26 NEOTROPICAL BEES 



less fit than a local or endemic form. Yet, if it becomes oligo- 

 tropic, and anything happens to the necessary plant or plants, 

 the species is in danger, whereas a loosely adapted form suffers 

 no inconvenience. 



Dr. J. C. Willis has recenth^ written much on the distribution 

 of flowering plants, urging that the most local species are gener- 

 ally those of most recent origin, whereas the older they are, the 

 wider will be the range. Naturally there are many exceptions, 

 such as that of the Sequoia species in California. Among the 

 bees, it is doubtless true that the precinctive species are nearly 

 always endemic, — that is, they have arisen in the general region 

 where we find them, and have not been formerly much more wide- 

 spread. Whether the wide-spread species are necessarily old, 

 may well be another matter. A species with strong flight or 

 migratory instincts ma}^ spread over a very large area in a short 

 time, as we see in the case of introduced insects. The evolution, 

 or as it were liberation, of a species capable of spreading widely 

 from a relativel}- local type may well follow the natural lines of 

 advantage at the boundaries of the original species-area; but it 

 is probably more common for the new adaptations to lead to 

 other species just as specialised, but living different lives. Doubt- 

 less the very resistance to modification, the stabilitj^ of type seen 

 in such butterflies as Euvanessa antiopa, Pyrameis cardui and 

 atalanta, has had something to do with their wide range. They 

 could not follow the path of local adaptation, being without the 

 necessary variability; hence there was no conflict between the 

 two different tendencies. Thus it appears that the range of a 

 species has more to do with its variability than its age. There 

 are, indeed, especially among the bees, some species said to be 

 very wide-spread and very variable. Whenever I have been able 

 to get good series of such forms from different localities (e. g. in 

 Crocisa and Xylocopa) I have found that they were composite, 

 and consisted in fact of numerous local species which had been 

 lumped together. In the older collections, with frequently 

 slight indications of localities, the long series gathered together 

 from many places give the impi'ession of single species showing 

 an enormous range of variation. 



