Systematic melittology: where to from here?

‘Do not go where the path may lead; go instead where there is no path and leave a trail.’ – Ralph Waldo Emerson (1803–1882) Those whose interest resides among the bees are truly fortunate. Among Hymenoptera, only the ants and some of the social wasps can boast of receiving comparable attention, but even these cannot compete if the full enormity of literature on the honey bee, specifically the western honey bee Apis mellifera L., is taken into consideration. Naturally, this is not to say that bees as a whole have been studied evenly, as is more apparent when the same comparison is made between Apis and the remainder of the bees. Most work has gravitated to the conspicuously social lineages, to which Apis may be joined by its close relatives the bumble bees and stingless bees. Nonetheless, melittologists have a historical intensity of research, even excluding Apis, that is to be envied. In particular, a majestic, comprehensive and current account of the global supraspecific diversity is available (Michener, 2007), supplemented and enhanced by excellent regional treatments (e.g. Michener et al., 1994; Pauly et al., 2001; Silveira et al., 2002; Eardley et al., 2010). Such important guides, alongside annotated checklists for large regions or clades (e.g. Cardale, 1993; Gusenleitner & Schwarz, 2002; Moure et al., 2007; Ungricht et al., 2008; Ascher & Pickering, 2010; Eardley & Urban, 2010), render accessible a fauna that might otherwise be challenging to grasp. Hand-in-hand with these works has been a plethora of phylogenetic studies, rooted in a tradition of detailed comparative work, continually enhancing, expanding and revising the body of available data. Furthermore, bees are of such considerable biological interest and importance that melittological systematics has been enriched further by an organic synthesis with diverse and voluminous investigations into their behaviour, ecology and almost any conceivable aspect of their biology. It is no surprise then that bee phylogenetics is such a robust and active field. Hardly a season goes by without a new treatment of some component of melittological diversity being considered from a phylogenetic perspective, and revisions or descriptions of new species are continually appearing. Certainly many genera remain to be investigated carefully, and for those already treated new data, studied either in isolation or more frequently with prior analyses, continue to revise perspectives on relationships. Indeed, the familiar corbiculate bees highlight such dynamics as these tribes have received a disproportionate amount of inquiry. But such intense scrutiny has served to improve our estimates of relationships among these four tribes by gradually accumulating numerous diverse sources of data (Cardinal & Packer, 2007). Morphological and molecular studies have together provided several improvements at both broad (e.g. Roig-Alsina & Michener, 1993; Alexander & Michener, 1995; Danforth et al., 2006a) and more constrained (e.g. Fuller et al., 2005; Larkin et al., 2006; Gonçalves, 2010) scales of inquiry, and these complementary forms of data are repeatedly and wisely synthesized for their maximum explanatory power (e.g. Sedivy et al., 2008; Michez et al., 2009). In contrast to most insect groups, if one takes a macroscopic view across the bees there are few major lineages that have not received some degree of phylogenetic inquiry, and some examples seem particularly noteworthy (this list is by no means exhaustive of papers from the last decade): among families (Danforth et al., 2006a, b; Davis et al., 2010), within Colletidae (Michener, 2002; Magnacca & Danforth, 2007; Packer, 2008; Almeida et al., 2008; Almeida & Danforth, 2009; Kuhlmann et al., 2009), within Halictidae (Engel, 2000a; Janjic & Packer, 2003; Coelho, 2004; Danforth et al., 2008; Patiny et al., 2008; Gonçalves & Melo, 2009; Gonçalves, 2010), within Andrenidae (Engel, 2000b; Ascher, 2003; Roig-Alsina & Compagnucci, 2003; Larkin et al., 2006; Ramos & Melo, 2010), within Melittidae (Michez et al., 2009), within Megachilidae (Engel, 2004a; Praz et al., 2008; Sedivy et al., 2008) and among the Apidae (Engel, 2000c, 2001a, b; Schwarz et al., 2003; Camargo & Pedro, 2003a, b; Rightmyer, 2004; Anjos-Silva et al., 2007; Straka & Bogusch, 2007; Rasmussen & Cameron, 2007; Cameron et al., 2007a; Cardinal & Packer, 2007; Dubitzky, 2007; Kawakita et al., 2008; Schaefer & Renner, 