Abstract

Michener, C. D. (Departments of Entomology and of Systematics and Ecology, University of Kansas, Lawrence, Kansas 66045) 1977. Discordant evolution and the classification of allodapine bees. Syst. Zool. 26:32-56.-The allodapine bees constitute a part of the tribe Ceratinini. The immature stages have evolved a variety of adaptations to life in a common burrow instead of each isolated in its separate cell. This is a unique environment for bee larvae. These adaptations, which result in greater morphological diversity among allodapine larvae than among all other bee larvae together, involve often nonhomologous structures and behavior which serve for supporting the young in the nest burrow and for manipulating larval food masses, as well as various structures which appear to be sensory in nature, and are presumably related to the frequent interactions among individuals. In contrast, most adult allodapines are monotonously similar; most of the exceptions are social parasites which live in the nests of related nonparasitic species. Larvae of the parasites closely resemble those of their nonparasitic relatives; in this case it is the adults which have diverged sharply, the various species or groups of parasites having independently entered an environment that is unusual for adult bees. They rarely or never visit flowers and in effect replace the egg laying female (queen) of the host with which they live. A cladogram was developed; then four mutually exclusive sets of variables (larval, pupal, external adult, and male genitalic) were superimposed on the cladogram. These sets included all of the known and readily codified variables, a total of 269, not merely the few useful in preparing the cladogram. Presumably because of the very different and independent environmental factors impinging upon these sets of variables, phenetic classifications based on them would be exceedingly different. Other sets of variables (biochemical, internal structure, etc.) may well convey equally different information about resemblances. A phenetic classification, based of necessity only on known sets of variables, is always likely to be altered when new sets become accessible. A cladistic hypothesis should be more stable. To base a classification on a cladogram alone, however, seems undesirable. In the allodapines it would result either in lumping very dissimilar forms or in placement of very similar forms in different taxa; these procedures are contrary to usefulness and convenience as well as to our biological knowledge other than the sequence of cladistic branching points. [Phylogenetic relationships; phenetic classification; discordant evolution; allodapine bees.] Systematic and behavioral studies of allodapine bees (Michener, 1962 to 1976 except 1970c; Michener and Scheiring, 1976) have shown that there are striking variations in degree of resemblance between different stages. Thus two species may be very different in larval characters but similar in adult characters while two other species may be similar in larval features but strikingly different in the adult stage. This means that rates of morphological change have been different for different stages-sometimes larvae have di' Contribution number 1598 from the Department of Entomology, University of Kansas, Lawrence, Kansas, U.S.A. This is a product of National Science Foundation Grant GB-37301X. verged strongly while adults have not and sometimes the reverse has taken place. This paper concerns such discordant evolution, not only of larval and adult external characters, but also of pupal and male genital characters, together with a kind of adaptive radiation, primarily of larval characters. Allodapine bees are members of the tribe Ceratinini and are most richly developed in Africa, but found throughout the Old World tropical and austral regions, northward well into the temperate zone in Asia Minor and into the southern edge of that zone across the lowlands of Asia from India eastward. Like their non-allodapine relatives, the cosmopolitan Ceratina (in-

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call