Friday, November 26, 2010

Aculeata Agonistes: Bombus Squared & Gone?

Herein, I begin my bumbling through the subgenera and species of Bombus in the Edmonton area with what seemed like an easy start: Bombus (Bombus). Thanks to the online resources that Paul Williams has created at the British Museum of Natural History website, I know that only 11 of the 250 or so recognized species of bumble bee belong to the nominate subgenus – less than 5%. Also, selecting the Nearctic Region from the key, I can reduce this to 4 possible species, albeit with 21 different colour forms. However, according to Richards (1978 – see citations at end) Alberta is abuzz with 21 species of Bombus – how can I tell if any of my bees are Bombus (Bombus)?
The best solution would be to ask a specialist on Bombus and preferably one with knowledge of the local fauna. Alas, no one I know will admit to such knowledge. I could try Calgary – there are bumble bee ecologist there - but here the response from the hymenopterists here has been to bugger off and learn them myself, they have more than enough to do already (which is probably what the Calgary ecologists would say). This is called a ‘taxonomic impediment’ (TI) – a well-known problem to overworked taxonomists and frustrated people who need identifications, but seemingly of limited interest to most of the world. For example, Google comes up with less than 60 thousand hits for ‘taxonomic impediment’, but 27 times as many (1,610,000) for ‘bee decline’.
For those who haven’t heard of TI before, here’s a definition from an Australian source that is more than 15 years old: “The taxonomic impediment to progress in the study of biodiversity is linked to a worldwide shortage of taxonomists who can be called upon to identify species, describe species that are new to science, determine their taxonomic relationships, and make predictions about their properties.” And, 15 years later, if you are interested in how Canada is dealing with TI, then you are in luck: last week the Council of Canadian Academies released Canadian Taxonomy: Exploring Biodiversity, Creating Opportunity. You can download the 4mb pdf for free – but to make a long story short: “Job openings in taxonomy have virtually ceased’ ‘Canada’s ranking [in taxonomic publications] dropped from 6th in the 1980s to 14th in the 2000s’ ... [and most relevant to this post] ‘pollinators provide a crucial ecosystem service (via fertilization of crops) to agriculture, yet there is a growing taxonomic expertise gap in pollinator identification.”
Okay, I get the point – the Government may tut-tut about the shortage of taxonomists every decade or two, but better look elsewhere for a solution. Is there a generous specialist from some more enlightened elsewhere with a pro bono bombology streak? Well, one could submit pictures to BugGuide where John Ascher seems to have the talent to name that bee. For example, The Bugwhisperer’s recent spectacular photo of a Bombus melanopygus ménage à trios now resides name and all on BugGuide. The BugGuide route is tempting, but Dr Ascher probably has more than enough to do as it is. In theory, I have some skill with arthropods, small though it may be, and it’s not like I’ll be seeing any new bumble bees before May. If I could make a small, local contribution to the pollinator expertise gap, then it would be a pleasant break from shoveling snow.
So, what lies ahead? Richards (1978) reported that 15 species of Bombus could be found in Aspen Parkland in southwest Alberta. That is a pretty good diversity considering that Donovall & van Engelsdorp (2010) found records for only 17 species of Bombus in the entire state of Pennsylvania and Grixti et al. (2009) only 16 for Illinois (and four of these appear to be locally extinct). I don’t know if a similar diversity survives in the lower elevation (~700 m) former Aspen Parkland of the Home Bug Garden, but this is mostly because we HBGs have been loathe to collect and kill. Instead, we flutter around flowers with our cameras capturing many a mysterious mugshot. I’ve spent much of my spare time the last few weeks perusing BugGuide for matches, but I’ll be buggered if I can identify many pictures with any certainty. That leaves the Laverty & Harder, Discover Life, and BMNH keys (see previous post for links). The authors of these keys, however, expect you to know all the characters they think important, not just those you happen to have photographed clearly.
