CONTINUOUS VERSUS DISCONTINUOUS VARIATION 



this is an error, and that the species is a vahd, natural unit, even though 

 we cannot define it adequately. The quotation from Bateson, with which 

 this discussion was introduced, expresses this viewpoint nicely. Dobzhan- 

 sky * has said that "Some biologists, lacking familiarity with the subject, 

 have, in fact, fallen into this error ( that species are arbitrarily determined 

 units ) . In reality, no category is arbitrary so long as its limits are made to 

 coincide with those of the discontinuously varying arrays of living forms. 

 Furthermore, the category of species has certain attributes peculiar to 

 itself that restrict the freedom of its usage, and consequently make it 

 methodologically more valuable than the rest." Goldschmidt found that 

 species which he studied have been so sharply, if not necessarily widely, 

 differentiated, that he felt justified in speaking of a "bridgeless gap" sepa- 

 rating species from one another. 



Rassenkreise and Speciation. Dobzhansky believes that this confusion 

 results from the mode of origin of new species. According to the neo- 

 Darwinian conception, a widely distributed species should break up into 

 partially isolated subspecies, which become differentiated by selection of 

 different characters and by genetic drift. Thus a Rassenkreis, or circle of 

 races, is formed, the terminal members of which should eventually become 

 sufficiently differentiated that the sheer weight of difference would raise 

 a sterility barrier. These would now be good species. With such a gradual 

 origin, there would be no sharp break in morphological characters, hence 

 taxonomists, who must depend mainly upon morphological characters, 

 would have difficulty in determining whether the specific level of differ- 

 entiation had been attained. All of this is based upon the assumption that 

 the processes of evolution are the same both on the microevolutionary 

 ( changes within a species ) and macroevolutionary ( origin of new species 

 or higher groups ) levels, a plausible but unproven assumption which has 

 been the center of much controversy. 



Perhaps no more strongly suggestive study of a Rassenkreis has been 

 published than that of Ensatina eschscholtzii, a Pacific coast salamander, 

 by R. C. Stebbins. This species ranges very widely in the mountains of 

 California and the Pacific Northwest ( Figure 102 ) . It is absent, however, 

 from the great central valley of California, a low, hot, arid or semiarid 

 region. Thus the distribution of this species forms an oval, with one seg- 

 ment corresponding to the coastal mountains, the other to the Sierra 

 Nevada's. The two are connected at either end. As might be expected, 

 this is not a homogeneous population, but it is broken up into seven sub- 

 species, with intergrading populations usually being found wherever the 

 ranges of two subspecies meet. It appears probable that the species origi- 

 nated in the northwest, then spread southward along the coastal and 

 Sierra Nevada ranges, breaking up into subspecies along the way. The 

 four coastal subspecies are uniformly dark brown to reddish brown dor- 

 sally, while the three interior subspecies show an increase in orange or 

 yellow spotting from north to south. In southern California, the ends of 



* Dobzhansky, Th., "Genetics and tlie Origin of Species," 1st Ed., Columbia Univer- 

 sity Press, 1937. 



267 



