ECOLOGY AND GENETIC VARIATION 



603 



demic mortality (p. 273); from a reduced 

 eflFective interbreeding population in cyclo- 

 morphic species in which the major in- 

 crease in numbers is the outcome of cyclical 

 parthenogenesis (p. 686); or from the 

 establishment of a portion of the population 

 in a new, somewhat isolated locality by 

 territoriality, colonization, emigration, or 

 dispersion. 



The isoagglutinogen in human red blood 

 cells caused by the gene I'" which, in the 

 homozygous condition, produces blood 

 group A, is completely absent in the 

 Indians of Peru, but has a high incidence 

 in the Blackfoot Indians of the northwest- 

 em United States (Strandskov, 1941). It is 

 thought that small emigrating groups, by 

 chance, carried a widely different percent- 

 age frequency of this gene. The gene in 

 this case would seem to be neutral so far as 

 either positive or negative selection is con- 

 cerned. If a large group emigrates, the 

 gene frequencies in the group should, 

 according to chance, be close to those of 

 the original population. As an example, the 

 blood group frequencies of American and 

 West African Negroes are similar. 



If a small population becomes isolated, 

 random combinations of genes may become 

 fixed with consequent nonadaptive differ- 

 entiation (Wright, 1937). Wright (1941) 

 concludes, concerning the fixation of recip- 

 rocal translocations (transfer of the 

 chromosome ends between two chromo- 

 somes), that "such fixation can hardly occur 

 under exclusive sexual reproduction ex- 

 cept in a species in which there are numer- 

 ous isolated populations that pass through 

 phases of extreme reduction of numbers. 

 The most favorable case [for fixation] 

 would seem to be that in which there is 

 frequent extinction of the populations of 

 small isolated localities, with restoration 

 from the progeny of occasional stray mi- 

 grants from other localities." 



Excessive inbreeding in a small popula- 

 tion might result in the fixation of delete- 

 rious genes with consequent weakening or 

 extinction of the isolated population. Ran- 

 dom mutations are more likelv to be de- 

 generative than adaptive (Wright, 1942). 

 If the selection coefficient (p. 649) (s) is 

 0.001, the critical size of the breeding 

 population would be 500. If s is 0.01, the 

 critical number would be only 50. Popula- 

 tions of intermediate size produce condi- 



tions of random variation that act some- 

 what like changes in the direction of selec- 

 tion. "The system of gene frequencies is 

 kept continually on the move and this gives 

 a trial and error process which at times 

 may lead to adaptive combinations which 

 would not have been reached by direct se- 

 lection." Conditions for adaptive evolution 

 are more favorable in populations of inter- 

 mediate size than in small or large ran- 

 domly breeding populations. If a large 

 population is subdivided into numerous 

 small, almost but not completely isolated 

 groups (Fig. 229), random divergencies in 

 gene frequencies and intergroup selection 

 seem to provide the most favorable condi- 

 tions for evolutionary advance (Wright, 

 1937, 1945, p. 416, 1948a; Erickson, 1945; 

 also see p. 407). The breeding system is an 

 adaptive character of the group as a whole 

 and is subject to selection pressure (Ma- 

 ther, 1943; Wigan, 1944). 



Fixation of genetic variations through 

 cyclical parthenogenesis and asexual repro- 

 duction in such species as the aphids, ma- 

 larial parasites, rusts, bryophytes, and 

 pteridophytes, with occasional cross breed- 

 ing between populations in the sexual 

 phase, has evolutionary consequences some- 

 what similar to those in sexual populations 

 partially isolated by geographic or ecologic 

 factors (Banta, 1939a). 



Elton (1930) uses the Arctic fox (Alo- 

 pex lagovus) to illustrate aspects of the re- 

 lation of numbers to s[ene frequency in 

 natural populations. This species is cir- 

 cumpolar in distribution and has two color 

 phases, the white and the blue, which are 

 particularly marked in the winter season. 

 So far as we know, these color phases are 

 not adaptively differentiated, although the 

 color of the species as a whole may be 

 adaptive. (Color phases of the red fox, 

 Vtilpes fiilva, show some indication of dif- 

 ferential survival according to Butler, 

 1945.) In general, the white and the blue 

 phases of the arctic fox are found through- 

 out the rans;e of the species, but on certain 

 islands only the blue occurs, and on 

 peninsular Kamchatka only the white is 

 found. In the extreme north the blue oc- 

 curs in the proportion of one in three hun- 

 dred. Animals of both phases, when they 

 are in the same geographic area, live to- 

 gether, have similar habits, and inter- 