The keys also are written with the expectation that you to have a bumble bee in hand, and not some faker. Discover Life does have a Bumblebees & Mimics Key with some of the likely mistakes. Robber flies (Diptera, Asilidae, Laphria spp.) and hover flies (Syrphidae) such as Volucella bombylans are the mimics with the mostest – but as with the drone flies discussed in the previous post, the wings, head, and antennae give away their true fly identity. Also, any ‘bumblebee’ eating another insect is not a Bombus. If you live further south, you could be fooled by large carpenter bees in the genus Xylocopa especially the Eastern Carpenter Bee X. virginica, but they do not make their homes in Alberta. We do, however, have several bumblebeeish solitary bees, e.g. some digger bees (Andrena) and mason bees (Osmia). All I can say is that most bees do not resemble Bombus except in the most bee-general way, and those that do are only as large as the smallest bumble bee workers, their fur isn’t as brightly coloured, and they never have orange bands (but a red thorax is possible).
Bumble bees themselves indulge in mimicking each other which leads to convergence in colour patterns between species: mimicry complexes that emphasize the “don’t mess with me” message of the boldly contrasting bands of black and brightly coloured hairs (the skin itself is invariably black). Self-mimicry is probably the primary non-taxonomic impediment to bumble bee identification. So, if one wants to pursue the species identity of the local Bombus, you need to take one of two routes: (1) collect, kill, and pin a good selection or (2) take lots of photographs of individuals from every angle. If one really wants to get serious about Bombus identity, then you need to consider being even more intrusive than the TSA and pulling out the genitalia of the males for inspection. Male claspers are a useful guide to Bombus subgenera and probably to species. Williams at the BMNH provides a rather Andy Warholish gallery of dissected male claspers. I’m sure this is a worthy project, but at my level of understanding, I can’t say that the images are of much help. However, the Home Bug Garden has now become a dangerous place for lazy, shiftless drone bumble bees (the corpses of whom are destined to repose in the Royal Alberta Museum – where irrespective any names I rightly or wrongly put on them, they will contribute to a record of what lived here at this time).
So on to our first subgenus Bombus (Bombus) – a mere 11 species worldwide and only 4 in the Nearctic. Range and unique colour patterns can be used to eliminate three of the four species from my female Bombus. That leaves Bombus (Bombus) terricola looking like some of my bees. A good colour pattern character is that the upper side (tergum) of first and fourth segments (T1 and T4) of the ‘abdomen*’ are covered in black hairs and the second and third (T2, 3) with yellow hairs (remember the skin underneath is black). This contrasting yellow-black pattern, I suppose, accounts for the common name Yellow-banded Bumble Bee. No other local Bombus have this specific pattern, although many are black and yellow. The only local Bombus with the first abdominal segment black belong to the subgenus Psithyrus – cuckoo bees that usurp young colonies of species in other subgenera of Bombus – and these kleptoparasites also have black-haired second abdominal terga.
Among those bees pinned and pictured are a few with a black-yellow-yellow-black abdominal pattern and also a yellow-black thorax and other characters consistent in all three keys = Yellow-banded Bumble Bee Bombus terricola. One of my pinned specimens is a TSA’d male and the genitalia match the BMNH picture. However, on double checking for somewhere I may have gone wrong, I found a species on BugGuide - Bombus occidentalis – that is not present in the Williams’ key. A little searching on other pages of the website discloses that current research suggests that occidentalis and terricola may be one highly variable species, which helps explain the 13 different colour form pictures in the Williams’ key for terricola. This would be a bit annoying if I had colour morphs attributable to occidentalis (especially since Bertsch et al. 2010 think they are both good species) but I don’t. The reason that I don’t is probably because (a) occidentalis was restricted to southern Alberta (Hobbs 1968) and (b) this once common species has essentially disappeared in the last two decades (Evans et al. 2008).
