Changes in Populations 139 



likely to be a significant factor. Seven collectors once captured 400 

 snakes on an island in 5 hours, and three collectors at another time 

 took 234 snakes in 4 hours. The juvenile-adult shift is toward un- 

 handed individuals on all the islands studied, indicating a system- 

 atic pressure rather than random drift. Therefore, by a process of 

 elimination, selection seems most likely to be responsible for the 

 change in pattern-type frequencies. 



To prove that selection has taken place, one does not have to dis- 

 cover its mode of operation. However, it is interesting to speculate 

 on the factors producing the observed situation. To the human eve, 

 unhanded snakes are cryptically colored on the flat limestone rocks 

 of the island peripheries and banded individuals are very conspicu- 

 ous. It is likely that banding would help to break up the outline of 

 the snakes in their more typical, less uniform swamp habitat. There 

 are visual predators present which will eat snakes (gulls, herons, 

 hawks, etc.), and man kills many with firearms. 



The selective force is obviously very strong. If the pattern spec- 

 trum is divided arbitrarily into two halves (banded and unhanded), 

 the snakes in the banded half have only about 25 percent of the 

 chance of survival of the unhanded half ( 5 for the banded phenotype 

 equals approximately .75). This raises the question why, with such 

 heavy selection, any banded individuals at all remain in the popu- 

 lation. The answer appears to be that migration brings a steady in- 

 flux of genes for banding into the gene pools of the island popula- 

 tions. Snakes have been observed swimming far from land on many 

 occasions, and the distance from the shore to the islands is not too 

 great to be spanned by migrating individuals. Thus the mainland 

 populations form a reservoir of banded individuals, some of which 

 periodically reach the islands. The resultant interaction between 

 selection and migration has produced a situation unusually amen- 

 able to analysis. 



Chromosomal Polymorphism in Drosophila 



Classic examples of microevolution are found in the work pioneered 

 by Dobzhansky on inversion frequencies in some 30 species of Dro- 

 sophila that show polymorphism in chromosome type. This type of 



Fig. 7.5 I ( see opposite page ) Natural selection in water snakes on 

 the islands of Lake Erie, as shown by the differential in frequencies of 

 banding types in young and old snakes. 



