130 I The Process of Evolution 



There have been numerous experiments testing the viabihty of 

 larvae of the melanic form. Ford, working with the moth Cleora 

 repandata, produced backcross broods by mating melanic hetero- 

 zygotes with typical recessive homozygotes. As in all test crosses, 

 a 1:1 ratio of melanics to typicals was expected, and when the 

 broods were well fed there was no significant deviation from this 

 ratio. However, when the caterpillars were starved every day (put 

 under physiological stress), the ratio found was 51 melanic to 31 

 typical (P < .02), a significant departure from the expected 41.5 of 

 each type. Kettlewell exposed six backcross broods to stress and 

 found 108 melanic survivors as opposed to 65 typical individuals 

 (P=.01). 



There can be little doubt that, under certain conditions of stress, 

 the larvae of the melanic moths are better able to survive, but the 

 most recent work on the subject indicates that the situation is 

 more complex than was previously thought. In some recent ex- 

 periments, the expected deficiency of nonmelanic individuals was 

 not found. In other experiments the results showed interbrood 

 heterogeneity: While the offspring from some matings showed 

 a significantly higher proportion of melanics, the offspring from 

 others did not. Furthermore, a study of backcross broods of 

 Biston hetularia raised between 1900 and 1905 showed no surplus 

 of melanics but rather a slight (not statistically significant) defi- 

 ciency of them. It is possible that, early in the evolution of industrial 

 melanism, melanic larvae were not physiologically superior and 

 that this superiority, where it now exists, is a rather recent develop- 

 ment. Perhaps it was only with the easing of the severe selection 

 against melanic adults that melanic individuals increased sufficiently 

 in populations to permit selective reorganization of the melanic 

 genotype to gain the physiological advantage. Just as selection for 

 modifier genes increased the dominance of the genes producing 

 melanism (Chap. 3), so could selection enhance the efiFects of the 

 melanic genes on viability. 



Kettlewell tried experiments to see whether melanic moths tended 

 to settle on dark surfaces and typical moths on light surfaces. He 

 painted the inside of a barrel with alternate black and white surfaces 

 and then released in it an assortment of moths to see which svirfaces 

 they chose. Seventy-seven moths selected the noncontrasting back- 

 ground, while forty-one selected the contrasting background. It has 

 been suggested that this choice is possible because the moth can de- 

 termine the degree of contrast between the scales around its eyes 

 and the background on which it is resting. 



In some areas, melanics are becoming predominant where the 



I 



