4Q O. E. PLATH. 



207-215) believes that the representatives of the genus Psithyrus, 

 because of their "perfect homogeneity," could not have come from 

 several B ran us species, and, on the basis of structure, tries to show 

 that they have all arisen from a single form which was closely 

 related to Brcinus mastrucatus Gerstsecker. He believes that in 

 the course of time Ps. barbutellus, cauipcstris, and vest alls became 

 more differentiated from this original type than the other PsitJiyrus 

 species. Hoffer ('88) and Sladen ('12), on the other hand, be- 

 lieve that there is no evidence for such an assumption, and the last- 

 named author suggests that such resemblances as regards coat- 

 color, etc., between several species of Psithyrus and their Brcinus 

 hosts are " clearly attributable to mimicry or exposure to the same 

 conditions of life and not to ancestry." If we accept this explana- 

 tion, it is difficult to account for those cases where there is not the 

 slightest similarity between Psithyrus and host e.g., Ps. campcs- 

 tris and Brcinus agronim Fabricius. 



It is no doubt true, as Schmiedeknecht ('07) has pointed out, 

 that in the case of the Bremidse other criteria, such as length of 

 head, are of greater taxonomic importance than coloration. That 

 the latter is nevertheless of considerable value in determining rela- 

 tionship is indicated by the striking similarity between many of the 

 species of the various English and American Brcinus and Psithyrus 

 groups, as they have been established by Franklin ('12) and Sladen 

 ('12, p. 152), in the one case on the basis of structure, and in the 

 other on the basis of habit (pollen-primers, pollen-storers, and 

 carder-bees). 



Using coloration as a basis, the various European and American 

 Psith\ri can be divided into three groups. The first of these, 

 which may be called the rupcstris group, is restricted to the Old 

 W T orld. The members of this group strikingly resemble a number 

 of Brcinus species, likewise restricted to the Old World, with one 

 of which, B. mastrucatus, they are very similar in structure, as 

 Perez ('83) has shown. Ps. rupestris, the chief representative of 

 this group, is parasitic on B. lapidarius, from which it is difficult 

 to distinguish it. 



theory, we should expect the parasitic Hymenoptera to be the ancestors of 

 their hosts, a view which, as Saunders ('03) has pointed out, is difficult to 

 harmonize with certain well-established facts. 