Now that we have an id, what can we say about our bee? Well, Laverty & Harder (1988) list it as an underground nester (places like rodent burrows are good) and with early spring emerging queens. Like all bumble bees, Bombus (Bombus) terricola is considered a long-tongued bee, but all is relative: our humble bee is rather short-tongued. This means that terricola can pollinate only relatively shallow flowers or those that require buzz pollination. But as Laverty & Harder point out “queens and workers commonly bite holes in flowers with long corolla tubes”. Technically, this makes terricola a nectar robber and not necessarily the best pollinator in an ecosystem. Also, terricola seems to be a bit lazy when it comes to foraging from shallow flowers – it prefers to walk when it can, rather than fly.
Plowright & Plowright (1998) noticed that at a site in Ontario where B. terricola and B. ternarius (one of the orange banded bees in the subgenus Pyrobombus) were foraging at the same time, terricola was confined to the milkweed Asclepias syriaca L. – a plant with a dense cluster of upright flowers – and ternarius to the dogbane Apocynum androsaemifoliurn L. – a plant with loose clusters of pendulant flowers. At another site without milkweed, however, both species foraged together in dense patches of dogbane. Their observations and experiments demonstrated that B. terricola is relatively inefficient at flying between flowers, and so tends to visit dogbane only in areas where the plants are dense and closely spaced. Also, Williams et al. (2008) note that Pyrobombus species are noted for their ability to forage from hanging flowers, no doubt giving them an advantage when there is no easier alternative.
All this is interesting, but the place that B. terricola most often shows up in studies is where it isn’t showing up at all – this is one of the species that appears to be disappearing. What could account for the decline? Perhaps as a consequence of emerging early in the spring, Franklin & Sampson (1992) report that the cuckoo bee Bombus (Psithyrus) ashtoni will usurp terricola nests. Cuckoo bees enter a young nest, usually kill the resident queen, and then take over the workers to raise new queen and drone cuckoos. B. (P.) ashtoni, however, is also in decline and may be on the edge of extinction (Evans et al. 2008), presumably because of the decline of its hosts B. (B.) terricola and the related B. (B.) affinis.
Well, if not a cuckoo, then perhaps a predator is knocking off our B. terricola. There are lots of predators and parasites of bumble bees and Morse & Myles (2005) found that B. terricola workers showed no tendency to avoid patches of milkweed with predatory crab spiders. However, mice, shrews, ants, parasitic conopid and syrphid flies, crab spiders, and ambush bugs seem to be problems with which all Bombus must deal, so too the habitat destruction and pesticide encounters that are the suspected causes of much native bee decline. A more insidious cause, and one that seems a likely explanation for the rapid decline of the species in the subgenus Bombus is pathogen spillover (Colla et al. 2006, Rao & Stephen 2007, Otterstatter & Thomson. 2008).
Members of the subgenus Bombus are good pollinators, including of plants that require buzz pollination, and these bees are easy to rear and handle. As a result, Bombus (Bombus) species such as the European terrestris and the North American occidentalis have been mass reared and shipped around the World. As part of these commercial operations, New World bees have been exposed to Old World pathogens and then shipped back to North America for use in greenhouses. Invariably, some escape and intermingle with native bees foraging at flowers around greenhouse facilities. The result is the spread of Old World diseases for which our New World bees have little or no resistance. What a bummer: the first bumble bee that I feel comfortable identifying is now on its way to the Red List of Threatened Species (Mann 2010). I guess I can take some comfort in knowing that, although rare, they are still here around Edmonton. Perhaps some of these survivors are resistant to the introduced diseases and with luck the populations will eventually recover.