2008; Michez et al., 2008a; Ramírez et al., 2010; Rehan et al., 2010). Although relationships among the families have shifted between studies (e.g. Alexander & Michener, 1995; Danforth et al., 2006a, b; Davis et al., 2010), the families themselves have proven relatively well circumscribed, with certain refinements here and there (e.g. demotion of Oxaeidae, removal of the paraphyletic Anthophoridae, proper association of Ctenoplectridae and Fideliidae): authors generally recognize six or seven families (e.g. Engel, 2005; Michener, 2007) (Fig. 1). Bee phylogenetics is assuredly healthier and more robust today than it ever has been. Representative diversity of the families of bees (after Grimaldi & Engel, 2005). Monophyly of the bees is of little debate. A few authors have considered the bees as polyphyletic (e.g. Robertson, 1904), but melittological monophyly has been recovered repeatedly in studies critically examining myriad forms of data (e.g. Börner, 1919; Michener, 1944; Danforth et al., 2006a; Davis et al., 2010; among many other citations). Contrary to the assertions of some authors (e.g. Börner, 1919; Lanham, 1960, 1979, 1980), the bees are a derived lineage (clade Anthophila) among the Apoidea, a larger group encompassing Anthophila and the grade of apoid wasps formerly known as ‘Sphecoidea’ (or Spheciformes) (Müller, 1872; Ashmead, 1896; Michener, 1944, 2007; Bradley, 1958; Malyshev, 1968; Brothers, 1975, 1999; Königsmann, 1978; Lomholdt, 1982; Alexander, 1992; Brothers & Carpenter, 1993; Alexander & Michener, 1995; Prentice, 1998; Melo, 1999; Engel, 2001b; Danforth et al., 2006a, b; Ohl & Bleidorn, 2006; Pilgrim et al., 2008; Davis et al., 2010). Non-molecular characters supporting Anthophila as a natural group include: the presence of branched or plumose setae (although this also occurs elsewhere in Aculeata); the subantennal sulci internally connected to a fan-shaped sheet of the tentorium; the labrum thickened basally and articulated at the clypeal apical margin; the mesocoxa is dorsoventrally elongate, about equal to the distance from the mesocoxal summit to the hindwing base (hemicryptic in short-tongued bees except melittids); the presence of a mesotibial comb (reversed or vestigial among oxaeines, some melittids and long-tongued bees; Jander, 1976); the presence of the metatibial plate (reversed or frequently vestigial among the bees); the metabasitarsus broader than the subsequent tarsomeres; the absence of the metabasitarsal (posterior) strigil; the cleft pretarsal claws (reversed to simple claws frequently among the bees, and with some minute teeth among apoid wasps); the presence of an alar fenestra cutting across 1 m-cu and not M in the forewing; cu–a shorter than the second abscissa M + Cu (except in Caupolicana) in the hindwing; the mid-dorsal reduction of the female seventh metasomal tergum and formation of lateral hemitergites; the seventh and eighth metasomal sterna in males modified and concealed by the sixth sternum, or with only extreme apical portions exposed; the larva maxilla with a single apical papilla (secondarily reversed in some melittids and apids); the cleaning of the foreleg by drawing it through the flexed midleg at the femorotibial joint; and most notably the consumption by the larva of pollen mixed with nectar or floral oil, or glandular secretions of adults who consume such plant materials [reversed in one derived clade of Trigona (Camargo & Roubik, 1991), and apparently in Lisotrigona and Pariotrigona, which collect mammalian tears, and probably provision their nests with this protein-rich fluid (Bänziger et al., 2009)]. An additional feature discussed at times is the development of a single sperm cell from each spermatocyte (Lomholdt, 1982), but there remains insufficient sampling to determine the validity of this attribute for supporting Anthophila. The sister group to the bees has been controversial but is considered to be the digger wasp family Crabronidae (e.g. Ashmead, 1896; Lomholdt, 1982; Alexander, 1992; Melo, 1999), although monophyly of this family relative to Anthophila is of some question (Ohl & Bleidorn, 2006). It is interesting that among the Crabronidae there exists at least one species, Krombeinictus nordenae Leclercq, that has become independently a ‘bee’, provisioning its brood with pollen and nectar (Leclercq, 1996; Krombein & Norden, 1997a, b; Krombein et al., 1999). It is greatly hoped that investigation of crabronids and other apoid