Bumblebees (Vol. 6) (Naturalists' Handbooks, Vol. 6) [3rd Revised & enlarged] 1907807063, 9781907807060

Citation preview

6

Bumblebees

An indispensable guide to identification, ecology and study of bumblebees.

ISBN 978-1-907807-06-0

90000 >

PUBLISHING

www.pelagicpublishing.com

9 781907 807060

OLIVER E. PRŶS-JONES & SARAH A. CORBET Plates by Tony Hopkins and foreword by Mark Avery

Pelagic Publishing

B. N. K. Davis Insects on nettles Valerie K. Brown Grasshoppers Peter F. Yeo & Sarah A. Corbet Solitary wasps Margaret Redfern Insects and thistles Francis S. Gilbert Hoverflies Oliver E. Prŷs-Jones & Sarah A. Corbet Bumblebees Peter L. Miller Dragonflies Trevor G. Forsythe Common ground beetles Peter J. Hayward Animals on seaweed Michael Majerus & Peter Kearns Ladybirds Graham E. Rotheray Aphid predators Marjorie Guthrie Animals of the surface film Janet Harker Mayflies Keith R. Snow Mosquitoes D. M. Unwin & Sarah A. Corbet Insects, plants and microclimate M. G. Morris Weevils Margaret Redfern & R. R. Askew Plant galls William D. J. Kirk Insects on cabbages and oilseed rape D. H. S. Richardson Pollution monitoring with lichens Marjorie Hingley Microscopic life in Sphagnum Peter J. Hayward Animals of sandy shores C. Philip Wheater & Helen J. Read Animals under logs and stones Zakaria Erzinçlioğlu Blowflies Gary J. Skinner & Geoffrey W. Allen Ants William D. J. Kirk Thrips David T. Salt & John B. Whittaker Insects on dock plants Simon R. Leather & Keith P. Bland Insects on cherry trees C. Philip Wheater & Penny A. Cook Studying invertebrates Tony Dixon & Thomas Thieme Aphids on deciduous trees

PELAGIC

3rd edition

Bumblebees

The Naturalists’ Handbooks series: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

Prŷs-Jones & Corbet

This new edition embraces the wealth of information published on bumblebee life history, ecology, foraging, parasites and conservation in recent years. It includes a new chapter on the very real threats to bumblebees; their crucial role as pollinators of our native flora and crops; ways to promote their survival; advantages and problems posed by their commercial use; as well as updated colour plates, keys and distribution maps of all British species (including Bombus hypnorum). The book introduces techniques and approaches to original work so that anyone with an interest can usefully contribute to furthering our understanding and appreciation of these wonderful and important insects.

PE

PUBLISHING

PELAGIC PUBLISHING

Naturalists’ Handbooks 6 Ecology and identification

Naturalists’ Handbooks 6

Bumblebees OLIVER E. PRŶS-JONES Ysgubor Ystlum, Waen, Bodfari, Denbigh, North Wales SARAH A. CORBET St Loy, St. Buryan, Penzance, Cornwall With illustrations by Tony Hopkins

Pelagic Publishing www.pelagicpublishing.com

Published by Pelagic Publishing www.pelagicpublishing.com PO Box 725, Exeter, EX1 9QU

Bumblebees (3rd Edition) Naturalists’ Handbooks 6 ISBN 978-1-907807-06-0 Reprinted 2014 with corrections and revision of the identification keys Text © Pelagic Publishing 2011 Key Illustrations © David Alford Other illustrations © Tony Hopkins All rights reserved. Apart from short excerpts for use in research or for reviews, no part of this document may be printed or reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, now known or hereafter invented or otherwise without prior permission from the publisher. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library.

Contents Preface Acknowledgements Foreword by Mark Avery

iv v vi

1 Introduction

1

2 Distribution and recognition

2

3 The natural history of true bumblebees

8

4 Nests and their establishment in captivity

27

5 Cuckoo bumblebees, parasites and nest associates

34

6 Foraging behaviour

42

7 Threats, conservation and commercial use

62

8 Identification

73

Chart A. Is the specimen a true bumblebee or a cuckoo bumblebee?

76

Chart B. Is the bumblebee a male or a female (queen or worker)?

77

Quick-Check Key to the commonest species of true bumblebees

78

I Female true bumblebees Bombus II Male true bumblebees Bombus III Female cuckoo bumblebees Bombus (subgenus Psithyrus)

79 85 91

IV Male cuckoo bumblebees Bombus (subgenus Psithyrus)

92

9 Approaches to original work 10 Further reading

Synonymy

11 Index Distribution maps

94 108 115

116 119

The cover illustration is designed to be used in conjunction with the Quick-Check Key (p. 78)

Preface The first edition of this book appeared over 20 years ago, and since then a great many dedicated individuals have added a wealth of information to what we know about the lives of bumblebees. A strand that emerges from these studies is one of affection and also, increasingly, of concern - that we run a grave risk of losing admired, respected and very important ‘friends’. Such are the feelings these insects generate in us as individuals, and in the public mind: of beauty, industry and utility; the often unrecognised agents of security, through pollination, of much of the fabric of our native green environment. We lose them at our peril. In this new edition the text and reading list have been revised and brought up to date, species new to the British list have been included in the keys and plates, a new chapter (on conservation) has been added, and the distribution maps have been updated. The book was based around the research of OP-J; we are retaining that as the background in order to continue the sense that anyone with a personal interest can contribute usefully to furthering our understanding of the lives of bumblebees.

Acknowledgements OP-J acknowledges the late Professor Sir James Beament, who kindly provided facilities in the Department of Applied Biology, as it was then, in the University of Cambridge, and the Natural Environment Research Council and Gonville & Caius College, Cambridge, who gave financial support. Our appreciation of general bumblebee biology was initially enhanced by the writings of Sladen (1912, republished 1989), Plath (1934), Free & Butler (1959), Alford (1975) and Brian (1980), and then improved by many subsequent writers, including Kearns & Thomson (2001), Benton (2006) and Goulson (2010). The keys were developed around those by Alford (1975), and we are grateful to him for permission to re-use many of his drawings. Stuart Roberts and Mike Edwards of BWARS (The Bees, Wasps and Ants Recording Society) gave generously of their time in producing up to date distribution maps that include records from the public, BWARS, the RSPB, the Highland Biological Recording Group, BBCT (The Bumblebee Conservation Trust) and the National Biodiversity Data Centre – we are very grateful to them all. George Else, Paul Williams and the late G. M. Spooner commented on the original keys, and we have adapted the supplements to our keys, for B. hypnorum, from the paper by Dave Goulson and Paul Williams (2001). Pat Willmer, Nigel Collar and the late John Free kindly read and commented on the text of the original edition, Robert Prŷs-Jones and Judith PrŷsJones commented on both the original and third editions, and Ted Benton made valued suggestions for the third edition. We are grateful for discussions with Andreas Bertsch, Sydney Cameron, Erling Ólafsson, Kristján Kristjánsson and Paul Williams. We thank Tony Hopkins for his meticulous and beautiful artwork, and for adding the illustration of Bombus hypnorum for this new edition. O.P-J. S.A.C.

Foreword As a simple birdwatcher I have always been a bit scared by the bewildering world of insects.  Bumblebees provide an inviting introduction to this world, and this book is an excellent gateway into the fascinating life of bumblebees.  Found right across the UK, including in my back garden, bumblebees are familiar as a group, but this book provides the means to identify those species and to understand much more about their variety of behaviour, ecology and lifestyles. The threats to bumblebees are very similar to those to birds – changes in agricultural practice are of great importance. Recognition of this provides opportunities for the promotion of practices that should benefit the survival of healthy populations of both groups. I’m delighted that this excellent book has been revised, enlarged, updated and republished.  So delighted that, as I write these few words with snow on the ground outside my home, I am waiting with some impatience for the first sunny days of spring when I resolve to banish my fears and start getting to grips with these wonderful little beasts.  Why don’t you too? Dr Mark Avery Conservation Director, RSPB January 2011

1 Introduction

queen: a female who becomes the main egglaying member of the colony. She ceases to forage once sufficient workers are produced worker: a female who forages and/or looks after the nest. She usually lays few or no eggs female: develops from a fertilised egg male: develops from an unfertilised egg

Bumblebees are likeable creatures, and are among the most attractive of British insects and the most amenable to study. Friendlier than honeybees, they do not sting unless severely molested. Furrier, more rotund and colourful, and often larger than honeybees, and conspicuous by their deep buzz and their habit of working in gardens, they are a familiar sight in summer in town and country. In spring a bumblebee colony is founded by a queen, who has overwintered. Initially she lays eggs that give rise to workers. These look after the nest, defend it and collect food for it. Usually many workers are produced before eggs are laid which develop into males and young queens, who leave the nest and mate (see section 3.2). The appeal of bumblebees as subjects for study is partly due to their predictable behaviour. Most animals are forever compromising between multiple objectives such as feeding, seeking a mate, laying eggs, and defending a territory. In contrast many of the bumblebees we see are foraging workers, whose main task is the collection of nectar and pollen to supply themselves and their colony. Because we can follow just what these foragers are doing we can begin to ask how well they are doing it, and this quantitative approach is facilitated by the ease with which the energy and water content of a flower’s nectar can be measured (see sections 6.1 and 9.4). (Pollen collection is at least equally important, but more challenging to study.) Bumblebees are interesting too for their social behaviour. Their colonies, rather small and lasting less than a year in temperate regions, are simpler to work with than those of honeybees and can be managed quite easily in nest-boxes. Perhaps the most important practical element of interest is the role of bumblebees as pollinators, often underestimated and still poorly understood. Much more needs to be known about them as pollinators of crops and our native flora, and about their nesting requirements and biology. The need is urgent, as pressures on land use are intensifying, population sizes and the number of bumblebee species are declining markedly, and valuable beepollinated crops are put at risk of reduced yields.

2 Distribution and recognition 2.1 Distribution and decline The world fauna of bumblebees, about 250 species (Williams, 1998*), is centred on the North Temperate Zone, extending through Europe, Asia and North America. Apart from some species in South America the only bumblebees native to the southern hemisphere are in the East Indian archipelago. British species (B. ruderatus, B. hortorum, B. terrestris and B. subterraneus) were established in New Zealand in the nineteenth century for clover pollination (Hopkins, 1914); and recently Bombus ruderatus has become established in South America, having been introduced to Chile from New Zealand for the same purpose (Arretz & Macfarlane, 1986). In addition, there has been an increasing number of accidental or intentional introductions, including B. lucorum, B hypnorum and B. pascuorum to Iceland (Prŷs-Jones & others, 1981; Olafsson and Kristjánsson, personal communication) and B. terrestris to Tasmania (Semmens & others, 1993, Buttermore, 1997), Israel and Japan (Dafni, 1998), among others; while non-native species and subspecies continue to be incidentally introduced on a large scale as a result of escapes from commercial bumblebee colonies used for glasshouse pollination (see Chapter 7). Out of their own environment such species have unforeseen and usually bad effects, outcompeting or interbreeding with local species (Kondo & others, 2009) and subspecies (Ings & others, 2010), thereby reducing biodiversity (Goka, 1998), and introducing diseases (Goka & others, 2001; Colla & others, 2006). There are very few sensible arguments in favour of the intentional introduction of a species to an area where it has not previously existed. The true scope of adverse effects of an introduction on the native flora and fauna is almost always unforeseeable. For pollination purposes local species should always be used in preference to imported non-native ones. A few countries, such * References cited under the authors’ names in the text appear in full in Futher Reading on p. 108.

Distribution and recognition | 3

Fig. 1. Pre-1960 and post1960 distribution of Bombus species in England, Wales and Scotland. Widespread throughout (broad and narrow lines): B. pascuorum, B.lucorum complex, B. terrestris, B. pratorum, B. hortorum and B. lapidarius. Local and southern (broad lines): B. ruderarius, B. sylvarum, B. humilis, B. subterraneus and B. ruderatus. Local but widespread (dots): B. muscorum, B. monticola, B. soroeensis and B. jonellus. Even in the areas where they occur, these last two groups of bees are not usually common. From Williams (1982), by permission of the International Bee Research Association.

as Norway and Israel, have taken a lead in banning the import of bumblebees. International legislation to further this aim is urgently needed, with particular controls on the commercial use of non-native species and subspecies. The British bumblebee fauna of, currently, 24 species includes 18 ‘true’ bumblebees (Bombus) and 6 parasitic ‘cuckoo’ bumblebees (previously in the genus Psithyrus, which is now regarded as a subgenus within the genus Bombus, Williams, 1998). It contains representatives of a wide range of subgeneric groups, which have recently been revised into a simpler scheme (Cameron & others, 2007; Williams & others, 2008), and we are fortunate in having a correspondingly wide diversity of ecology and behaviour to study. Since the last edition of this book we have gained B. hypnorum (Goulson & Williams, 2001), almost certainly through human agency; lost B. subterraneus to extinction; and also ‘gained’ B. cryptarum (Bertsch & others, 2004, 2005; Bertsch, 2009; Macdonald, 2006; Murray & others, 2008), a cryptic species that we already had but, embarrassingly, had not recognised! When we compare the present distribution of true bumblebees in mainland Britain with records made before 1960, it is apparent that there have been marked changes over recent decades (fig. 1; Williams, 1982). Six species are widespread over much of Britain, and you

100 km

Pre-1960

Post-1960

Queen supplementation

Colony introduction

400

20

300

15

200

10

100

5

1976

1977

1978

1979 Year

1980

1981

Area planted (hectares)

Fig. 2. Effect of releasing queen bumblebees in spring, and introducing B. hortorum colonies in nestboxes at flowering, on yields of red clover seed in New Zealand. After Macfarlane & others (1983).

Yield (kilograms per hectare)

4 | Bumblebees

1982

can probably see most of them in any year: B. pascuorum, B. lucorum, B. hortorum, B. pratorum, B. terrestris and B. lapidarius. The newly arrived B. hypnorum is spreading, rapidly (see map p. 127), and may soon join this group. For convenience these widespread species are illustrated on the cover. Four species are very local and restricted to southern Britain, and their distributions have contracted: B. ruderatus, B. sylvarum, B. humilis, and B. ruderarius. Four others are widespread but very patchy, and have disappeared from many localities: B. muscorum, B. monticola, B. soroeensis and B. jonellus. B. magnus is one of the B. lucorum complex, along with B. cryptarum, and is associated with heath and moorland areas in the north and west. B. distinguendus is now rare and mainly restricted to coastal sites; although it was once widespread. Apart from B. subterraneus, last recorded in 1988 (but now being re-introduced at Dungeness in southern England), two other species, which were always uncommon, are presumed to be extinct: B. cullumanus and B. pomorum. Neither has been seen for many years. The map (fig. 1) shows where the three main biogeographic elements occur in Britain, and will give you some idea of how many bumblebee species to expect around your home. Over much of the country there are only six or seven species. Although this impoverishment of the fauna makes identification easier, it must be regarded as a conservation problem (Chapter 7). Direct information on the impact of declines in bumblebee populations for the pollination of wild flowers and crops is hard to come by. Work in New Zealand (Macfarlane & others, 1983) has shown that supplementing bumblebee populations

Distribution and recognition | 5

dramatically increases seed yields (fig. 2): the implication is that declining bumblebee populations will have the opposite effect. As more members of the public become interested and submit records, our knowledge of the changing distribution of British bumblebees is improving, but it deserves further study (pp. 77 and 119–130).

2.2 Recognising bumblebees

inquiline: an animal living in the home of another species and using its food

There should be no difficulty in recognising a bumblebee by sight and by sound. Plates 1–4 show you what they look like. Bumblebees are all members of the genus Bombus (meaning ‘booming’), and they are divided into a number of distinct subgenera, which differ slightly in their biology and characteristics. The most distinct of these, the cuckoo bumblebees (genus Bombus, subgenus Psithyrus, meaning ‘murmuring’), resemble the other, more conventional Bombus species, the ‘true’ bumblebees, but have a softer buzz, a sparser coat of hair showing the shiny black cuticle through it, and no pollen-collecting apparatus on their legs. Cuckoo bumblebees are inquilines in the nests of true bumblebees (section 5.1). With a little experience it soon becomes possible to recognise most of the common species of true bumblebees in the field, without disturbing them, on the basis of colour pattern. This is a valuable asset in ecological studies. The Quick-Check Key (p. 78) and the colour plates should help to make this possible. Initially, though, it will be necessary to catch and examine a few bees of each type in order to name them (technique, p. 94–95). There are several distinct colour patterns among British bumblebees. Surprisingly, each is adopted not by one species only, but by a group of species that may not be closely related to each other at all (e.g. B. lapidarius, B. ruderarius, B. pomorum, B. cullumanus and B. (Ps.) rupestris). These are apparently examples of Müllerian mimicry (see Williams, 2007 & 2008, for a fuller consideration). A predator that gets stung whenever it tries to eat a bee with a particular colour pattern will eventually learn to associate the colour pattern with a painful sting and therefore to avoid it. The commoner the colour pattern that a bee wears, the sooner predators

6 | Bumblebees

will learn to avoid it. A bee can therefore reduce the risk of getting eaten by sharing a common uniform with many other bees of the same or different species. These Müllerian mimicry groups involve cuckoo bumblebees as well as true bumblebees, and can confuse ecologists as well as predators.

2.3 Reading about bumblebees Since the start of the new millennium a range of excellent new books on bumblebees has been published. Goulson’s Bumblebees - behaviour, ecology, conservation (2010) and Benton’s Bumblebees (2006) provide up-to-date coverage of the expanding literature on bumblebee biology and natural history, and Benton also includes keys to species. A useful Field guide to the bumblebees of Great Britain & Ireland has been produced by Edwards & Jenner (2009). Books with a more regional emphasis include Benton’s The bumblebees of Essex (2000), Macdonald’s Bumblebees – naturally Scottish (2003) and Macdonald & Nesbit’s Highland bumblebees (2006). Some earlier classics remain well worth reading. The first major book on British bumblebees was F.W.L. Sladen’s The humble-bee, its life history and how to domesticate it, first published in 1912. With superb coloured illustrations and delightful anecdotal accounts of the species, reflecting acute biological observation, this excellent book was republished in 1989. Free & Butler’s Bumblebees, published in 1959, and also now available again, was a worthy successor, and included a field key by Yarrow which, for the first time, made it possible for beginners to name their bees. Alford’s Bumblebees (1975) covers the biology of bumblebees and their associates and parasites, and includes a key, more critical but harder to use than Free & Butler’s. That key was the basis for the Bumblebee Distribution Maps Scheme that gave so much valuable information on bumblebee distribution before it ended in 1976. A brief popular account of bumblebee biology appears in Alford’s The Life of the bumblebee (1978). Bumble bees for pleasure and profit edited by Matheson (1996) outlines the beginnings of the commercial use of bumblebees. North American bumblebees are considered in Plath’s Bumblebees and their ways (1934), which has long

Distribution and recognition | 7

been out of print; while Heinrich’s Bumblebee economics (1979) describes some fascinating aspects of bumblebee ecology and physiology. Kearns & Thomson The natural history of bumblebees – a sourcebook for investigations (2001) is a practical guide to North American bumblebees and outlines their biology, and useful topics and techniques for further research. British bumblebees’ names have changed from time to time, and some of the names in use now are different from those in earlier books. In this book we use the names given in the checklist by Fitton & others (1978), but we retain B. magnus (Alford, 1975), add B. hypnorum (Goulson & Williams, 2001) and split B. lucorum into B. lucorum and B. cryptarum (Bertsch & others, 2004). Names and synonyms are listed on p. 115. There is a rapidly expanding range of websites containing up-to-date information on bumblebee biology and distribution. Some links to this are given on p. 107.

3 The natural history of true bumblebees 3.1 Natural history The different species of true bumblebee are alike in some aspects of their biology, for example in the main features of their life cycle (section 3.2), but very different in others. Some of the idiosyncrasies of British species begin to emerge from intensive studies of their ecology (table 1). To illustrate this point we shall introduce four of the commonest species. Bombus terrestris (of the earth; from the Latin terra) In spring the gigantic queens of B. terrestris are among the first to emerge from hibernation (fig. 3). In the past, if cold weather was followed by a warm spell, some queens would occasionally emerge in midwinter, and attempt to nest if flowers were available (Prŷs-Jones, 1982). Since the 1990s B. terrestris has more regularly been starting a winter generation, on a limited scale, in some parts of southern Britain. Lack of males observed in winter may mean these colonies are not, at least as yet, successfully producing sexuals; and up to now winter nesting activity has depended entirely on introduced plants, flowering in urban situations in gardens and parks (Stelzer & others, 2010).

10

Maximum soil temperature (oC)

Fig. 3. Maximum soil temperature (at a depth of 30 centimetres) on the day spring queens were first observed. Letters indicate species: B. pratorum (P), B. terrestris (T), B. lucorum complex (L), B. ruderarius (R), B. pascuorum (A), B. lapidarius (D) and B. hortorum (H). Data from Prŷs-Jones (1982). Values are mean + standard error; for an introduction to statistics see Wheater & Cook (2003).

9

H

8

R

A D

L

7 T P

6 5

March

April

May

The natural history of true bumblebees | 9

In most species the queen and worker have the same coloured tail, but British B. terrestris queens have brownish tails and workers usually have white or buff ones (pl. 1.3 and 1.4). Towards the end of the nesting period some workers have queen-like coloration. The reason for this is not clear, but it may be that declining vigour of the

Average corolla depth visited (mm) (nectar-only and nectar + pollen visits) No. of observations Average tongue length (mm) Pocket-maker (M) or pollenstorer (S) Predominant nest-site (U, underground; S, surface) Colony size

B. terrestris

B. pratorum

B. pascuorum

B. hortorum

Pollen-only visits as a % of all pollen visits % visits according to orientation of flower entrance: Down Horizontal Up % of visits to separate (as opposed to clustered) flowers No. of observations

B. lucorum

% visits for: Nectar-only Nectar + pollen Pollen-only No. of observations

B. lapidarius

Table 1. Life history information for six common British bumblebees

75 13 12 216

71 1 28 85

80 9 11 192

23 67 10 184

60 33 7 258

38 57 5 159

48

96

56

13

17

8

0 35 65

1 38 61

7 40 53

39 28 33

13 61 26

16 71 13

46

61

62

81

82

93

210

84

188

177

254

159

5.1 139 8.1 S

5.1 59 7.2 S

6.3 168 8.2 S

7.4 100 7.1 S

8.3 351 8.6 M

8.7 133 13.5 M

U

U

U

S

S

S

Large

Large

Large

Small

Medium

Small– medium

May–June First workers observed Aug. Max. nos. of workers observed Medium Length of life cycle Evidence for 2nd cycle of No summer colony activity Relative degree of size variaLittle tion (foraging workers) None Worker/queen size overlap

May

May

Aug. Long

July–Aug. Long

Apr.–May May–June June Short

Aug. Long

June–July Short

May

No

No

Yes

No

Yes

Little

Little

Moderate

Large

Large

None

None

Slight

Moderate Moderate

Summarised from Prŷs-Jones (1982). Based on regular observations throughout the life cycle of each species. Numerical data apply to foraging workers.

10 | Bumblebees

corolla: tube or crown of petals

Fig. 4. B. terrestris worker taking nectar ‘illegally’, through a hole she has bitten in a honeysuckle flower.

queen eventually allows the expression of at least some queen-like characteristics in all the female offspring (see p. 19, ‘complex’ species). Most workers of B. terrestris look quite similar to those of B. lucorum, and small individuals can be difficult to separate. Both species have a short broad face and a relatively short tongue, features associated with their habit of collecting nectar from rather short open flowers. They also have the ability to bite a hole in the corolla tube of deeper flowers (fig. 4), such as honeysuckle Lonicera periclymenum, comfrey Symphytum officinale and field bean Vicia faba. This enables them to extract nectar which they would be unable to reach from the front of the flower. In Britain only queens and workers of B. terrestris and B. lucorum have been observed to rob flowers regularly in this way, but other bumblebees (and even honeybees) often re-use the holes, acting as secondary robbers. Sometimes almost every flower in a field bean crop may be robbed. It is hard to evaluate the impact this has on crop pollination. On the one hand one might expect that bees removing nectar in this way, without contacting the anthers and stigma, would impair pollination. On the other hand it has been suggested that robbery may sometimes cause long-tongued bumblebees, which do not rob, to visit more flowers for each load of nectar, so enhancing pollen transfer. Hole-biting may attract more honeybees to a clover crop, but if most of these are gathering nectar via rob-holes they may fail to effect cross-pollination. Probably the most important consideration is that pollen-collecting bees are unaffected by hole-biting and remain successful pollinators. This may account for the fact that robbery of red clover does not appear to decrease seed set (Hawkins, 1961). Certainly in New Zealand, where introduced bumblebees are very important pollinators of red clover, robbery by B. terrestris is common, yet in areas where this species is the main visitor to the crop it is still a useful pollinator (Gurr, 1975). In situations where there is a wide choice of flower types B. terrestris rarely uses pendulous flowers, probably because it is not very agile. Instead it usually selects flowers that face upwards or horizontally and provide a substantial landing platform (table 1). As its name implies, this bumblebee often nests in

The natural history of true bumblebees | 11

Fig. 5. Seasonal cycles of flight activity of four common Bombus species at Cambridge in 1978. From Prŷs-Jones (1982).

the ground, commonly adopting the disused nest of a small mammal (see section 4.1). It starts nesting early in the season and has a long cycle, reaching peak numbers of workers in late July (fig. 5), when colonies may contain several hundred individuals. Sladen (1912) was among the first to note that workers of B. terrestris are more defensive than those of most other species, so although its nests are among the easiest to find, they are not the easiest to work with.

Census days: 10

Queens

Census days: Bombus pratorum

0

10

Bombus hortorum

Queens

0

No. of observations

40 Workers

20

20

10

0 20

0 10 0

Males

10 Mar.

Apr. May

June

July

Queens 10

Bombus terrestris

0 10

0

Mar.

Apr. May

June

July

Aug. Sept. Oct.

Bombus pascuorum

Queens

0 Workers 40

No. of observations

Males

Aug. Sept. Oct.

20

40

Workers

Workers

20 20 0 100 0 40 Males 60

Males 20

20 Mar.

Apr. May

June

July

Aug. Sept. Oct.

0

Mar.

Apr. May

June

July

Aug. Sept. Oct.

12 | Bumblebees Fig. 6. Species differences in tongue length.

B. hortorum ♀

B. pascuorum ♀

B. pratorum ♀

Fig. 7. B. hortorum visiting nasturtium Tropaeolum majus L.

B. hortorum (of the garden; from the Latin hortus) Another white-tailed bumblebee with yellow bands (pl. 1.5), B. hortorum differs from B. terrestris most obviously in a detail of pattern - it has a yellow band at the rear of the thorax - and in its long narrow face and long tongue, longer than that of any other common British species (fig. 6). With a long tongue comes the ability to take nectar from flowers with long tubular corollas, and B. hortorum specialises on these. It is among the best pollinators of field bean (Free, 1993) and red clover (Holm, 1966; Gurr, 1975). Most often it visits relatively nectar-rich flowers such as honeysuckle, Delphinium species, and Indian balsam Impatiens glandulifera, which have openings facing horizontally, and are arranged singly, so that a bee must fly from one flower to the next (fig. 7). In this respect B. hortorum contrasts markedly with, for example, B. lapidarius (pl. 2.1 and 2.2) which shows a strong tendency to visit massed flowers such as members of the family Asteraceae (common knapweed Centaurea nigra is a favourite). The tiny individual florets are often nectar-poor, but a bee can probe many of them between flights. Differences between species in enzyme activity of the flight musculature may partly account for such differences in foraging behaviour (section 6.1). Although B. hortorum is Britain’s most widespread bumblebee species, individuals are seldom very abundant, possibly because the long-tubed flowers that they visit are relatively uncommon. Queens are late to emerge (fig. 5) and start nest-building in spring; colonies are quite small, often producing no more than about 30–80 workers. Males and young queens are sometimes reared from only the second or third mass of eggs to be laid, and will have been produced in most nests between late June and early July (fig. 5). Related to the short life cycle of the colony are indications that this species possesses the unusual ability to complete a second nesting cycle within a single season in some areas and/or years (Prŷs-Jones, 1982; see fig. 11), whilst most bumblebee species can complete only one. Other short cycle species, including B. jonellus (pl. 1.6 and 1.7; Meidell, 1968) and B. pratorum (p. 14; pl. 2.5 and 2.6), almost certainly do so too on occasion. As yet the evidence is very suggestive, but not conclusive, coming

The natural history of true bumblebees | 13

as it does from indirect information concerning the seasonal timing of flight activity. More direct information, on the natural history of individual colonies, is needed to help us understand what is actually happening, which individuals undertake these second cycles (see also pp. 24 and 25) and under what circumstances. Patient observational studies may well reveal interesting and unexpected aspects of life history. Nests of B. hortorum are hard to find because there are so few workers going in and coming out to draw attention to them. Nests are usually placed among plant roots and litter just above or just below the soil surface. B. pascuorum (of the pastures; from the Latin pascuum) B. pascuorum is distinctive as the only common species, apart from B. hypnorum (which is rapidly becoming commoner), to have a uniformly ginger-coloured thorax. The abdomen, too, is gingery, but there is much variation in the colour (ranging from foxy-red to almost grey) and in the intensity of the darker markings (pl. 3.5–3.7). This is another long-tongued bumblebee (fig. 6). It resembles B. hortorum in taking nectar from long, tubular flowers, but B. pascuorum is more catholic in its flower choice. As in B. hortorum the workers tend to visit flowers that face sideways (table 1) - white dead-nettle Lamium album is a favourite. Unlike B. hortorum, this bumblebee shows a sex-related difference in foraging habits: males visit clustered or compound flowers, such as marsh thistle Cirsium palustre, much more often than workers do. B. pascuorum is one of five British species known as ‘carder bees’ (subgenus Thoracobombus) because of their distinctive habit of carding (combing) together moss and other material from around the nest to form a covering for the cells (Sladen, 1912). Nests are built among vegetation on or just below the soil surface. The bees gather together the covering material with their mandibles and legs, often standing facing outwards from the colony, passing the moss or grass stems backwards between their legs in a scrabbling fashion. Colonies vary in size, but can be relatively large, containing up to about 200 workers; with peak worker numbers in August, they are slow to develop to their full strength (fig. 5). Colonies are also long-lived, often lasting into October. A patch of water figwort Scrophu-

14 | Bumblebees

laria auriculata or marsh woundwort Stachys palustris may be alive with foragers of B. pascuorum, and they are potentially valuable pollinators of deep-flowered crops. Unfortunately, along with other surface-nesters, the species suffers badly when areas of rough grass are mown. Ploughing also removes hibernation sites, as well as underground nesting places used by other species. This serves to emphasise the importance of leaving some suitable areas undisturbed (see chapter 7).

Fig. 8. Outline life cycle of true bumblebees in a temperate region, such as Britain (variations that are sometimes found in the arctic, the tropics and at high altitude are mentioned on pp. 25 and 26).

B. pratorum (of the meadows; from the Latin pratum) A small black bumblebee with yellow bands and an orange tail, B. pratorum is one of the simplest species to recognise in the field (pl. 2.5 and 2.6). Its tongue is of only modest length (fig. 6), but its flower selection includes a wider range of corolla depth than this would suggest. At one extreme it will visit very short, open flowers; at

Workers produced Colony development Nest initiation

Nest may be taken over by a true or cuckoo bumblebee queen Young queens and males produced

Queen foraging

Mating

Males die

Second colony cycle (sometimes, in some species: see text)

Queen overwintering

The natural history of true bumblebees | 15

the other, it makes use of some long tubular flowers by thrusting the whole of its relatively narrow head into the corolla. A very agile bumblebee, it commonly visits flowers that hang downwards, and is often seen acrobatically hanging upside down to work a flower of comfrey or snowberry Symphoricarpos rivularis. B. pratorum is good at working in low temperatures and it can be seen early in the morning, adopting a characteristic flight pattern in which hovering alternates with darting flight. The species appears early in the spring, and colonies are established in April and May. At maturity, in early to mid June, nests may contain only a handful of workers or they may occasionally be large, with up to about 200 workers. Compared with other species, workers begin to be seen foraging relatively soon after queens emerge from hibernation, and worker numbers peak well in advance of other species. The cycle is short, and the first generation ends in midsummer (fig. 5). Most of the young queens then begin their period of overwintering, at soil temperatures well above those at which they will emerge the following spring. However, as in B. hortorum, which also has a short cycle, a few young B. pratorum queens appear to go on to produce a second cycle in some years. The nests of B. pratorum are constructed in a wide variety of sites above, on and below ground. It is not uncommon for disused birds’ nests to be occupied, particularly those in nest-boxes (see section 4.1).

