Family Colletidae

(5 subfamilies, 5 tribes, 54 genera, ~2000 spp.)

Colletidae is widely considered to be the most primitive bee family (Michener 1944, 1979, 2000; Alexander & Michener 1995). Colletids are a diverse group ranging from small, slender, relatively hairless bees (such as Euryglossinae) to large, robust, hairy bees (such as Diphaglossinae). Females carry pollen either externally, in well developed trochanteral and femoral scopae (Colletinae and Diphaglossinae) or scopae formed by sparse hairs on the hind legs and long hairs on the first abdominal sternum (Xeromelissinae), or internally (in Euryglossinae and Hylaeinae) and there is only one known group of cleptoparasites (in Hawaiian members of the Hylaeus subgenus Nesoprosopis; Daly & Magnacca 2003). Floral relationships in Colletidae range from polylectic (generalist) to oligolectic (specialist), with some taxa showing very narrow host plant preferences (Houston 1999, Krombein et al. 1979, Michener 1966, and Scott 1996). Colletids are most abundant and diversified in the temperate parts of Australia (>50% of all Australian bees are colletids! [Michener 1965a, Cardale 1993]) and South America. Among the five subfamilies, only Colletinae are potentially paraphyletic (Michener 2000). All colletid bees (excluding males in some derived genera) possess a bifid, rather than acutely pointed, glossa.

The results from Alexander & Michener (1995) did not strongly support monophyly of Colletidae, even though there are a number of potential synapomorphies for this family. The cellophane-like nesting cell lining (Hefetz et al. 1979, Batra 1980, Espelie et al. 1992) has been considered to be the strongest evidence for the family monophyly for being an unique and unreversed character found among colletid bees. There are morphological characters in the colletid mouthparts are also unique among bees, but many of them are controversial because their interpretation can drastically change depending on how those characters are polarized (see Alexander & Michener 1995, Michener 2000 for discussions). While the bifid glossa is seemingly a good character for uniting the diverse bees in this family, questions remain about whether to consider the bifid glossa a symplesiomorphy (shared with spheciform wasps; Ulrich 1924) or a synapomorphy of Colletidae (Alexander & Michener 1995). Torchio (1984) found a developmental character that can also be interpreted as a synapomorphy for Colletidae. The most recent evidence for the monophyly of this family was given by a molecular character: the presence of a unique intron in the F1 copy of EF-1alpha in all colletids sampled, but not in Stenotritidae or any other bee family (Brady & Danforth 2004). In summary, while questions remain about colletid monophyly, accumulating evidence from morphology, behavior, and molecular data are all indicating that Colletidae is likely monophyletic and probably not the basal branch of the bees.

Eduardo Almeida (Cornell University, Ithaca) is currently working on a combined molecular and morphological data set to try to resolve the higher level relationships among the colletid subfamilies, and looking for evidence for monophyly of the family, subfamilies, and tribes. Preliminary results confirm that Colletidae is monophyletic but the subfamily Colletinae is mostly likely paraphyletic, and will not be maintained as delimited by Michener (2000). In addition, the relationships among the Australian and South American genera of the formerly recognized tribe Paracolletini will be analyzed. Karl Magnacca (Cornell University) is working on a molecular data set for the generic and subgenetic relationships within Hylaeinae. Lawrence Packer (York University, Toronto) is developing molecular and morphological data sets for the relationships within the subfamily Xeromelissinae.

Download

Data sets:

none