No abstract

The subgenus Onagrandrena was first recognized and described by Linsley and MacSwain (1956) to include those black Andrena that are oligolectic on plants of the family Onagraceae and have pollen-collecting hairs modified to collect the specialized pollen from those plants. The males are more difficult to recognize than the females and most males are very similar to those of Melandrena. Since first described, two species of Onagrandrena have been recognized that have pale vestiture in both sexes. However, the pollen collecting hairs of both of these are of the Onagrandrena type, both sexes have well-formed pronotal angles and lateral ridges, and the males have relatively narrow, long mandibles with reduced or absent subapical teeth. The species of Onagrandrena are very similar and are difficult to tell apart. Populations seem to be relatively isolated in desert locations with habitats amenable to the host plants. This, we believe, has led to a proliferation of species and we can detect slight average differences between populations from different geographic locations within some species. A few of these microgeographic races have been recognized in the literature as subspecies, but the present authors prefer not to formally recognize these races with names. The reader is referred to earlier sections of this revision (LaBerge l967, l969, l97l, 1973, l977, l980, l986, l989; LaBerge and Bouseman 1970, 1987; LaBerge and Ribble 1972, 1975; Bouseman and LaBerge 1979; Thorp, 1969; Donovan, 1977) for details of morphology and for a more complete bibliography of the literature on Andrena. No new morphological terms have been introduced in the present work and the bibliography presented includes only references cited in the text or not listed in earlier parts of the revision. Published locality and floral records are included in the appropriate sections near the end of each species account. Maps showing the known distributions of species (Figs. 2-6) do not have all listed localities spotted on them. Localities that could be located only in a general way, such as county, or could not be found on maps or in gazetteers are omitted. Considerable detailed information is available concerning the floral activity of several species of Onagrandrena in papers by Linsley, MacSwain, Raven and Thorp (1963a and b, 1964) and MacSwain, Raven, and Thorp (1973). These papers also provide brief notes on nesting burrows and an earlier paper by Linsley, MacSwain and Smith (1955) gives details on the nesting biology of a few species of Onagrandrena.

Abstract•

Detailed biological information on bees of the genus Andrena is scant and for most species nothing is known. Several valuable papers concerning Andrena biology are available (Davis & LaBerge 1975;

Abstract. 1. In northern Michigan 72% of the high fidelity dipterous Batesian mimics of bumblebees and vespoid wasps occurred in spring while they still outnumbered their models but before the great majority of fledgling birds left their nests to begin foraging for insects on their own. 2. This extends the known occurrence of this sort of phenology to several additional mimetic species and to an area which is radically different climatically and ecologically from the central Illinois areas in which this sort of phenology was first noted. 3. Our observations confirm the hypothesis that Batesian mimics may be selected to avoid the midsummer period when newly fledged insectivorous birds are abundant but have not yet learned to shun their stinging hymenopteran models. 4. The shortness of the warm season in Michigan reduces opportunities for temporal separation and may have forced the other 28% of the high fidelity mimics to occur when naive birds are abundant. The threat from naive birds was, however, presumably somewhat ameliorated by the abundance of models at that time.