3.2 The annual cycle A honeybee colony is potentially immortal, and its queen does no foraging. Honeybees overwinter as a cluster of workers with their queen and a store of honey. On the first warm days of spring these workers emerge from the hive to forage for nectar and pollen to feed themselves and to support the development of new workers, building up the colony towards summer. The colony can persist through the winter because of the honey store. Workers are therefore present in spring so the queen herself need not forage, but can stay at home laying eggs. Bumblebees do things differently (fig. 8). In most temperate regions colonies die out in autumn, and only the

16 | Bumblebees (sample size) 31

41

34

47

6

22

10

100 7 80

% observations

Fig. 9. Behaviour of B. terrestris queens after emergence from overwintering sites. 1, inactive; 2, sunning; 3, eating pollen; 4, nestsearching; 5, collecting pollen; 6, collecting nectar; 7, flying. From Prŷs-Jones (1982).

60 6 40

20

5 1

2 3

4

0 April

fat body: a mass of food storage cells, mainly in the abdomen, containing fat, glycogen and protein

May

June

young mated queens overwinter. Alone, therefore, they establish new colonies in spring. In mid to late summer a young queen may be seen exploring the ground. When she finds a suitable site often a north-facing bank which will not be warmed by the winter sun - she burrows several centimetres into the soil. Here she overwinters. Dissection of a queen caught seeking a hibernation site reveals a massive clear or white fat body, which may occupy most of the abdominal cavity. Most of this will be used up as a food reserve during winter; in spring the fat body is almost exhausted. When the soil temperature rises in spring the queen emerges from hibernation (fig. 3), and may be seen sunning herself or foraging on early flowers such as pussy willow Salix species, white dead-nettle or flowering currant Ribes sanguineum. Her behaviour will change in a predictable way as colony foundation proceeds (fig. 9). At first she forages only for herself, eating large quantities of pollen and nectar while her ovaries develop, and roosting at night under moss and other vegetation. She then seeks a place to establish her nest. Nest-searching queens fly to and fro, low over banks and rough uncultivated land, sometimes investigating dark holes, crawling

The natural history of true bumblebees | 17 Fig. 10. Bombus pupa in an opened cocoon.

briefly into cavities and tussocks. The search often ends in the disused nest of a small mammal or bird. It appears that Bombus queens quite often search out and fight for established nests - which represent a valuable resource in terms of time and energy - particularly if nest-sites are scarce or the season is short (for example, see Gjershaug, 2009). It is not unusual to find several dead queens, of the same or different species, in the entrance of the newly established nest (see p. 28). Having found a suitable site, the queen adjusts the nesting material to form a small chamber. She then goes out to collect pollen, which she brings back to the nest, packed in the pollen baskets on her hind legs. The pollen is moulded into a mass that forms the base of the egg clump. In an untidy cell made of wax (extruded from between the plates of her abdomen) and pollen, built on the pollen mass, the queen lays her first batch of eggs. Within convenient reach of the egg clump she constructs a thimble-like wax honeypot. This she stocks with nectar. Taking occasional sips of nectar, she broods the egg clump, warming it by contact with the lower, less hairy surface of her abdomen. Bumblebees have considerable control over their body temperature (see pp. 43–46); a brooding queen can keep her body at about 30–35 °C, and maintain her egg clump at about 25 °C, despite low outside temperatures (Heinrich, 1979). During this stage of colony development the queen spends much of her time in the nest, incubating the eggs, and makes only occasional foraging trips. Eggs hatch after 4–6 days. The resulting larvae feed on the pollen mass, and are supplied by the queen with nectar and pollen, in a manner that depends on the species (see below). At first, the larvae remain together in their cell, which the queen enlarges to accommodate their growth as they progress through a series of larval moults. Later, each spins a delicate silken cocoon, creating a chamber of its own. After 10–20 days as a larva, a much tougher, neater, cylindrical cocoon is formed, in which the larva discharges faecal material (the meconium) which has accumulated in its gut throughout life. It then pupates (fig. 10). At this stage the queen scrapes the pollen/wax mixture from the pupal cocoons. She re-uses it to construct one or more chambers, between or on top of the cocoons,

18 | Bumblebees

in which she lays her second batch of eggs. In some species (such as B. hortorum, B. pratorum) these egg clumps, like the first, are primed with a pollen mass; in others they are not (Free & Butler, 1959). Adult workers emerge after about two weeks as pupae. Each bites her way out of her cocoon, often helped by the queen or other workers. At first pale silvery-grey, tousled and soft-winged, the worker acquires her full colours and fluffy appearance after a few hours, and her wings harden within about a day. New workers soon begin foraging, as well as helping the queen to tend the brood. Once there are enough of them to take over foraging duties the queen remains within the nest, where she devotes herself to housework and egg-laying. Workers of the very first brood are almost always relatively small, probably because they are fed solely by the efforts of the queen. As worker numbers increase so the average size of individuals tends to increase, up to a maximum after, perhaps, the first few broods; even then there will be variability in size between workers hatching from any given brood clump. Species differ in the way they feed pollen to the larvae. ‘Pocket-makers’ construct waxen pouches or pockets near the base of the larval chamber or brood clump. Into these the returning foragers deposit their loads of pollen; the larvae feed from the resulting mass. Later, the queen (or workers) may supplement this diet by regurgitating a nectar/pollen mixture into the brood chamber, through a temporary hole in the wax envelope. The larvae of a batch remain together, sharing a common chamber and the pollen supply that goes with it. Some pocket-making species prime each egg cell with pollen; others - the carder bees - do not. In the other group of species, the ‘pollen-storers’, pollen brought into the nest is not put into pockets in the brood clump, but instead is stored in empty pupal cocoons and specially constructed wax cells or cylinders. The queen or workers then feed it to the larvae, bit by bit, squirting a regurgitated mixture of nectar and pollen into the larval cell through a hole in the wax covering. Which species are pocket-makers, and which are pollenstorers, is summarised in the synonymy table (p. 115) and Table 1 (p. 9). Instead of continuing to share a common chamber

The natural history of true bumblebees | 19

pheromone: substance produced by an animal that influences the behaviour or physiological development of another individual of the same species

as they mature, the larvae of pollen-storing species spin separate compartments, and they are fed individually. This has two consequences. First, it is much harder to ‘read’ the history of a colony from the arrangement of the chambers in pollen-storers than it is in pocket-makers. Secondly, it appears to influence pathways of development. Workers from a single colony may vary widely in size, probably because of differences in the amounts of food each received as a larva. Larvae of pollen-storers, fed individually, may receive fairer shares of food. In contrast, larvae of a pocket-maker may do better or worse according to both the degree of larval competition and position in relation to the pocket (Sladen, 1912; Cumber, 1949). Thus size variability is said to be less marked in pollen-storers (such as B. terrestris, B. pratorum) than in pocket-makers (such as B. hortorum, B. pascuorum). When the colony is mature some eggs develop not into workers, but into males and queens. Just what influences the changeover to sexual production is not well understood. Adequate food stores, a particular worker/ larva ratio, chemical cues, nest temperature stability and bee density in the nest are probably all influential factors. Their relative importance appears to vary with the species concerned. At least two groups of species can be distinguished: ‘simple’ and ‘complex’ (Brian, 1980). This is probably an oversimplification. In simple bumblebees there may be less influence of the queen over brood rearing. Whether a female larva becomes a worker or a queen does not appear to be determined until just before the larval/pupal transition; possibly it is not always fixed even then. Larvae that become queens have usually eaten for longer and grown more than those that become workers. Such species include, in Britain, B. pratorum and probably B. hortorum. In contrast, in complex bumblebees, such as B. terrestris and perhaps B. lapidarius, the queen can, probably with the aid of pheromones, stop the workers feeding the larvae with sufficient quantity and/or frequency for them to develop into queens, even at high worker/larva ratios when ample food is available. Thus complex species can delay queen production, allowing the build-up of large worker populations. One can speculate that in general species with large colonies and long life cycles (many pollen-storers) may well be complex species, while those with small, short-cycle colonies (many

20 | Bumblebees

spermatheca: the sac in which sperm are stored after mating

pocket-makers) are probably simple species. As a result of the differing methods of queen production in the two groups, queens are usually distinctly larger than workers in complex species (having fed better for a longer period), but in simple species they cannot be distinguished on size alone (having merely fed a little longer at the end of their larval life). The real differences between a queen and a worker are physiological and behavioural. A queen can mate and develop her ovaries, and is the colony member who is primarily active in egg-laying. Unlike a worker, a queen may also hibernate and build up in autumn the abundant reserves in her fat body that will keep her alive through the winter. To be sure whether an individual is a queen or a worker, it is necessary to know how she functions in the colony; an indication of this may be gained by dissecting her. The presence of sperm in her spermatheca indicates that she has mated successfully (fig. 28, and technique, p. 97). It is probably uncommon for workers to mate, and therefore a fertilised bumblebee with mature eggs in her ovary is almost certainly a queen. Further, in autumn, a female bumblebee with an abundant fat body must be a young queen. In bees, unfertilised eggs give rise to males and fertilised eggs produce females (queens and workers). An unmated bee can lay only male eggs, but a mated female can determine the sex of her egg by controlling the release of sperm from her spermatheca. A few males are produced early in the season, perhaps from eggs laid by workers or unfertilised queens. The proportion of males (and, if fertilised workers are present, the proportion of females) derived from worker-laid eggs remains to be determined. Probably it varies with the species and the conditions. Inbreeding can result in fertilized (diploid) eggs developing into males. This may occur in small populations of rare bumblebees restricted to fragments of their previous range (Takahashi & others, 2008; Whitehorn & others, 2009) and may seriously reduce reproductive success - hastening the loss of species (see Zayed & Packer, 2005). Although some colonies produce only male or female sexual forms, or neither, most produce both, the males somewhat in advance of the young queens. Males play no part in looking after the colony except for a little in-

The natural history of true bumblebees | 21

cidental brooding; they leave the nest and seldom or never return. Their sole function is to mate with young queens. They forage rather slowly, for themselves alone, sleeping out at night, often clinging under the heads of flowers such as thistles and knapweeds. Young queens go out foraging, but they do not normally help to provision the colony; unlike males, they often return to the nest at night. Those who collect pollen appear to do so only when their mother has died, or lost her dominant status. Such young queens usually contain developed eggs (Prŷs-Jones, 1982) and are probably laying in their parental nest. The reproductive behaviour of male bumblebees is conspicuous and unusual. Several methods have been described. The majority, including all the British species, undertake ‘patrolling’; however, ‘cruising’, ‘racing’ and ‘territorial’ systems have also been described, the exact method relating, it is thought, to the type of habitat in which the bee species lives (Williams, 1991). Among British species individual males patrol characteristic routes, circuiting the same flight path again and again, pausing on the wing at intervals at certain features such as a particular leaf, stone or area of tree trunk. These features, visited repeatedly through the day, sometimes carry a species-specific scent detectable by the human nose (Sladen, 1912). They are places where, early in the morning, the male bumblebee has deposited a scent mark, a fragrant mixture of compounds secreted by a gland - once thought to be the mandibular gland, but now known to be primarily the labial gland (Kullenberg & others, 1973) - that opens between the mandibles. Many of these compounds have been identified and found to be fatty acid derivatives and terpene alcohols and esters (table 2). How can a queen find a male of her own species? The species differ from one another in the mixture of compounds that make up their characteristic scents and in the nature of the route the males patrol - along a low hedgerow or at treetop level for instance. It is assumed that queens visit the scent-marked trail, and that mating takes place here, but bumblebees have rarely been seen mating in the wild. Work done in Scandinavia (Bringer, 1973; Svensson, 1979) shows the flight levels of some species that are also found in Britain (table 3). There is a

22 | Bumblebees

Geraniol Citronellol Geranyl acetate Citronellyl acetate Farnesene isomers All-trans-farnesol 2,3-dihydro-6-trans-farnesol 2,3-dihydro-6-trans-farnesal All-trans-farnesyl acetate 2,3-dihydrofarnesyl acetate Geranylgeraniol Geranylcitronellol Geranylgeranyl acetate Tetradecanal Tetradecanol Hexadecenol Hexadecanal Hexadecenal Hexadecanol Octadecenol Eicosenol Hexadecenyl acetate Hexadecyl acetate Octadecyl acetate Eicosyl acetate Docosyl acetate Ethyl decanoate Ethyl dodecanoate Ethyl tetradecenoate Ethyl tetradecanoate Ethyl hexadecatrienoate Ethyl hexadecadienoate Ethyl hexadecenoate Ethyl octadecatrienoate Ethyl octadecadienoate Ethyl octadecenoate Ethyl octadecanoate Tetradecenoic acid

¡ ¡ ¡ ¡ ¡ l ¤

l ¤

B. (Ps.) sylvestris

B. (Ps.) barbutellus

B. (Ps.) campestris

B. (Ps.) bohemicus

l

l ¡ ¤

l

¤ ¡ ¡

¡ l l ¤ ¤

B. (Ps.) rupestris

B. lucorum

¤

¡ ¤

B. terrestris

B. monticola

B. pascuorum

B. lapidarius

B. soroeensis

B. jonellus

B. pratorum

Table 2. Composition of the male marking substances

¤

l

¤ ¡ ¡ l ¤

¡ ¤

¡ ¡ l l ¡

¤ l ¡ ¤

¡ ¤ l

¡ ¡ ¡ ¡ ¡ ¤ l ¡ l ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

¡

¡

¤

Symbols indicate relative amount present: minor component (¡), major component (¤) and main component (l). Adapted from BergstrÖm & others (1981), Cederberg & others (1984) and Descoins & others (1984).

The natural history of true bumblebees | 23

10–11 8–9 6–7 4–5 2–3 0–1

¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

¡ ¡ ¡ ¡ ¡ ¡

B. (Ps.) sylvestris

B. (Ps.) rupestris

B. pratorum

¡ ¡ ¡ ¡ ¡

B. sylvarum

¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡ ¡

B. terrestris

B. pascuorum

B. (Ps.) campestris

12–13

B. (Ps.) bohemicus

14–15

B. hortorum

16–17

B. lucorum

Flight level (metres above ground)

B. lapidarius

Table 3. Flight levels of various species of true bumblebees and cuckoo bumblebees in coniferous forest in Sweden

¡ ¡ ¡ ¡ ¡

Adapted from Bringer (1973).

little information about patrolling behaviour in Britain (Fussell & Corbet, 1992b). It would be interesting to know to what extent flight levels are characteristic of a species, and to what extent they vary with the habitat. Even after mating, a young queen may continue to return for a while to the nest where she grew up. By eating large quantities of nectar and pollen she builds up her fat body, and as an additional food store she fills her highly distensible honeystomach with thick honey. She then seeks a site in which to overwinter. Queens and males are the colony’s contribution to the next generation. They are the last brood to be reared; when they have left, the old nest has no further role to play. The few remaining workers forage only for themselves and behave more like males when visiting flowers, moving slowly and lethargically. Like the males, they will have died off by the end of the season. Seasonal timing of events in the colony cycle varies with the species, with the geographical area, and with

24 | Bumblebees Fig. 11. B. hortorum: a young queen that has not overwintered brooding her first batch of eggs.

weather conditions in a particular year (Prŷs-Jones, 1982; Goodwin, 1995). We have seen that the large queens of B. terrestris are among the first to emerge from hibernation in spring, but their colonies are not among the quickest to develop. It is often little B. pratorum whose workers appear first. B. pratorum, a short-cycle species, is also one of the earliest to produce males and new queens: these frequently appear early in the summer in May or June, occasionally before some of the other species have produced their first workers. By about August the latest species have begun to produce males and queens, and by October even the slow-maturing nests of B. pascuorum are over or on the decline (fig. 5). Overwintered spring queens found in colonies cannot be confused with new virgin queens, produced late in the summer, because they appear at different seasons and they look and behave differently. An overwintered spring queen has often worn away the fine hairs (microtrichia) on her wings, and her wingtips rapidly become ragged as she forages in a business-like way for nectar and pollen for her brood. Her fat body is by now almost exhausted and looks yellowish, but her ovaries contain mature or ripening eggs. A newly emerged queen in late summer is, by contrast, fresh and unworn. Her fat body appears white or clear and is building up for hibernation, and her ovaries are undeveloped (she will not lay eggs until the next year). Generally she forages only for herself and therefore does not carry pollen loads (see p. 21). We have mentioned that a few species (B. hortorum, B. pratorum and B. jonellus) may sometimes produce a second colony cycle within one season. This means that some of the new queens that emerge from the nest and mate in summer do not hibernate, but instead at once establish nests (or take over another nest, possibly the one they were reared in) and begin worker production (fig. 11). If you see a fresh unworn queen collecting pollen, at a season when old spring queens are confined to their nests, she may be a young queen founding a second-generation colony. This exciting possibility can be investigated by dissection: she would have mature eggs in her ovary, sperm in her spermatheca, and sparse yellowish-brown fat (see techniques, p. 97). However, instead of killing her, it would, if possible, be most informative to follow her home and gently inspect the nest.

The natural history of true bumblebees | 25

Has she started a new nest of her own, is she nesting and laying her eggs in the remains of her mother’s nest, or is she simply foraging for her parental nest? Techniques of genetic analysis offer one means to help distinguish between these alternative possibilities. Alternatively, you can encourage a young queen to found her colony in a nest-box (fig. 12; see sections 4.3 and 4.4) and follow events that way. This procedure has the disadvantage of being conducted under unnatural conditions, but the advantage that you know for certain that you are dealing with a queen that has not overwintered, and, if successful, it at least shows that overwintering is not a necessary preliminary to nesting in that species. Be careful not to anaesthetise young queens with carbon dioxide (see technique, p. 96) as this will artificially induce nesting behaviour. A queen was defined earlier (p. 1) as the main egglaying member of a nest. This is a definition based on her function, and up until the time she achieves the role of ‘queen’ she is sometimes better, and more correctly, called a ‘gyne’. For simplicity, the term ‘new queen’ is commonly used here instead of gyne, but there are times when it is important to be clear what role is actually being described. Being queen-sized does not always mean that a bee functions as a queen within the nest, or ensure that she will go on to found a nest of her own. The term gyne therefore refers to a female individual who is queen-sized and may mate, but who has not yet become a functional queen. In bumblebees only gynes can enter diapause and overwinter. As we have seen, in Britain, the bumblebee foraging season (and, because they do not store much honey, the active season) differs between species, in some cases to a considerable degree (see B. pratorum and B. terrestris in fig. 5). This may, in part, reflect competition for food. However, the extent of any temporal segregation of species is limited by a shortage of flowers in winter. Further north and at higher altitudes the season is even shorter, and it is the short-cycle species that extend furthest towards the North Pole. Nearer the equator flowers are sometimes available throughout the year. In the South American tropics bumblebee colonies may last several years, producing enormous numbers of workers (see Michener, 1974). In middle latitudes (about 40° N or S),

26 | Bumblebees

such as New Zealand (Cumber, 1954; Gurr, 1973), Corsica (Ferton, 1901), and, more recently, in Britain (Stelzer & others, 2010), colonies may survive through the winter. Length of the colony cycle can also vary with latitude within species (Prŷs-Jones, 1982). Some bumblebees which in Britain have long cycles (such as B. terrestris, B. ruderatus), appear capable of extending their life cycles at lower latitudes. Species with short cycles (such as B. hortorum, B. pratorum, B. jonellus) apparently do not do this, although they may complete a second cycle if the season is long enough. B. pascuorum is interesting in that it has a long cycle in Britain (50°–60° N) and yet it occurs as far north as 70° N. Perhaps different populations of B. pascuorum (which may correspond with the many described subspecies; Løken, 1973) are adapted to different climatic conditions. Benton (2006, p. 374) speculates that the apparently long cycle of B. pascuorum in Britain may perhaps be interpreted as variability in the date of colony initiation, or in the rate of colony development.

4 Nests and their establishment in captivity 4.1 Natural nests A great variety of nest-sites chosen by queen bumblebees are structures built by other animals, which are taken over when the original owner has finished with them. Most species commonly use nests of small mammals. The more adaptable species (B. terrestris, B. lucorum, B. pratorum, B. lapidarius, and, perhaps most noticeably, B. hypnorum) take advantage of man’s artefacts. B. terrestris will nest in surprising places. We have found nests in a rolled-up carpet and a disused armchair, under an old lawnmower and an upturned sink, in a heap of coal, and often in or under a garden shed. B. pratorum is probably the most versatile British species: a disused robin’s nest suspended over a pond, bird boxes, an old cushion, lagging on a water pipe and an old fishing-net float filled with wadding are just a selection of nesting places that we have recorded. B. hypnorum, the tree bumblebee, appears to make particular use of bird nest boxes, as its English name might suggest. All these sites have one thing in common: they provide a dry, wellinsulated home. There are few reliable differences between species in their nest placement, but most species predominantly favour one of two types of site. Some build a nest on or just below the soil surface, covering it with fine roots, grass and moss (B. ruderarius, B. hortorum, B. pascuorum, B. sylvarum, B. humilis and B. muscorum). Others normally make a nest underground, approached by a tunnel varying in length from a few centimetres to more than a metre (B. terrestris, B. lucorum, B. ruderatus and B. lapidarius). Nest-sites of rarer species are not well known; more information would be useful. If a small mammal or bird nest has been taken over it may be possible to identify the original owner by examining nest debris (for a useful key to hair and feathers see Day, 1966). Occasionally this is unnecessary: in one nest of B. hortorum that we examined the comb was built directly on the decomposing body of a bank vole. It is

28 | Bumblebees

unlikely, although possible, that the queen killed the vole. One way to acquire a colony is to collect an established nest. Apart from some ‘difficult’ colonies of B. terrestris, B. muscorum and, perhaps, B. hypnorum, British bumblebees are defensive rather than aggressive when their nests are disturbed. Ease of collection therefore depends mainly on nest size and situation. A useful collecting kit includes five or six jam jars with perforated lids, a postcard, a box or biscuit tin to receive the nest, a pair of thick gloves, trowel, torch, insect net, a pair of forceps and a few specimen tubes with tops. For added confidence a beekeeper’s veil and hat are ideal, but not usually necessary. Evening is the best time for nest collection as most bees are then at home. Before beginning to excavate a subterranean nest it is useful to mark the course of the tunnel with a flexible stick. Gently remove and agitate material covering the nest; this will encourage the occupants to come out to defend it in manageable numbers. It is a simple task to place a tube over each and transfer them to a jar, one after another. Sladen (1912) suggests covering the jar with a card cover, which can be slid aside quickly to let in the next bee. When no more bees appear the comb can be exposed and lifted carefully into a box. It should be kept upright, supported on a bed of material from the nest-site. Collect enough nesting material to allow the bees to re-cover and insulate the nest. Take great care to capture the old queen, who will usually look more worn than any daughter queens that she may have produced. A ball of dry grass left in place of the nest will provide returning bumblebees with somewhere to settle and congregate; they can be collected later the same evening or early the following morning. As you remove the nest look for the remains of dead queens, which are often to be found amongst nest debris. One may be the original foundress, another her successor or they may all be unsuccessful intruders. Several species may be found (see p. 17). Nest odour appears to be used by cuckoo bumblebee females to locate true bumblebee nests, and true bumblebee queens may also use this odour as a cue, allowing some of them to find already established nests, which they may then be able to take over. Bumblebees should be returned to the comb as soon

Nests and their establishment in captivity | 29

as possible after the nest is collected: if you are interested in following the behaviour of individuals, this is a convenient time to mark the bees (technique, p. 96). If there is only a little nectar stored in the comb some concentrated sugar solution can be added to empty cells. When choosing a new nest-site make sure that it is well ventilated, protected from rain and direct sun and not exposed to extremes of temperature. Once the nest is in the new site, keep the entrance plugged for a few hours to allow the bees to settle into their new quarters. When the plug is removed the bees that emerge and take off will probably hover facing the nest entrance. This initial phase of orientation lasts a few seconds. Each bumblebee then begins to fly slowly back and forth in front of the nest in an ever-widening pattern, as she inspects and remembers the new nest location and the character of the surroundings. Do not startle bumblebees at this stage or some of them may fly off before learning their way home. Often the quickest way of obtaining nests is to advertise in your local paper. In addition, local pest-control officers may be prepared to pass on information about nests reported to them. Many people are only too happy for you to remove nests from their gardens, perhaps forgetting that bumblebees are very useful when it comes to pollinating their runner beans and soft fruit. Bumblebees are increasingly under threat from more intensive use of the land, from drainage, and from the spread of modern ‘clean’ farming practices (Chapter 7). Reversal of the decline in their numbers and distribution requires the concern and action of all of us. There are a variety of steps you can take to encourage bumblebees to nest in gardens and on agricultural land (fig. 26, p. 66). Shelter in the form of a suitable nest-site, and a continuous succession of flowers, are both essential. Therefore always discourage unnecessary tidying-up and socalled improvement of rough land, removal of banks and hedgerows, and the use of herbicides to kill ‘weeds’ (wild flowers). Derelict land can be made more attractive for bees (and people) by planting an appropriate mixture of wild flower seeds (check they are of native origin; see sections 7.3 and 9.8) which will ensure a food supply throughout the nesting season (March to September). Rotting wood, piles of cut vegetation, mounds of earth and rubble, can all be valuable nesting and hibernation

30 | Bumblebees

sites. Old stone walls and compost heaps provide potential nest-sites, and nest-boxes can be put out. Garden plants such as those traditionally grown in a cottage garden provide a good mixture of bumblebee flowers (table 6, p. 65). Most members of the dead-nettle family (Lamiaceae) are useful (including many of the herbs used in cooking and salvias and lavender), as well as delphiniums, snapdragons and honeysuckle. For early spring, when queens have just emerged, pussy willow, lungwort, white dead-nettle and flowering currant provide good sources of forage.

4.2 Making a nest-box Two types of nest-box are useful: a starter box for nestfounding and small colonies, and another for housing larger nests. The smaller box (internal dimensions about 20 cm long, 10 cm wide and 5 cm deep) should be divided into a nest compartment and an outer chamber. The partition requires a hole (about 2.5 cm diameter) for traffic between the two. On each side wall a similar-diameter hole covered in wire gauze should give adequate ventilation. Make an entrance hole, at floor level, in the end wall of the outer chamber. Provide each compartment with a glass or preferably Perspex lid which can be stuck down with tape or otherwise secured. If a gravity feeder is to be used to supply sugar solution, a further hole will be needed in the side or end wall of the outer chamber. A bigger version of the same basic design is suitable for larger colonies (fig. 12). The outer chamber can be relatively smaller, but is still useful as a place to feed the bees and for them to defecate away from the nest. Ventilation holes should be about 5 cm in diameter and are only required in the nest chamber; the degree of ventilation can be varied with a keyhole-type cover. A detachable floor is often useful when housing a large colony and cleaning the box at the end of the season. For outdoor use a runner on either side of the floor will keep the nest off the ground, preventing it from getting damp, while a piece of untreated felt carpet underlay, or similar insulation, covered by a roofing tile or slate, will serve as a weatherproof roof. Do not treat wood with preservative or paint as it may upset the bees. New boxes may be more acceptable if they are left outside to

Nests and their establishment in captivity | 31

weather for several days before being used. Similar nest-boxes are described in Plowright & Jay (1966) and Alford (1975). Pomeroy & Plowright (1980) give two useful designs, one of which is for a more complex heated nest-box. Intenthron & Gerrard (1999) have produced a useful guide to making nests for bumblebees (available from the IBRA, see link 9.13.21, p. 107). Fussell & Corbet (1992c, 1993) give the results of a public survey on the nesting places of different bumblebee species.

4.3 Starting a colony in captivity In the field, nesting can be encouraged by putting out, early in spring, nest-boxes containing cellulose wadding or upholsterer’s cotton (so long as this has not been treated with insecticide). Do not use cotton wool because bumblebees tangle their feet in it. Other methods of preparing sites in the field are reviewed in Holm (1966) and described in Alford (1975), Macfarlane & others (1983) and Fussell & Corbet (1992c). A more reliable way of starting a nest is to catch a queen after she emerges from hibernation. Whenever possible use nest-searching queens (p. 16). Those carrying pollen loads have already founded a colony and are unlikely to begin again. Several methods of proceeding are reviewed in Holm (1966). A method that we have used with some success involves introducing queens into

Fig. 12. Bumblebee nestbox (based on a type used by J. B. Free).

35 cm 8 cm

outer chamber

22 cm

nest chamber

20 cm

16 cm 12 cm

entrance hole (2.5 cm)

feeder hole

detachable floor

32 | Bumblebees Fig. 13. B. pratorum worker taking sugar solution from a small dish.

a garden summerhouse, or a small greenhouse. If this can be heated (to 25–30 °C) so much the better. Windows may need to be covered inside with loose-woven cheesecloth or net-curtain material if bees persist in flying against the glass. Flowers are provided as a nectar and pollen supply; for example trays of growing white deadnettle for nectar, or cut flowers such as broom Cytisus scoparius and gorse Ulex europaeus for pollen. Cut flowers must be changed regularly. Queens will quickly learn to take honey or scented sugar solution (1:1 wt sugar/wt water) set out in small shallow dishes (see fig. 13). Left relatively undisturbed the queens should become quite accustomed to the unusual surroundings and begin to nest-search, exploring corners and crevices, and even cuffs and trouser legs. If nest-boxes containing upholsterer’s cotton are put on shelves and the floor, some queens may begin nesting in them of their own accord. A less time-consuming method is to confine queens in a small nest-box (see sections 4.2 and 4.4). A ball of pollen dough (see p. 17) the size of a large pea should be provided, together with a supply of sugar solution (to obtain pollen, see techniques, p. 101). Chances of success are increased by putting two or more queens together. Initially they may fight, but they usually come to tolerate one another, even if they are of different species (Sladen, 1912; Free & Butler, 1959). Often they cooperate for quite long periods, although one or more may eventually be killed. Another method, which has been used with success, is to provide the bumblebee queen with a few honeybee workers: these help to stimulate the queen in the initial stages of nest development (Ptacek, 1991; Velthuis & van Doorn, 2006).

4.4 How to keep a colony A starter box can be lined with corrugated cardboard. This will soak up moisture and allow the developing nest to be picked up and transferred to a larger box later on. The entrance hole is not needed initially and can be corked. Bumblebees work hard to keep their nest warm. You can help them by providing insulation (such as upholsterer’s cotton), or by keeping the nest chamber at a temperature of about 30 °C. Warmth will generally improve a colony’s chances of developing successfully. Between

Nests and their establishment in captivity | 33 Fig. 14. A simple gravity feeder for supplying sugar solution. syringe artificial flower cork nest-box

periods of observation keep the nest dark with a cover. Sugar solution should be supplied in the outer chamber, either in small plastic dishes or from a gravity feeder. A feeder can be improvised from a syringe, by shortening the nozzle (fig. 14). Avoid spillages; sticky bumblebees are not enthusiastic about nesting. Make sure the supply of sugar solution does not run out, and always clean the containers and make up fresh solution every couple of days. A very small quantity of clove oil mixed into the solution will help the bees to find and use a new feeder. Honey diluted 1:1 with water is also attractive. Once a queen has become broody and laid her first eggs, more pollen will be needed. Small pieces of pollen should be placed nearby. These should be replenished as required or regularly exchanged for fresh material if they are not used. In both pocket-making and pollen-storing species (p. 18) pollen may be placed in empty cells as these are built or become vacant. In pocket-making species pollen can also be placed directly into the feeding pockets on each larval clump. Fed in this way a colony can be kept confined. Alternatively, when about 10 workers have hatched they can be allowed out to collect the required nectar and pollen. Once it has begun to grow, the nest can be transferred to a more spacious box. Bumblebees from indoor boxes will forage outside if a piece of hosepipe, or transparent tube, is passed from the nest entrance out through a window. On the outside, make sure there is an adequate landing platform for easy access to the tube. A coloured surround will help the foragers to locate the correct spot and prevent them from wandering in through other windows. To transfer an established colony to a nest-box, follow the guidelines in section 4.1. Avoid dampness and mould from accumulated faeces by lining the outer chamber with absorbent paper, or a little dry soil, which can be replaced periodically. Finally, try not to disturb the nest unnecessarily. A helpful tip is not to breathe on bumblebees; cultivate the habit, as Sladen (1912) did, of breathing from the side of the mouth when looking at them.

