Increase in Number of Species 185 



years of emergence of the more important broods form an irregular 

 seventeen-year clock composed, in succession, of two eastern, two 

 western, one central, one widespread, three central, one western, 

 and a widespread brood (Fig. 80a). Here are the elements of 

 at least 11 interbreeding populations isolated from each other by 

 time and perhaps each possessing the potential of developing into 

 a separate species. 



In the thirteen-year species of M. tredecim there are only two 

 large broods, 19 and 23, separated in time of emergence by nine 

 and four years (Fig. 80b). Only one other abundant brood is 

 known, 22, confined to a small southern area. Although the num- 

 ber of broods is smaller in this species, there seems an excellent 

 probability that at least broods 19 and 23 are sufficiently isolated 

 in time to evolve eventually into distinctive species. 



The results of these implications are as yet unknown. The different 

 broods may have come into existence because of local freakish 

 weather, but how old they are and how much they differ genetically 

 remains to be discovered. The taxonomic complexities of individual 

 broods are only now being analyzed extensively (Alexander and 

 Moore, 1958), and this analysis will undoubtedly shed much light 

 on their evolution. 



Species fission by cyclic isolation of another sort is suggested 

 from the study of sawflies and crickets (Ghent and Wallace, 1958; 

 Bigelow, 1958). A monophyletic cluster of four species of nearctic 

 sawHies belonging to the genus Neodiprion illustrates this type of 

 temporal isolation. All four species feed on pine, have a short adult 

 life span, and a rigid life history of one generation a year. Three of 

 the species {N. virginianus, N. taedae, and N. pratti) overwinter as 

 fully mature larvae, and the adults emerge and mate in spring, 

 whereas the only other species of the cluster (IV. maurus) over- 

 winters in the egg stage, the larvae pupate in late summer, and 

 the adults emerge and mate in the autumn (Ross, 1955). Most 

 other species in the genus overwinter as larvae, indicating that this 

 is the ancestral condition for the N. virginianus cluster. At present 

 N. maurus occurs in the same area and on the same host species as 

 a subspecies of N. pratti. The circumstances suggest very strongly 

 that the species N. maurus arose as a colony of IV. pratti (its sister 

 species) in which for some reason the larvae pupated in autumn 

 instead of going into hibernation and that the new seasonal timing 

 became established in succeeding generations. The species Neodi- 

 prion hetricki, belonging to another group of four species, exhibits 



