311 skills, refining initial level in New World 49, 50, 53 in Spain 50 slag 53, 55, 57, 74, 75, 85, 95, 97, 100 See also grasas slaves, African 1, 49, 63, 100, 353 exposed to solimán 357 in mining 48, 355 punishment for 82

438 role in cotton, sugar 40, 41, 310 technical role in smelting 52 slurry, of silver ore 115, 120, 135, 159, 229 heating of 137, 138n106, 153, 239 mixing of 122n65, 153, 160n16, 163, 188, 234n10, 349n44 separation of amalgam 128, 161, 208, 210, 211 smelting 44n, 55–57, 58n45, 66n, 192 balance with patio process 312 Cajas dominated by 335 challenge in New Spain 313, 358 charcoal role of 93 weight ratio 93, 94 (see also under Regla (smelting)) as complex process 53, 57 costs in Europe 295 in Mexico (see under Regla (smelting)) in New Spain 263–266 projections 16c to 18c 287–293 counterfactual scenario, New Spain 365 early history, New Spain 48–53 claims of early demise 45 and environment 78, 86, 97, 100, 344, 346, 347, 360, 363, 364 evolution of 192 furnaces 69, 70, 72f, 74f chimneys 350 See also hornos castellanos; Regla: hornos castellanos; reverberatory furnace and lead consumption 79–82 per kg silver refined 85, 100 percentage 84, 85 lead fume from 79n, 354, 364 ground deposition of 88 lead to silver ratio, in recipe 82–84 minor role, claims of 4 in New Spain, limited sources on 98, 99 origins of 55 profit, role of base metals in Europe 36, 58, 294–296, 313, 361 proto-smelting period in New Spain 53 ratio of consumables to silver refined 339–340 and ‘rich’ silver ores 264 silver content

index technical threshold 58 threshold for profit 36, 263, 291, 293, 295 silver loss 262 silver produced by New Spain and Mexico 363 skill, level required 53, 54 slag. See grasas smoke, from 88f, 89f, 350, 364 complaints against 86, 87 (see also bellanding) occupational hazards 87–89, 100, 354 as two stage process 55–58 as universal refining process 57, 58, 264 and water requirements 69 See also under hacienda, smelting; Regla (smelting) sodium chloride. See salt Soetbeer, Adolf 318 solimán toxicity of 117, 134n90, 144, 358 trials in mercury refining recipe 357 See also sublimate; mercuric chloride Sonneschmidt on barrel process 304 on lead as impediment to patio process 123n67 on losses of mercury 127, 129n83 on mercurialism 356n67 on refining costs 261 soroche 47 Spanish Crown and alchemy 109 assistance to miners 303n94 debt to Fuggers 297–299 debt by refiners 298n83 geopolitical advantage in silver 6, 9, 12, 18, 37, 42, 43, 361 interest on refining with mercury 143n121 lack of skilled silver refiners 362 and lead mining 359 profit on mercury 301 and refining hazards 355 silver refining policy 300, 302n92, 304n95, 311, 313 smelting option 304 Spanish Treasury 143, 300 revenues from mining in Spain (16c) 50 specie 311, 314, 316, 321

index spice trade 5, 40, 314 displacement of sources 41 mark-up 310, 314 stamp mills 73n82, 135, 154, 157n6, 188, 290 See also morteros; Regla: milling Sternberg, Pedro 109 stoichiometric ratio 123, 125 stoichiometry 125, 152 subduction 13, 14, 16, 19, 22, 24, 42, 193, 361 and global silver deposits 15, 43 longevity in Americas 15 sublimate 115, 117 See also mercuric chloride; solimán subsidy 302, 303 sugar 5, 10n4, 41, 42, 170n42, 310, 311, 314 displacement of sources 40 sulphur 26n45, 87, 104–106, 125, 141 supergene sulphide enrichment 27, 43 sweet mercury 120 See also calomel tacana 28 tahona 157, 178, 182, 188 See also arrastres tailings 1, 30n59, 197n11, 289, 362 effect of mercury price 303 factor in peak production 338 and gold amalgamation recipe 151 silver content 337 sunk cost of 258, 259, 312, 337 tax revenues 255n3, 256, 260, 301–304, 318, 321 as base line 317 conversion to silver production, mercury sales 374–398 and tk set 320 See also royalties Taxco, and early mercury recipe 142 Taylor, John 216, 305n99, 313 technology, at colonial source 148, 311, 361 tectonic plates 13, 14n11, 15, 24, 361 temperature range of camp fire 54 of smelting furnaces 56 Terreros, Pedro de 197, 200 planner of Regla 202n24 See also Conde de Regla (First) Theophilus 102 Tilman, E. 180n52

439 tk set 320, 321 Toledo, Francisco (Viceroy Peru) 71n80, 350 tolling 239n17, 260n11, 264, 292, 314 toneles 304 See also barrel process Tordesillas Line 9n3 Torres y Portugal, Fernando de (Viceroy Peru), 136 tortas 153, 159, 179, 188, 229, 232, 234n10 transmutation 103n5, 104, 107, 108, 113, 137n102 and amalgamation 110 and mercury 106 trials and Philip ii 109 tuyère 70 Twain, Mark 44, 244 United States of America (u.s.a.) 42, 310 invasion of Mexico 193 leading silver producer 16, 42, 193, 269 mercury production 274 and Mexican refining know-how 53n30, 138n106 mills and mercury loss 129 and subduction 16, 193 See also Comstock Lode Upper Peru 8, 16, 134, 137, 142, 153 u.s. Mint 270 Uyuni 38, 43 Valle de Pozos 54 Variscan orogen 19, 20n29 vasos 59, 73, 74, 88 Velasco, Luis de, (Viceroy, New Spain) 143n121, 357 Venice 6, 22, 48, 298 and metallurgy, alchemy 113 patents on silver refining with mercury 113, 117 refining experts and Philip ii 114 verdigris 115 composition of 117 early recipe New Spain 141 Vetagrande 39, 83, 97n50, 150, 187 Vilanova, Arnaldo de 105 Villaseñor y Sánchez, José Antonio de 257, 261n13, 262n18, 277, 281n57, 285 breakeven pricing for mercury 260

440 defense of mercury price extraction cost 312 vinegar, role of 115 Virgen de Regla 197 voladora 157, 227n

index 259

Ward, H.G. 205, 209, 220 Washoe process. See pan process waste, mineral 115, 163, 187, 210, 245, 252, 363 flooding due to 187, 188 and landfills 188, 191, 346, 347 in Mexico 345 in New Spain 341, 342, 344, 345 in streams, waterways 128, 161, 339, 363 water, role in patio process 187 smelting 69 water table 20n31, 27–29, 43

weathering. See under silver ores: deposits Wedgwood, Josiah 41, 255 Welsers 51 women, role in silver refining 1, 30n59, 45n3, 100, 161, 240 woodlands charcoal ratio 94, 95 depletion of 341–345 estimated per kg silver refined 364 and population 347 regeneration 95, 98, 339, 346, 347, 359n73 resources, Mexico 358 Ymporte de la Memoria de la Mina Yocalla 38

369

zangarros 170, 172, 182, 190 Zimapán argentiferous lead mines 198, 216, 235 lithograph of silver smelter 88f