5 Cuckoo bumblebees, parasites and nest associates 5.1 Cuckoo bumblebees Bombus (subgenus Psithyrus)

inquiline: an animal living in the home of another species and using its food

Cuckoo bumblebees are members of the subgenus Psithyrus, and their name is abbreviated here as ‘B. (Ps.)’. They look very much like true bumblebees, as we have seen (pl. 4), and this resemblance is more than a coincidence. Structural similarities come from shared ancestry: cuckoo bumblebees are thought to have originated from true bumblebees (Williams, 1991, 1998). Similarity of colour patterns may have more immediate ecological significance. Each of the six British species of cuckoo bumblebee is an inquiline of one or a few species of true bumblebee (see key III), and most of our species show some resemblance to their usual hosts. The function, if any, of this similarity remains uncertain. Are cuckoo bumblebees simply additional members of Müllerian mimicry groups (p. 5) that protect bumblebees from predators (Williams, 2008)? Cuckoo bumblebees do not work for the colony; they cannot secrete wax and have no pollen baskets. There is no worker caste, all individuals developing into reproductive females or males. The cuckoo larvae are fed and reared by the workers of their true bumblebee host. In extreme environments, such as the Arctic, and probably also at high altitude in some of the mountains of southern Europe, there are true bumblebee species that are capable of behaving just like cuckoos. That is, they do not produce a worker caste of their own, but instead take over nests of another true bumblebee species, the workers of which then rear queens and males for the intruder (Yarrow, 1970; Richards, 1973; Gjershaug, 2009). Mated female cuckoo bumblebees hibernate over winter, emerging rather late in the spring when true bumblebee queens have already established nests. After a period of flower-feeding and ovarian maturation the cuckoo females begin nest-searching, quartering the ground like true bumblebees, but searching for established bumblebee nests. When the cuckoo finds the

Cuckoo bumblebees, parasites and nest associates | 35

nest of her host species she crawls in through the entrance. The true bumblebees are likely to defend their nest against invaders, and the very thick cuticle of female cuckoo bumblebees is presumably an adaptation that helps protect them against defensive stings. The female cuckoo bumblebee is said to hide among the nest material for a few days, until, perhaps, she begins to smell like a nest-mate and therefore excites less aggression from her hosts. She then emerges into the nest cavity. She may act aggressively towards the host workers, but permits most of them to survive; they will rear her own young. She may kill the host queen. Whether or not she does so seems to vary with her species, and may be related to the method of control of queen production by the host bumblebee species (see p. 19). More information is needed on this subject. There are some indications that a true bumblebee queen is more likely to be tolerated by the cuckoo female if she belongs to a ‘simple’, rather than a ‘complex’ species (see p. 19). The cuckoo female often destroys host larvae and eggs, and uses the pollen/wax mixture taken from cells of the host to construct egg cells of her own, building them, as a true bumblebee does, on top of the cocoons. In them she lays numerous eggs. As her ovaries have more branches (ovarioles) than those of a true bumblebee queen, she can lay more eggs in a batch. Interestingly, cuckoo bumblebee species do not show the large differences in tongue length found between some of their true bumblebee hosts. This is presumably because cuckoo females do not compete directly with other bumblebees for nectar, to supply the requirements of their offspring, but instead rely on their hosts to perform this service. When cuckoo bumblebees develop into adults, and leave the nest, they feed lethargically on flowers. The males are particularly slow-moving and obvious, and sometimes give the impression that almost every thistle head is draped with sleepy bumblebees. Like male true bumblebees, male cuckoo bumblebees patrol particular routes, investigating, at intervals, scent-marked features to which females are thought to come in order to mate. Mated cuckoo bumblebee females develop their fat bodies and then hibernate in the soil, while males die off before winter.

36 | Bumblebees

Cuckoo bumblebees are often locally common and may take over a substantial proportion of true bumblebee colonies. In nest-boxes they can cause severe disappointment, cutting short the colony’s development and causing premature production of their own reproductives. Nevertheless, they are interesting bees in their own right and much remains to be discovered about their biology. It is not even certain to what extent each species of cuckoo is limited to its known true bumblebee host species. Does the relationship between the host queen and the cuckoo female vary from one species to another? If so, is it associated with the method of queen production by the true bumblebee host? How does the cuckoo female defend herself and her brood against attack by the host bumblebees?

5.2 Nest associates and parasites A bumblebee’s nest is a rich store of food and cuckoo bumblebees are not the only other animals to take advantage of it. The nest contains a complex community including parasites, predators and large numbers of small insects and mites, of many species: they live in its protected environment, often as scavengers on fragments of wax, pollen and other nutritious material that accumulates there. Many of these nest-mates are tiny and poorly known, and there is abundant scope for research on their natural history, and on their impact on the bumblebees’ fitness. Here we concentrate on a few of the more conspicuous forms that seem to have a specific association with bumblebees. A review of information about bumblebee infections and parasites is given in Macfarlane & others (1995). For a detailed account of the parasites of social insects, including bumblebees, see Schmidt-Hempel (1998). One of the most destructive enemies of bumblebee colonies in the wild and in nest-boxes is the wax moth Aphomia sociella. In summer, the adult moths find and enter bumblebee nests, particularly those built on the ground surface, and lay clusters of eggs. These soon hatch, giving pale larvae which remain in groups, each spinning a silken gallery in which it lives, feeding at first on scraps and rubbish. As the larvae grow they begin to invade the comb, eating cells, food stores and even

Cuckoo bumblebees, parasites and nest associates | 37 Fig. 15. Antenna of Volucella bombylans.

bumblebee larvae. In this way they quickly destroy the colony, leaving it riddled with their silken galleries. Still moving in a group, the mature larvae then leave the nest, to spin tough silken cocoons in some sheltered place nearby. In these they overwinter. During spring they pupate, to emerge as moths in summer. Wax moth caterpillars are not the only caterpillars to be found in bumblebee colonies. Other species, which appear less frequently and do less damage, are described by Alford (1975). Wax moth caterpillars, yellowish with olive-green on the back, are fairly distinctive, but for certain identification it would be necessary to rear the adult moths and identify these from Goater (1986). Many species of two-winged flies (Order Diptera) are associated with bumblebee nests. Some are generally harmless scavengers. One of the most interesting of these is a large hoverfly (Family Syrphidae) called Volucella bombylans. The adult flies, fatter and furrier than most hoverflies, are amazingly good mimics of bumblebees and can easily be mistaken for them in the field, even emitting a similar buzz when caught, and raising the middle leg in the typical bumblebee defensive posture. On close inspection they are known to be flies because they have only one pair of wings (bumblebees have two, but often zip the hind wings to the fore wings with a row of hooks) and distinctive feathery antennae (fig. 15). This species of hoverfly has several colour forms: one has a white tail and yellow bands, resembling B. hortorum and other white-tailed bumblebees, and another is black with a red tail, resembling B. lapidarius and B. ruderarius. Presumably flies such as Volucella are taking advantage of the Müllerian mimicry groups which protect bumblebees from predators (p. 5); experimental evidence is described in Evans & Waldbauer (1982). A Volucella female enters a bumblebee nest and lays her eggs there. A newly killed mature female will sometimes lay her eggs in the killing bottle. This reflex egglaying probably occurs in nature, allowing females to leave progeny in a nest even if they are killed by bumblebee workers (Sladen, 1912). The legless larvae feed on debris at the bottom of the nest, probably doing no harm to the bumblebees. As in all higher flies the mature larva forms a tough brown puparium; the soft white pupa forms inside the rounded-off, hardened, darkened cuticle

38 | Bumblebees Fig. 16. A conopid fly, Conops quadrifasciatus

Fig. 17. Anchor-like structure on a conopid egg (Sicus species). After Smith (1969).

of the last larval stage. The puparium overwinters in the nest and the adult hoverfly emerges the following summer. Coloured pictures and descriptions of the adults of V. bombylans can be found in Gilbert (1986), and Alford (1975) describes and illustrates the immature stages. Brachicoma devia of the Family Sarcophagidae is a smaller fly, less flamboyant but commoner and more harmful. The adult, looking rather like a house fly, lays her larvae (rather than eggs) in the clumps of bumblebee larvae in the nest. The fly larvae wait until the bumblebees have finished feeding and spun cocoons in which to pupate. Then they begin to suck out their contents, living as external parasites. At this stage if you open up a pupal cocoon you find, instead of a pupa, a shrinking bumblebee larva with perhaps three or four little white fly larvae plugged into it, feeding through the cuticle on body fluids. When fully grown each fly larva leaves the cocoon in which it has fed and forms a brown puparium, which may give rise to an adult in a few weeks. There may be several generations a year. B. devia is very common, parasitising other bees and wasps as well as bumblebees, but it rarely destroys a whole colony. Critical identification is not easy; it is necessary to rear adults and use van Emden’s (1954) key. Conopid flies are distinctive parasites, more likely to attract attention in the field. They are large-headed flies (fig. 16) and many species adopt a very characteristic hunched posture as they sit on flowers, feeding, or ambushing bumblebees. The female fly lays her eggs inside the body cavity of an adult bumblebee, inserting each egg by piercing the bee’s body wall with a special structure at the tip of the abdomen (Alford, 1975). The fly sits on or near a flower and if a bumblebee approaches, the fly shifts around, always facing the bee, perhaps swaying its head from side to side as if taking a fix on the bee. When a bumblebee comes near enough the fly takes to the wing, darts out and grapples with it briefly in flight. The episode is over very quickly, but if you now catch and dissect the bumblebee you will probably find inside it a long, narrow egg with a complex anchor-like or filamentous structure at one end - a conopid egg (fig. 17). The conopid larva that hatches from it (fig. 18) will develop as an internal parasite, eventually growing large enough to fill the host’s abdomen. The bumblebee dies,

Cuckoo bumblebees, parasites and nest associates | 39 Fig. 18. The conopid Physocephala: third instar larva. After Smith (1969).

Fig. 19. Parasitellus species deutonymph.

0.5 mm

often in her nest, and the fly larva forms a puparium which overwinters within the husk of the host’s abdomen, to emerge the following summer. There are 24 species of conopid flies in Britain and the adults can be named using Smith’s (1969) key. Little is known about their biology and distribution in Britain. Observations on conopids in the field or at the nest could make a useful contribution. In Switzerland they have been extensively studied by Schmidt-Hempel and his co-workers, who have shown that conopids can have a very significant effect on the biology and behaviour of their hosts. The parasite can affect a large proportion of foragers (Schmidt-Hempel & Durrer, 1991), altering their foraging behaviour on flowers (Schmidt-Hempel & Müller, 1991), and causing the bees to dig into the soil before they die – which favours the survival of the parasite (Müller, 1994). A bumblebee’s nest is often alive with tiny mites and the bees themselves are often found to carry numerous mites, especially lodged at the back of the thorax. Identifying mites is a job for an expert; therefore they are not easy animals to work on, although much remains to be discovered about them. One of the commonest species found on foraging adult bumblebees is Parasitellus fucorum (fig. 19). Its young stages develop, probably as scavengers, in bumblebee nests. When nearly mature, the mites attach themselves to adult bumblebees, particularly young queens. In this way they remain with the queen throughout the winter and can invade her new nest when she establishes it in spring. When a mite-laden queen visits a flower, some of the mites may disembark, and wait in the flower to board another bee, perhaps from a different colony, and be carried back to her nest (Corbet & Morris, 1999). When you probe a flower with a micropipette to withdraw nectar (p. 99), you may occasionally see such a mite scamper onto the micropipette, as it would onto a bee’s tongue. If you examine a few mites from a bumblebee’s thorax, mounting them on a microscope slide and using a compound microscope, you may see an even smaller mite that lives on them, a bizarre tortoise-like mite with huge claws: Scutacarus acarorum (fig. 20). Mites of another species are sometimes found inhabiting the respiratory system of bumblebees, feeding on the bumblebee’s blood

40 | Bumblebees

trachea: a tube of the respiratory system

Fig. 20. Scutacarus acarorum: a mite that lives on mites that live on bumblebees.

0.1 mm

corpora allata: small secretory organs behind the brain

through the wall of the trachea. These and many other mites are described and illustrated by Alford (1975). One of the most important and unusual bumblebee parasites is a microscopic roundworm, or nematode, Sphaerularia bombi. This infects only queens, but it can seriously disrupt the behaviour and physiology of its hosts, and an understanding of its effects is necessary for the interpretation of field observations on queen bumblebees. A proportion of the queens caught in spring contain in the abdomen a white, bobbly, sausage-shaped structure about 10–20 millimetres long (fig. 21). This is an adult female roundworm - or rather it is the reproductive system of the worm, turned inside out, with the rest of the worm attached as a tiny strand at one end. It produces large numbers of eggs which are released into the blood of the bumblebee. Inside each, a roundworm larva develops through its first two stages. In the third stage it hatches; examination of the bumblebee’s blood under a compound microscope at this time reveals thousands of larvae as little white strands about 1 millimetre long. From the blood these larvae move into the gut and reproductive system, and eventually reach the outside world, often in the bumblebee’s faeces. A surprising feature of this parasite’s life history is that it centres not on the bumblebee’s nest but on the hibernation site. Here mated adult worms infect new queens when these arrive and burrow into the soil at the beginning of winter. Within the infected queen in spring the female roundworm grows, turns its reproductive system inside out, and starts releasing eggs. The effect on the host is remarkable. In a healthy queen in spring, the corpora allata release a hormone in the presence of which the ovaries develop. When this happens the queen begins to search for a nest-site and establish a nest. In a parasitised queen the roundworm somehow prevents the development of the bee’s corpora allata: the ovaries fail to develop and the queen does not search for a nestsite and establish a nest, or load her pollen baskets with pollen. Instead she forages for herself alone, in a desultory way, and may return to the hibernation site. Here numbers of parasitised queens may spend their time. When worm larvae are discharged in their faeces the soil of the hibernation site becomes infected, and it is here that the worms develop, mate and infect new queens as

Cuckoo bumblebees, parasites and nest associates | 41 Fig. 21. Sphaerularia bombi: a nematode parasite of bumblebees. body of a female

enlarged uterus

these arrive to hibernate in autumn. The life history of S. bombi is described and illustrated by Poinar & van der Laan (1972) and Alford (1975). In early spring there may be no obvious behavioural differences between infected and healthy queens. However, the healthy individuals soon begin establishing colonies, and by summer most of these queens will be confined to their nests; the majority of queens still seen at flowers will be those carrying the roundworm.

5.3 Predators

1 mm

Bumblebees have many parasites but few predators, perhaps because of their sting. We have seen that convergence in colour pattern has been interpreted as Müllerian mimicry (p. 5). A system like this will give protection only against predators that hunt by sight, and the accuracy of the mimetic patterns implicates predators with good visual acuity. Apart from spiders, most bumblebee predators are vertebrates. Spotted flycatchers Muscicapa striata destroy bumblebees’ stings by wiping the insects against a branch (Davies, 1977). Blue and great tits Cyanistes caeruleus and Parus major are probably quite important predators at times (Benton, 2006), especially as they prey on bumblebee queens in spring (Badmin, 2009; Redhead, 2009), perhaps benefiting from times when the bees are torpid, and unlikely to sting. Further south in Europe, bee-eaters Merops apiaster can also remove bees’ stings. Shrikes, particularly the red-backed shrike Lanius collurio, collect bumblebees and impale them, along with other prey, on the thorns of a ‘larder’ bush. Badgers Meles meles dig out nests and feed on their contents. In Iceland bumblebees and their brood sometimes form a large part of the diet of mink Mustela vison (see Prŷs-Jones, Ólafsson & Kristjánsson, 1981). Other than man, the vertebrates causing the most trouble for bumblebees are probably small mammals. These may invade nests and destroy them; they probably eat the bees and larvae. But of course bumblebees also benefit from small mammals, as their abandoned nests provide the sites for many bumblebee colonies.

6 Foraging behaviour “No one cares for the humble-bee. But down to the flowering nettle in the mossy-sided ditch, up to the tall elm, winding in and out and round the branched buttercups, along the banks of the brook, far inside the deepest wood, away he wanders and despises nothing”. Richard Jefferies, Pageant of Summer, Longman’s Magazine, June 1883.

6.1 Economics of foraging Jefferies may have mistaken the gender of the bee, but the sentiment was right. Before the queen has founded a nest, and once workers are produced who can begin to forage for the colony, the over-riding task of all female true bumblebees involves collection of nectar and pollen to feed themselves, their offspring and their nest-mates. Above everything else, this job is vital to building up a forager population that can successfully rear as many queens and males as possible. Adaptations that may have evolved to make this process more efficient can be studied in a whole variety of ways: observational, structural, physiological, biochemical or behavioural, to name but a few. In collaboration with ecologists, true bumblebees have played a large part in the development and testing of economic ideas about foraging: these enable the behaviour of a real bumblebee to be compared with behaviour that might be expected if the bee were trying to achieve a particular aim. A good introduction to this approach to studying behaviour can be found in Krebs & Davies (1993). The ‘aim’ of a foraging worker might be to maximize her rate of gain of some necessary ‘currency’, commonly taken to be energy, given certain constraints. A constraint might be, say, the necessity to sample the available flower types, or the requirement for water or a particular nutrient. This approach has allowed the incorporation of mathematical ideas, and dignified bumblebee natural history with the sta-

Foraging behaviour | 43

tus of theoretical ecology. It can sometimes allow us to make testable hypotheses, which help us get a clearer understanding of why bees do what they do. However, although popular with biologists, it is only one way of investigating bumblebee behaviour, and it has its limitations; it is important to bear in mind that the mathematical treatment of biological systems requires simplifying assumptions. The selection of assumptions that are at once realistic and productive requires a careful understanding of bumblebee biology (see also Thomson & Chittka, 2001 and Benton, 2006). We should not underestimate the extent to which major theoretical advances will depend on patient observational studies of natural history. Further observations of the type outlined below should help us gain a better understanding of the ways in which bumblebees enhance their foraging efficiency. Energetic costs of foraging can be estimated from the work of Heinrich (1979). Insects can only fly if their flight muscles – which constitute most of the thorax – are warm enough. Bumblebees require a higher temperature than most insects: the thorax is normally maintained at between 30 and 40 °C during flight, and if it cools below 27 °C the bee can only crawl. Despite this requirement bumblebees can often be seen flying at air temperatures well below this; in the case of queens, down to about 4 °C or even less. They can do so because of their ability to generate and retain heat. As in all insects, the muscular flight motor is not perfectly efficient, and only about 20% of its power output does useful work; the rest goes as heat, which warms the thorax during flight. So a bee that keeps flying keeps warm. In preparation for flight, and in order to incubate a brood, bumblebees can warm up by shivering. This involves repeatedly contracting the flight muscles, while functionally uncoupling them from the wings. No useful work is done, but heat is produced. The existence of this form of heat production is well established (Heinrich, 1979; Esch & others, 1991). It has been suggested that bumblebee flight muscles may also be able to produce heat without contracting, using an energy-releasing biochemical cycle – a ‘substrate cycle’ – under the bee’s control (Newsholme & others, 1972; Newsholme & Crabtree, 1976; Storey, 1978; Surholt & others, 1990, 1991). The hypothesis is that substrate

44 | Bumblebees

cycling may enable bumblebees to warm up, or at least reduce their rate of cooling, between periods of flight. It might also be anticipated to be a ‘cheaper’ means of heat generation than shivering, because none of the energy generated would be wasted in unused muscular contraction. The very existence of substrate cycling has proved contentious (Heinrich, 1993; Staples & others, 2004). Thus far the arguments for and against have themselves generated, perhaps, more heat than the cycle itself! Differences of opinion may in part relate to the biological context in which the measurements were made, and they are not necessarily irreconcilable. We now know that bumblebee species vary considerably in the activity of the enzyme required for a substrate cycle to function (fructose bisphosphatase). Measurement of the levels in four common British bumblebee species, B. pascuorum, B. terrestris, B. hortorum and B. lapidarius (table 4), showed much the highest values in B. lapidarius (PrŷsJones & Crabtree, 1983). North American bumblebee species have also been found to have levels that vary with species but are generally lower than those in some of the British species that we studied (see Staples & others, 2004). As Heinrich (1993) acknowledges, the biochemical data are not in question, and an explanation needs to be found for them. B. lapidarius tends to specialise in foraging on massed flowers, such as members of the daisy family, in which individual florets are often nectar-poor, but their clustered arrangement allows a bee to probe many flowers between flights. While some bumblebee species can become torpid on massed flowers, allowing their body temperature to drop, B. lapidarius workers commonly remain active. By enabling the bees to keep warm cheaply, high enzyme activity may make foraging on such flowers (which generally have very small nectar rewards) more profitable for B. lapidarius than it is for other bumblebee species, which may need to rely more on shivering in order to keep warm enough to fly on to the next flower (Prŷs-Jones, 1986). Whatever the role of substrate cycling turns out to be, physiological variation between species may account for a number of differences in observed foraging behaviour, and deserves further study. Although substrate cycling should be cheaper than shivering, the fuel to power both systems is costly. Insulating body hair helps bumblebees economise by reduc-

Foraging behaviour | 45 Table 4. Foraging behaviour and the activity of fructose bisphosphatase in the flight muscles

Species B. lapidarius B. lucorum B. pratorum B. terrestris B. pascuorum B. hortorum

Fructose bisphosphatase activity (μmol min-1 g-1 muscle, mean ± standard error)a

Proportion of visits to massed flower arrangementsb

131 ± 7 (8) 80 ± 16 (5) 73 ± 10 (13) 59 ± 13 (7) 45 ± 6 (20) 23 ± 1 (11)

0.54 (210) 0.39 (84) 0.19 (177) 0.38 (188) 0.18 (254) 0.07 (159)

Correlation coefficient, r = 0.88, df = 5, P < 0.02 a

Based on data supplied by B. Crabtree (summarised in Newsholme & others, 1972), and on Prŷs-Jones & Crabtree (1983). Numbers in brackets are sample size. b From Prŷs-Jones (1986). Numbers in brackets are sample size.

ing their rate of heat loss (Church, 1960; Heinrich, 1993), and in cool weather they often use supplementary solar heating, basking on sunny surfaces. Sometimes they will even flatten themselves against the warm, sunlit jersey or sock of an observer. This ability to generate a high body temperature enables bumblebees to fly in cold weather and to maintain a steady high temperature in the nest. It permits them to forage very early in the morning and late in the evening, when there is little competition from other flower-visiting insects (see p. 46 for implications regarding forager numbers at various times of day) and it enables them to thrive at high latitudes and high altitudes. Although they can regulate cooling rates to some extent, by controlling the flow of warm blood past the hairless ‘heat window’ on the lower surface of the abdomen (Heinrich, 1993), a sturdy, furry flying machine the size of a bumblebee will always be above ambient temperature when active, and in hot weather overheating is probably a serious problem. Bumblebees often show a lull in flower-visiting activity in the hottest part of the day, and at low latitudes they are commonest in cool, mountainous regions. Generation of a high body temperature accounts for most of the energetic cost of activity. By measuring the rates of oxygen consumption of flying bumblebees Ellington & others (1990) have shown that flight costs a

46 | Bumblebees

bumblebee about 1.2 kilojoules per hour. Crawling can be cheaper, if the bee allows passive cooling to occur, the exact cost depending on the temperature of the thorax and the air. Clearly, in order to estimate a bumblebee’s foraging costs one needs to know what proportion of its time is spent on the wing, its weight, and its body temperature (Heinrich, 1979). A bumblebee crawling slowly over the massed inflorescences of goldenrod Solidago species may have allowed its thorax to cool down, thereby cutting costs. One foraging rapidly, or pausing briefly between short flights, needs to keep its thorax warm so that it can fly again without delay, and so must expend more energy. In this way it is possible to estimate a bumblebee’s expenses while it is working a patch of flowers. If one adds to these the costs of flying to and from the nest, again calculated on the basis of observed flight times, one gets the overall cost of a return foraging trip from the nest to the flowers and back. Profits to be set against these costs can be computed from the number of flowers visited per trip and the mean reward per flower. However it is not easy to follow a bumblebee for an entire trip to count its flower visits. Instead, one can record the rate of flower visitation (from timings of, say, 10 or 20 visits), and estimate the duration of a whole foraging trip by noting the times when marked individuals (technique p. 96) are seen at the foraging patch, or when they are seen to enter or leave the nest; thus the number of flowers visited per trip can be estimated. At this stage it is worth making a point, the significance of which is often overlooked. A female bumblebee must transport each load to her nest: she is a ‘central place’ forager. As a result, when nectar is abundant she may spend quite a small proportion of a foraging trip actually on flowers compared to the time spent travelling to and from the colony and depositing the load (PrŷsJones, 1982; Cresswell & others, 2000). Conversely, when nectar is scarce - often in the middle of the day - a large proportion of her time will be spent on flowers. This has paradoxical consequences. First, if one makes counts of the number of bumblebees on a patch of flowers at various times, numbers of foragers may appear to increase as resources decrease. Secondly, if one attempted to estimate the extent of competition between species just by

Foraging behaviour | 47

counting bees on flowers, one would seriously overestimate the contribution of times when large numbers of bees are present, but little food is available to any of the foragers. For example, the extent to which bumblebees experience competition from honeybees Apis mellifera for nectar would almost certainly be overestimated. Both forage during the middle of the day, when the amount of nectar is often quite small. But bumblebees also forage early in the morning, late in the evening and under more adverse weather conditions than honeybees – times when nectar is often relatively abundant (Corbet & others, 1995). They may therefore gather the majority of their requirements at times when honeybees are not present. Interpretation of the relative abundances of bumblebees and honeybees (for example, Forup & Memmott (2005)), may be influenced by these considerations. In an increasingly fragmented habitat, nest placement in relation to suitable foraging areas will become an increasingly important influence on colony foraging success. The need to transport each honeystomachful of nectar to the nest has implications that relate to the volume and concentration of nectar per flower. The decisions that enable a forager to gather the maximum amount of sugar per unit time for her nest may often depend more on nectar volume and concentration than on the amount of sugar per flower. A bumblebee’s honeystomach holds only a limited volume of nectar (for workers between about 60 and 200 microlitres depending on their size). The more concentrated that nectar is, the more sugar the nest gains from each load. Foraging a given distance (travel time) from the colony, a bumblebee may be able to bring home a greater weight of sugar per unit time by visiting flowers containing concentrated nectar, than by visiting flowers supplying larger volumes of more dilute nectar, even though the latter contain an equal or even greater weight of sugar per flower. Alternatively, the choice may be between flowers close to the nest and others at a distance. In this case foragers may make a greater rate of gain of sugar to the colony by visiting the closer flowers, even if these provide more dilute and/or smaller amounts of nectar per flower (Prŷs-Jones, 1982). Since bumblebees are not known to communicate the location of food, initially finding nectar sources takes time, thereby increasing their ‘effective’ distance (Wad-

48 | Bumblebees

dington, 1977), and decreasing the rate of energy return. At the same time the area in which to locate food increases as the square of the distance from the nest. Thus honeybees, relying on received information, will become increasingly efficient at exploiting new food sources relative to bumblebees, the further away these are from the colony (see for an example Beekman & Ratnieks, 2000). Given some of the slightly counterintuitive considerations mentioned above, about what sort of nectar resources may be preferred, how far from the colony would we expect a bumblebee to forage? One must remember that the experience of any individual has developed in relation to scattered and constantly changing patterns of floral resources. Releasing bumblebees at different distances from their nest shows that some species can return from distances of several kilometres (Rau, 1924; Goulson & Stout, 2001); and ingenious experiments using harmonic radar to track the movements and speed of travel of individual bumblebees (Carreck & others, 1999; Osborne & others, 1999), together with studies of the genetic identity of individual foragers (Chapman & others, 2003; Darvill & others, 2004), have confirmed that workers will, in fact, travel considerable distances to forage. A simple and effective observational study by Dramstad (1996) reaches the same conclusion, and considers a number of possible reasons why workers may avoid foraging close to the nest: such as resource availability and the risk of predation on the nest or individual foragers. Other less obvious features may be important too. Nectar is not just an energy source: unlike honeybees, which will collect water on its own, bumblebees rarely do so and must normally (see below) achieve water balance for themselves and their brood from the water contained in nectar. What compromises are involved so as to balance the need for energy with the requirement for water? So far, this important point has not been addressed in much detail when considering queen and worker foraging, either in optimal foraging models or in field studies. If bumblebee larvae do not thrive on highly concentrated nectar, as is the case with honeybee larvae, foragers may sometimes select lower concentration nectars than those that would be most profitable energetically. In contrast, male bumblebees forage only for themselves, and an

Foraging behaviour | 49

elegant experimental study by Bertsch (1984) has shown that they may gain water, from flight metabolism and from nectar, faster than they can lose it. Thus bumblebees’ activity patterns may sometimes be structured by the need to discharge excess water. Although collection of nectar and pollen is by far the major preoccupation of foragers, they will also collect honeydew from aphids, sometimes extensively. On occasion they will also visit faeces, urine, carrion (Herrera, 1990) and sweat (Prŷs-Jones, personal observation), presumably for salts or nitrogen, and they will also collect water (Ferry & Corbet, 1996). Our own observations over many years suggest that these latter activities are rare, at least in Britain, and may relate only to specific stressful conditions. The energetic reward per flower can be calculated from the volume and concentration of the nectar in each, on the assumption - which is almost, but not quite true - that all the solutes in nectar are sugars. Nectar volume can be measured by allowing nectar to run up into a microcapillary tube, and measuring the length of the nectar column (technique, p. 99; and see Corbet, Unwin & Prŷs-Jones, 1979; Corbet, Willmer & others, 1979; Corbet, 2003). From such measurements made on, say, 10 or 20 flowers, it is possible to calculate the mean weight of sugar per flower, and so discover how much sugar the bee acquires during her whole foraging trip. Variability in the amount of sugar per flower is also important: the same mean weight of sugar per flower may be based on, for example, one full flower in 10, or 10 partially full flowers. If a bee samples only five flowers before deciding whether to stay or leave it will make a different profit, and may therefore behave differently, in the two cases. A word of caution is needed here. Close observation will probably reveal that bumblebees are selective in the flowers they visit; perhaps choosing flowers of a particular age, or in a particular position on each plant or inflorescence. As far as possible, nectar samples should be taken from flowers on the same basis. Even then, sampling may underestimate the quantity of sugar a bee can collect, because bumblebees may be avoiding recently visited – and therefore empty – flowers in a way that an observer cannot do. For example, although different individual bumblebees sometimes visit the same flower

50 | Bumblebees

Fig. 22. B. terrestris worker dissected to show the honeystomach.

in quick succession, it has been shown that each leaves a chemical mark, which can influence the decision of a second bee whether to visit or avoid the flower (Stout & others, 1998; Stout & Goulson, 2001). Although they may have a variety of functions, these marks may simply be the incidental consequence of visiting a flower. If so, they may act as ‘cues’ rather than ‘signals’ (Saleh & others, 2007), and may therefore just provide more information that a subsequent visitor can attend to, or ignore, depending on the foraging conditions at the time: in other words, be ‘context specific’ (Thompson & Chittka, 2001). Measurements of nectar quantity and timings of bumblebee visits must be done quickly, because both may change markedly from hour to hour and from day to day. Nectar content changes as rates of secretion, and rates of removal by flower visitors, vary with the weather. Nectar solute concentration changes with ambient relative humidity, especially in open flowers, where the nectar may become concentrated and viscous by evaporation in dry air (Corbet, Willmer & others, 1979; Corbet, Unwin & Prŷs-Jones, 1979). Bee visiting patterns, and the time taken to empty each flower, will also change through the day, as weather influences the bees’ behaviour directly, and also influences the amounts and quality of nectar per flower (Willmer, 1983) A nectar-collecting bumblebee usually leaves the nest carrying only a small reserve of nectar, so that the nectar she carries in her honeystomach when she returns represents the net profit from her foraging trip. The honeystomach, a cuticle-lined swelling in the foregut, lies in the front of the abdomen (fig. 22). A bumblebee fills (or partially fills) it with nectar when she is foraging, flies back to the nest, and regurgitates its contents into an empty pupal cell or specially constructed wax cell. When it is full, simple dissection reveals the honeystomach as a clear bag of nectar (technique, p. 97). If the oesophagus has been cut by first removing the bee’s head, the whole honeystomach can be removed from the abdomen with a pair of forceps. If it is punctured the nectar flows out, and the volume and concentration can be measured. In this way the load can be estimated. An alternative technique, preferable because it need not damage the bee, is to anaesthetise it (technique, p. 96), unfold the tongue, and press gently on the tip of the abdomen using