wasps continues to receive intense scrutiny, as it is among these close relatives that the greatest insights into bee origins shall be revealed. Despite the low of available relative to other lineages, has gradually and has and on species (e.g. Michener, 1982; Michener & & Michener, Engel, 2001a, 2006; Camargo et al., Michez et al., 2008a; et al., (Fig. Indeed, the of insect has been across insect and has become in of & Engel, 2005). such data are for clades and for the of molecular estimates of although in the of such is et al., 2010). has generally an for bees and prior to the of in the (Engel, Ohl & Engel, 2007). The of apoid wasps has also in the last greatly estimates for the and a of data for apoid wasp (e.g. 2010; 2004; & Engel, 2006; Ohl & 2009; Ohl & 2010). The corbiculate in (after Engel, the of molecular and work on bees and their close and an of to the for the of new of and the of new is from on and the of continues to be and the same of are of data are of the or are being in other of insect such work has not a intensity among is by to a of and available is not to the of in the and few to the have well within such a even more than most of more than most have to the some are provided here for to be of investigation that have received little from the of bee but that of and or even Naturally, this is not and there are in which melittologists have already into Nonetheless, that by some examples of from such work might receive and become of the among the of in any the of new or and sources of data not the of more Indeed, such new or their more when with those already well and melittologists have not to the to data a or not in not to that or have been or that and molecular studies be in of any other of inquiry. the of new and in otherwise data sources an already robust field. there have been of frequently not have to phylogenetic studies at (e.g. Lanham, & Renner, Michener, Michener & & Jander, 2002; Packer, and these characters are into (e.g. Roig-Alsina & Michener, 1993; Rightmyer, 2004; Cardinal & Packer, 2007; Packer, these have or of familiar such as the of the have not received detailed the that the of the is familiar to Such characters have been discussed in numerous comparative studies (e.g. Michener, and have been to some degree in revisions and of bee (e.g. Prentice, Alexander, 1992; Roig-Alsina & Michener, 1993; Alexander & Michener, 1995; in melittologists have already been such in their to relationships. Nonetheless, the of has been relatively and a of The work by on apoid wasps a of on as well as and there is no to that a comparative specifically on across the bees not the body of data available for more is the of for phylogenetic data. there are examples of characters being from bee Among these the most familiar is the of the and & 2003; Alexander & The of and to be important in supporting the of bees. The diverse family Apidae from other bees in a of four in social and in other families of bees the is (Michener, 2007). several other have received comparative in the and work that is has investigations of the of the and or of the have that is generally with studies on other data 1968; & et al., 1993; 2007; & 2001; & one of the most interesting of data from the investigated the secretions of the Naturally, the and with a of even a single species to the of in such Nonetheless, these studies the of the and for data to of Although the of data from is more relative to the of in some lineages of it has not received among the diversity of may from a in for and in comparative studies, and the are recovered with those of or in which the systematics of such is Among the of in of and are more familiar (e.g. & 2009; also the in on in and is as a of in the of such data among that of and this is a of the relatively of the many (although even among these lineages is well and receive more detailed and the in among the the are more in their and a of for detailed of the of bee the with the many from in and the of bee by Michener studies and the and of have been through the of one has that the of bees can be if a is their to the of data on the of these same species, if the of is made a then the and be there for the these and studies, data from has been phylogenetic and has proven repeatedly to be of considerable (e.g. & Michener, et al., 1978; & Alexander, 1993; Indeed, some of the of to relationships among of bees from the of characters (e.g. A of some works by the plethora of in bee and even (e.g. 