*All aculeate Hymenoptera have a ‘wasp-waist’ or petiole that divides the first segment of the abdomen from the remainder. Although there are technical terms for this reorganization of the abdomen, aculeate workers seem happy to use abdomen for the rump and start numbering with the first apparent segment, and so shall I.

Literature Cited

Bertsch A, de Angelis MH & Przemeck GKH. 2010. A phylogenetic framework for the North American bumblebee species of the subgenus Bombus sensu stricto (Bombus affinis, B. franklini, B. moderatus, B. occidentalis & B. terricola) based on mitochondrial DNA markers (Hymenoptera: Apidae: Bombus). Beitraege zur Entomologie 60: 229-242.

Colla SR & 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 SR, Otterstatter MC, Gegear RJ & Thomson JD. 2006. Plight of the bumble bee: Pathogen spillover from commercial to wild populations. Biological Conservation 129: 461-7.

Donovall LR & van Engelsdorp D. 2010. A Checklist of the Bees (Hymenoptera: Apoidea) of Pennsylvania. J Kansas Ent Soc 83: 7–24.

Evans E, Thorp R, Jepsen S & S. Black H. 2008. Status Review of Three Formerly Common Species of Bumble Bee in the Subgenus Bombus. Prepared for the Xerces Society of Invertebrate Conservation [Online]:

Grixti J C, Wonga LT, Cameron SA, & Favret C. 2009. Decline of bumble bees (Bombus) in the North American Midwest. Biol Conserv 142: 75-84)

Fisher RM & Sampson BJ. 1992. Morphological specializations of the bumble bee social parasite Psithyrus ashtoni (Cresson) (Hymenoptera: Apidae). Can. Ent. 124: 69-77.

Hobbs GA. 1968. Ecology of species of Bombus Latr. (Hymenoptera: Apidae) in southern Alberta. VI. Subgenus Bombus, Canadian Entomologist 100: 156-164.

Laverty TM. & LD Harder. 1988. The bumble bees of eastern Canada. The Canadian Entomologist 120: 965-987.

Mann A. 2010. Plight of the bumblebee. Nature News.

Otterstatter M C & J D Thomson. 2008. Does pathogen spillover from commercially reared bumble bees threaten wild pollinators? PLoS ONE 3(7):e2771.

Plowright, CMS & Plowright, R.C. 1998. Floral use by two sympatric bumble bee species (Bombus terricola and Bombus ternarius): efficiency considerations. Canadian Entomologist 130: 595 – 601.

Richards KW. 1978. Nest site selection by bumble bees (Hymenoptera: Apidae)in southern Alberta. Can. Ent. 110: 301-318.

Rao S & Stephen WP. 2007. Bombus (Bombus) occidentalis (Hymenoptera: Apiformes): In decline or recovery? The Pan-Pacific Entomologist 83:360–362.

Williams PH, Cameron SA, Hines HM, Cederberg B & Rasmont P. 2008. A simplified subgeneric classification of the bumblebees (sic) (genus Bombus). Apidologies 39: 1-29.

Sunday, November 14, 2010

Aculeata Agonistes: Bumbling among the Bombus

 Although a few flowers are still defrosting on sunny days, it has been several weeks since the last greedily feeding queen bumblebee has been active. Or rather, I should say ‘bumble bee’ not ‘bumblebee’, because I don’t want to upset any entomologists. Bug people have a simple rule about names with a bug in them: if the common name is technically correct, then keep them separate; if not, merge them. Hence the dragonflies and damselflies in the last post are merged. They are not true flies (= Diptera), but bumble bees and honey bees are true bees (= Apoidea), so keep them separate.
 I admit that I’m not always properly pedantic about bees (bumblebee has ruled on this blog) and sometimes it can get perplexing. For example, what to do about a ‘bee fly’: a fly true, but a bee not. I guess the noun used as a noun trumps the noun used as an adjective, so bee fly it is. Then there is the ‘fly bee’ that adorns Small Farm Canadas otherwise exemplary editorial “Bee-ware” (Nov-Dec 2010, p. 13 in hard copy) about the importance of bees as pollinators and their apparent decline in abundance. Unfortunately, the picture of a bee in a pasture is actually a hover fly, probably the Drone Fly (Eristalis tenax), a mimic of honey bees. A mistake like this really bugs entomologists, but probably makes life easier for the fly.