Foraging behaviour | 51

one’s thumbnail (to avoid the sting). The honeystomach contents will be regurgitated and can be collected into a microcapillary tube for measurement of volume and concentration (technique, p. 99). Measuring the weight of the bee on leaving and returning to the colony provides another alternative method of estimating the amount of nectar, or nectar and pollen, which has been collected. Using these techniques it should be possible to explore the energetic profitability of bumblebee foraging in different circumstances, and to see how profitability is influenced by, say, the distance of the flower patch from the nest, or the distance apart of flowers in a clump, or the presence of honeybees. Unlike honeybees, which receive precise directional instructions from other foragers about the location of foraging areas, bumblebees learn where to forage largely by their own initiative. They may be ‘recruited’ into searching for a particular forage plant in a non-directional way, by excited movements, and release of chemical compounds, by returning foragers, but so far as is known no directional information is imparted (Granero & others, 2005; Dornhaus & others, 2003). As already mentioned above, tongue length varies between Bombus species and influences which flowers are visited for nectar. Similar patterns are found within species: larger workers with longer tongues learn to visit a slightly different range of flowers than their smaller, short-tongued sisters (table 5; and see Brian, 1957; and Prŷs-Jones 1976 & 1982; Peat & others, 2005). In the case of long-tongued species in particular, it seems probable that this has a significant influence on overall colony foraging efficiency. By timing the flower visits of newly emerged and more experienced workers, Heinrich (1976) has shown how important learning is as a component of successful foraging. On complicated flowers such as monkshood Aconitum, true bumblebees take some time to learn the best way of acquiring nectar, and increase their foraging efficiency very much by doing so. The experience of individual foragers is paramount: an experienced bumblebee working a familiar and productive patch of flowers may adopt a regular foraging route, which she follows repeatedly with only minor variations (Manning, 1956; Thomson, 1996). The advantage of this ‘traplining’ to the

52 | Bumblebees Table 5. Tongue length and foraging behaviour within species Species and size of worker B. terrestris small medium large B. pratorum small medium large B. pascuorum small medium large a b

Tongue length (mm, mean ± standard error)a

Corolla length (mm, mean ± standard error)b

6.9 ± 0.1 (11) 8.5 ± 0.2 (15) 9.3 ± 0.1 (9)

5.6 ± 0.3 (40) 6.3 ± 0.2 (112) 7.7 ± 0.4 (16)

6.2 ± 0.2 (8) 7.3 ± 0.1 (9) 7.8 ± 0.8 (9)

6.9 ± 1.3 (9) 7.1 ± 0.4 (76) 9.0 ± 1.0 (15)

6.9 ± 0.1 (6) 8.4 ± 0.1 (15) 10.1 ± 0.2 (10)

7.3 ± 0.4 (23) 8.0 ± 0.2 (203) 8.6 ± 0.3 (125)

Numbers in brackets are sample size. The average weighted by frequency of visitation (% × corolla length/100). Based on flower visits for nectar, and nectar + pollen. Numbers in brackets are sample size. From Prŷs-Jones (1982).

bee is reduced search time, and ensuring she does not revisit flowers she has just emptied. For long-tongued bumblebees, which have more exclusive use of long-tubed flowers, this method of foraging may be particularly profitable (Prŷs-Jones, 1982). If a video recorder, or an observer with a stopwatch and a notebook, is focused on a patch of flowers that are being visited by marked bees, it is sometimes possible to detect traplining, and to discover the interval between successive visits to a flower by a particular bee (Kearns & Thomson, 2001). If the rate of secretion of nectar is measured at the same time, it should be possible to calculate how much of an energetic profit she is making under a variety of conditions. Other systematic visiting patterns may also serve to reduce revisits. Bumblebees foraging on flowers arranged in vertical spikes usually work upwards on a spike, flying downwards between plants to start near the bottom of the next spike. This is easy to confirm by counting the proportion of flower-to-flower movements, on a spike or between spikes, that are upwards, downwards or level. There has been much discussion about the significance of this habit to the bees (see Benton, 2006). Pyke (1978) considered the movements from an optimal foraging perspective in which the bees visited the low-

Foraging behaviour | 53

est, most rewarding flowers first, and moved upwards to progressively less rewarding flowers before leaving. In contrast, Corbet & others (1981) found that common toadflax Linaria vulgaris produced most nectar at the top of the inflorescence, but nevertheless bumblebees still began at the bottom and worked upwards. Whether or not this pattern of flower visiting always fits strictly into an optimal foraging model, it may still be advantageous as a means of avoiding recently visited flowers. It is likely to be of adaptive significance to the flowers too, favouring cross-pollination. In species such as rosebay willowherb Chamerion angustifolium, whose flowers open in sequence from the bottom upwards, ripening their anthers before their stigmas, the lower flowers will be effectively female and the upper ones effectively male. The upward directionality of the bumblebees will therefore mean that a bee, arriving newly dusted with pollen from another plant, deposits this on the receptive stigmas of the lower flowers before picking up a fresh load of pollen from the anthers of the upper flowers, and transferring this to another plant. Flower scent, colour, orientation and size may all provide recognisable cues which result in further patterns of foraging behaviour and differences between bumblebee species, and individuals, in their flower choice. A recently emptied flower may be recognisable by a scent mark, or smell less strongly than a full one, and thereby save the bee the trouble of alighting. Colour changes may provide equally useful information: for example petals of young flowers of horse-chestnut Aesculus hippocastanum are marked with yellow, whereas in old flowers the marks are red. Some colours can be seen better by some bees than others (Raine & Chittka, 2005) and larger foragers may find flowers more easily than small ones because their eyes are more sensitive (Spaethe & Chittka, 2003). As a result of this, and together with their greater control of body temperature, large bees may forage earlier in the morning and finish later in the evening. Although one thinks of nectar as an attractant, encouraging visitors, it may have deterrent effects on some visitors in some circumstances. An example is the highly alkaline nectar of the purple toothwort (Lathraea clandestina), which seems to act as a deterrent to nectar thieves, such as ants (Prŷs-Jones & Willmer, 1992). Collection of

54 | Bumblebees

some nectars may also be a means of self-medication: Manson & others (2010) found reduced levels of infection by a protozoan gut parasite in bumblebees collecting a nectar alkaloid from the bee-pollinated plant Gelsemium sempervirens. Some nectar can, in some circumstances, ‘intoxicate’ bees (see Adler, 2000, for more information). Lime Tilia and Rhododendron nectars are well known for this: in certain situations it may be an incidental side effect of the production of sugars that bees cannot metabolise. Accumulations of dead or dying queen and worker bumblebees are sometimes found under flowering shrubs and trees, especially lime Tilia, Rhododendron and willow Salix (Alford, 1975; Benton, 2006; Badmin, 2009). If the corpses are intact, poisoning may be the cause of death, especially in association with lime and Rhododendron. But if the bodies are dismembered, great tits and possibly blue tits may have been removing the abdomen to feast on nectar in the honeystomach. In spring, when willow is flowering, tits are often energy restricted; at this time blue tits will systematically collect nectar directly from flowers of both willow (Kay, 1985) and flowering current Ribes sanguineum (Swynnerton, 1916; Fitzpatrick, 1994). Indirect collection of nectar, by killing bumblebees, may have developed from this habit. The holes found in the flowers of flowering current are usually made by blue tits, rather than, as one might assume, by bumblebees (Swynnerton, 1916). On some flower visits a bumblebee collects nectar only. In such cases her pollen baskets may be empty, or she may, rarely, carry persistent pollen masses left over from an earlier trip. Often she will collect both nectar and pollen, probing with her tongue at the base of the flower, becoming dusted with pollen meanwhile, and periodically hovering or pausing on a flower to groom her body hairs and comb the resulting pollen into the pollen baskets on her hind legs. Sometimes a bee collects pollen alone, even when that flower type also produces nectar; bumblebee species differ in fairly consistent ways in the frequency with which they do this (table 1). B. lucorum does so often; the habit is less common among the other species that we studied (Prŷs-Jones, 1982), and B. hortorum, at the opposite extreme, nearly always collects nectar and pollen together when pollen-gathering. In addition, B. pratorum and B. hortorum, both of which

Foraging behaviour | 55

have a short colony cycle, collect pollen on a much higher percentage of flower visits than other common species (table 1). With information of this type, patterns of ecological differences between species begin to emerge (fig. 23). The pattern in fig. 23 presumably reflects colony demands on the forager population, resulting from species differences in colony size and the length of the colony cycle, and individual abilities, which vary with tongue length and thermoregulatory efficiency. Some flowers lack nectar, and pollen is the only resource available; meadowsweet Filipendula ulmaria, broom Cytisus scoparius and some St John’s-worts Hypericum are examples. A pollen-collecting bumblebee may scramble over the anthers, coating her body in pollen which is later groomed off; or she may dislodge the pollen by vibration, emitting a distinctive high-pitched buzz (Corbet & others, 1988). This buzz pollination is frequently used when visiting flowers with anthers from which pollen escapes at the tip, such as woody nightshade Solanum dulcamara. Tomato flowers have anthers like this too, and commercial growers used to use a vibrating ‘electric bee’ to assist hand pollination, before the development of much cheaper methods, based on buzz pollination by commercially produced colonies of bumblebees (see p. 70). (Honeybees cannot be used H

A

P 100 90 80 70 60

Av e de rage pt h ( coro mm lla )

Flower visits for nectar (% to separate flowers)

Fig. 23. Patterns of flowervisiting behaviour (see text). Letters indicate species: B. hortorum (H), B. pratorum (P), B. pascuorum (A), B. terrestris (T), B. lucorum complex (L) and B. lapidarius (D). Based on information collected throughout the life cycle of each species. From Prŷs-Jones (1982).

9

8

50 40

T

10

L

7

D

6 5

20

40

60

80

Flower visits (% nectar-only)

100

56 | Bumblebees

Fig. 24. B. pascuorum queen grooming.

because they do not buzz-pollinate.) Even when a bumblebee does not take pollen and nectar from the same flower, she may collect both on the same trip. In such cases the forager, though probing a flower for nectar only, may be seen to carry pollen loads on her legs. Absence of pollen loads and pollen-collecting behaviour indicates that a bumblebee is collecting nectar alone. Pollen collection is dictated partly by the needs of the nest, but the time at which flowers release their pollen must be important too. The splitting open of the anthers depends critically on the weather. A species of flower that releases its pollen early in the morning on fine days may release it at quite a different time of day in bad weather. Some flowers, such as hollyhocks Alcea rosea, produce pollen that seems to be unacceptable to some bees. Further study of this is needed. In some circumstances it could be adaptive for a plant to produce pollen of this kind. Instead of grooming it away to their pollen baskets, bumblebees leave the pollen for some time, daubed on the body hair, where it may contact a stigma before they eventually groom it off and reject it, kicking it off the ends of their legs (fig. 24). The nutritional quality of different pollens varies, and this may be a significant, but as yet little studied, influence on foraging (Brian, 1951; Hanley & others, 2008). For example, B. terrestris can successfully use the pollen of the strawberry tree Arbutus unedo, whereas honeybees Apis cannot (Rasmont & others, 2005). Species differences in the suitability of certain pollens may also exist within the genus Bombus. We tend to think that bumblebees can use most different pollens, but that might be because most of the bumblebees we see are the common species, which are common because they have more catholic tastes. Many solitary bee species have specific pollen requirements, and this possibility remains under-explored as a factor influencing the decline of our native bumblebee species (see chapter 7). Particularly important pollen sources for most bumblebees include members of the Fabaceae (peas, vetches and clovers) and Lamiaceae (labiates), whose pollen seems notably rich in protein. A plausible, but as yet unexplored, reason for the overwhelming importance of the Fabaceae to bumblebees, throughout the world, may relate to the ability of these plants to fix atmospheric nitrogen (due to a symbiotic relationship with

Foraging behaviour | 57

mycorrhiza: (Greek: fungus roots) is a symbiotic (generally mutualistic but occasionally weakly pathogenic) association between a fungus and the roots of a vascular plant.

Fig. 25. Pollen grains from some bumblebee-visited plant species.

Rhizobium bacteria in their roots). As a result they should not be nitrogen limited, as many flowers are, and may be able to provide more nutritious pollen. One can speculate that other families of bumblebee plants may also benefit from nitrogen produced by the Fabaceae, via common mycorrhizal* associations, especially in permanent pastures, where long-standing root/fungus relationships might develop. The nutritional quality of the pollen of species that do not fix nitrogen might therefore be variable, depending on whether Fabaceae are also growing nearby. By examining a bee’s pollen loads it is possible to make an (incomplete) list of the flowers she has been visiting. When the pollen has a characteristic colour, like the blue of viper’s bugloss Echium vulgare or the orange of mullein Verbascum species, this can be done on sight, without disturbing the bee; the colours of various pollens, as seen in honeybee loads, are illustrated by Kirk (2006) and Hodges (1974). More usually the pollen grains must be examined under a compound microscope (technique, p. 101), when their elaborate sculptured shapes (fig. 25) make it possible to identify them with the aid of a key (Moore & Webb, 1978; Sawyer, 1981), or, more conveniently, by comparison with pollen samples collected from flowers known to be available to the bees. If one finds a nest it is possible (with practice) to identify the almost indestructible husks of the pollen grains that have been eaten and egested by the larvae. Just before pupation each larva empties its gut inside its newly spun pupal cocoon. Traces of this larval faecal material can be scraped from within the empty cocoons and the pollen grain husks can be identified (Brian, 1951; Yalden, 1982). If you can ‘read’ the history of the colony, recognising successive batches of cocoons from their relative positions, it may be possible to reconstruct the seasonal succession of pollen sources. Work of this type may prove extremely valuable, in that it can help to show which plants are important in maintaining local bumblebee populations, and which pollens, amongst those available, are of significance to different species of bee. Likewise, pollen present on bumblebee specimens in museum collections can provide very useful information on patterns of plant use in the past, when conditions for many species * Brundrett (1991; 2009) gives a comprehensive introduction to the mycorrhizal literature.

58 | Bumblebees

were much more favourable (Kleijn & Raemakers, 2008). Bumblebees, like most social bees, are generalists – nearly all British species visit a wide range of flower species. This is fortunate for us because it means that bumblebees readily forage in crops and gardens, even where the native flowers are suppressed as weeds and replaced by introduced species from all over the world. With the possible exception of the handsome B. monticola, which inhabits moorland where Vaccinium species (bilberry, cowberry, cranberry) grow, our bumblebees do not appear to depend on particular plant species. Do any of our plant species depend on bumblebees for pollination? We may speculate that among the flowers best adapted for pollination by bees of this shape and size are members of the mint family (Lamiaceae), particularly white dead-nettle Lamium album, and the figwort family (Scrophulariaceae), including the foxglove Digitalis purpurea, and flowers such as common toadflax Linaria vulgaris which require a bumblebee’s strength to open the lips of the flower, and the long tongue of B. hortorum to reach the nectar in the spur.

6.2 Studying foraging behaviour and pollination If we are to understand the causes and the potential consequences of the decline of bumblebees in Britain and how they can be remedied, we need more information on the importance of particular flowers for the various species of bumblebee, and on the importance of the bumblebee species as pollinators of both wild flowers and crops. We have seen that useful information on the various plant species used by bees as pollen sources can be gained from analysis of pollen loads and faecal material from nests. Unfortunately there is no corresponding way of unravelling the seasonal history of a colony’s nectar collection: there is no substitute here for direct observation. As we have outlined, evaluating the importance of different nectar sources for bees is not as straightforward as it seems. One problem is that the relative value of different flowers changes through the day, with weatherrelated changes in the concentration and volume of the nectar available in them. This means that a thorough

Plate 1

Plate 1 1.

Bombus lucorum ♀

2.

B. lucorum ♂

3.

B. terrestris ♀

4.

� B. terrestris ♀

5.

B. soroeensis ♀

6.

B. soroeensis ♂

7.

B. hortorum ♀

8.

B. jonellus ♀

9.

B. jonellus ♂

2

1

10. B. ruderatus ♀ (pale) 11. B. ruderatus ♀ (dark) 3

12. B. ruderatus ♀ (intermediate)

4

scale x1.15

5 6 7

8

9

12

10 11

Plate 2

Plate 2 1. Bombus lapidarius ♀ 2. B. lapidarius ♂ 3. B. ruderarius ♀ 4. B. ruderarius ♂ 5. B. pratorum ♀ 6. B. pratorum ♂

1 2

7. B. sylvarum ♀ 8. B. monticola ♀ scale x1.15

3 4

5 6

7

8

Plate 3

Plate 3 1. Bombus hypnorum ♀ 2. B. distinguendus ♀ 3. B. muscorum ♀ 4. B. pascuorum ♀ � 5. B. pascuorum ♀ 1

6. B. pascuorum ♀ 7. B. humilis ♀ 8. B. subterraneus ♀ 9. B. subterraneus ♂ scale x1.15

2

3 4

5

6

7

8 9

Plate 4

Plate 4 1. Bombus (Psithyrus) rupestris ♀ 2. B. (Ps.) rupestris ♂ 3. B. (Ps.) sylvestris ♀

1

4. B. (Ps.) sylvestris ♂ 5. B. (Ps.) bohemicus ♀ 6. B. (Ps.) bohemicus ♂

2

7. B. (Ps.) vestalis ♀ 8. B. (Ps.) vestalis ♂ 9. B. (Ps.) barbutellus ♀ 10. B. (Ps.) campestris ♀

3

11. B. (Ps.) campestris ♂ (dark) 12. B. (Ps.) campestris ♂ (light)

4

scale x1.15

5 7

6 8

9

10 11

12

Plate 5

Plate 5 Sting sheaths of female true bumblebees 1. B. lapidarius 2. B. cullumanus

1

3. B. ruderarius (similar to B. sylvarum)

2

4. B. pascuorum 5. B. muscorum 6. B. humilis 7. B. pratorum (similar to B. monticola) 8. B. lucorum (similar to B. magnus and B. terrestris)

3

9. B. soroeensis

4

10. B. jonellus (similar to B. hypnorum) 11. B. ruderatus (similar to B. hortorum) 12. B. distinguendus (similar to B. subterraneus)

5

6

7

8

11

9

10

12

Plate 6

Plate 6 Genital capsules of male true bumblebees 1. B. hortorum (similar to B. ruderatus) 2. B. ruderarius

3

1

3. B. sylvarum 4. B. pascuorum

2

5. B. humilis 6. B. muscorum 7. B. pomorum 8. B. cullumanus 9. B. jonellus (similar to B. hypnorum and 13)

6

4

10. B. distinguendus similar to B. subterraneus)

5

11. B. soroeensis 12. B. terrestris (similar to B. lucorum, B. magnus and probably also B. cryptarum)

7

9

13. B. pratorum (similar to B. monticola) 14. B. lapidarius

8

10

12

11

13

14

Plate 7

Plate 7 Callosities on the last (6th) ventral plate of female cuckoo bumblebees 1. B. (Ps.) rupestris 2. B. (Ps.) sylvestris 3. B. (Ps.) campestris 4. B. (Ps.) barbutellus 5. B. (Ps.) vestalis 6. B. (Ps.) bohemicus

1

2

3

4

5

6

Plate 8

Plate 8 Genital capsules of male cuckoo bumblebees 1. B. (Ps.) campestris 2. B. (Ps.) rupestris 3. B. (Ps.) sylvestris 4. B. (Ps.) bohemicus 5. B. (Ps.) barbutellus 6. B. (Ps.) vestalis

1

2

3

4

5

6

Foraging behaviour | 59

investigation of flower usage in a particular locality at a particular time should really span the whole foraging period of a day (or, better still, several days). Because bumblebees can forage at low ambient temperatures, this may mean starting early (perhaps as early as 4 a.m. in summer) and continuing late (perhaps after 10 p.m.). A dawn-to-dusk session of this kind can give useful results, particularly if combined with studies of nectar availability (p. 99) and microclimate (p. 103), but it is best done by a group of people, and needs careful planning beforehand. Counts are made in a standard way at regular intervals of, say, l½–2 hours through the day. They usually involve making a ‘bee walk’ along a pre-selected route, often a hedgerow or bank providing a defined strip of flowers perhaps 1–2 metres wide and up to about 100 metres long, and recording, for each bee seen visiting a flower, the species of flower, species, sex and caste (whether queen or worker) of bee, and whether the bee is taking nectar, pollen, or both. Alternatively, when bees are very abundant, a bee count might involve spot counts of the numbers of bees active at a given time within, say, five quadrats; or counts of the numbers entering (or leaving) a quadrat during a period of, say, 10 minutes. The interpretation of spot counts highlights the second problem in a study of this kind, in that a high count may mean one of two things. It may mean that the flowers have attracted a large number of bumblebees because nectar is abundant. Alternatively, it may mean that each bee is having to spend a long time in the patch because each flower contains very little nectar and a bee must visit many flowers to fill its honeystomach. To distinguish between these two possibilities one requires information on nectar availability and foraging rate. An obvious prerequisite for work of this kind is that every observer should be able to recognise the species of flowers and the species, sex and caste of the bumblebees, without taking specimens. This requires some work beforehand. The plates and keys in chapter 8 of this book should enable bumblebees to be named. It is sometimes useful to take into the field a named reference collection, or photographs, of the local bumblebee species. Flowers can be named with the aid of a book such as Stace (2010) or Streeter & others (2009). Studies of this kind are particularly valuable if they

60 | Bumblebees

relate to crop plants, or wild plants of species that might be encouraged on waste land or road verges; and if they relate to seasons (early spring and the ‘June gap’) when nectar can be in critically short supply. Even when observers cannot identify bumblebees reliably to species, as will usually be the case in studies involving the public, useful results can sometimes be obtained by dividing observations up by the colour pattern of the bumblebee (Fussell & Corbet, 1992a and table 7, p. 104. See the Quick-Check Key p. 78, for colour groupings). If particular species of bumblebee continue to decline, will crops and wild flowers suffer reduced pollination? As mentioned above, to explore the role of bees for particular plants is worthwhile, especially if those plants are species of importance to agriculture, horticulture or conservation. One way to see whether insect pollination influences seed set is to enclose flowers, from the bud stage onwards, in loose bags of petticoat netting or bridal veiling, of mesh fine enough to exclude bees but not fine enough to have serious effects on the flowers’ microclimate. Control flowers at the same stage of development are tagged at the same time and left exposed to pollinators. After a period of perhaps 3–10 days it should be possible to see whether or not seed is setting in the bagged and exposed flowers. The experiment should be left longer if the seeds are to be allowed to mature, and be counted and weighed. If bagging substantially reduces seed set, bees, or other insects too large to get through the mesh, may be significant pollinators. If not, the plant may need no pollination, or it may be self-pollinated or pollinated by smaller insects or by the wind. If bagged flowers set no seed, one way to see which visitors are effective pollinators is to bag a flower until it is fully open, then remove the bag and watch patiently until the flower has been visited once (or twice, or three times) by the suspected pollinator, then re-bag it at once and later score seed set. Another, more direct way to get information is to examine the stigma with a lens; it may be possible to see whether or not pollen is present on it. A bumblebee visit may result in the appearance of pollen on a hitherto pollen-free stigma. If the stigma is sampled at once, the presence of pollen can be checked microscopically (technique, p. 102); but if the flower is bagged and left for a few hours, it may be possible to

Foraging behaviour | 61

show that the pollen has germinated, forming a pollen tube (technique, p. 102), and was therefore viable pollen of the same species. Other approaches involve discovering the type of pollen being carried by flower-visiting bumblebees. Bees can be confined (technique, p. 96) and ‘swabbed’ gently with a cotton bud, which can then be rinsed in alcohol and the grains filtered out and examined (technique, p. 102). Another method involves the ‘Sellotape peel’ technique (p. 102) to map the position of pollen of different species on the bee’s body hair. This may help to identify which parts of a bee contact the stigma of each flower species that it visits. Many other aspects of foraging offer scope for further work. The possibility that pollen from certain plants is rejected by certain bumblebee species (p. 56), and the way bees select flowers to visit within a patch, both need much more investigation. If each flower is marked, mapped or collected after it is visited, it may be possible to see whether the bees are preferentially visiting, say, the larger flowers; or the lower flowers; or those at a particular stage of flowering; or those with or without other insects in them, such as thrips or beetle larvae. Do flowers of the selected type contain more nectar than the others? The possibility that individual bees have regular foraging routes, perhaps visited day after day, is also worth further investigation, using bumblebees that have been marked so that they can be recognised individually (technique, p. 96). In a study of this kind it is important for the observer to sit quietly to avoid disturbing the bees; it may take them a little while to return to their normal foraging behaviour after an observer’s arrival. Is there any evidence for positive or negative interactions between bumblebees of the same or different species? Brian (1957) has suggested that bumblebees are more likely to approach a foraging site if other bumblebees are already there (see also Leadbeater & Chittka, 2005; Kawaguchi & others, 2006). Brian also suggested that some bumblebees avoid other species: B. pascuorum, in particular, is said to be more likely to leave a foraging patch when other species arrive. Is there a hierarchy of species, or size of worker, each capable of displacing the one below it in the pecking order? These and many other questions are wide open for investigation.

7 Threats, conservation and commercial use “Only the individual can think, and thereby create new values for society” Albert Einstein, The world as I see it, 1949

7.1 Threats, agriculture and conservation It is a paradox that agriculture desperately needs bumblebees, but many bees face extinction through unsympathetic agricultural practices and land development. A third of the food we eat depends on bees as pollinators (Williams, 1995; Kearns & others, 1998; Dias & others, 1999), together with 60–90% of the world’s flowering plants and two-thirds of the world’s crop species (Kremen & others, 2007). Conservation of bumblebees is a necessity that we cannot afford to ignore. The answer seems straightforward: those in a position to influence events must make bee conservation compatible with land management. Government has to take the initiative, but there is much that we as individuals can do to make the issues more clearly and widely understood. Dramatic declines in bumblebee numbers, and in species at risk, have been documented both in Britain, since the 1950s (Free & Butler, 1959; Williams, 1982; Williams, 1995), and in Europe and beyond. A recent study of 11 European countries found at least 30% of bumblebee species to be threatened throughout their range, with 80% in danger in at least one country (Kosior & others, 2007; see also Berezin & others, 1996; Xie & others, 2008; Williams & Osborne, 2009; Williams & others, 2009; Cameron & others, 2011). Seven species out of the UK’s 25 (now 24) have become so rare that they are listed as priority species on the U K Biodiversity Action Plan; one (B. subterraneus) has since become extinct, and although there are plans for reintroduction (Gammans & others, 2009), little of the countryside is currently in a fit state to support it. Over the past 50 years we have lost 97% of hay meadows that support the types of flowers needed by bumblebees: less than 1500 hectares remain (see link

Threats, conservation and commercial use | 63

9.13.16, p 107). Fragments of suitable habitat persist, but they are probably too small to sustain viable populations of many bumblebee species in the long term. Currently, within the whole of lowland mainland Britain, only three extensive areas retain, for the moment, lower intensity agricultural environments and practices conducive to the survival of a broad range of bumblebee species. These are Salisbury Plain Military Ranges, Dungeness, and some areas of machair in the northwest of Scotland and the west of Ireland. Change in the environment is not always bad: we must remember that the landscape of Britain and much of continental Europe is essentially man-made or modified, but until quite recently it supported thriving bumblebee biodiversity, and it could do so again. Before the Second World War agricultural practices throughout Europe were, largely by chance, broadly sympathetic to flourishing bumblebee populations – much as they remain, for the moment, in parts of eastern Europe. Since then agriculture has intensified. Small scale unploughed hay meadows and clover leys, which supported the flowers on which bumblebees depend, have been replaced by larger, more monotonous, highly fertilized silage and cereal fields, where ‘weeds’ are suppressed with herbicides and intensive cutting regimes (see Carvell & others, 2006; Fitzpatrick & others, 2007). The profit motive has always been there; it is the methods of management that determine whether or not we finally encourage or eliminate our pollinator populations, the key natural resource for crop production and for the very existence of wild flower species. It is very hard for any of us now to imagine just how much richer the landscape of the UK was before 1945, in both abundance and diversity of bumblebees and flowers, and just what we have lost since. Some idea can be gained by visiting parts of former eastern bloc countries of Europe, notably Transylvania in Romania, whose agricultural practices were, until they joined the EU (some in 2004 and some in 2007), much like those of Britain a century ago (Akeroyd & Page, 2006; Jones, 2010). Do so soon: disastrously for bees, and farmland birds, the new accession states of Eastern Europe are adopting European Common Agricultural Policy (CAP) practices. It is likely that they will rapidly lose much of what they

64 | Bumblebees

have, as we have done, unless policies change very soon (see, for example, New challenges, new CAP, link 9.13.13, and 9.13.14, p. 107). Determining which factors are significant to bumblebees is difficult, but vital to their conservation. There are factors common to all, such as a continuous succession of flowers through the season; nesting and hibernation sites that will not be disturbed; and, without doubt, individual species requirements that must be met. Most British species were formerly found over much of the country, which suggests that the relict populations - of an increasingly large number of threatened species survive where they do because they can no longer live where they used to, rather than because these remaining habitats are necessarily ideal (for example, see Maps, p. 126; also Edwards & Williams, 2004; Goulson & others, 2006). Williams (1982; section 2.1, fig. 1) showed that compared to pre-1960 distributions there is a ‘central impoverished region’ in mainland Britain, associated with areas of intensive agriculture, where only six ‘mainland ubiquitous’ species are reasonably well represented (B. pascuorum, B. lucorum, B. hortorum, B. pratorum, B. terrestris and B. lapidarius – illustrated on the cover). For these species, areas of undisturbed land with rough grass and occasional willow trees (for spring forage), hedge bottoms and roadside verges, are of absolute importance to their survival. These habitat remnants need to be appreciated and sensibly managed. Road verges form an extensive ‘nature reserve’, under public management, with enormous potential to conserve and increase biodiversity, or, if poorly managed, to undermine it. Usually underappreciated, and viewed as just an additional cost in the council maintenance bill, there is great scope for verge cutting regimes compatible with maintaining and improving plant diversity, allowing flowering and seed set to occur. Verges should be the hay meadows of the future, alive with flowers. These can be simple, often cost-saving options, and they need urgent implementation before irreversible losses increase. Other bumblebee species have persisted in coastal, marshy and western areas, less conducive to intensive agriculture, and in brownfield (previously developed) sites that benefit from low levels of disturbance or soil enrichment. Kent and Essex marshes and adjacent brown-

Threats, conservation and commercial use | 65

field sites form one such species-rich complex that remains undervalued and threatened by development (see Benton, 2006; p. 519). While availability of suitable habitat is ultimately of over-riding importance, there continues to be a productive debate about other influences on the degree of rarity of the various species. Significant factors may include position of the particular species within its natural latitudinal range, along with interactions of climate, particular limiting types of food source, and the timing of emergence of queens in spring in relation to available forage (for example see Fussell & Corbet, 1992a; Osborne & Corbet, 1994; Goulson & others, 2005; Williams, 2005; Benton, 2006; Williams & others, 2007; Charman & others, 2009). An attempt to summarize significant common factors of importance in maintaining and encouraging bumblebees, in both farmed and seminatural areas, is given in fig. 26. Some useful bumblebee forage plants are listed in table 6 and table 7, p. 104). Table 6. Flowers favoured by bumblebees Native plants

Garden plants

Willows

Lungwort

Red and white clovers

Flowering currant

Vetches (many species)

Raspberries, blackberries, gooseberries

Birds-foot-trefoil (in particular)

Fruit tree blossom

Sycamore

Berberis

Blackthorn and hawthorn

Fuchsia magellanica

Woundworts

Penstemons

Ground-ivy

Beans

Red and white deadnettles

Most culinary herbs

Yellow archangel

Geranium species

Dandelions

Cotoneasters

Comfrey

Jerusalem sage

Yellow iris

Poppies (single)

Foxglove

Delphinium (single)

Knapweeds Yellow-rattle

Montbretia

Viper’s-bugloss

Columbine

Honeysuckle

Snowberry

Willowherbs

Borage

Cranebills

Buddlejas

Brambles

Lavenders

Thistles

Snapdragon (single)

Irises

66 | Bumblebees Fig. 26. Factors to be promoted, provided or considered in order to encourage bumblebee species and populations, in both farmland and semi-natural or conservation areas.

Semi-natural / Conservation Areas (Rarer species)

Farmland Areas (Commoner species) Minimize herbicide and pesticide use

Avoid ground disturbance to promote nesting, hibernation sites & perennial plants

Adequate seasonal floral succession Plant field margins e.g. agricultural legume mixes

Maintain extensive unimproved flower-rich grasslands (especially for ‘late-emergers’)

Promote large-scale restoration projects (>10 square km)

Suitable pollen types

Factors promoting bumblebee survival Incentives to increase traditional hay field flora & reduce dependence on silage Perennial and selected annual flowers Adequate size and distribution of habitat areas or ‘partial’ habitats (e.g. nesting and foraging)

Willows for ‘early emergers’

Perennial flowers

Manage cutting regimes

Manage grazing – winter/cattle/minimize sheep

Manage woodland edges, road verges, hedge bottoms (especially for ‘early emergers’)

Early emergers’ and ‘late emergers’ refer to species of bumblebee whose queens emerge from hibernation either relatively early or relatively late in spring. The significance of other factors, such as cutting regimes, winter grazing by cattle, minimizing sheep grazing and encouraging hay production, is mentioned in Section 7.1.