2003; & Certainly many have to the of but no one has it as a natural of Although exists between the by and (e.g. Michener, et al., there is also and no for not to as and as the one by and more characters from are in analyses, and it is that this data, in with molecular work, have lineages that from et al., have in the of origins of among & Bogusch, 2007), have to relationships among honey bees & and have among the major of Halictidae In considerable has been for among the apoid wasps (e.g. & b; Ohl & is an important and frequently of and one for which melittologists are not only in the of their data but in the biological the of such A and on the of bees also the for the of molecular and to the and formation of with (e.g. and already for in which this has taken more frequently and (e.g. Danforth, 1999), with work on the development of these same have already been but the not the same this is and by a continually expanding of It is known and that and can of characters for phylogenetic 1992; & 1993; & 1995; & 1996; 1996; & 2007). is assuredly no to as the of bee and is particularly in the of the phylogenetic of and is well among bee although more in the of such across derived from other sources (e.g. & 2007). Such work has been of considerable importance for social (e.g. Schwarz et al., 2007) and (e.g. Patiny et al., 2008; Michez et al., Sedivy et al., the attribute bees is of the consumption of pollen mixed with nectar and floral by the and the of such materials by the a and has been no more investigated in bees than among the corbiculate data its into several corbiculate but to the degree of (e.g. Michener, Prentice, & Carpenter, and biological have been for in a comparative and frequently of such have been in a several characters of discussed and even by & Michener although on the no made to a or comparative provided by in to the of bumble bees, and by et for bees. and Michener et numerous in the of bee and pollen that in the of but these as be for with or Michener a of comparative on the social bee The has been the by in which many of the studies and for an that provided robust with investigations on or molecular data. Naturally, studies continually new that can be not only in a phylogenetic but can be into data on the of nectar in corbiculate bees an clade & alongside the of An of diverse that of for the & Engel, and the of the by & it may the or a larger clade of Apidae relative to phylogenetic on of also by by 1991), & Michener & et and & among many these are a few and a of the of bees is well this Michener is an excellent into this body of as is the work of have been the that characters into (e.g. Michener, & Alexander, 1996; & or on such (e.g. et al., et al., 2001; such as or can also be even in the absence of body to clade the presence of lineages or diversity in (e.g. & 1998; Engel, et al., 2002; & Engel, 2003; & 2006; et al., 2008; et al., 2009; that of bees have been by et al., 2010). and the studies repeatedly the to some of biological and in the of bees, and it is hoped that such characters be more in investigations to the considerable interest in bee and a of be and it be not to these data and melittologists to on comparative investigation remains an and a natural to molecular and other forms of data, the of new are to and this of In the same that provided through which to familiar are to the and Among the new and accessible the provided by and have the greatest et al., 2002; et al., 2003; & 2005; et al., 2005; & is and some examples of their to the of are have enhanced the of not only the by but have the of or (e.g. the An of to both males and and the of this et al., 2005). The minute and frequently of as with or in their natural with such as to revise some of these in a of for et al., 2009). cannot within larger or from particularly large and it is to Among the bees, the of the or of particularly among the more species, a new on otherwise familiar can be provided by Such even can of the the of even the most of characters & even more are the provided by a more of The of such data is of a and the of such the of or even minute and to (e.g. in Such by a or through a have proven to be of considerable in and characters among as well as within lineages (e.g. et al., 2002; et al., 2008; et al., 2009; & 2010). the of with it is also in insect (e.g. 2003; et al., 2006; et al., 2009), and as the continues to it become more Among the bees the of has been only within the of a account of the stingless