576 A CENTURY OF PROGRESS IN THE NATURAL SCIENCES 



sylvania, most of his studies were based upon specimens brought back by numer- 

 ous collectors in the West, for the scientific exploration of western America was 

 in full swing in Cresson's time. Many of the bee species which could be collected 

 in Philadelphia itself went undescribed during Cresson's activities and it re- 

 mained for Charles Robertson to discover and name them in Illinois, mostly 

 during the last decade of the last century and the first decade of the present one. 



Probably because of the activities of Cresson and Robertson in this country 

 and of I'Abbe Provancher in Quebec (many of whose species were incorrectly 

 placed generically and still remain to be elucidated), European authors avoided 

 work on North American forms after Smith's time. They studied material from 

 all other parts of the world, and H. Friese in particular described thousands of 

 species from all faunal regions except the Nearctic. His work extended over a 

 very long period, at least from 1891 to 1935. Friese's counterpart in America 

 was T. D. A. Cockerell, who first published on bees in New Mexico in 1894 and 

 whose last work, on bees from Honduras, appeared in 1949. During this long 

 period Cockerel described bees from all parts of the globe, and he himself col- 

 lected them in many countries. 



In addition to the publications of these workers who have studied bees from 

 all parts of the world, notable contributions in collecting and naming bee species 

 have been made by a number of students whose interests or opportunities have 

 been more localized, for example Tarlton Payment in Australia, E. L. Holm- 

 berg, Padre J. Moure, and C. Schrottky in South America, and P. H. Timber- 

 lake in California. Others have specialized on certain groups of bees, and have 

 often contributed more of lasting value than those whose work has been of a 

 faunal nature. Examples are II. J. Franklin (Bombini), T. B. Mitchell {Mega- 

 chile), P. Bliithgen (Halictinae), and H. F. Schwarz (Anthidiini and Meliponini) . 



The result of all this activity has been a very large number of described 

 species of bees. In the recent catalogue of Hymenoptera of America North of 

 Mexico by C. F. W. Meusebeck et ah, 3,285 species and subspecies of bees are 

 listed. Some of these will prove to be synonyms, but at least as many new ones 

 will probably be described. Assuming that there may be 4,000 species in the entire 

 North American continent and that the other major continents (South America, 

 Eurasia, Africa) average 4,000 additional species each, while in Australia and 

 insular regions another 3,000 species exist, we reach a total of 19,000 species. 

 This is remarkably close to an estimate of 20,000 made many years ago by Friese. 



Phylogeny 



As large numbers of bee species were described, increasing attention was 

 given to their relationships and to the manner in which they may be grouped in 

 a classification. Earlier authors (e.g., Friese in 1895, W. H. Ashmead in 1899) 

 arbitrarily divided bees into those which are parasitic and those which are not. 

 The resulting classifications were highly artificial for they separated such obvi- 

 ously close relatives as Bonibus and Psithyrus. 



An entirely new and carefully considered classification of bees was proposed 

 by Robertson, a Carlinville, Illinois, schoolteacher and botanist, in 1904. Rob- 

 ertson observed that the seventh abdominal tergum of many female and the 

 eighth of many male bees bears a flattened, bare, margined pygidial plate. He 