To encourage and maintain suitable foraging areas for bumblebees there needs to be an emphasis on haymaking and autumn grazing, which permits flowering and seed set to occur without interference. Cattle are often preferred to sheep as they graze more selectively and less intensively; they also cause some ‘poaching’ (disturbance) of the ground, producing small bare patches, within which seeds can germinate. So long as ploughing is infrequent, such conditions are conducive to biennial and perennial flowers, which are generally richer in nectar than annuals (see Corbet, 1995; Dramstad &

Threats, conservation and commercial use | 67

Fry, 1995). Traditional farming systems, where cattle are taken in at night and back out in the day (‘pendulation’ see Jones, 2010), tend to result in the removal of nutrients from grazed areas, which in turn promotes perennial flowers, rather than grasses. Over the past 30 years one of the most significant changes to the suitability of the environment has been the vast change from hay-making to silage production. Re-seeding with prolific grasses, which are then heavily fertilized and repeatedly cut, has effectively sterilized fields, removing feeding opportunities and nesting sites for pollinators and ground-nesting birds, in particular. Considerable areas of previously good bumblebee habitat have thus been rendered unusable. Such changes were promoted by the pattern of grants and incentives pertaining at the time. More holistic agri-environment schemes are urgently needed to help reverse such trends. B. subterraneus, which became extinct in the UK in about 1988, and has become increasingly rare across much of its European range, seems to have fairly demanding habitat requirements; at least when viewed in the context of much current landscape management, where small, enclosed areas are intensively managed in a wide variety of ways. Reintroduction may, nevertheless, be a worthwhile aim, even though achieving the required degree of environmental change will be hard to effect and sustain over a reasonable area and time scale. The potential for introduction of new bumblebee diseases will need to be adequately addressed. The attempt will prove most notable if it also helps focus emphasis and funding on improvements in habitat quality for the majority of existing species, rather than diverting attention away from them. Many crops and wild flowers throughout the world depend on bees for pollination, and ultimately, for wild flowers at least, for their survival (see Corbet & others, 1991; Kearns & others, 1998). Loss of this pollination service would inevitably result in decreased crop yields, and the gradual but progressive extinction of wild flowers dependent on bee pollination. In an intensively farmed landscape there comes a point where the islands of habitat suitable for nesting, foraging and hibernating become so small and widely scattered as to be unusable, and pollinator species are lost. To maintain

68 | Bumblebees

and encourage bumblebee species that can cope in such borderline and therefore demanding situations requires a degree of consistency over time, and involves management of habitats at a regional, rather than just at a farm, scale (Carvell, 2002; Corbet, 1995; Pywell & others, 2006). Thought needs to be given to the connections between suitable areas or ‘partial habitats’ (habitats suitable for nesting, foraging or hibernating; see Westrich, 1996), and to the needs of commoner species, not just the rare ones (Corbet, 2000). Agri-environment schemes can be beneficial, restoring factors important to bees, such as undisturbed areas for nesting, or planted field margins to improve the amount and seasonal succession of forage plants (Pywell & others, 2006). The finding that not all the schemes undertaken so far have proved successful or cost effective (Kleijn & others, 2001) serves to emphasise the importance of coordinating management over suitable time-scales, and beyond individual farms. This, in turn, requires agencies with the understanding and influence to effect the required changes. We have the beginnings of an understanding of the forage plants and habitat features of importance to bumblebees in agricultural and semi-natural landscapes (see fig. 26, and e.g. Fussell & Corbet, 1992a; Kells & Goulson, 2003; Edwards & Williams, 2004; Goulson & others, 2006; Williams & others, 2009; Darvill & others, 2010; Redpath & others, 2010) and we have a structure – developed through the International Pollinator Initiative (Dias & others, 1999) – to help direct the response of government and its agents. That response is needed now, and should recognise, foster and coordinate the experience of researchers and bodies that already exist, such as Hymettus (the Bumblebee Working Group) and the Bumblebee Conservation Trust (see links, 9.13.8 and 9, p 107).

7.2 Involvement of the public Undoubtedly the greatest ally of bumblebees is the public. Through sympathetic appreciation of their beauty and importance we are all, individually, in a position to significantly influence the future of bumblebees: directly, through our gardens and our influence on our local area; and indirectly, by educating others to their importance; by financial support of conservation bodies;

Threats, conservation and commercial use | 69

and by lobbying government and local authorities for the protection and good management of the environments on which they depend. Bumblebee conservation goes hand in hand with the protection of birds, something the British have already embraced. At a mundane level, in the lowlands of southern Britain areas that were once suitable for shrikes were also good for bumblebees. Likewise, farmland birds and bumblebees have been in decline for many of the same reasons (Wilson & others, 2009). A joint venture between the Royal Society for the Protection of Birds (RSPB) and bumblebee researchers aims to develop means to ensure the survival of the machair environments of north west Scotland and the west of Ireland, and, thereby, the nationally important bird and bumblebee communities that depend upon them (link 9.13.19, p. 107). Cooperative projects such as this could be very greatly extended. The Bumblebee Conservation Trust (BBCT) is instrumental in raising interest and support for bumblebee research and conservation, and has recently received financial support towards a cooperative project aimed at improving flower-rich grasslands around the Castlemartin Military Range in south Wales (see Rayner, 2010). Their efforts deserve encouragement. Members of the public have also been influential in studies of the distribution and habitat requirements of bumblebees. The original atlas of the distribution of bumblebees in Britain (ITE, 1980) was the culmination of a national recording scheme, and continuing studies of distribution, coordinated by BWARS (link 9.13.2, p. 107), welcome recordings made by the public. A number of research studies on forage plant usage and nest site selection (Fussell & Corbet, 1991, 1992a, c; and see table 7) have also successfully employed public surveys.

7.3 Pollination webs and plant populations Threats to bumblebees endanger the plant species that depend on them for pollination. In attempting to conserve bumblebee diversity we must be mindful of their role in the ecology of the species they pollinate. While in agricultural situations schemes that supplement forage plants for bees focus on the needs of the bees, in the wider countryside we must remember that it is the

70 | Bumblebees

native flowers and their unique genotypes that need bees and other pollinators for their survival. Careless use of planted forage (such as some wildflower seedmixes that include genotypes of non-British origin) may itself threaten these populations (Akeroyd, 1994, and chapter 9.8). We would do well to promote what is present, encouraging native plant populations with their unique genotypes, before introducing others that may themselves be a threat. Joined up thinking is required. One large seed bank of native species remains in the verge flora of our older roads, as mentioned above. While this remains we have a chance to encourage it by sensible cutting and management regimes. Once gone, it is gone for good.

7.4 Commercial use Bumblebees have been used commercially both in field crops and in glasshouses. Initially British species were introduced into the wild in New Zealand in the 1880s for clover pollination. They were incredibly effective in terms of increasing seed yield, although we do not know what effect they may have had on the native flora and pollinator fauna. Two of these species, B. ruderatus and B. terrestris, have recently been exported from New Zealand to Chile, where they, in turn, have become established (Arretz & Macfarlane, 1986), and B. ruderatus has now spread to Argentina (Abrahamovich & others, 2001). Since the mid 1980s bumblebees have become extremely important in commercial pollination, particularly of tomatoes, in glasshouse production. This mainly involves the use of B. terrestris and a North American species, B. impatiens. A million colonies were being sold annually by 2004, with a pollination value of several billion euros (Velthuis & van Doorn, 2006). This is set to increase further. It became possible through development of techniques that allow year-round production of colonies on a commercial scale - in particular the treatment of new queen bumblebees with carbon dioxide, which alters their behaviour, allowing them to start colonies without the need for hibernation (Röseler, 1985; see also techniques, p. 96). It has proved a lucrative trade, but there is urgent need for an immediate ban on the introduction of non-native species and subspecies (Goka, 1998; Kondo

Threats, conservation and commercial use | 71

& others, 2009), along with their parasites and diseases (Colla & others, 2006; Goka & others, 2001), which affect native bumblebee populations and further threaten their security. Diseases derived from imported bumblebees appear to have contributed to the recent precipitous decline of at least three North American bumblebee species (B. affinis, B. terricola and B. occidentalis) and the near extinction of another (B. franklini) (Thorp & others, 2002; Colla & Parker, 2008; Evans & others; 2008; Cameron & others, 2011: see link 9.13.11, p. 107). Some regulations exist to help prevent inadvertent release of imported bumblebees from greenhouse colonies, such as netting ventilators and freezing colonies that have ceased to be useful, but there is little scope to enforce such rules, and great potential for accidental release of fertilized queens. Escapees, as well as hibernating queens inadvertently imported in produce from abroad, may quickly result in interbreeding with native subspecies (see Ings & others, 2005), mating with other species (Kondo & others, 2009) or establishment in new areas (for example, Prŷs-Jones & others, 1981; Arretz & Macfarlane, 1986; Semmens & others, 1993; Abrahamovich & others, 2001). B. hypnorum has established itself in the UK since 2001 (Goulson & Williams, 2001) and in Iceland since 2008 (Erling Olafsson & Kristján Kristjánsson, personal communication): in both these cases a plausible route involves fertilized queens hibernating amongst the compost of imported pot plants. Due to the distances involved natural colonisation seems improbable in the case of the UK, and impossible in the case of Iceland. Within most countries there is therefore a pressing need for work on the domestication, management and production of commercial colonies using local genotypes, while thought needs to be given to ways of reducing risks associated with the unintentional importation of species. Within the UK, valuable work could be undertaken on the commercial use of our own subspecies of B. terrestris, reared in this country, as well as investigating the usefulness of B. hypnorum (now that it is established here, and seems at home in nest-boxes), and of longer-tongued, and potentially more versatile species, such as B. pascuorum. There is great scope here for the interested amateur, and those keen to avoid the

72 | Bumblebees

genetic homogenisation of bumblebee populations around the world. For purposes of commercial pollination, it should rapidly become an absolute requirement that only bumblebee species local to the area concerned be licensed, with a further requirement that their colonies be produced locally, to prevent the further introduction of non-native bees, and the spread of their diseases. Where bumblebees do not as yet exist, such as on the Australian mainland, there are many good reasons – including the great threat to native biodiversity - to be extremely wary of introducing them (see Hingston, 2007), if it is not already too late.

8 Identification ♂ male ♀ queen � worker ♀

Most areas of Britain harbour only a few bumblebee species and with practice these can be recognised on sight. In order to gain experience it is necessary to take a number of specimens through the Main Keys (pp. 79–93). These keys are not difficult to use, but if they look forbidding at first, start by making a guess at the name of your specimen using the colour plates and the Quick-Check Key (p. 78). Terms used in the main keys are explained in the glossary diagram (fig. 27). The left-hand side of the Quick-Check Key deals only with workers and queens of the common species of true bumblebees, and uses only their most obvious features. It is therefore unreliable, but it may help give you a sense of direction before you work through the main keys. It is relatively easy to separate the few species of British bumblebee that are really common and widespread, but separating them from the similar-looking less common species requires careful attention to detail, and critical identification is not always possible without catching a specimen. The right-hand side of the Quick-Check Key deals with bumblebee colour groups. For situations in which observers do not undertake critical identification to species, such as public surveys involving large numbers of children and amateurs, there is a significant risk of misidentification; the few individuals of rarer species that turn up may be recorded as common species, and vice versa. This risk can be reduced by asking for records relating to colour groups, rather than individual species. Caution must be exercised in the interpretation of such records, but given that each colour group will mainly include only one or two of Britain’s widespread and ubiquitous species, this procedure can still give useful results. It formed the basis of a public survey of flower usage by British bumblebees conducted in 1987 and 1988 (Fussell & Corbet, 1992a; see table 7). The Main Keys include all the species that have been recorded in Britain, and whenever possible they separate similarly coloured species on the basis of structural

74 | Bumblebees

thorax

abdomen

interalar band scutellum

collar

dorsal plates (tergites)

head flagellar segments

tail antenna(-ae)

ventral plates (sternites) hind metatarsus hind tibia

sting extended

mid metatarsus pollen basket (in true bumblebee females) sagitta(-ae) volsella(-ae) squama(-ae)

sting sheath

gonocoxa(-ae) sting of female genital capsule of male

Fig. 27. Glossary diagram. The body of a bumblebee is divided into three obvious parts: the head, thorax and abdomen. The abdomen bears two sets of overlapping plates. The upper or dorsal plates, otherwise known as tergites, are referred to in the keys by the letter T (6 plates are visible in the female, and 7 in the male, numbered from the thorax end, T1–6 and T1–7 respectively). The lower or ventral plates are also known as sternites. The last few segments of the abdomen are loosely referred to as the tail, the tip of which houses the sting in the female and the genital capsule in the male. For the purposes of the keys three parts of the upper or dorsal surface of the thorax are distinguished: the collar at the front, the scutellum at the back and, in between, a band extending from one wing base to the other, the interalar band. The thorax bears the legs. In female true bumblebees the hind tibia is framed by long corbicular hairs, which form the pollen basket or corbicula. The two antennae are divided into segments, which are numbered (as in Chart B, p. 77). The two mandibles work from side to side, and between them is the tongue or proboscis, which can be folded away underneath the head when the bee is not feeding.

Identification | 75

characters, which are less variable, and therefore more reliable, than hair colour. Throughout the keys colour features refer to the coat of hairs and not to the underlying cuticle (which is always black). Bear in mind that hair colour may fade during life. Many species produce dark-haired specimens. For British populations of some species these are too rare to be considered in the keys, but the dark forms are illustrated for those species in which they occur quite frequently. In the plates, the male or worker of a species has been illustrated only if its colour markings differ appreciably from those of the queen. Bracketed statements in the main keys (other than instructions and special information) apply to additional information that is less easy to appreciate (for example, features that are best examined with a microscope), less reliable characters, and characters whose states are inconsistent in the alternative lead and therefore not stated there. First, is your specimen a bumblebee at all? The only other insects likely to be confused with bumblebees are solitary bees of the genus Anthophora, bee hawkmoths, and several hairy flies, including bee flies, warble flies, bot flies, bristle flies such as Tachina grossa and Servillia species, and hoverflies, some of which are very effective mimics (p. 37). Male Anthophora have yellow areas of cuticle (as opposed to hair) on the head: in bumblebees the cuticle is wholly black. Female Anthophora with black body hair differ from bumblebees most obviously in having a dense covering of stiff, orange hair on the outer surface of the hind tibia. Flies have only one pair of wings, whereas bees have two, but beware - bees zip their fore wings to their hind wings; a row of hooks on the hind wing catches on a ridge on the fore wing. In fresh specimens a pin can be used to disengage the bee’s fore wing and hind wing, to prove that two pairs are present. Bees also differ from flies in the form of the antennae. Chart A shows how to recognise a bumblebee, and how to distinguish true bumblebees from cuckoo bumblebees. Chart B deals with the distinction between males and females. Note that hair colour varies geographically in some bumblebee species and for this reason features of coat colour used in the keys are not necessarily appropriate when identifying bumblebees from outside Britain and Eire.

76 | Bumblebees Chart A. Is the specimen a true bumblebee (Bombus), or a cuckoo bumblebee (Psithyrus), or neither? Hoverflies, and other hairy flies

Other bees

Cuckoo bumblebees

True bumblebees

Antennae Wings:

2 pairs

1 pair

one pair or 2? First submarginal cell divided

Yes a

No

from front to back by a pale narrow cross-bar? (thinner than veins in general)

Hairs on abdomen dense or sparse? Outer surface of hind tibia

dense female

male

Tips of male genital capsule dark and horny, or pale and pliable?

sparse

shiny, flat and hairless; framed with long hairs

dull, convex and hairy; not framed with long hairs

shiny, flattish, with few (unbranched) hairs

dull, convex, with many (branched) hairs b

dark

pale

Female mandibles square-ended or triangular-ended? square-ended a

Also present, but very faint, in Anthophora species Also applies to B. pomorum

b

triangular-ended

Identification | 77 Chart B. Is the bumblebee a male or a female (queen or worker)? Male Antenna: 12 or 13 segments?

Female

1

1

Abdominal segments visible from above: 6 or 7?

13

12

1 2 3 4 5 6

1 2 3 4 5 6 7

genital capsule:

sting:

side view

Tip of abdomen contains: a genital capsule or a sting (both shown extended)

top view

side view

top view

sting sheath

The Distribution Maps (pp. 119–130) can be used in conjunction with the keys in order to check a record for a species against what is already known. Some areas of the UK have not been well recorded; even in those that have, reliable records are still valuable as species distributions alter with time. Do consider contributing your records to your local biological records centre, or directly to the national recording scheme: for information on the Bees, Wasps and Ants Recording Society see the link in section 9.13.2, p. 107. Change in the distribution of a species over time can be explored using mapping facilities and records at NBN (link 9.13.7). We have used pre and post 1990 records from BWARS for maps presented here.

78 | Bumblebees

Quick-Check Key to queens and workers of the commonest species of true bumblebees. Before using this key, use Chart A and Chart B to ensure that your specimen is the queen or worker of a true bumblebee. Beware: this key cannot be used for critical identification.

Number of yellow stripes

B. lapidarius

B. hypnorum*

B. pascuorum

1

0

0

0

1

0 or 1

0

0

0

Two-banded white tail

Banded red tail

Black-bodied red tail

White-tailed brown

Brown

B. soroeensis, B. (Ps.) bohemicus, B. (Ps.) vestalis

B. monticola, B. sylvarum, some island forms of B. jonellus, males of B. lapidarius and B. (Ps.) rupestris

B. ruderarius, B. (Ps.) rupestris

None

B. muscorum, B. humilis, B. distinguendus

Other species in the colour group

Tail black or ginger

B. pratorum

1

Three-banded white tail

Colour group

Tail white

B. terrestris & B. lucorum complex

1

Thorax Abdomen

Thorax all black

Scutellum black and face short

2

Tail red or orange

These species are illustrated on the cover: clockwise from the top – B. hortorum, B. terrestris, B. hypnorum, B. lucorum, B. lapidarius, B. pascuorum, B. pratorum.

Thorax mainly brown or ginger

Thorax other colours

tail

collar scutellum

thorax

Thorax with some yellow or brown

B. hortorum

B. ruderatus, B. jonellus, B. distinguendus, B. subterraneus, males of the B. lucorum complex, B. (Ps.) sylvestris, B. (Ps.) bohemicus, B. (Ps.) barbutellus, B. (Ps.) campestris and B. (Ps.) vestalis abdomen

Tail white, buff or brown

Scutellum yellow face long

} }

* B. hypnorum is not yet very common, but it seems destined to become so fairly soon. It was first recorded in Hampshire, in southern England, in 2001 and has since spread as far north as the Midlands. It frequents urban and garden sites, which will make it relatively conspicuous.

Key I. Female true bumblebees | 79

I Female true bumblebees Bombus Many structural characters are seen most easily in large females: in particular, sting sheath characters are most obvious and well developed in queens. To help confirm your identification after using this key compare the sting sheath with the illustrations in pl. 5. To view the sting sheath, extend the sting with forceps, turn the shaft upwards so that the tip points towards the bumblebee’s head, then view the sting base from the bee’s rear, endon, using a binocular microscope (see Chart B, p. 77). The sting must be extended when the bee is still soft, soon after death; if the specimen has become rigid, relax it first (technique, p. 95). Terms used below are explained in the glossary diagram (fig. 27, p. 74). Distribution maps are at the end of the book.

I.1

I.2

I.3

1 Thorax all black (at most a few pale hairs at front and rear) 2 – Thorax not all black

7

2 Tail (T4–T6) red or orange red

3

–

Tail black

B. ruderatus (pl. 1.11) and B. hortorum (dark forms)

Dark forms are reasonably frequent in B. ruderatus and rare in B. hortorum, and the two species cannot always be distinguished reliably. In B. ruderatus the coat is shorter and more even, and the very long face is relatively broad; queens are relatively large and sculpturing on T6 is very marked (I.1). In B. hortorum the coat is longer and less even, and the very long face is relatively narrow; queens are somewhat smaller and sculpturing on T6 is less marked (I.2). The status of B. ruderatus is unclear due to difficulties distinguishing it reliably from B. hortorum; certainly uncommon and local, it declined after 1945, but it may be expanding its range at present (Stuart Roberts, personal communication). B. hortorum is common, widely distributed, and frequently found in gardens. Maps pp. 119 and 120.

3 Face short, about as long as wide (as I.3)

4

– Face long, about 1½ times as long as wide (as I.4); (hairs framing the pollen basket black, sometimes tipped by yellowish-red; T1 and T2 black; black on T3 merging into red of tail) B. pomorum (similar to B. lapidarius, pl. 2.1)

I.4

Not found in Britain since 1864. All specimens were from Deal on the Kent coast. Almost certainly a rare immigrant from the continent.

80 | Bumblebees

4 Pollen basket framed by black hairs; no spine at tip of mid metatarsus (I.5) 5 – Pollen basket framed by red hairs; mid metatarsus with spine at tip (I.6) 6

I.5 no spine

spine

I.6

5 Outer surface of hind metatarsus densely covered with short yellowish-white hair (I.7; use a dissecting microscope); (inner margins of sting sheath simple (pl. 5.1)) B. lapidarius (pl. 2.1) Common and widely distributed in Britain. Predominantly a lowland species; often found in gardens. Has spread north into Scotland over the past 35 years, probably in relation to changes in climate (Macdonald, 2001). Map p. 120.

– Outer surface of hind metatarsus shiny black with few hairs (I.8); (inner margins of sting sheath inwardly swollen (pl. 5.2)) B. cullumanus (similar to B. lapidarius, pl. 2.1) Once associated with chalkland habitats in south-eastern England. Unrecorded since ‘about 1941’ (specimen in the British Museum Natural History (Williams, P.H. website, for address see p. 107)). Almost certainly extinct in Britain, and severely endangered throughout its range.

6 Surface of 5th and 6th dorsal plates dull between the pits; (inner margins of sting sheath very broad, almost meeting in midline (pl. 5.3)) B. ruderarius (pl. 2.3)

I.7

Once fairly common, particularly in south and east England, but appears to be declining markedly throughout much of its range. Mainly a lowland species; sometimes found in gardens. Map p. 121.

– Surface of 5th and 6th dorsal plates shiny between the pits; (sting sheath as B. ruderarius (pl. 5.3)) B. sylvarum (dark form)

I.8

Only males of this continental colour form have been recorded in Britain, at Newhaven on the East Sussex coast.

7 Thorax buff, yellowish-brown or ginger all over, sometimes with black hairs mixed in, but without black areas (mid metatarsus with spine, as I.6) 8 – Thorax with black areas

10

Key I. Female true bumblebees | 81

8 At least some black hairs on upperside of abdomen. 19 – No black hairs on upperside of abdomen, except on T6*

I.9

I.10

9

9 No black hairs on upperside of thorax; T2 without a distinct brown band; (coat longer, more dense and of very even length, giving a velvety, crew-cut appearance; abdomen seen from above more rectangular (pl. 3.3); projection from inner margin of sting sheath undivided (pl. 5.5); hairs at sides of T3 arise from little bumps (I.9: use a microscope)) B. muscorum (pl. 3.3) Local and uncommon throughout the British Isles, especially in the south. Most frequent in damp or marshy inland and coastal sites, such as moorland, fens, salt marshes. Has declined throughout much of its range in Britain in recent years. Map p. 122.

– Usually a few black hairs on upperside of thorax, especially above the wing bases (use a lens); T2 with very characteristic brown band (pl. 3.7); (coat shorter, less dense and of less even length; abdomen seen from above more triangular (pl. 3.7); projection from inner margin of sting sheath divided at its tip (pl. 5.6); hairs at sides of T3 arise from pits (I.10)) B. humilis (pl. 3.7) Local and uncommon. Mainly southern England particularly coastal and chalkland areas. Has become much less common in central England. Map p. 123. Sometimes it may be difficult to distinguish whether a specimen is B. muscorum or B. humilis.

10 Scutellum having black hair only

11

– Scutellum with at least some yellowish or brownish hairs 14

* B. muscorum subspecies allenelus, found on the Aran Islands, Eire, and some individuals of B. muscorum subspecies liepeterseni, from the Outer Hebridean Islands, have black hair on T1 and the sides of T2. B. pascuorum occasionally lacks black hairs on the abdomen (Benton, 2006).

82 | Bumblebees

11 Tail orange-red (small species; yellow on 2nd dorsal plate (T2) may be interrupted in the middle and is often absent in workers; inner margins of sting sheath simple (pl. 5.7)) B. pratorum (pl. 2.5)

I.11

oblique groove notch

Common throughout Britain. Frequently found in gardens. Map p. 123.

– Tail white, buff or brown

I.12

12

12 Mandibles with notch and oblique groove, as in I.11 (use a dissecting microscope); (inner margin of sting sheath notched (as pl. 5.8)) 13 – Mandibles without notch or oblique groove (I.12); (small species; inner margin of sting sheath without notches (pl. 5.9)) B. soroeensis (pl. 1.5) Very local and often uncommon. Found mainly in the north and west of Britain. Absent from most of east and central England. Not recorded from Ireland. In southern England its range has become much more localised. Probably greatly under-recorded where it occurs, due to confusion with B. lucorum complex. Map p. 124.

13 Yellow stripes bright (lemon or creamy yellow); tail of queens and workers white or pinkish, without any trace of brownish at the junction of black and white on T4 B. lucorum (pl 1.1), B. magnus & B. cryptarum (referred to as the B. lucorum complex) Queens can sometimes be difficult to separate with confidence; at present workers cannot be distinguished reliably on structural characters alone. Until more information is available these three species are mapped together as the B. lucorum complex, p. 124.

I.13

I.14

I.15

B. lucorum has a narrower yellow collar, ending level with, or just below the wing bases (I.13); the yellow of T2 is narrower, and queens are often small relative to B. magnus, and perhaps slightly smaller than B. cryptarum. Common throughout Britain and frequently found in gardens. B. cryptarum was first distinguished in Britain as a distinct species by Bertsch (and others, 2004; 2005). Bertsch also recognised it among Welsh specimens of ‘B. lucorum’ during a visit to OP-J in 1993. The collar extends slightly lower than in B. lucorum, and, on each side, at the level of the wing bases, the yellow of the collar is crossed by a characteristic S-shaped line of dark hairs (I.14). Probably common and widely distributed, at least in the north and west of the British Isles (Macdonald, 2006; Macdonald & Nesbit, 2006; Murray & others; 2008; Waters & others, 2010), its overall distribution remains to be determined. B. magnus has a wider yellow collar, extending well below the level of the wing bases (I.15); the yellow band on T2 is wider;

Key I. Female true bumblebees | 83 and queens are often larger than those of the two other species. B. magnus is distributed in the north and west of Britain in exposed habitats, such as heath and moorland. It emerges later in spring than the other two species, and may be adapted to a shorter season, or differing climatic conditions.

– Yellowish stripes dark (brownish-yellow); queen tail buff or brownish, worker tail buff, or white with a thin brownish band next to the black of T4 B. terrestris (pl. 1.3 and 1.4)

I.16

Common, particularly in the south of Britain. Local in Scotland, where it appears to be spreading northwards, perhaps in relation to changes in climate (Macdonald, 2001). Frequently found in gardens. Map p. 125. Using the characters given above most females can be identified as B. lucorum complex or B. terrestris, but some workers are very difficult to distinguish, even for experts.

I.17

14 Tail red or orange*

15

– Tail white, brownish-yellow or greyish

16

15 Second and 3rd dorsal plates (T2–T3) reddish-orange, (T1 black or yellow; T2–T6 reddish-orange, inclining to yellow at sides of T4–T5; abdominal hair long but even; yellow on rear of thorax C-shaped, being of similar width in the middle and at the sides; inner margins of sting sheath simple (as pl. 5.7)) B. monticola (pl. 2.8)

I.20

Found on upland moors in the north and west of mainland Britain, particularly where Vaccinium species grow. Probably benefits from moorland-edge plants, particularly in spring, such as willows and bird’s-foot-trefoil Lotus corniculatus (traditionally found in unimproved hill pastures). Recently recorded from the north and east of Ireland. Declining, almost certainly due to overgrazing by sheep, lack of moorland management, and re-seeding and fertilizer use on upland pastures. Map p. 125.

– Second dorsal plate (T2) largely pale greenish-grey, with traces of black, at least on the sides; T3 black, edged with pale greenish hairs; (T1 pale greenish-grey; T4–T6 orange, edged with narrow band of greenish-white; coat rather thin and uneven; yellow on rear of thorax wider in the middle than at the sides; inner margins of sting sheath as in pl. 5.3) B. sylvarum (pl. 2.7) A lowland species. Has declined massively and is in danger of extinction. Persists, in small numbers, in a few undisturbed floristically rich sites. Map p. 121. * On the islands of the Outer Hebrides and Shetland B. jonellus (couplet 16) occurs in an orange-tailed form, rather than the usual white-tailed (mainland) form (pl. 1.8).

84 | Bumblebees

I.18 oblique groove

I.21

16 Face long or medium length as in I.16 or I.17; mandibles with oblique groove but without notch (I.18; use a dissecting microscope); mid metatarsus with spine (as I.19; use a dissecting microscope or strong lens) 17 – Face short as in I.20; mandibles without oblique groove but with notch (I.21); mid metatarsus without spine (as I.22). (Small species; inner margins of sting sheath simple (pl. 5.10)) B. jonellus (pl. 1.8) Widely distributed throughout much of Britain, especially in areas of heath and moorland. Reasonably common, particularly in the north, but quite local in the south. Map p. 127.

notch

17 Sixth ventral plate without keel; (face long (as I.16); inner margins of sting sheath with characteristic notches (as pl. 5.11)) B. hortorum (pl. 1.7) and B. ruderatus (pl. 1.10 and 1.12) spine

In some cases, particularly small workers, it may not be possible to distinguish whether a specimen is B. hortorum or B. ruderatus. Typically, in B. hortorum the yellow of thorax and abdomen is bright; the rear thoracic band is C-shaped (that is, of about equal width along its length) and noticeably narrower in the middle than the collar (pl. 1.7); sculpturing on the 6th dorsal plate (T6) is shallow (I.23); the coat is long and uneven; and T1 and the base of T2 are usually yellow. Queens are normally smaller than those of B. ruderatus.

I.19

I.22 no spine

In B. ruderatus, workers are usually undarkened or completely black (the latter are dealt with in couplet 2, p. 79), whereas in queens intermediate forms, showing variable degrees of darkening, are also frequent (pl. 1.12). In undarkened specimens the yellow of the thorax and abdomen is golden or brownish-yellow, the rear thoracic band is wider in the middle than at the sides and, at its middle, about as wide as the yellow collar (pl. 1.10); sculpturing on T6 (which is best developed in queens) is deep (I.24); the coat is short and even; and on the abdomen yellow is usually restricted to T1 and often replaced in the middle by black. Queens are usually larger than those of B. hortorum. In darkened specimens of both B. hortorum and B. ruderatus the tail may be grey or brownish (as in pl. 1.12), the relative width of the thoracic bands is variable, and one or both of the bands may be absent. For information on distributions and abundance see couplet 2, p. 79. Maps p. 119 and 120.

– Sixth ventral plate with pronounced keel, or ridge, along the middle, as in I.25; (face medium length (as 1.17); inner margins of sting sheath without notches (as pl. 5.12)) 18

Key II. Male true bumblebees | 85

18 Abdomen brownish-yellow all over; (thorax brownishyellow with a black or dark grey band between the wings; coat long) B. distinguendus (pl. 3.2)

I.23

Rare and very local. Probably restricted to the Hebrides, Orkney and the north of Scotland, but formerly present locally throughout much of the UK and Ireland. Map p. 126.

– First three dorsal plates (T1–T3) of abdomen black, often with a narrow fringe of brownish or pale hairs at the rear edge of each; T4–T5 whitish; (coat very short, especially on front segments of abdomen) B. subterraneus (pl. 3.8)

I.24

Last seen at Dungeness in 1988 and probably extinct. Previously restricted to the south of England. Reintroduction plans are in progress, using queens from Sweden (see www.bumblebeereintroduction.org). Map p. 126.

19 Tail white; T1-T3 covered in black hair (inner margins of sting sheath simple (pl. 5.10)) B. hypnorum (pl. 3.1)

I.25

First recorded near the New Forest, Hampshire, in July 2001, it is spreading rapidly throughout mainland Britain. Particularly associated with gardens, it often uses bird nest-boxes. Map p. 127.

– Tail brownish, occasionally black, but not white; hairs of T1-T4 partly or largely pale buff or brownish, but with some black hairs present, at least on the sides; (inner margin of sting sheath has small projection (pl.5.4)) B. pascuorum (pl. 3.4, 3.5 and 3.6) Common throughout Britain. Frequently found in gardens. Map p. 122.