bee Trigona and an et al., The of to studies of bees the from and of the the on relationships among bees has on Certainly excellent revisions and continue to (e.g. 2003; & 1992; & b; & 2003; & Engel, 2003; & 2003; Coelho, 2004; 2004; & Brothers, 2006; 2008; Rightmyer, 2008; Gonçalves, 2010), but detailed work within genera by the of and investigations and on the among the bees. Such a as it is to that a or is and to the most for any more certain in the of bees have a perspective, such as the of origins of among corbiculate bees, or the of bees from among the apoid the supraspecific of many is more and more the of into relationships is Such with of and where can to and Such have historical (e.g. Janjic & Packer, 2003; Coelho, 2004; Fuller et al., 2005; Ramos & Melo, 2010), of (e.g. Michez et al., Sedivy et al., 2008) and particularly important insights into in social (e.g. Danforth et al., 2003; Schwarz et al., 2007; et al., as more are a natural of the voluminous work being in it be of not to the of from in molecular It goes without that molecular systematics has been a to phylogenetic research, and an on molecular data has alongside the of and other forms of data et al., 2010), which that not is in the of might be et al., its to bees has proven to be of considerable interest and of either in isolation or when with (e.g. et al., 2008; Michez et al., or any of the other But even here to which has become and of the of phylogenetic research, to even are it to more larger and larger of the from more in and for are more to an of have been for insect (e.g. & 2005; Cameron et al., 2009), and such become more in the works there are to that can the and as with the of any of data, the means of be Furthermore, the apparent of of such data to both and more not only its for a group as relatively as the bees least among or even within but for many in the (e.g. Cameron et al., data for bees continue to in the of species as well as the diversity of an from the of today to can be more truly as analyses, either on or on a diversity of the bee in it is in its most (e.g. the relatively large of by Kawakita et al., already bee its greatest when with large of et al., systematics is more than the of and the of those to and is in with the of continue to be to are among the most important that and Gusenleitner & Schwarz for the of a Moure et for bees, Ungricht et for Eardley & for bees and Ascher & for the not annotated as in the are examples of such our current of to and and to a and historical are the at the of any inquiry, and their is more such to their be and it is hoped that may become more to a of the bees, with and to studies of continue to be (e.g. 2001; & 2003; Coelho, 2004; & 2005; 2006; & 2009; Engel, 2010; Gonçalves, 2010; Ramos & Melo, 2010), and continually that is as and as today as ever & Engel, 2007). are to a in the of In the same the from to and from the of to the and of current in and data are this into such as checklists to data from of on (e.g. Ascher & Pickering, 2010), and accessible of data that challenging to to determine (e.g. et al., 2007). In in the of natural descriptions from data (although not characters for are in to and and even the of species revisions and et al., 2010; et al., 2010). or in with their descriptions and among other the of bees. Such are for many of or such as studies important in and natural and the of and with or and it to be considered in any phylogenetic to bees are are challenging and can to but by the are Nonetheless, some to on molecular to and has for some but is on a minute of the that may or may not or among a this has some (e.g. & 2005; et al., 2005; et al., 2006; et al., 2006; 2010). the most sampling of a broader sampling of the than a single particularly that a feature can be in a sampling of either through characters or through the sampling of as many as is more for the and of species than The most is the of and for particularly in which to then to as many other forms of data as can be be behaviour, and diverse data with be to that the systematics of is relative to many of Hymenoptera, a is of the be to more than the and or molecular The to the be It be to as the of in other insect lineages more among bee in which some have already into or to and for other that the for of these have been made by the alongside those who have in our of bees. to and for to this for the to and for their Michener, Ascher, and an provided particularly and for the is a of the of of