Increase iri Number of Species 183 



peculiar situations in aquatic animals. Lake Baikal has a number of 

 monophyletic groups of species, or species flocks, especially mem- 

 bers of the Amphipoda, which evidently have evolved within the 

 lake. The African rift lakes Nyasa and Tanganyika each have 

 several species flocks; in particular the cichlid fishes of Lake Nyasa 

 contain flocks of 23, 24, and 101 species each, a remarkable num- 

 ber. At present the sympatric species show considerable ecological 

 segregation, and thus ecological isolation has been cited as the 

 cause of species fission. Brooks (1950), however, has pointed out 

 that there is evidence for assuming that the levels and bottom con- 

 tours of the lakes changed sufficiently to bring about geographic 

 isolation on a small scale within the present lake boundaries. 



Almost all cited instances of species fission due to presumed 

 ecological isolation prove on analysis to be explained better on the 

 basis of fission caused by geographic isolation, followed by range 

 movements resulting in overlapping ranges. A few circumstances 

 do seem to support the concept of ecological factors as isolating 

 mechanisms, as Emerson (1949) pointed out. To be effective, these 

 circumstances need to provide a sharp, sudden break either in 

 behavior or location within the habitat which, in turn, results in the 

 genetic isolation of two populations. Two circumstances affording 

 such conditions are changes in temporal cycles and host relation- 

 ships. 



TEMPORAL ISOLATION 



The type of change suggested here would divide a population so 

 that, in the same locality, part of the population would breed at one 

 time and part of it at another. Furthermore, the time differences 

 would be perpetuated so that the two breeding populations would 

 be continuously isolated genetically. 



Banta and Wood (1928) found a mutant Cladocera which had 

 its temperature tolerance raised 10° C. Such a mutant could con- 

 ceivably give rise to a strain which matured at a different season 

 than the original population and thus set the stage for dividing 

 the population into two time units. Muller ( 1942), however, beheved 

 that characters producing this result would be polygenic in nature 

 and would become balanced in the population. They would not 

 produce two sharply separate modes. Even if this did not occur, 

 presumably the heterozygotes would be intermediate in nature and 

 would destroy any sharp timing differential. 



Insects having rigid developmental patterns covering a year or 

 more do have the possibilities of a time change which would 