II Male true bumblebees Bombus To help confirm your identification after using this key, compare the genital capsule with the illustrations in pl. 6. The genital capsule should be extended gently with forceps while the bee is still soft, soon after death. If the bee has become rigid, relax it first (technique, p. 95). If the genitalia do not correspond with those of the species name reached via the key, then reject the key identification in favour of the best match in pl. 6 and go through the key again to see what went wrong. For information on species distributions and abundance see relevant couplets (referred to below) in key I, and the maps at the end of the book. The terms used below are explained in the glossary diagram (fig. 27, p. 74).

86 | Bumblebees

1 Thorax entirely black, or at most with scattered brownish-yellow hairs at front and rear

2

– Thorax with yellow, whitish, buff or ginger areas, at least at front

4

2 Face short, about as long as wide (as II.1)

3

– Face long, about l¼ times as long as wide (as II.2); (genital capsule as pl. 6.1) B. hortorum and B. ruderatus

II.1

(dark forms; similar to ♀, pl. 1.11) See couplet 2, p. 79, for distribution and abundance. Maps pp. 119–120. Dark forms occur frequently in B. ruderatus and infrequently in B. hortorum, and the species are not easy to distinguish: colour of the beard on the mandibles may be useful (black in B. hortorum, reddish in B. ruderatus), as well as the length and position of hairs on the rear edge of the hind tibia (continued round the end, and longer at the base in B. hortorum (II.3), stopping short of the end, and shorter at the base in B. ruderatus (II.4)).

3 Third antennal segment much longer than 4th (II.5); (genital capsule pl. 6.2) B. ruderarius (pl. 2.4)

II.2

See couplet 6, p. 80, for information on distribution and abundance. Map p. 121.

– Third antennal segment only a little longer than 4th (II. 6); (genital capsule pl. 6.3) B. sylvarum (dark form; similar to pl. 2.4)

II.3

Males of this continental form have been recorded, on one occasion, from Newhaven, East Sussex.

4 Thorax buff, brown or ginger all over, sometimes with black hairs mixed in, but without black areas 5 – Thorax with black areas

II.4

5 Tail white; top of thorax uniformly buff, yellowish brown or ginger, without a black band (but may have a few black hairs intermixed); genitalia similar to B. jonellus (pl. 6.9) B. hypnorum (similar to ♀, pl. 3.1)

3 4

Dark individuals occur occasionally. See couplet 19, p. 85,for information on distribution and abundance. Map p. 127.

–

II.5

8

II.6

Tail not white

6

Key II. Male true bumblebees | 87

II.7

6 Upperside of abdomen with at least some black hairs among the brownish ones (use a lens); mid antennal segments swollen underneath, more so at their apical ends (II.7); (genital capsule pl. 6.4)

B. pascuorum (similar to ♀ and♀ � , pl. 3.4, 3.5 and 3.6) Common throughout Britain. Map p. 122. Frequently found in gardens.

II.8

– Upperside of abdomen without any black hairs among the brownish ones;* mid antennal segments less swollen underneath, swellings symmetrical (II.8) 7 7 Thorax with at least a few black hairs, especially above the wing bases (use a lens); (brownish band on T2; coat shorter, uneven and less dense; genital capsule pl. 6.5)

B. humilis (similar to ♀ pl. 3.7) See couplet 9, p. 81, for information on distribution and abundance. Map p. 123.

– Thorax without any black hairs; (coat longer, more even; genital capsule pl. 6.6)

II.9

B. muscorum (similar to ♀, pl. 3.3) See couplet 9, p. 81, for information on distribution and abundance. Map p. 122.

8 Face long, about l¼ times as long as wide (as II.9)

II.10

9

– Face short, about as long as wide (as II. 10)

11

9 Tail white or grey-brown; (genital capsule pl. 6.1)

10

– Tail red; (mandibles without a beard; genital capsule pl. 6.7) B. pomorum (similar to B. (Ps.) rupestris ♂, pl. 4.2) For information on previous records see couplet 3, p. 79.

10 Yellow front and rear bands on thorax broad, sharply separated from black, and equally broad in midline; coat shorter and more even; (mandibles with reddish beard; hairs on rear edge of hind tibia stopping short of its tip, and shorter at its base (II.11) than in B. hortorum (II.12))

II.11

B. ruderatus (similar to ♀, pl. 1.10 and 1.12) Banded and dark forms frequent. Dark forms sometimes with grey-brown tail. Intermediate forms rare. See couplet 2, p. 79, for information on distribution and abundance. Map p. 119.

II.12

* B. muscorum subspecies allenelus, which is found on the Aran Islands, Eire, and some individuals of B. muscorum subspecies liepeterseni, from the Outer Hebridean Islands, have black hair on T1 and T2.

88 | Bumblebees

– Yellow front and rear bands on thorax often narrow, less sharply separated from black; rear band narrower than the front one; coat longer and less even; (mandibles with blackish beard; hairs on rear edge of hind tibia continuing round its tip, and longer at its base (II.12) than in B. ruderatus (II.11))

B. hortorum (similar to ♀, pl. 1.7) Banded forms common, dark forms (see couplet 2, p. 79) infrequent. Intermediate forms occur, with variable degrees of darkening of yellow bands and white tail. See couplet 2, p. 79, for information on distribution and abundance. Map p. 120 In some cases it may not be possible to distinguish whether a specimen is B. hortorum or B. ruderatus.

11 Front and rear bands on thorax of roughly equal width in midline 12 – Rear band absent or much narrower in the middle than the front band 18 12 Tail reddish or orange, at least on T6

13

– Tail white or a greenish- or yellowish-brown

16

13 Genital capsule (refer to fig. 27 for explanation of terms): tips of volsellae sharply pointed, tips of sagittae pointing slightly outwards (as pl. 6.2 and 6.3) 14 – Tips of volsellae blunt, tips of sagittae pointing inwards and hook-like (pl. 6.8 and 6.9) 15 14 Third antennal segment much longer than 4th (II.13); (genital capsule pl. 6.2) B. ruderarius (similar to ♂ B. lapidarius (pl. 2.2) but with less yellow hair) 3 4

See couplet 6, p. 80, for information on distribution and abundance. Map p. 121.

– Third antennal segment only a little longer than 4th (II.14); (genital capsule pl. 6.3)

B. sylvarum (similar to ♀, pl. 2.7) See couplet 15, p. 83, for information on distribution and abundance. Map p. 121.

II.13

II.14

15 Volsellae extending well beyond tips of sagittae (pl. 6.8); tips of volsellae elongated, and flattened on their inner surfaces (pl. 6.8); (flagellar segments of antennae somewhat curved in profile)

B. cullumanus (similar to ♂ B. pratorum, pl. 2.6) For information on previous records see couplet 5, p. 80.

Key II. Male true bumblebees | 89

– Volsellae extending only just beyond tips of sagittae (pl. 6.9); tips of volsellae not elongated or flattened on their inner surfaces (pl. 6.9); (flagellar segments almost straight-sided) B. jonellus (as pl. 1.9, but with an orange tail) This colour form of B. jonellus occurs on the Outer Hebridean Islands and on the Shetlands.

16 Abdomen with at least some black hairs, which may be restricted to the sides of T2 17 – Abdomen without any black hairs; (thoracic bands and abdomen brownish-yellow; facial hairs mainly pale; genital capsule pl. 6.10)

B. distinguendus (similar to ♀, pl. 3.2) See couplet 18, p. 85, for information on distribution and abundance. Map p. 126

17 Abdomen largely greenish-yellow; (hair fringe on hind tibia at most equal to tibial width; body hair short; abdomen elongate; facial hairs mainly dark; genital capsule as pl. 6.10) B. subterraneus (pl. 3.9) See couplet 18, p. 85, for information on distribution and abundance. Map p. 126.

II.15

II.17

– T3 and T4 black, tail white or yellowish; (hair fringe on hind tibia much longer than tibial width; body hair long; abdomen rounded; facial hairs mainly pale; thoracic bands, T1 and base of T2 yellow; genital capsule pl. 6.9) B. jonellus (pl. 1.9) See couplet 16, p. 84, for information on distribution and abundance. Map p. 127.

3

18 Tail white or buff-brown

19

– Tail orange or reddish

21

19 Hind metatarsus with slender base and long hair fringe on rear edge (II.15); 3rd antennal segment shorter than 5th (II.16); (hair long; genital capsule pl. 6.11) B. soroeensis (pl. 1.6)

5

See couplet 12, p. 82, for information on distribution and abundance. Map p. 124.

– Hind metatarsus with broad base and short hair fringe on rear edge (II.17); 3rd and 5th antennal segments of about equal length (as II.18); (hair short; genital capsule as pl. 6.12) 20

II.16

II.18

90 | Bumblebees

20 Tail pure white; yellow of thorax and T2 pale lemon or creamy yellow; coat long and uneven; facial hair yellow black or a mixture of the two; yellow hair may be present on the rear of the thorax B. lucorum (pl. 1.2), B. cryptarum and B. magnus (referred to collectively as the B. lucorum complex) So far males of these three sister species can be distinguished by the chemical components of their labial gland secretions, but not morphologically; characters that will allow us to separate them need to be identified. For information on distribution and abundance see couplet 13, p. 82. Map p. 124.

– Tail off-white or buff-brown; yellow of thorax and T2 dark golden or brownish-yellow; coat short and even; facial hair black; no yellow hair on the rear of the thorax

B. terrestris (similar to ♀, pl. 1.3) For information on distribution and abundance, see couplet 13 p. 83. Map p. 125. Using the characters given above most males can be distinguished as B. lucorum complex or B. terrestris, but some specimens are very difficult to identify, even for experts.

21 Hair on 2nd and 3rd dorsal plates (T2 and T3) black or yellow 22 – Hair on T2 and T3 red; (genital capsule as pl. 6.13)

B. monticola (similar to ♀, pl. 2.8)

For information on distribution and abundance see couplet 15, p. 83. Map p. 125.

22 Genital capsule (refer to fig. 27 for explanation of terms) with hooked end to sagitta (pl. 6.13); (hair long and uneven; hair on T1 and T2 yellow (may be reduced or rarely absent); T3 black; T4 black or orange; remainder of abdomen orange) B. pratorum (pl. 2.6) Common throughout Britain. Frequently found in gardens. Map p. 123.

– Genital capsule with spiky end to sagitta (pl. 6.14); (hair short and even; hair on T1 black (at most a trace of yellow); T2 and T3 black; remainder of abdomen red) B. lapidarius (pl. 2.2) For information on distribution and abundance see couplet 5, p. 80. Map p. 120.

Key III. Female cuckoo bumblebees | 91

III Female cuckoo bumblebees Bombus (subgenus Psithyrus) In all cases confirm your identification by comparing the two bulges or ridges (callosities) on the last (6th) ventral plate with the illustrations in pl. 7. The terms used below are explained in the glossary diagram (fig. 27, p. 74). 1 Tail (T4–T6) entirely red; thorax and rest of abdomen black; bulges (callosities) on 6th ventral plate very large, fin-like (pl. 7.1) and visible from above; wings dark smoky-brown B. (Ps.) rupestris (pl. 4.1) The preferred host is B. lapidarius. Uncommon. Declined last century but seems to be increasing again, perhaps in response to climatic changes. Mainly found in the southern part of England and Wales. Not recorded from Scotland. According to Løken (1984) it may occasionally usurp nests of B. sylvarum and B. pascuorum. Map p. 128.

– Tail not entirely red; callosities on 6th ventral plate smaller (pl. 7.2–6) than those of B. (Ps.) rupestris (pl. 7.1) 2 2 Callosities on 6th ventral plate very small and inconspicuous (pl. 7.2); (tip of abdomen strongly curved under; T3–T4 mainly white or yellowish-white) B. (Ps.) sylvestris (pl. 4.3) Takes over nests of B. pratorum, and sometimes B. jonellus. Widely distributed, and locally not uncommon. Not thought to occur in Ireland. Map p. 128.

– Callosities on 6th ventral plate large and conspicuous (pl. 7.3–6) 3 3 Callosities form a shallow U-shape (pl. 7.4); (last dorsal plate (T6) dull and closely pitted, with a distinct keel along the middle (III.l)) B. (Ps.) barbutellus (pl. 4.9)

III.1

Takes over nests of B. hortorum. Not uncommon; distributed throughout Britain. Probably more widespread than its map distribution (see p. 129) suggests. Possibly an occasional parasite of B. hypnorum (Pouvreau (1973)). Map p. 129.

– Callosities form a V-shape (pl. 7.3, 7.5 and 7.6)

III.2

4

92 | Bumblebees

4 Tail without white hairs; (T6 as III.2; 6th ventral plate pl. 7.3) B. (Ps.) campestris (pl. 4.10) Takes over nests of B. pascuorum, and possibly B. humilis. Darkened specimens are common; completely black specimens are rare. Local, but not uncommon; widely distributed. Map p. 129.

III.3

III.4

– Tail with white hairs

5

5 Callosities end short of the tip of 6th ventral plate (pl. 7.5); T6 not shiny (III. 3); (yellow of thorax and abdomen darker; larger bee with shorter coat; T3 lemon-yellow at the sides; scutellum (rear part of the thorax) black) B. (Ps.) vestalis (pl. 4.7) Takes over nests of B. terrestris. Common in the southern half of Britain. Has now been recorded in Ireland and Scotland. It appears to be spreading northwards, along with its host, probably in response to changes in climate. Map p. 130.

– Callosities end near tip of 6th ventral plate (pl. 7.6); T6 shiny (III.4); (yellow of thorax and abdomen paler (and rapidly fading); usually smaller bee, with longer shaggy coat; T3 pale yellow at the sides; scutellum often has pale hairs) B. (Ps.) bohemicus (pl. 4.5) Takes over nests of B. lucorum, and possibly also B. cryptarum and B. magnus. Common, particularly in northern and western Britain. Map p. 130.

IV Male cuckoo bumblebees Bombus (subgenus Psithyrus) To help confirm your identification after using this key, compare the genital capsule with the illustrations in pl. 8. The genital capsule should be extended gently with forceps while the bee is still soft, soon after death. If the bee has become rigid, relax it first (technique, p. 95). If the genitalia do not correspond with those of the species name reached via the key, then reject the key identification in favour of the best match in pl. 8 and try the key again to see what went wrong. For information on species distributions and abundance see relevant couplets (referred to below) in key III, and the maps at the end of the book. The terms used below are explained in the glossary diagram (fig. 27, p. 74).

Key IV. Male cuckoo bumblebees | 93

1 Last ventral plate bearing a black hair-tuft on each side (IV.1); (genital capsule pl. 8.1) B. (Ps.) campestris (pl. 4.11 and 4.12) Entirely black specimens are not uncommon. For information on distribution, abundance and host species see couplet 4, p. 92. Map p. 129.

IV.1

– Last ventral plate without these tufts (as IV.2)

IV.2

2

2 Tail mainly red or red-brown; (genital capsule pl. 8.2) B. (Ps.) rupestris (pl. 4.2) For information on distribution, abundance and host species see couplet 1, p. 91. Map p. 128.

– Tail mainly yellow or white

3

3 Seventh dorsal plate (tip of tail seen from above) usually red-haired (occasionally yellow-haired; genital capsule p1. 8.3) B. (Ps.) sylvestris (pl. 4.4) For information on distribution, abundance and host species see couplet 2, p. 91. Map p. 128.

– Seventh dorsal plate black-haired

3 5

4

4 Third and 5th segments of antenna about equal in length (IV.3); (T3 with yellow to whitish patch on each side, black in the middle; genital capsule pl. 8.4) B. (Ps.) bohemicus (pl. 4.6) For information on distribution, abundance and host species see couplet 5, p. 92. Map p. 130.

– Third segment of antenna obviously shorter than 5th (as IV.4) 5 5 Last ventral plate with a small mound on each side of midline, near the tip (IV.2); (T3 without obvious patches of white or yellow on each side; genital capsule pl. 8.5)

B. (Ps.) barbutellus (similar to ♀, pl. 4.9) 1V.3

1V.4

For information on distribution, abundance and host species see couplet 3, p. 91. Map p. 129.

– Last ventral plate without a small mound on each side of midline, near the tip (IV.1); (T3 yellow (fading to white) at the sides, usually black in the middle, but sometimes yellow; genital capsule pl. 8.6) B. (Ps.) vestalis (pl. 4.8) For information on distribution, abundance and host species see couplet 5, p. 92. Map p. 130.

9 Approaches to original work: techniques and web resources 9.1 Catching and handling bumblebees for identification It is usually quite easy to catch a bumblebee, either by placing a specimen tube over it while it is perched on a flower, or by sweeping it up with a butterfly net. A collapsible pocket net is convenient, and can be purchased from Watkins and Doncaster or B&S Entomological Services (addresses p. 95 and section 9.13, 26 and 27). To transfer the specimen to a tube, hold the net bag up by its tip - which will encourage the bumblebee to move upwards - then carefully slide the tube up inside. There should be a strong presumption against killing bees, especially queens. Good reference collections of the commoner bumblebee species can be made by looking out for dead bees, especially along the sides of roads, as, unfortunately, they are often hit by cars. If it is necessary to kill a bee, it can be put in a tube in the freezer. Alternatively, ethyl acetate or chloroform may be used. Ethyl acetate is preferable because it leaves the bee relaxed. Apply the fluid to 1 centimetre depth of plaster of Paris previously set into the base of a glass jar or tube. To prevent the specimen from getting damp, keep the plaster and container dry and use just a few drops of killing fluid. Leave the bumblebee in the jar for at least half an hour to make sure it is dead. Mount the bee on an entomological pin passed through the centre of the thorax. If you want to position the body parts (as in the plates) press the pin into a sheet of cork or polyethylene foam (see p. 95). The bee’s wings can be held forward by a pin positioned between the bases of the fore and hind wings, on each side. Each hind wing has a row of small hooks on the leading edge; these can be attached to a ridge on the back of the fore wing (as in life) to keep the hind wing in position. Claws on the tips of the legs will grip the surface to which the bee is pinned, and can be used to keep the legs in position. Before setting the bee, it may be desirable to extend

Approaches to original work: techniques and web resources | 95

the sting or genital capsule with forceps, to enable it to be examined later on. After several days the body parts will remain rigidly set in position. Sometimes it is necessary to relax and reset a bumblebee. This can be done by placing the specimen in a small airtight container in which there is a dish of water or a wet piece of tissue. After 2–3 days in this high humidity the membranes become soft and pliable, and with care the bee can then be reset. Do not leave the specimen at high humidity for too long, or it will go mouldy. Every specimen must have a label attached to its pin indicating when, where and by whom it was caught. A second label bearing the species name of the bee may be placed below the first. When identifying a specimen always use the best light source available. Vary the angle of observation: this can be done easily if the bumblebee is pinned at various angles on an L-shaped stand improvised from polyethylene foam. A more versatile ‘hinge and bracket’ stand is illustrated in Yeo & Corbet (1995, p. 57). Pinned bumblebees may be stored in a conventional cork-lined insect box (expensive) or in a sandwich box with a tight-fitting floor of polyethylene foam. To avoid damage by insects and mites, these boxes should have close-fitting lids. Wooden boxes can be treated with paradichlorobenzene or Rentokil Woodworm Fluid. Suppliers of entomological equipment including nets, pins, insect boxes, etc. are: Watkins and Doncaster, Four Throws, Hawkhurst, Kent (see 9.13.26, p. 107) B&S Entomological Services, 37 Derrycarne Road, Portadown, Co. Armagh BT62 1PT, N. Ireland (see 9.13.27) Polyethylene foam (Plastazote) is available (as sheets 1000 mm x 750 mm x 12 mm) from: Watkins and Doncaster (above), or Polyformes Ltd., Cherrycourt Way, Stanbridge Rd., Leighton Buzzard, Bedfordshire, LU7 4UH (tel 01525 852444) as well as many internet suppliers. This material is ideal for pinning and storing insects. Expanded polystyrene ceiling tiles are a cheaper but less satisfactory alternative.

96 | Bumblebees

9.2 Marking bumblebees and sampling honeystomach contents It can be useful to mark bees individually in order to make detailed observations on foraging behaviour, particularly traplining (p. 51). Small coloured and numbered discs, designed for marking queen honeybees, are useful for marking bumblebees. A disc is glued to the bumblebee’s thorax with a small amount of resin. Alternatively, bees can be marked with coloured correction fluid or quick-drying paint such as Humbrol model aeroplane dope. One or a number of small spots of paint can be applied to the thorax with a sharpened matchstick. Use a circular motion to flatten the hair and take great care to avoid the wing bases. Make sure that the paint or resin is dry before the bumblebee is free to groom itself. Anaesthetics can cause long-term changes in bee behaviour. Their use can be avoided if the bumblebee is restrained during marking, using a cage such as those used by beekeepers when marking honeybee queens (supplier, see below). Alternatively, a bumblebee can be immobilised by cooling it in a refrigerator for about half an hour, although this may also affect subsequent behaviour. If anaesthetics are necessary a convenient field method involves using carbon dioxide from a pocket dispenser. A sparklet-operated wine bottle opener such as the Sparklets Corkmaster is very useful for this: although no longer sold in the UK used ones can sometimes be found on ebay. An equivalent, the Cork Pops Wine Bottle Opener, remains available from the US on ebay. The bumblebee should be placed in a tube, a dose of carbon dioxide added and the tube immediately closed. As soon as the bumblebee ceases to move it should be removed. This method allows the honeystomach contents to be sampled without killing the bee (see p. 50–51). Suppliers of bee marking outfits (containing discs, resin and applicator) and marking cages: E. H. Thorne (Beehives) Ltd, Wragby, Lincoln (see 9.13.28, p. 107). Other suppliers can be found on the web. The Bumblebee pages (link 9.13.20) describe simple inexpensive methods of marking bees.

Approaches to original work: techniques and web resources | 97

9.3 Dissection and measurements

Fig 28. Spermatheca of a fertilised bumblebee (Bombus) sperm mass accessory gland

valve

duct

0.3 mm

As mentioned above, killing bumblebees should be avoided on conservation grounds unless it is strictly necessary. The information below may be helpful when required (also see Dade, 2009). It is a simple matter to remove the honeystomach from a dead or anaesthetised bee. The edges of two adjacent ventral plates (see fig. 27, p. 74) at the front end of the abdomen can be grasped, each with a pair of fine watch-maker’s forceps, and gently pulled apart. This exposes the underlying honeystomach. If the head of the bee is then removed, thus severing the oesophagus, the honeystomach is no longer attached in front. It can then be detached from the midgut behind, and removed from the abdominal cavity (see fig. 22, p. 50). The fat body and ovaries are seen best in specimens that are fresh or have been stored deep-frozen. Having removed some of the ventral plates both these structures can be dissected, again using two pairs of watchmaker’s forceps, in a little insect saline solution (7.5 grams sodium chloride made up to 1 litre with distilled water). For most other purposes bumblebees may be preserved and dissected in 70% ethanol. For a detailed examination of internal structures a dissecting microscope is essential (see below for suppliers). Most of the fat body lies within the abdominal cavity. Its colour varies with age and activity: white or clear in newly hatched bumblebees and in queens before and immediately after hibernation, it gradually turns yellow, and finally brown in old individuals. A female contains two ovaries, each consisting of several ovarioles (four in true bumblebees, up to l5 or so in cuckoo bumblebees). Near the union of the oviducts a thin tube connects the dorsal surface of the vagina to a small spherical sperm storage organ, the spermatheca. To determine whether or not a female has mated successfully and so contains sperm, the spermatheca may be removed (using fine forceps), mounted under a coverslip on a slide and examined with a compound microscope. If sperm are present they appear as a dense opaque mass (fig. 28); if not, the lumen of the gland appears transparent.

98 | Bumblebees Fig 29. Mating scars (arrowed) on the sting base of a female B. hortorum.

side view

ventral view

In B. hortorum, B. pratorum and B. jonellus (and probably in some other species as well, such as B. monticola, B. ruderatus, B. cullumanus) it is possible to see whether a female has mated (successfully or unsuccessfully) by the presence of two or four prominent black marks on the sting apparatus, caused by the male genital capsule. In B. pratorum and B. jonellus there are two marks, one on each inner projection of the sting sheath (see pl. 5.7 and 5.10). In B. hortorum there is also a mark on each side of the base of the sting apparatus (fig. 29). An ovariole contains a string of follicles, each of which represents a developing oocyte, alternating with groups of nutritive ‘nurse’ cells. As it enlarges the oocyte draws nutriment from its attendant nurse cells. These shrink and eventually degenerate into a ‘yellow body’ shortly before the egg is mature. Several yellow bodies may be present in one ovariole, each lying above the corresponding egg (Palm, 1948). A yellow body at the base of a duct therefore suggests that an egg has recently been laid. The mature egg is about 3 millimetres long and covered by a shiny skin (the chorion). If it is not laid the egg may degenerate and be resorbed in the ovariole. The length and number of developing oocytes can be used as indicators of a bumblebee’s reproductive condition. In studies of foraging behaviour it may be desirable to determine tongue length. This is difficult to do in a direct way on living bumblebees, but can be estimated quite accurately on an immobilised bee (see section 9.2), by measuring a body part that varies in a consistent way with tongue length, such as one of the cells of the forewing (see the illustration in Chart A, p. 76). This method requires one to plot how the two measures vary with body size for each species (see Morse, 1978). It will necessitate measuring some freshly dead specimens that have not dried, and therefore shrunk. Measurements can be made with a low power dissecting microscope with an eye-piece graticule (see below for suppliers). A measurement that is easy to make, and important functionally, is the length of the glossa together with the prementum (fig. 30). Take care not to stretch the tongue accidentally while measuring its length. Wing wear can provide a useful indicator of age or activity, and is readily scored (e.g. 0 = no wear, 1 = single nick in wing edge, 2 = most of margin slightly worn, and so on).

Approaches to original work: techniques and web resources | 99 Fig 30. A bumblebee tongue (glossa + prementum).

prementum

When determining body weight, remember to allow for the weight of pollen loads and contents of the honeystomach. To assess accurately the unloaded body weight, both should be removed. Suppliers of dissecting microscopes include: Bioquip, Unit 8, Marbury House Farm, Bentleys Farm Lane, Higher Whitley, Nr. Warrington, Cheshire WA4 4QW (see 9.13.33, p. 107) ALS, Station Road, Hindolveston, Norfolk, NR20 5DE (see 9.13.32)

9.4 Measuring nectar concentration and energy content glossa

Nectar may be extracted from a flower by carefully probing to the base of the corolla with a 1 or 5 microlitre microcapillary tube (available from Camlab or SigmaAldrich, see 9.13.30 and 29, p. 107). The length of the column of nectar in the tube can be used to determine the volume present. Nectar solute concentration can be measured in the field with a pocket sucrose refractometer modified by the makers to take volumes down to less than 1 microlitre. Two refractometers are required to measure the whole range of nectar solute concentration (one measures 0–50%, the other 40–85%). Blow the nectar sample out of the tube onto the lower prism of the refractometer and close the instrument immediately so that the drop has no time to evaporate. Readings of concentration are given directly in % weight/weight (w/w) as grams sucrose/100 grams solution. The relationship between concentration (% w/w) and refractive index is very similar for the sugars common in nectar: sucrose, glucose and fructose. It is convenient to make the assumption that nectar is simply a solution of sugars. This is not quite correct, as small amounts of other substances, such as amino acids and proteins, may be present, and affect the readings slightly. The amount of sugar per flower can be calculated by multiplying the volume of nectar by the weight of solute per unit volume of solution. To convert milligrams of sugar to joules, multiply by 15.5. The weight of sucrose per unit volume of solution that corresponds with the weight of sucrose per 100 grams solution measured by

100 | Bumblebees

the refractometer can be found from tables in the CRC Handbook of Chemistry and Physics (Weast, 1978; Lide, 2009); or calculated, by multiplying the weight of sucrose per 100 grams solution by the density of a solution of that concentration. The appropriate value for density can be estimated very simply, using the following formula (suitable for use with a pocket calculator): d = 0.0037291C + 0.0000178C² + 0.9988603 where d is the estimate of density for a given value of C, and C is the weight of sucrose per 100 grams solution (the refractometer reading). (Between C = 0 and 84 this formula gives an estimate of density to within 0.05% of the values given in Weast, 1978.) The calculation can be illustrated by an example. A flower contains 4 microlitres of nectar giving a refractometer reading of 48%. The density of a solution of 48 grams sucrose per 100 grams solution is (from the above formula) 1.2189 milligrams per microlitre of solution. Thus the nectar weighs 4.0 × 1.2189 = 4.88 milligrams, of which 0.48 (48%) is sugar. The nectar therefore contains 4.0 x 1.2189 × 0.48 = 2.34 milligrams sugar, or 4.0 × 1.2189 × 0.48 × 15.5 = 36 joules of energy. In a given plant species at a given time the amount of sugar may vary greatly between flowers. Some flowers may be full of nectar while others are empty, having recently been visited. Such variability necessitates careful use of statistics in any detailed analysis of the results. To work out the average amount of sugar per flower you should take measurements from at least 10 flowers on each sampling occasion. Corbet (2003) gives more details on measuring the standing crop and the secretion rate of nectar. Suppliers of microcapillary tubes (disposable micropipettes): Sigma-Aldrich Company Ltd, tel. 01202 712300, e-mail [email protected] (see 9.13.29, p. 107). Camlab Ltd, Nuffield Road, Cambridge (see 9.13.30) For all but the largest flowers, the most useful sizes are 0.5 microlitre, 1 microlitre and 5 microlitres. Suppliers of refractometers: Bellingham and Stanley Ltd, Polyfract Works, Longfield Road, Tunbridge Wells, Kent TN2 3EY (see 9.13.31). Ask for the special instrument for small volumes of nectar.

Approaches to original work: techniques and web resources | 101

9.5 Obtaining and identifying pollen The simplest method of obtaining pollen for nest-founding (p. 32) is to contact a sympathetic beekeeper. Pollen may be scraped from honeybee combs (remove wax and other debris), or harvested with a pollen trap fitted to the front of a honeybee hive. Design and use of pollen traps are described in Synge (1947) and IBRA (1975). Pollen should be stored, deep-frozen, in an airtight container. Pollen dough can be prepared by mixing pollen with a very small amount of honey solution, giving a thick, dry paste. Pollen identification can sometimes be based on colour (Kirk, 2006; Hodges, 1974), but usually requires the use of a compound microscope to examine structure. Helpful guides to identification are to be found in Moore & Webb (1978) and Sawyer (1981). The following is a suitable method of preparing slides. (i) Place the pollen on a slide and wash it with 95% ethanol to remove waxes and oils. (ii) Evaporate off the ethanol and add a drop of basic fuchsin in glycerol. Observe the process of staining under the microscope. When slightly over stained, stop further staining and remove any excess, by carefully washing with more ethanol. (iii) Dry the slide and place a small piece of glycerin jelly and a coverslip over the pollen. Warm the slide gently until the jelly melts. Use a piece of jelly small enough to leave a gap around the edge of the coverslip. Melted paraffin wax should be run into this gap to provide a support for the coverslip. (iv) When the wax has set, the edge of the coverslip can be sealed with nail varnish, to prevent the jelly drying out. Using this method a reference collection of slides of different pollens can be prepared, with which unknown species may be compared. More elaborate and effective procedures for preparing pollen for microscopic examination are described by Kearns & Inouye (1993).

102 | Bumblebees

9.6 Removing pollen from bumblebees Pollen intentionally collected by a female bumblebee will be gathered onto the outer surface of the hind tibia of each hind leg, where it is retained by the fringing corbicular hairs (see fig. 27, p. 74). These loads are easily identified and removed. If a more complete list of the pollen types that a bee has been in contact with is required, the following method may be adopted. Washing with ethanol (i) Kill the bee and then wash it thoroughly in 70% ethanol. Repeat, washing with water. (ii) Centrifuge (or filter) the two sets of washings. Discard the liquid, keeping the pollen that is in the bottom of the centrifuge tube (or on the filter). (iii) Dry the pollen (hasten this process by mixing with a little acetone) and proceed to stain (as from (ii) in previous section). In studies of pollen transfer and pollination it may be desirable to determine whether pollen is deposited on a particular part of a bee’s body, and for this purpose the following method may be used. Sellotape peel technique Kill or anaesthetise the bumblebee. Cut several tiny pieces (about 2 millimetres square) of Sellotape. Hold one in fine forceps and dab its sticky side against the dorsal surface of the bee’s thorax. Lay the Sellotape on a slide, sticky side up, and add a few drops of stain (such as basic fuchsin) and a coverslip. The distribution on the bee of the various species of pollen can be mapped by dabbing squares at various positions on the bee’s body.

