464 INTRODUCTION TO EVOLUTION 



out" entirely, the TL chromosome virtually disappearing from the popula- 

 tion. Evidently, then, the ST/TL heterozygote enjoys no advantage over 

 the ST/ST homozygote (or, of course, over the TL/TL homozygote), 

 and hence the TL chromosome eventually disappears. Why then does the 

 TL chromosome persist in appearing with a constant though low fre- 

 quency in the Pifion Flats population (Fig. 20.4)? The answer was 

 given by another experiment in which the TL chromosome was placed in 

 competition with the AR chromosome. In this case an equilibrium was 

 established, at a point where about 80 percent of the third chromosomes 

 were AR, about 20 percent TL, indicating that AR/TL heterozygotes have 

 an advantage over AR/AR homozygotes and TL/TL homozygotes. This 

 advantage of the heterozygotes would be sufficient to keep the TL 

 chromosome from disappearing from the Pifion Flats population, even 

 though ST/TL heterozygotes had no advantage over their respective 

 homozygotes. TL chromosomes "form adaptively valuable heterozygotes 

 with AR but not with ST chromosomes present in the same populations." 



One aspect of natural selection evident in this experiment of Dobzhan- 

 sky's, as well as in that of Reed and Reed (pp. 459-460), deserves especial 

 emphasis. That is the rapidity with which changes are produced. Tradition- 

 ally students of evolution have assumed that natural selection must 

 operate very slowly — that time in the bountiful quantities provided by 

 geologic history must be available for detectable results to be achieved. 

 This was an assumption, but one generally accepted. In contrast to this 

 view are these recent experiments in natural selection, in one of which 

 (Reed and Reed) natural selection accomplished the greater part of its 

 work within two months, and in the other of which natural selection 

 operated with comparable rapidity, to produce changes with changing 

 seasons of the year. Such rapidity and delicacy of control by natural 

 selection came as a distinct, though welcome, surprise. To be sure, these 

 experiments and observations deal with one particular, rapidly breeding 

 organism. But even so, the number of generations required for production 

 of significant change is small. 



We have described Dobzhansky's experiment in some detail, partly 

 because of its trail-blazing nature, partly because it is typical of results 

 obtained subsequently by other investigators. Succeeding investigations 

 have illuminated many facets of heterozygote superiority. What factors 

 are involved? We have seen that temperature is one; food supply has 

 been found to be another. The adaptive values of different gene combina- 

 tions differ with differing foods (differing species of yeasts and bacteria). 