 Technically, calling a fly a bee is a big time taxonomic failure that you can quantify if you’d like using the Myrmecos Taxonomy Fail Index. Perhaps more importantly, if you can’t tell a hover fly from a bee, then you may not notice if your bees start disappearing. Many flies are pollinators, especially of easily accessed flowers like daisies or apples, but in general bees are what you want to see visiting flowers, at least if you want to maximize your production. Also, many structurally complex flowers require bees for pollination. For example those interesting pea-like flowers that adorn legumes (e.g. alfalfa, clover, peas, beans, lentils, peanuts) usually need a bee that knows how to pry apart the petals to get at the goodies. Then there are those crops with specialized anthers that require buzz-pollination (e.g. tomato, egg plant, blueberries). Nature is full of strange pollination syndromes and bees are integral to many. Even seemingly simple flowers, e.g. those of squash, have bees that specialize on them.
 There are lots of different kinds of bees – perhaps 25,000 different species sharing the Earth with us at this very moment. That seems like a lot to keep track of and, in fact, fewer than 20 thousand have been given names and we know little about more than a few species. Foremost among these favoured few are the members of the genus Apis – barely more than a half dozen species. In much of the World it is one species, the honey bee (Apis mellifera), that we depend on for honey and pollination. Here and there in far parts of the World or here in specialized jobs (e.g. alfalfa bees) people have developed mutually beneficial relationships with other kinds of bees, but for the most part bees go about their business with more hindrance than help from us.
 How much we may be hindering wild bees is a question now being asked, but one that is difficult to answer. A recent study using museum specimens (Grixta et al. 2009. Biological Conservation 142: 75-84) came to the conclusion that half of the bumble bee (Bombus) species historically present in the state of Illinois had become locally extinct or highly reduced. Colla & Harder (2008, Biodivers Conserv 17:1379–1391) also found an apparent general decline in bumble bees in southern Ontario. These results would seem to mirror the declines and local extinctions of bumble bees that have been reported in Europe.
 So what is happening in the Edmonton area? To know whether a bee is doing well or is in decline, one must first be able to put a name on it. In my experience, most entomologists can tell a bee from a fly, and many can make a good guess at family or genus, but usually only those that have specialized on bees can make a good guess at the species identity. This is true even of large, colourful, and noisy bees like bumble bees. A bumble bee proclaims its genus to anyone who cares to look, but keeps its species identity closely guarded.
 So is it possible to identify the Home Bug Garden bumble bees to species – especially without killing them? Well there is a key, Laverty & Harder (1988)*, and it is worth quoting what they think: “The single most useful character for species identification is colour pattern.” (Can. Ent. 120: 966). Sounds like bumble bees and birds may have something in common. However, there is a caveat: “It should be noted that although colour pattern is usually a reliable guide to species, colour variants do occur and misidentification of some specimens is inevitable. We estimate that about 5% of specimens identified by the key will be incorrect... (p. 967).” Also, alas, the key is to the bumble bees of eastern Canada and does not include some of the most common ones in the Edmonton area. However, it does have black & white cartoon-like images of the colour patterns in the 26 species of Bombus covered in the key.

Discover Life also has a key to North American bumble bees. I’ve found this key to be a bit difficult to use and the answers I get don’t always match up to pictures (mostly really sad looking pinned specimens) or distributions (which seem to be more museum collection records than actual distributions). My getting a wrong answer may say more about me than the key, but I’m willing to share the blame. Also, the Discover Life key is clearly under construction and it seems to be improving all the time. For example, it now has Canadian Provinces as one of the characters that can be used to reduce the number of taxa.
Continuing with the Worldwide Web, perhaps the most definitive of all bumble bee keys is the British Museum of Natural History’s Paul Williams’ Lucid interactive key to queens and workers (males are not included) of the World. As with any image-rich interactive key, this requires a fast internet connection and a reasonably powerful computer to avoid giving up in frustration. The complete key contains 1070 diagrammatic images of various colour patterns, but one can select only those in the Nearctic Region – only 220 colour patterns! After that one must have some idea of the subgenus to have much hope of making progress. Subgenera should provide some grist for the Aculeate Agonistes mill and I think Bombus (Bombus) will be the first up – a nice simple, small subgenus. The colour pattern diagrams in Williams’ key and the early, black & white examples in Laverty & Harder seem pretty useful. I’m not much of an artist, but I can handle computer drawing programs and will try work up a key using cartoons and pictures to the Edmonton bumble bee fauna. Here’s a first start comparing two common black & yellow local bumble bees. Any not too caustic suggestions for improvement appreciated.

I'm sure I will be making lots of errors as I bumble through the Bombus. Corrections are always appreciated, but I hope I don't end up on Arthropoda Blog's Friday Fail!

*Laverty TM. & LD Harder. 1988. The bumble bees of eastern Canada. The Canadian Entomologist 120: 965-987.