9.7 Staining germinating pollen and pollen tubes If a bumblebee’s visit to a flower transfers compatible pollen that subsequently germinates, the resulting pollen tubes, which grow within the stigma of the flower, can be revealed by staining. Several methods are described by Kearns & Inouye (1993). Many require a fluorescence microscope, but one of the simplest, due to Levin (1990), does not. Pollen tubes are fixed in 3:1 (by volume) alcohol:acetic acid for at least 24 h, and then stained in 1% basic fuchsin: 1% fast green (4:1) for at least 24 h. De-stain

Approaches to original work: techniques and web resources | 103

and soften the tissue in lactic acid for 12 h, squash under a coverslip and examine with a compound microscope. Pollen tubes stain maroon against a white background.

9.8 Flowers for bumblebees Gardens, field margins, road verges and disused or derelict land can be improved for bumblebees by avoiding disturbance such as ploughing (to encourage perennials as opposed to annuals) or application of herbicides or fertilizers (which would promote rank vegetation), or by sowing or planting wild flowers. Section 7.3 and Flora Locale (section 9.13.17, p. 107) discuss responsible planting, using native seeds from approved suppliers. Plantlife (section 9.13.18) is concerned with the conservation of wild plants, and campaigns to conserve and protect threatened plants and habitats. Books that list suitable forage plants for bees include Garden plants valuable to bees (IBRA, 1981), published by the International Bee Research Association (section 9.13.21), and Gardening for butterflies, bees and other beneficial insects (Miller-Klein, 2010). Fussell & Corbet (1992a) describe the results of a public survey of the forage plants used by common species of bumblebees in Britain (table 7; and see table 6, p.65).

9.9 Microclimate recording The construction and operation of simple equipment for recording microclimate is described in Unwin (1978; 1980) and Unwin & Corbet (1991). Corbet (1978a, b); Corbet, Willmer & others (1979) and Unwin & Corbet (1991) give examples of the use of microclimate records.

9.10 Recording distribution In the 1970s volunteers throughout the country took part in the Bumblebee Distribution Maps Scheme (BDMS), which collected records with the aim of determining the distribution of all British bumblebee species: the resulting published maps (ITE, 1980) provided a good summary of knowledge at the time at the scale of the 10 km square. Since then the Bees, Wasps and Ants Recording Society (BWARS) has taken over the promotion of recording of all British species in each of these groups.

104 | Bumblebees Table 7. Results summarised from a national survey of flower visitation by different colour groups of bumblebees in Britain, undertaken in 1987 and 1988. Colour groups (p. 78) are used where specific identification is beyond the scope of the participants. (a) The top 10 types of flowers visited by each colour group ranked in descending order of ‘simple selectivity’, an index of flower preference that takes account of the number of visits to each plant species available. 2-banded white tails

Black-bodied red tails

Banded red tails

Browns

3-banded white tails

Rhododendron

Knapweeds

Knapweeds

Vetches

Rhododendron

Cotoneaster

Sedums

Snowberry

Indian balsam

Foxglove

Buddleja

Campanulas

Cotoneaster

White dead-nettle

Delphinium/ larkspur

Bramble

Chives

Raspberry

Lavender

Red clover

Willowherbs

Bird’s-foot-trefoil

Ragwort

Woundworts

Woundworts

Lavender

White clover

Lavender

Comfrey

Honeysuckle

Heather

Cotoneaster

Cranesbills

Catmint

Buddleja

Comfrey

Buttercups

Wild thistles

Red clover

White dead-nettle

Sedums

Rhododendron

Comfrey

Raspberry

Catmint

Knapweeds

Thistles

Rhododendrons

Borage

Sage

(b) The top 10 types of flowers visited by each colour group ranked in descending order of ‘groupspecific selectivity’, an index of flower preference that illustrates dietary differences between groups of bumblebees. It compares the proportion of visits to each plant type by each colour group to the proportion of visits to that plant type made by all the groups combined. 2-banded white tails

Black-bodied red tails

Banded red tails

Browns

Hollyhock

Hawksbeards

Loganberry

Primrose

Delphinium/ larkspur

Hawthorn

Bluebell

Snowberry

Purple loosestrife

Foxglove

Nasturtium

Gorse

Heuchera

Vetches

Pansy

Michaelmasdaisy

Bird’s-foot-trefoil

Raspberry

Apple

Red clover

Flowering currant

Selfheal

Forget-me-nots

Wood sage

Bindweeds

Azalea

Chives

Flowering currant

Golden rod

Teasel

Lupins

Aubrieta

Knapweeds

Columbine

Buttercups

Ragwort

Meadow vetchling Sallows

Indian balsam

Hawkweeds Knapweeds (garden)

Cranesbills

White dead-nettle

Columbine

Alkanet

Hyssop

Red campion

Golden rod

Adapted from Fussell & Corbet (1992a)

3-banded white tails

Toadflax Honeysuckles

Approaches to original work: techniques and web resources | 105

The records it collects are published, in conjunction with the Biological Records Centre (BRC), as new and updated maps (Provisional atlases of the aculeate Hymenoptera of Britain and Ireland; available from the Centre for Ecology and Hydrology, link 9.13.5, p. 107). Most of this information is available and updated online, via BWARS (link 9.13.4), and the National Biodiversity Network Gateway (link 9.13.7). Anyone interested in furthering our knowledge of the distribution of bumblebee species is strongly encouraged to join BWARS and contribute information. Distributions change with time, and therefore all records are valuable. There are also significant gaps in our knowledge; particularly in parts of Ireland (see Maps, p. 119–130). BWARS ‘recording’ page, on their website (link 9.13.2), will tell you how to submit a record. Unless it is absolutely necessary, please avoid collecting queens, in particular, and especially those of rarer species, as each represents a potential colony. If you are not certain of a bee’s identity a photograph may enable you to get help with identification - see BWARS ‘forum’, accessible from the ‘identification’ page of their website (see 9.13.2). If close inspection is required, the bee can be confined temporarily in a specimen tube, or jam jar, so that critical diagnostic features can be examined with a lens before it is released.

9.11 Entomological societies Some readers may wish to join a society such as the Amateur Entomologists’ Society (see 9.13.23, p. 107) or the British Entomological and Natural History Society (see 9.13.24). Both publish attractive journals which contain articles on entomological topics and keep members in touch with various meetings and recording schemes. The Royal Entomological Society (see 9.13.22) organises meetings and workshops in St Albans and elsewhere; publishes several journals; and holds a library from which members may borrow by post. The International Bee Research Association (see 9.13.21) deals mainly with honeybees, but its publications include material on bumblebees and pollination; and the Eva Crane / IBRA Library - which has been transferred to the National Library of Wales (NLW) - will soon become searchable with their online catalogue. The

106 | Bumblebees

Bees, Ants and Wasps Recording Society, mentioned above (section 9.13.2), has a newsletter and holds field and indoor meetings.

9.12 How to present the results of research Writing up is an important part of a research project. A really thorough, critical investigation may be worth publishing; particularly if it has established new information of general interest, and if the animals on which it is based can be identified with certainty. Journals that publish short papers on insect biology include the Entomologist’s Monthly Magazine, Bulletin of the Amateur Entomologists’ Society (links 9.13.39 and 9.13.23, below) and, for material with an educational slant, Journal of Biological Education (link 9.13.25). Those unfamiliar with publishing conventions are advised to examine current numbers of these journals to see what sort of thing they publish, and then to write a paper along similar lines, keeping it as short as is consistent with the presentation of enough information to establish the conclusions. It is then time to consult an appropriate expert, who can give advice on whether and in what form the material might be published. It is an unbreakable convention of scientific publication that results are reported with scrupulous honesty. Hence it is essential to keep detailed and accurate records throughout the investigation, and to distinguish in the write-up between certainty and probability, and deduction and speculation. In many cases it will be necessary to apply appropriate statistical techniques to test the significance of the findings. A book such as Wheater & Cook (2003) or Fowler & Cohen (1995) will help, but this is an area where expert advice can contribute much to the planning as well as the analysis of the work.

Approaches to original work: techniques and web resources | 107

9.13 Web-based information There is a rapidly expanding range of websites containing up-to-date information relating to, or relevant to, bumblebees. Useful examples are given below: Bumblebee identification - Britain and the world:

1. Dr Paul Williams, Natural History Museum, London: http://www.nhm.ac.uk/research-curation/research/projects/bombus/

Recording British bumblebees (and other ants, bees and wasps):

2. Bees, Wasps & Ants Recording Society (BWARS): http://www.bwars.com/ 3. Biological Records Centre: http://www.brc.ac.uk/recording_schemes.asp

British bumblebee distribution: 4. 5. 6. 7.

BWARS maps: http://www.bwars.com/NBN_Maps.htm Bees, Wasps and Ants Provisional Atlases: http://www.ceh.ac.uk/products/publications Highland Biological Recording Group: http://www.hbrg.org.uk/MainPages/BeeAtlasUpdate.html National Biodiversity Network Gateway: http://www.searchnbn.net/

Conservation of bumblebees, other organisms, and habitats

8. Hymettus/ Bumblebee Working Group: http://www.hymettus.org.uk/index.htm 9. Bumblebee Conservation Trust: http://www.bumblebeeconservation.org.uk/ 10. Buglife: http://www.buglife.org.uk/ 11. The Xerces Society for invertebrate conservation (USA): http://www.xerces.org/bumblebees/ 12. UK Biodiversity Action Plan: http://www.ukbap.org.uk/ 13. New challenges, new CAP: http://www.birdlife.org/eu/pdfs/CAP%20Brochure.pdf 14. Birdlife: http://www.birdlife.org/action/campaigns/farming_for_life/campaign.html 15. Fundatia ADEPT: http://www.fundatia-adept.org/ 16. Lowland Hay Meadows in the UK: http://www.jncc.gov.uk/publications/JNCC312/habitat.asp?FeatureIntCode=H6510 17. Flora locale: http://www.floralocale.org/ 18. Plantlife: http://www.plantlife.org.uk/ 19. RSPB: http://www.rspb.org.uk/ourwork/conservation/biodiversity/keyspecies/invertebrates/ bumblebee/index.aspx 20. The Bumblebee pages: http://www.bumblebee.org See also The Short-haired bumblebee project: http://www.bumblebeereintroduction.org

Bee, and other insect, organisations:

21. International Bee Research Association: http://www.ibra.org.uk/ 22. Royal Entomological Society: http://www.royensoc.co.uk/ 23. Amateur Entomologists’ Society: http://www.amentsoc.org/ 24. British Entomological & Natural History Society: http://www.benhs.org.uk/portal/ 25. The Society of Biology: http://www.societyofbiology.org/home

Equipment and book suppliers 26. Watkins and Doncaster: http://www.watdon.co.uk/the-naturalists/ 27. B&S Entomological Services: http://www.entomology.org.uk/ 28. E. H. Thorne (Beehives) Ltd: http://www.thorne.co.uk/index.htm 29. Sigma-Aldrich: http://www.sigmaaldrich.com/united-kingdom.html 30. Camlab Ltd: http://www.camlab.co.uk/ (search the site using the term ‘DMP’ or ‘disposable micropipette’) 31. Bellingham and Stanley: http://www.bellinghamandstanley.com/ltd/hh_overview.html 32. ALS (Anglian Lepidopterist Supplies): http://www.angleps.com/ 33. Bioquip: http://www.bioquip.net/acatalog/Microscopes.html 34. Abe Books: http://www.abebooks.co.uk/ 35. NHBS- Everything for wildlife, science & environment: http://www.nhbs.com 36. Northern Bee Books: http://www.groovycart.co.uk/cart.php?c=533 37. Field Studies Council: http://www.field-studies-council.org/publications/index.aspx 38. British Wildlife Publishing: http://www.britishwildlife.com/ 39. Entomologists’ Monthly Magazine: http://www.pemberleybooks.com 40. Pelagic Publishing: http://www.pelagicpublishing.com

10 Further reading 10.1 Finding books and journals Many journal articles are now available on the internet, either complete or in abstract form. It is worth searching there first. Where references have a ‘doi’ number this can be copied into a computer browser to locate the relevant article online. It is often possible to make arrangements to see or borrow books or journal articles by visiting the library of a local university, or by asking your local public library to borrow the work (or a photocopy of it) for you via the British Library Document Supply Centre (see http://www.bl.uk/articles). This may take some time, and it is important that your librarian, or online request, has a reference that is correct in every detail. References are acceptable in the form given here, namely the author’s name, the date of publication, followed by (for a book) the title and publisher or (for a journal article) the title of the article, the journal title, the volume number, and the first and last pages of the article. The Handbooks for the Identification of British Insects, published by the Royal Entomological Society, can be bought online (see link 9.13.22, p. 107). In print books are available from NHBS - Everything for wildlife, science & environment (see link 9.13.35), out of print books may be available via Abebooks (link 9.13.34). Abrahamovich, A.H., Tellería, M.C. & Díaz, N.B. (2001). Bombus species and their associated flora in Argentina. Bee World 82: 76–87. Adler, L.S. (2000). The ecological significance of toxic nectar. Oikos 91: 409–420. Akeroyd, J. R. (1994). Seeds of destruction? Non-native wildflower seed and British floral biodiversity. Plantlife. For an online version see http:// www.floralocale.org/content. asp?did=23788. Akeroyd, J. R. & Page, N. (2006). The Saxon villages of Southern Transylvania: conserving biodiversity in a historic landscape. In Nature conservation: concepts and practice (eds Gafta, D. & Akeroyd, J.R.), pp.199–210. Heidelburg: Springer Verlag. Alford, D.V. (1975). Bumblebees. London: Davis-Poynter. Alford, D.V. (1978). The life of the bumblebee. London: DavisPoynter. (Republished in 2009 by Northern Bee Books, link 9.13.36)

Arretz, P.V. & Macfarlane, R.P. (1986). The introduction of Bombus ruderatus to Chile for red clover pollination. Bee World 67: 15–22. Badmin, J. (2009). Are birds the cause of bumblebee decline in Britain? British Journal of Entomology and Natural History. 22: 205–206. Beekman, M & Ratnieks, F.L. W. (2000). Long-range foraging by the honeybee, Apis mellifera L. Functional Ecology 14: 490–496. Benton, T. (2000). The bumblebees of Essex. Saffron Walden: Lopinga. Benton, T. (2006). Bumblebees, The natural history and identification of the species found in Britain. The New Naturalist Library. London: Collins. Berezin, M.V., Beiko, V.B. & Berezina, N.V. (1996). Analysis of structural changes in the bumblebee (Bombus, Apidae) population of Moscow Oblast over the last forty years. Entomological Review 76: 115–123, translated from Zoologicheskii Zhurnal 75 (1996).

Bertsch, A. (1984). Foraging in male bumblebees (Bombus lucorum L.): maximizing energy or minimizing water load? Oecologia (Berlin) 62: 325–36. Bertsch, A. (2009). Barcoding cryptic bumblebee taxa: B. lucorum, B. cryptarum and B. magnus, a case study. Beiträge zur Entomologie 59: 287–310. Bertsch, A., Schweer, H. & Titze, A. (2004). Discrimination of the bumblebee species Bombus lucorum, B. cryptarum and B. magnus by morphological characters and male labial gland secretions. Beiträge zur Entomologie 54: 365–386. Bertsch, A. Schweer, H., Titze, A. & Tanaka, H. (2005). Male labial gland secretions and mitochondrial DNA markers support species status of Bombus cryptarum and B. magnus (Hymenoptera, Apidae). Insectes Sociaux 52: 45–54. Brian, A.D. (1951). The pollen collected by bumble-bees. Journal of Animal Ecology 20: 191–4. Brian, A.D. (1957). Differences in the flowers visited by four

Further reading | 109 species of bumblebees and their causes. Journal of Animal Ecology 26: 69–96. Brian, M.V. (1980). Social control over sex and caste in bees, wasps and ants. Biological Reviews 55: 379–415. Bringer, B. (1973). Territorial flight of bumble-bee males in the coniferous forests on the northernmost part of the island of Öland. Zoon, Supplement 1: 15–22. Brundrett MC. (1991). Mycorrhizas in natural ecosystems. In: Advances in Ecological Research (eds Macfadyen, A., Begon, M. & Fitter, A. H.) Vol. 21. Academic Press, London. pp. 171–313. Available from: http://mycorrhizas.info. Brundrett, M.C. (2009). Mycorrhizal associations and other means of nutrition of vascular plants: understanding the global diversity of host plants by resolving conflicting information and developing reliable means of diagnosis. Plant Soil 320: 37–77. (http:// www.springerlink.com/ content/g220m41186052g74/ fulltext.pdf) doi 10.1007/ s11104–008–9877–9. Buttermore, R.E. (1997). Observations of successful Bombus terrestris (L.) (Hymenoptera: Apidae) colonies in southern Tasmania. Australian Journal of Entomology 36: 251–254. Cameron, S. A., Hines, H.M. & Williams, P.H. (2007). A comprehensive phylogeny of the bumble bees (Bombus). Biological Journal of the Linnean Society 91: 161–188. Cameron, S.A., Lozier, J.D., Strange, J.P., Koch, J.B., Cordes, N., Solter, L.F. & Griswold, T.L. (2011). Patterns of widespread decline in North American bumble bees. Proceedings of the National Academy of Science, U.S.A. 108: 662–7. Carreck, N.L., Osborne, J.L., Capaldi, E.A. and J.R. Riley (1999). Tracking bees with radar. Bee World 80: 124–131. Carvell, C. (2002). Habitat use and conservation of bumblebees (Bombus spp.) under different grassland management regimes. Biological Conservation 103: 33–49. Carvell, C., Roy, D.B., Smart, S.M., Pywell, R.F., Preston,

C.D. & Goulson, D. (2006). Declines in forage availability for bumblebees at a national scale. Biological Conservation 132: 481–489. Chapman, R.E., Wang, J. & Bourke, A.F.G. (2003). Genetic analysis of spatial foraging patterns and resource sharing in bumblebee pollinators. Molecular Ecology 12: 2801–8. Charman, T.G., Sears, J., Bourke, A.F.G. & Green, R.E. (2009). Phenology of Bombus distinguendus in the Outer Hebrides. The Glasgow Naturalist 25: Supplement. Machair conservation: successes and challenges 32–42. Church, N.S. (1960). Heat loss and the body temperature of flying insects. II. Heat conduction within the body and its loss by radiation and convection. Journal of Experimental Biology 37: 186–212. Colla, S. & Packer, L. (2008). Evidence for decline in eastern North American bumblebees (Hymenoptera: Apidae), with special focus on Bombus affinis Cresson. Biodiversity and Conservation 17: 1379–1391. Colla, S., Otterstatter, M.C., Gegear, R.L. & Thomson, J.D. (2006). Plight of the bumble bee: pathogen spillover from commercial to wild populations. Biological Conservation 129: 461–467 Corbet, S.A (1995). Insects, plants and succession: advantages of long-term set-aside. Agriculture, Ecosystems and Environment 53: 201–217. Corbet, S.A. (1978a). Bee visits and the nectar of Echium vulgare L. and Sinapis alba L. Ecological Entomology 3: 25–37. Corbet, S.A. (1978b). Nectar, insect visits and the flowers of Echium vulgare. In The pollination of flowers by insects. Ed. A.J. Richards, pp. 21–29. Corbet, S.A. (2000). Conserving compartments in pollinator webs. Conservation Biology 14: 1220–1231. Corbet, S. A. (2003) Nectar sugar content: estimating standing crop and secretion rate in the field. Apidologie 34: 1–10. Corbet, S.A., Cuthill, I., Fallows, M., Harrison, T., & Hartley, G. (1981). Why do nectar-foraging bees and wasps work upwards

on inflorescences? Oecologia 51: 79–83. Corbet, S.A. & Morris, R.J. (1999) Mites on bumble bees and bluebells. Entomologist’s Monthly Magazine 135: 77–83 Corbet, S.A., Saville, N.M., Fussell. M., Prŷs-Jones, O.E. & Unwin, D.M. (1995). The competition box: a graphical aid to forecasting pollinator performance. Journal of Applied Ecology 32: 707–719. Corbet, S.A., Unwin, D., & PrŷsJones, O.E. (1979). Humidity, nectar and insect visits to flowers, with special reference to Crataegus, Tilia and Echium. Ecological Entomology 4: 9–22. Corbet, S.A., Williams, I.H. & Osborne, J.L. (1991). Bees and the pollination of crops and wild flowers in the European Community. Bee World 72: 47–59. Corbet, S.A., Willmer, P.G., Beament, J.W.L., Unwin, D. & Prŷs-Jones, O.E. (1979). Postsecretory determinants of sugar concentration in nectar. Plant, Cell and Environment 2: 293–308. Corbet, S., Chapman, H. & Saville, N. (1988). Vibratory pollen collection and flower form: bumble-bees on Actinidia, Symphytum, Borago and Polygonatum. Functional Ecology 2: 147–156. Cresswell, J.E., Osborne, J.L. & Goulson, D. (2000). An economic model of the limits to foraging range in central place foragers with numerical solutions for bumblebees. Ecological Entomology 25: 249–255. Cumber, R.A. (1949). The biology of humble-bees, with special reference to the production of the worker caste. Transactions of the Royal Entomological Society of London 100: 1–45. Cumber, R.A. (1954). The life cycle of humble-bees in New Zealand. New Zealand Journal of Science and Technology, Series B, 32: 95–107. Dade, H.D. (2009). Anatomy and dissection of the honeybee. Cardiff: International Bee Research Association. (Available from IBRA: see link 9.13.21, p 107). Dafni, A. (1998). The threat of Bombus terrestris spread. Bee World 79: 113–114. Darvill, B., Knight, M.E. & Goulson, D. (2004). Use of genetic markers to quantify bumblebee

110 | Bumblebees foraging range and nest density. Oikos 107: 471–8. Darvill, B., O’Connor, S., Lye, G.C., Waters, J., Lepais, O. & Goulson, D. (2010). Cryptic differences in dispersal lead to differential sensitivity to habitat fragmentation in two bumblebee species. Molecular Ecology 19: 53–63. Davies, N.B. (1977). Prey selection and the search strategy of the spotted flycatcher (Muscicapa striata): a field study on optimal foraging. Animal Behaviour 25: 1016–33. Day, M.G. (1966). Identification of hair and feather remains in the gut and faeces of stoats and weasels. Journal of Zoology 148: 201–217. Dias, B.S.F., Raw, A. & Imperatriz-Fonseca, V.L. (1999). International pollinators initiative: the Sao Paulo declaration on pollinators. (Report on the recommendations of the workshop on the conservation and sustainable use of pollinators in agriculture with emphasis on bees.) Brasília: Brazilian Ministry of the Environment. 99 pp. http:// www.bfn.de/fileadmin/MDB/ images/themen/bestaeuber/ agr-pollinator-rpt.pdf Dornhaus, A., Brockman, A. & Chittka, L. (2003). Bumble bees alert to food with pheromone from tarsal gland. Journal of Comparative Physiology A 189: 47–51. Dramstad, W.E (1996). Do bumble bees (Hymenoptera: Apidae) really forage close to their nest? Journal of Insect Behavior 9: 163–182. Dramstad, W.E. & Fry, G. (1995). Foraging activity of bumblebees (Bombus) in relation to flower resources in arable land. Agriculture, Ecosystems and Environment 53: 123–135 Edwards, M. & Jenner, M. (2009) Field guide to the bumblebees of Great Britain & Ireland. Eastbourne: Ocelli. Edwards, M. & Williams, P. (2004). Where have all the bumblebees gone, and could they ever return? British Wildlife 15: 305–312. Ellington, C.P., Machin, K.E. & Casey, T.M. (1990). Oxygen consumption of bumblebees in forward flight. Nature 347:

472–473. Esch, H. & Goller, F. and Heinrich, B. (1991). How do bees shiver? Naturwissenschaften 78: 325–328. Evans, D.L. & Waldbauer, G.P. (1982). Behaviour of adult and naïve birds when presented with a bumblebee and its mimic. Zeitschrift für Tierpsychologie 59: 247–259. Evans, E., Thorp, R., Jepsen, S, & Black, S.H. (2008). Status review of three formerly common species of bumblebee in the subgenus Bombus. Report for the Xerces Society for Invertebrate Conservation, Portland, Oregon. See http:// www.xerces.org/bumblebees/ . Ferry, C. & Corbet, S.A. (1996). Water collection by bumblebees. Journal of Apicultural Research 35: 120–2. Ferton, C. (1901). Sur l’époque du réveil des bourdons et des psithyres à Bonifacio. Annales de la Societe Entomologique de France 70: 84–85. Fitton, M.G., & others (1978). A check list of British Insects. Part 4: Hymenoptera. Handbooks for the identification of British insects 11. London: Royal Entomological Society of London. Fitzpatrick, S. (1994). Nectarfeeding by suburban blue tits: contribution to the diet in spring. Bird Study 41: 136–145. http://www.informaworld.com/ smpp/content~db=all~content =a912706660 Fitzpatrick, U., Murray, T.E, Paxton, R.J., Breen, J., Cotton, D., Santorum, V. & Brown, M.J.F. (2007). Rarity and decline in bumblebees. A test of causes and correlates in the Irish Fauna. Biological Conservation 136: 185–194. Forup, M. L. & Memmott, J. (2005). The relationship between the abundances of bumblebees and honeybees in a native habitat. Ecological Entomology 30: 47–57. Fowler, J. & Cohen, L. (1995). Statistics for ornithologists. BTO Guide 22. Thetford: British Trust for Ornithology. Free, J.B. (1993, 2nd edition). Insect pollination of crops. London & New York: Academic Press. Free, J.B., & Butler, C.G. (1959). Bumblebees. The New Naturalist Library. London & Glasgow:

Collins (see http://www.newnaturalists.com/titles/46888/ bumblebees-john-b-free-cg-butler-9780007318124 for availability). Fussell, M. & Corbet, S.A. (1991). Bumblebee habitat requirements: a public survey. Acta Horticulturae 228: 159–163. Fussell, M. & Corbet, S.A. (1992a). Flower usage by bumble-bees: a basis for forage plant management. Journal of Applied Ecology 29: 451–465. Fussell, M. and Corbet, S. A. (1992b). Observations on the patrolling behaviour of male bumblebees (Hym.). Entomologist’s Monthly Magazine 128: 229–235. Fussell, M. and Corbet, S. A. (1992c). The nesting places of some British bumble bees. Journal of Apicultural Research 31: 32–41. Fussell, M. & Corbet, S.A. (1993). Bumblebee (Hym., Apidae) forage plants in the United Kingdom. Entomologist’s Monthly Magazine 129: 1–14. Gammans, N. Banks, B. & Edwards, M. (2009). The return of the native: loss and repatriation of the short-haired bumblebee Bombus subterraneus. British Wildlife 21: 116–118. Gilbert, F.S. (1986). Hoverflies. Naturalists’ Handbooks no. 5. Slough: The Richmond Publishing Company Ltd. Gjershaug, J. O. (2009). The social parasite bumblebee Bombus hyperboreus Schönherr, 1809 usurps nest of Bombus balteatus Dahlbom, 1832 (Hymenoptera, Apidae) in Norway. Norwegian Journal of Entomology 56: 28–31. Goater, B. (1986) British pyralid moths. Essex: Harley Books Goka, K. (1998). Influences of invasive species on native species – will the European bumblebee, Bombus terrestris, bring genetic pollution into Japanese native species? Bulletin of the Biogeographic Society of Japan 53: 91–101. Goka, K., Okabe, K., Yoneda, M. & Niwa, S. (2001). Bumblebee commercialization will cause worldwide migration of parasitic mites. Molecular Ecology 10: 2095–2099. Goodwin, S.G. (1995). Seasonal phenology and abundance of early-, mid- and long-season

Further reading | 111 bumble bees in southern England, 1985–1989. Journal of Apicultural Research 34: 79–87. Goulson, D. (2010, 2nd edition). Bumblebees. Behaviour, ecology, and conservation. Oxford: Oxford University Press. Goulson, D, Hanley, M.E., Darvill, J.S., Ellis, J.S & Knight, M.E. (2005). Causes of rarity in bumblebees. Biological Conservation 122: 1–8. Goulson, D., Hanley, M. E., Darvill, B. & Ellis, J.S. (2006). Biotope associations and the decline of bumblebees (Bombus spp.). Journal of Insect Conservation 10: 95–103. Goulson, D. & Stout, J.C. (2001). Homing ability of the bumblebee Bombus terrestris. Apidologie 32: 105–112. Goulson, D. & Williams, P.H. (2001). Bombus hypnorum (Hymenoptera: Apidae), a new British bumblebee? British Journal of Natural History 14: 129–131. Granero, A. M., Sanz, J.M.G., Gonzalez F.J.E., Vidal, J.L.M., Dornhaus, A., Ghani, J., Serrano, A.R., & Chittka, L. (2005) Chemical compounds of the foraging recruitment pheromone in bumblebees. Naturwissenschaften 92: 371–374 Gurr, L. (1973). Evidence of overwintering nests of Bombus hortorum and Bombus ruderatus (Hymenoptera: Apidae) in New Zealand. New Zealand Entomologist 5: 339–341. Gurr, L. (1975). The role of bumblebees as pollinators of red clover and lucerne in New Zealand: a review and prospect. Proceedings of the New Zealand Grasslands Association 36: 111–122. Hanley, M.E., Franco, M, Pichon, S., Darvill, B. & Goulson, D. (2008). Breeding system, pollinator choice and variation in pollen quality in British herbaceous plants. Functional Ecology 22: 592–598. Hawkins, R.P. (1961). Observations on the pollination of red clover by bees. I. The yield of seed in relation to the numbers and kinds of pollinators. Annals of Applied Biology 49: 55–65. Heinrich, B. (1976). The foraging specialisations of individual bumblebees. Ecological Mono-

graphs 46: 105–66. Heinrich, B. (1979). Bumblebee economics. Cambridge, Mass.: Harvard University Press. Heinrich, B. (1993). Hot-blooded insects: strategies and mechanisms of thermoregulation. New York: Harvard University Press/Springer-Verlag. Herrera, C.M. (1990). Bumble bees feeding on non-plant food sources. Bee World 71: 67–69. Hingston, A. B. (2007). The potential impact of the large earth bumblebee Bombus terrestris (Apidae) on the Australian mainland: lessons from Tasmania. The Victorian Naturalist 124: 110–117. Hodges, D. (1974). The pollen loads of the honeybee. London: Bee Research Association. Holm, S.N. (1966). The utilisation and management of bumble bees for red clover and alfalfa seed production. Annual Review of Entomology 11: 155–182. Hopkins, I. (1914). History of the bumblebee in New Zealand: its introduction and results. Bulletin of the New Zealand Department of Agriculture, Industry and Commerce 46: 1–29. IBRA (1975). Pollen and its harvesting. Reprinted from Bee World 56: 155–158 and 57: 20–25. IBRA (1981). Garden plants valuable to bees. Principal collaborators M.F. Mountain, R. Day, C. Quartley & A. Goatcher. (Available from IBRA: see link 9.13.21, p 107). Ings, T.C., Ings, N.L., Chittka, L. & Rasmont, P. (2010). A failed invasion? Commercially introduced pollinators in Southern France. Apidologie 41: 1–13. Ings, T.C., Schikora, J. and Chittka, L. (2005). Bumblebees, humble pollinators or assiduous invaders? A population comparison of foraging performance in Bombus terrestris. Oecologia 144: 508–516. ITE (1980). Atlas of the bumblebees of the British Isles. Published by the Institute of Terrestrial Ecology, 68 Hills Road, Cambridge CB2 1LA. Intenthron, M & Gerrard, J. (1999). Making nests for bumblebees. 36pp. (Available from IBRA: see link 9.13.21, p 107). Jones, A. (2010) Ghosts in our grasslands. Lessons from

abroad – a look at management of grasslands in Transylvania. British Wildlife 21: 339–344. Kawaguchi, L.G., Ohashi, K. & Toquenaga, Y. (2006). Do bumble bees save time when choosing novel flowers by following conspecifics? Functional Ecology 20: 239–244. Kay, Q. O. N. (1985). Nectar from willow catkins as a food source for blue tits. Bird Study 32: 40–44. Kearns, C.A. & Thomson, J.D. (2001). The natural history of bumblebees. A sourcebook for investigations. Boulder: University Press of Colorado. Kearns, C.A. & Inouye, D.W. (1993). Techniques for pollination biologists. Niwot, Colorado: University Press of Colorado. Kearns, C.A., Inouye, D.W. & Waser, N.M. (1998). Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29: 83–112. Kells, A.R. & Goulson, D. (2003). Preferred nesting sites of bumblebee queens (Hymenoptera: Apidae) in agroecosystems in the UK. Biological Conservation 109: 165–174. Kirk, W.D.J. (2006, 2nd edition). A colour guide to pollen loads of the honey bee. International Bee Research Association, Cardiff. Kleijn, D., Berendse, F., Smit, R. & Gillissen, N. (2001). Agrienvironment schemes do not effectively protect biodiversity in Dutch agricultural landscapes. Nature 413: 723–725. Kleijn, D. & Raemakers, I. (2008). A retrospective analysis of pollen host plant use by stable and declining bumble bee species. Ecology 89: 1811–1823. Kondo, N.I., Yamanaka, D., Kanbe, Y., Kunitake, Y.K., Yoneda, M., Tsuchida, K. & Goka, K. (2009). Reproductive disturbance of Japanese bumblebees by the introduced European bumblebee Bombus terrestris. Naturwissenschaften 96: 467–475. Kosior, A., Celary, W., Olejniczak, P., Fijał, J., Król, W., Solarz, W. & Płonka, P. (2007). The decline of the bumble bees and cuckoo bees (Hymenoptera: Apidae: Bombini) of Western and Central Europe. Oryx 41: 79–88.

112 | Bumblebees Krebs, J. R. & Davies, N. B. (1993). An introduction to behavioural ecology. Oxford: Blackwell Scientific Publications. Kremen, C., & others (2007). Pollination and other ecosystem services produced by mobile organisms: a conceptual framework for the effects of land use change. Ecology Letters 10: 299–314. Kullenberg, B., Bergström, G., Bringer, B., Carlberg, B. & Cederberg, B. (1973). Observations on scent marking by Bombus Latr. and Psithyrus Lep. males (Hym., Apidae) and localization of the site of production of the secretion. Zoon, Supplement 1: 23–30. Leadbeater, E. & Chittka, L. (2005). A new mode of information transfer in bumblebees. Current Biology 15: R447–R448. Levin, D.A. (1990) Sizes of natural microgametophyte populations in pistils of Phlox drummondii. American Journal of Botany 77: 356–363. Lide, D. R. (editor) (2009) CRC Handbook of chemistry and physics. Florida: CRC Press Inc. Løken, A. (1973). Studies on Scandinavian bumble bees (Hymenoptera, Apidae). Norsk Entomologisk Tidsskrift 20: 1–218. Løken, A. (1984). Scandinavian species of the genus Psithyrus Lepeletier (Hymenoptera: Apidae). Entomologica Scandinavica 23: 1–45. Macdonald, M. (2001). The colonisation of Northern Scotland by Bombus terrestris and B. lapidarius (L.) (Hym., Apidae) and the possible role of climate change. Entomologist’s Monthly Magazine 137: 1–13. Macdonald, M. (2003). Bumblebees. Naturally Scottish. Battleby: Scottish Natural Heritage. Macdonald, M. (2006). Recognising the Bombus lucorum group. Bees, Wasps and Ants Recording Society Spring newsletter, pp. 19–20. Macdonald, M. & Nesbit, G. (2006). Highland bumblebees – distribution, ecology and conservation. Inverness: Highland Biological Recording Group. Macfarlane, R.P., Griffin, R.P. & Read, P.E. (1983). Bumble bee management options to improve ‘Grasslands Pawera’ red

clover seed yields. Proceedings of the New Zealand Grasslands Association 44: 47–53. Macfarlane, R.P., Lipa, J.J. & Liu, H.J. (1995). Bumble bee pathogens and internal enemies. Bee World 76: 130–148. Manning, A. (1956). Some aspects of the foraging behaviour of bumble-bees. Behaviour 9: 164–201. Manson, J.S., Otterstatter, M.C. & Thomson, J.D. (2010). Consumption of a nectar alkaloid reduces pathogen load in bumble bees. Oecologia 162: 81–89. Matheson, A. (Ed.) (1996). Bumble bees for pleasure and profit. Cardiff: International Bee Research Association. Meidell, O. (1968). Bombus jonellus (Kirby) (Hym., Apidae) has two generations in a season. Norsk Entomologisk Tidsskrift 14: 31–32. Michener, C.D. (1974). The social behavior of the bees. Cambridge, Mass.: Harvard University Press (Belknap). Miller-Klein, J. (2010). Gardening for butterflies, bees and other beneficial insects. Holywell: Saith Ffynnon Books. See: http:// www.7wells.co.uk. Moore, P. D. & Webb, L.A. (1978). An illustrated guide to pollen analysis. London: Hodder & Stoughton. Morse, D. H. (1978). Estimating proboscis length from wing length in bumblebees (Bombus spp.). Annals of the Entomological Society of America 70: 311–315. Müller, C.B. (1994). Parasitoid induced digging behaviour in bumblebee workers. Animal Behaviour 48: 961–966. Murray, T.E., Fitzpatrick, U., Brown, M.J.F. & Paxton, R.J. (2008). Cryptic species diversity in a widespread bumble bee complex revealed by using mitochondrial DNA RFLPs. Conservation Genetics 9: 653–666. Newsholme, E.A. & Crabtree, B. (1976). Substrate cycles in metabolic regulation and in heat generation. Biochemical Society Symposia 41: 61–109. Newsholme, E.A., Crabtree, B., Higgins, S.J., Thornton, S.D. & Start, C. (1972). The activities of fructose diphosphatase in the flight muscles from the bumble-bee and the role of this enzyme in heat generation.

Biochemical Journal 128: 89–97. Osborne, J.L. & Corbet, S.A. (1994). Managing habitats for pollinators in farmland. Aspects of Applied Biology 40: 207–215. Osborne, J.L., Clark, S.J., Morris, R.J., Williams, I.H., Riley, J.R., Smith, A.D., Reynolds, D.R. & A.S. Edwards (1999). A landscape study of bumble bee foraging range and constancy, using harmonic radar. Journal of Applied Ecology 36: 519–533. Palm, N.-B. (1948). Normal and pathological histology of the ovaries in Bombus Latr. (Hymenoptera). Opuscula Entomologia Supplementum 7: 1–101. Peat, J., Tucker, J. & Goulson, D. (2005) Does intraspecific size variation in bumblebees allow colonies to exploit different flowers? Ecological Entomology 30: 176–181. Plath, E.O. (1934). Bumblebees and their ways. New York: Macmillan. Plowright, R.C. & Jay, S.C. (1966). Rearing bumble bee colonies in captivity. Journal of Apicultural Research 5: 155–165. Poinar, Jr, G.O. & van der Laan, P.A. (1972). Morphology and life history of Sphaerularia bombi (Dufour) (Nematodea). Nematologica 18: 239–252. Pomeroy, N. and Plowright, R.C. (1980). Maintenance of bumblebee colonies in observation hives (Hymenoptera: Apidae). Canadian Entomologist 112: 321–326. Pouvreau, A. (1973). Les ennemis des bourdons. 1. Apidologie 4: 103–148. Prŷs-Jones, O.E. (1976). Aspects of the feeding biology of five Bombus species, and related techniques. B.Sc. Honours Thesis, University of St. Andrews, Scotland. Prŷs-Jones, O.E. (1982). Ecological studies of foraging and life history in bumblebees. PhD Dissertation, University of Cambridge. Prŷs-Jones, O.E. (1986). Foraging behaviour and the activity of substrate cycle enzymes in bumblebees. Animal Behaviour 34: 609–11. Prŷs-Jones, O.E., Ólafsson, E. & Kristjánsson, K. (1981). The Icelandic bumblebee fauna and its distributional ecology. Journal of Apicultural Research 20: 189–97.

Further reading | 113 Prŷs-Jones, O. E, & Crabtree, B. (1983). Measurements of the activity of substrate cycle enzymes in bumblebees. Unpublished. Prŷs-Jones, O.E. & Willmer, P.G. (1992). The biology of alkaline nectar in the purple toothwort (Lathraea clandestina): ground level defences. Biological Journal of the Linnean Society 45: 373–388. Ptacek, V. (1991). Trials to rear bumble bees. Acta Horticulturae 288: 144–148. Pyke, (1978). Optimal foraging: movement patterns of bumblebees between inflorescences. Theoretical Population Biology 13: 72–98. Pywell, R.F., Warman, E.A., Hulmes, L., Hulmes, S., Nuttall, P., Sparks, T.H., Critchley, C.N.R. & Sherwood, A. (2006). Effectiveness of new agri-environment schemes in providing foraging resources for bumblebees in intensively farmed landscapes. Biological Conservation 129: 192–206. Raine, N.E. and Chittka, L. (2005). Colour preferences in relation to the foraging performance and fitness of the bumblebee Bombus terrestris. Uludag Bee Journal 5: 145–150. Rasmont, P., Regali, A., Ings, T.C., Lognay, G., Baudart, E., Marlier, M., Delcarte, E., Viville, P., Marot, C., Falmagne, P., Verhaeghe, J.-C. & Chittka, L. (2005). Analysis of the pollen and nectar of Arbutus unedo as a food source for Bombus terrestris (Hymenoptera, Apidae). Journal of Economic Entomology 98: 656–663 Rau, P. (1924). Notes on captive colonies and homing of Bombus pennsylvanicus de Geer. Annals of the Entomological Society of America 17: 368–81. Rayner, P. (2010). Bumblebees on the Pembrokeshire coast. Natur Cymru 36: 22–27. Redhead, D. (2009). Are bird kills of bumblebees beneficial to bumblebee populations? British Journal of Entomology and Natural History 22: 269–272. Redpath, N., Osgathorpe, L.M., Park, K. & Goulson, D. (2010). Crofting and bumblebee conservation: the impact of land management practices on bumblebee populations in

northwest Scotland. Biological Conservation 143: 492–500. Richards, K.W. (1973). Biology of Bombus polaris Curtis and B. hyperboreus Schönherr at Lake Hazen, Northwest Territories (Hymenoptera: Bombini). Quaestiones Entomologicae 9: 115–157. Richards, O.W. (1927). The specific characters of the British bumblebees (Hymenoptera). Transactions of the Entomological Society of London 75: 233–268. Röseler, P.F. (1985). A technique for year-round rearing of Bombus terrestris (Apidae, Bombini) colonies in captivity. Apidologie 16: 165–170. Saleh, N., Scott, A.G., Bryning, G.P. & Chittka, L. (2007). Distinguishing signals and cues: bumblebees use general footprints to generate adaptive behaviour at flowers and nest. Arthropod-Plant Interactions 1: 119–127. Sawyer, R. (1981). Pollen identification for beekeepers. Ed. R.S. Pickard. Cardiff: University College Cardiff Press. Schmidt-Hempel, P. (1998). Parasites in social insects. Princeton, New Jersey: Princeton University Press. Schmidt-Hempel, P. & Durrer, S. (1991). Parasites, floral resources and reproduction in natural populations of bumblebees. Oikos 62: 342–350. Schmidt-Hempel, R. & Müller, C.B. (1991). Do parasitised bumblebees forage for their colony? Animal Behaviour 41: 910–912. Semmens, T.D., Turner, E & Buttermore, R. (1993). Bombus terrestris (L.) (Hymenoptera: Apidae) now established in Tasmania. Journal of the Australian Entomological Society 32: 346. Sladen, F.W.L. (1912). The humblebee, its life history and how to domesticate it. London: Macmillan. (Republished including The humble bee (1892), in 1989, Woonton: Logaston Press.) Smith, K.G.V. (1969). Diptera; Conopidae. Handbooks for the Identification of British Insects X(3a). London: Royal Entomological Society of London. Spaethe, J. and Chittka, L. (2003). Interindividual variation of eye optics and single object resolution in bumblebees.

Journal of Experimental Biology 206: 3447–3453. Stace, C. (2010, 3rd edition). New flora of the British Isles. Cambridge: Cambridge University Press. Staples, J.S., Koen, E.L. & Laverty, T. M. (2004). ‘Futile cycle’ enzymes in the flight muscles of North American bumblebees. Journal of Experimental Biology 207: 749–754. Stelzer, R.J., Chittka, L., Carlton, M. & Ings, T.C. (2010). Winter active bumblebees (Bombus terrestris) achieve high foraging rates in urban Britain. PLoS One 5(3): e9559. doi: 10.1371/ journal.pone.0009559 Step, E. (1932). Bees, wasps, ants and allied insects of the British Isles. London & New York: F. Warne & Co., Ltd. Storey, K.B. (1978). Purification and properties of fructose diphosphatase from bumblebee flight muscle. Biochimica et Biophysica Acta 523: 443–453 Stout, J.C. & Goulson, D. (2001). The use of conspecific and interspecific scent marks by foraging bumblebees and honeybees. Animal Behaviour 62: 183–189 Stout, J.C., Goulson, D. & Allen, J.A. (1998). Repellent scent marking of flowers by a guild of foraging bumblebees (Bombus spp.). Behavioural Ecology and Sociobiology 43: 317–326. Streeter, D, Hart-Davies, C, Hardcastle, A., Cole, F. & Harper, L. (2009) Collins flower guide. London: HarperCollins. Surholt, B., Greive, H., Baal, T. & Bertsch, A. (1990). Nonshivering thermogenesis in asynchronous flight muscles of bumblebees? Comparative studies on males of Bombus terrestris, Xylocopa sulcatipes and Acherontia atropos. Comparative Biochemistry and Physiology 97A: 493–499. Surholt, B., Greive, H., Baal, T. & Bertsch, A. (1991). Warm-up and substrate cycling in flight muscles of male bumblebees, Bombus terrestris. Comparative Biochemistry and Physiology 98A: 299–303. Svensson, B.G. (1979). Patrolling behaviour of bumble bee males (Hymenoptera, Apidae) in a subalpine/alpine area, Swedish Lapland. Zoon 7: 67–94.

114 | Bumblebees Swynnerton, C. F. M. (1916). Short cuts to nectaries by blue tits. Journal of the Linnean Society (Botany) 43: 417–422. Synge, A.D. (1947). Pollen collection by honeybees (Apis mellifera). Journal of Animal Ecology 16: 122–138. Takahashi, J., Ayabe, T., Miitsuhata, M., Shimizu, I. & Ono, M. (2008). Diploid male production in a rare and locally distributed bumblebee, Bombus florilegus (Hymenoptera, Apidae). Insectes Sociaux 55: 43–50. Thomson, J.D. (1996). Trapline foraging by bumblebees: I. Persistence of flight-path geometry. Behavioural Ecology 7: 158–164. Thomson, J.D. & Chittka, L. (2001). Pollinator individuality: when does it matter? In: Chittka, L. & Thomson, J.D. (eds.) Cognitive ecology of pollination. pp. 191–213 Cambridge: Cambridge University Press. Thorp, R.W., Schroeder, P.C. & Ferguson, C.S. (2002). Bumble bees: boisterous pollinators of native California flowers. Fremontia 30, 3–4: 26–31. Unwin, D. (1978). Simple techniques for microclimate measurement. Journal of Biological Education 12: 179–189. Unwin, D. (1980). Microclimate measurement for ecologists. New York and London: Academic Press. Unwin, D. & Corbet, S.A. (1991). Insects, plants and microclimate. Naturalists’ Handbooks No. 15. Slough: The Richmond Publishing Company Ltd. van Emden, F.I. (1954). Diptera: Cyclorrhapha, Calyptrata (1) Section (a) Tachinidae and Calliphoridae. Handbooks for the identification of British Insects X(4a). London: Royal Entomological Society of London. Velthuis, H.H.W. & van Doorn, A. (2006). A century of advances in bumblebee domestication and the economic and environmental aspects of its commercialization for pollination. Apidologie 37: 421–451. Waddington, K.D. (1977). Studies on the foraging efficiency of bees. PhD thesis, University of Kansas. Waters, J., Darvill, B., Lye, G.C. & Goulson, D. (2010). Niche differentiation of a cryptic bumblebee complex in the

Western Isles of Scotland. Insect Conservation and Diversity 4: 46–52 doi: 10.1111/j.1752– 4598.2010.00101.x Weast, R.C. (1978). CRC Handbook of Chemistry and Physics, p. D-308. Florida: CRC Press Inc. Westrich, P. (1996). Habitat requirements of central European bees and the problems of partial habitats. In: The conservation of bees, edited by A. Matheson, S.L. Buchmann, C. O’Toole, P. Westrich, and I.H. Williams, pp. 2–16. London: Academic Press. Wheater, C.P. & Cook, P.A. (2003) Studying invertebrates. Naturalists’ Handbook No. 28. Slough: The Richmond Publishing Company Ltd. Whitehorn, P.R., Tinsley, M.C, Brown, M.J.F., Darvill, B. & Goulson, D. (2009). Impacts of inbreeding on bumblebee colony fitness under field conditions. BMC Evolutionary Biology 2009, 9: 152. doi:10.1186/14712148-9-152 Williams, C. S. (1995). Conserving Europe’s bees: why all the buzz? Trends in Ecology and Evolution 10: 309–310. Williams, P. (2005). Does specialization explain rarity and decline among British bumblebees? A response to Goulson et al. Biological Conservation 122: 33–43. Williams, P.H. (1982). The distribution and decline of British bumble bees (Bombus Latr.). Journal of Apicultural Research 21: 236–245. Williams, P.H. (1989). Why are there so many species of bumble bees at Dungeness? Botanical Journal of the Linnean Society 101: 31–44. Williams, P.H. (1991). The bumble bees of the Kashmir Himalaya (Hymenoptera: Apidae, Bombini). Bulletin of the British Museum (Natural History) (Entomology) 60: 1–204. Williams, P.H. (1998). An annotated checklist of bumble bees with an analysis of patterns of description (Hymenoptera: Apidae, Bombini). Bulletin of the Natural History Museum (Entomology Series) 67: 79–152. Williams, P.H. (2007). The distribution of bumblebee colour patterns worldwide: possible significance for thermoregulation, crypsis and warning

mimicry. Biological Journal of the Linnean Society 92: 97–118. Williams, P.H. (2008). Do the parasitic Psithyrus resemble their host bumblebees in colour pattern? Apidologie 39: 637–649. Williams, P.H., Araújo, M.B. & Rasmont, P. (2007). Can vulnerability among British bumblebee (Bombus) species be explained by niche position and breadth? Biological Conservation 138: 493–505. Williams, P.H., Cameron, S.A., Hines, H.M., Cederberg, B. & Rasmont, P. (2008). A simplified subgeneric classification of the bumblebees (genus Bombus). Apidologie 39: 46–74. Williams, P.H. & Osborne, J.L. (2009). Bumblebee vulnerability and conservation world wide. Apidologie 40: 367–387 (doi: 10.1051/apido/2009025). Williams, P.H., Colla, S. & Xie, Z. (2009). Bumblebee vulnerability: common correlates of winners and losers across three continents. Conservation Biology 23: 931–940 Willmer, P.G. (1983). Thermal constraints on activity patterns in nectar-feeding insects. Ecological Entomology 8: 455–69. Wilson, J.D., Evans, D.D. & Grice, P.V. (2009). Bird conservation and agriculture. Cambridge: Cambridge University Press. Xie, Z., Williams, P.H. & Tang, Y. (2008). The effects of grazing on bumblebees in the high rangelands of the eastern Tibetan Plateau of Sichuan. Journal of Insect Conservation 12: 695–703. Yalden, P.E. (1982). Pollen collected by the bumblebee Bombus monticola Smith in the Peak District, England. Journal of Natural History 16: 823–32. Yarrow, I.H.H. (1970). Is Bombus inexspectatus (Thalcu) a workerless obligate parasite? (Hym., Apidae). Insectes Sociaux 17: 95–112. Yeo, P.F. & Corbet, S.A. (1995, 2nd edition). Solitary Wasps. Naturalists’ Handbooks. No. 3. Slough: The Richmond Publishing Company Ltd. Zayed, A. & Packer, L. (2005). Complementary sex determination substantially increases extinction proneness of haplodiploid populations. Proceedings of the National Academy of Science USA 102: 10742–10746.

{

this work

Name used in

hypnorum

cryptarum lucorum magnus

Fitton &

{

others (1978)

monticola

lucorum

Alford (1975)

lucorum

{magnus } soroeensis

pascuorum muscorum Bombus (Ps.)

bohemicus sylvestris rupestris vestalis barbutellus campestris

a

and ITE (1980)

}

& Butler (1959)

sylvestris

humilis muscorum smithianus

Yarrow, in Free

Gipsy or Gypsy Cuckoo-bee Four-coloured Cuckoo-bee or Forest Cuckoo bee Hill or Red-tailed Cuckoo-bee Vestal or Southern Cuckoo-bee Barbut’s Cuckoo-bee Field Cuckoo-bee

Moss Carder-bee or Large Carder-bee

Stone, Large Red-tailed or Red-tailed Bumblebee Large Earth or Buff-tailed Bumblebee Cryptic Bumblebee (suggested name) Small Earth or White-tailed Bumblebee Northern White-tailed Bumblebee Broken-belted or Ilfracombe Bumblebee Early-nesting or Early Bumblebee Heath Bumblebee Mountain or Bilberry Bumblebee Tree Bumblebee Large Garden or Ruderal Bumblebee Small Garden or Garden Bumblebee Short-haired Bumblebee Great Yellow Bumblebee Red-shanked Carder-bee Shrill or Knapweed Carder-bee Common Carder-bee Brown-banded Carder-bee

English Namesb

ps ps ps ps ps ps ps ps ps ps pm/pp pm/pp pm/pp pm/pp pm/cb pm/cb pm/cb pm/cb pm/cb

groupingc

Sladen’s

A summary of generic and specific names for British bumblebees, and some standard works in which they appear. Presently used scientific names are in bold type. Generic naming of cuckoo bumblebees follows Williams (1998). English names used since Sladen (1912) are summarised.

bohemicus

solstitialis

ruderarius

subterraneus

Richards 1927

Synonymy Sladen 1912 Bombus: lapidarius terrestris lucorum soröensis pratorum jonellus lapponicus ruderatus hortorum latreillellus distinguendus derhamellus sylvarum agrorum helferanus muscorum Psithyrus: distinctus quadricolor rupestris vestalis barbutellus campestris

a (Ps.) = subgenus Psithyrus. b English name usage follows Sladen (1912); Step (1932); Macdonald (2003); Macdonald & Nisbet (2006); Edwards & Jenner (2009); and BBCT (link 9.13.9). Sladen (1912) and Step (1932) used the name ‘Humble-bee’ in place of ‘Bumblebee’. c Sladen’s grouping of true bumblebees: ps = pollen storer; pm/pp = pocket maker/pollen primer; pm/cb = pocket maker/carder bee.

116 | Bumblebees

Index Page numbers in bold type indicate species in identification keys. Page numbers in italics refer to figures and tables.

agriculture, 29, 62–64, 66, 66–67 annual cycle egg hatching, 17 egg laying and brooding, 17 emergence of queens, 16 end of colony, 23 larval stage, 17 latitude variations, 26 males and queens production, 20–21 mating, 21, 22, 23 nest establishment, 16–17 overwintering of queens, 16 pollen feeding, 18–19 variation, 23–24 winter preparations, 23 Aphomia sociella (wax moth), 36–37 Apis, 1, 10, 15, 47, 48 Bees, Wasps and Ants Recording Society (BWARS), 103, 105 behaviour, 1, 16, 16, 21 foraging behaviour see foraging parasites, impact of, 39, 40, 41 body weight, 99 Bombus cryptarum, 3, 4, 7, 82, 89–90 Bombus cullumanus, 4, 5, 80, 88 Bombus distinguendus, 4, 85, 89, 126 Bombus hortorum, 9, 12–13, 18, 24 colony cycle, 12–13, 24, 26 distribution, 2, 3, 4, 120 females, 79, 84 flight, 11, 23 foraging behaviour, 12, 12, 54–55, 55 fructose bisphosphatase activity, 44, 45 males, 86, 87 mating scars, 98, 98 nesting, 13, 27 queens, 4, 8, 19 tongue length, 12, 12, 58 Bombus humilis, 3, 4, 27, 81, 86–87, 123 Bombus hypnorum, 7, 27, 71, 85, 88, 127 introduction of, 2, 3, 4 Bombus impatiens, 70 Bombus jonellus, 3, 4, 12, 26, 85, 89, 98, 127 Bombus lapidarius, 8, 9, 23, 27, 55, 80, 90 distribution, 3, 4, 120 fructose bisphosphatase activity, 44, 45 Bombus lucorum, 8, 9, 10, 23, 27, 82, 89–90

distribution, 2, 3, 4 foraging behaviour, 45, 54, 55 Bombus lucorum complex, 7, 82–83, 89–90, 124 Bombus magnus, 4, 7, 82–83, 89–90 Bombus monticola, 3, 4, 83, 90, 125 Bombus muscorum, 3, 4, 27, 81, 87, 122 Bombus pascuorum, 8, 9, 13–14, 24, 56, 80–81, 86 colony cycle, 13, 24, 26 distribution, 2, 3, 4, 122 flight, 11, 23 foraging behaviour, 13–14, 52, 55 fructose bisphosphatase activity, 44, 45 nesting, 13, 27 tongue length, 12, 13, 52 Bombus pomorum, 4, 5, 79, 87 Bombus pratorum, 9, 14–15, 32, 82, 90, 98 colony cycle, 15, 24 distribution, 3, 4, 123 flight, 11, 15, 23 foraging behaviour, 14–15, 45, 52, 54–55, 55 nesting, 15, 27 queens, 8, 19 tongue length, 12, 14, 52 Bombus ruderarius, 8, 27, 80, 86, 88 distribution, 3, 4, 121 Bombus ruderatus, 26, 27, 70 distribution, 2, 3, 4, 119 females, 79, 84 males, 86, 87 Bombus soroeensis, 3, 4, 82, 89, 124 Bombus subterraneus, 62, 67, 85, 89 distribution, 2, 3, 3, 4, 126 Bombus sylvarum, 23, 27, 80, 83, 86, 88 distribution, 3, 4, 121 Bombus terrestris, 8–11, 9, 24, 50, 70, 83, 90 colony cycle, 11, 26 distribution, 2, 3, 4, 125 flight, 11, 23 foraging behaviour, 10, 52, 55 fructose bisphosphatase activity, 44, 45 nesting, 8, 10–11, 27 queens, 8, 8, 18, 19 robbing behaviour, 10, 10 tongue length, 10, 52 Bombus (Ps.) barbutellus, 91, 93, 129 Bombus (Ps.) bohemicus, 23, 92, 93, 130 Bombus (Ps.) campestris, 23, 91–92, 93, 129 Bombus (Ps.) rupestris, 5, 23, 91, 93, 128 Bombus (Ps.) sylvestris, 23, 91, 93, 128 Bombus (Ps.) vestalis, 92, 93, 130 Brachicoma devia, 38 Bumblebee Distribution Maps Scheme (BDMS), 103

Index | 117

callosities, 91 carder bees, 13 colour groups, 73, 78, 104 colour patterns, 5–6 commercial colonies, 2, 3, 70–72 'complex' species, 19–20, 35 conopid flies, 38, 38–39, 39 conservation, 62 factors, 64–65, 66, 66–68 native plant species, importance of, 69–70 public involvement, 68–69 crawling, 46 cuckoo bumblebees (subgenus Psithyrus), 3, 5, 22, 23, 34–35, 36 identification, 91–93

gynes, 25

Diptera, 37–39 distribution, 2–4, 3 distribution maps, 119–130 distribution records, 5, 77, 103, 105

land management, 62–64, 67–68 larvae, 17–19, 34 life histories, 9

eggs, 1, 17–18, 19, 20, 35, 98 entomological journals, 106 entomological societies, 105–106 extinct species, 3, 4, 62, 67 fat body, 16, 20, 23, 24, 97 females see also queens; workers cuckoo bumblebees, 91–92 gynes, 25 'true' bumblebees, 79–85 flight, 11, 43–46, 45 flight levels, 21, 23, 23 flower-bee dependency, 58, 60–61 forage sources, providing, 30, 103, 104 foraging flower cues, 53 flower visiting behaviour, 49–50, 55, 55 flowers favoured, 65 foods collected, 49, 54–56 measuring nectar, 50–51 and nectar availability, 46–47 season, 25 techniques, 51–53 timing flower visits, 46–47, 50 foraging economics, 42–43 costs, 43–45 foraging distances, 47–48 profitability, 51 rewards, 46, 49 foraging season, 25–26 foraging studies, 58–61, 96, 98 genital capsule, 74, 85, 92 grooming, 56, 56

honeybees, 1, 10, 15, 47, 48 honeystomach, 23, 47, 50, 50–51, 96, 97 identification, 73, 75, 76, 77, 77 cuckoo bumblebees: females, 91–92; males, 92–93 glossary diagram, 74 Quick Check Key, 78 'true' bumblebees: females, 79–85; males, 85–90 inbreeding, 20 inquilines, 5, 34 introductions, 2, 70–72

males, 1 cuckoo bumblebees, 35, 92–93 foraging behaviour, 48–49 patrolling, 21, 23, 35 production of, 19, 20 reproductive behaviour, 21, 35 role of, 20–21 scent marking substances, 21, 22 'true' bumblebees, 85–90 mating, 20, 21, 23 mating scars, 98, 98 microclimate recording, 59, 103 mites, 39–40 Müllerian mimicry, 5–6, 41 name changes, 7, 115 nectar, 53–54 see also tongue lengths collection, 42, 50, 54–55, 56 competition for, 46–47 and foraging behaviour, 46–47 measuring, 49–50, 99–100 queen's reserve of, 17 robbing, 10, 10 sources, 58–59 sugar concentration, 47, 49, 99–100 as water source, 48–49 nematodes, 40, 41 nest associates, 36–38 nest-box colonies feeding, 32, 32, 33, 33 maintenance of, 32–33 nest-box assembly, 30–31, 31 pollen provision, 32, 33 queen collection and introduction, 31–32 nest collection, 28–29

118 | Bumblebees

nest odour, 28 nesting encouraging, 29–30, 68 nest sites, 27–30 two-cycle species, 12–13 oocytes, 98 ovaries, 35, 97 ovarioles, 35, 98 Parasitellus fucorum, 39, 39 parasites, 36, 38–41 pheromones, 19, 21, 22 pocket-makers, 9, 18, 19–20 pollen, 57 collection, 1, 9, 42, 54–56, 61 egg clumps, 17, 17–18, 35 feeding, 18–19 identifying, 57–58, 60–61, 101 nutritional qualities, 56, 57 obtaining, 101 removing from bumblebees, 102 sources, 56–57 staining pollen tubes, 102–103 pollen-storers, 9, 18–19 pollination role, 1, 2, 4–5, 55–56, 67–68 populations decline of, 4, 5, 62–63 supplementing, 4, 4–5 predators, 5–6, 41, 54 Psithyrus see cuckoo bumblebees publication of research, 106 pupae, 17, 17 queens, 1, 17, 19–20, 20, 21, 24, 25 behaviour, 16, 16 colony establishment, 17–18 emergence, 8, 16 for nest-box colonies, 31–32 nest establishment, 16–17 non-hibernating, 24, 24–25 overwintering, 16, 23 supplementation with, 4 young (new) queens, 16, 21, 23, 24 Quick Check Key, 73, 78 reading resources, 6–7, 108–114 recognizing bumblebees, 5–6 records, distribution, 5, 69, 77, 103, 105 robbing behaviour, 10, 10 Scutacarus acarorum, 39, 40 'simple' species, 19, 20, 35 species loss, 3–4, 20 species, worldwide, 2

spermatheca, 20, 97, 97 Sphaerularia bombi, 40–41, 41 sting sheath, 74, 79 stings, 5, 74 studying bumblebees, 1 see also conservation catching, handling and storing, 94–95 dissecting and measuring, 97–99 distribution records, 5, 77, 103, 105 equipment suppliers, 95, 96, 99, 100 flowers for bumblebees, 103, 104 foraging behaviour, 58–61 further reading, 6–7, 108–114 marking, 96 microclimate recording, 103 nectar measurements, 50–51, 99–100 nest-boxes see nest-box colonies nest collection, 28–29 non-hibernating queens, 25–26 pollen collection and identification, 57– 58, 60–61, 101–103 presenting research results, 106 recognizing bumblebees, 5–6 sampling honeystomach contents, 96 web-based information, 107 substrate cycles, 43–45, 45 synonymy, 7, 115 temperature and emergence of queens, 8, 16 and flight, 43, 45 maintenance of body temperature, 17, 44–46, 45 of nest, 29, 32 threats, 29, 41, 62–64, 66–67 nest associates and parasites, 36–41, 37, 38, 39, 40, 41 tongue lengths, 12, 35, 98, 99 and foraging behaviour, 10, 12, 14, 51, 52 'true' bumblebees, 3, 5, 14, 22, 23 identification, 79–90; females, 79–85; males, 85–90 natural history, 8–15 two-winged flies (Diptera), 37–39 Volucella bombylans, 37, 37–38 water requirement, 48–49 websites, 107 wings, 75, 98 workers, 1, 18–19, 20, 23, 34