142 THE VARIATION OF ANIMALS IN NATURE 



I (a) and (b) . Seasonal occurrence and breeding season. 



In short-lived animals the breeding season of a species is 

 usually almost coextensive with the seasonal occurrence. 

 With longer-lived animals a definite season tends to be set 

 aside for breeding. In either case, one of the simplest ways 

 in which varying degrees of isolation may be brought about 

 is by specific differentiation of this season. Besides being 

 simple, separation in this way appears to be important because 

 the seasonal occurrence or breeding period is likely to express 

 the summed effect of the reaction of the organism to its environ- 

 ment. The various small characters by which we separate 

 species can be regarded as the visible expression of differences 

 in growth-rates and in various physiological processes. The 

 species must develop in a different way, and the length of 

 the period necessary to complete the life-cycle is one of the 

 most obvious ways in which developmental differences may be 

 expressed. Where the species live in more or less separate 

 habitats even greater disparities might be expected. It is, 

 indeed, surprising that species are not more often separated by 

 differences in breeding season, but it may be supposed that 

 the fluctuations of the environment make it difficult for any 

 species to have a sharply defined breeding season, and further 

 that the rhythm is much modified to fit in with other periodic 

 features in the environment, particularly the food-supply. 

 The latter factor becomes more important as species diverge 

 more and more widely from one another. Where an insect, 

 e.g., depends on one or a few species of plants there is often a 

 very close correlation between their life-cycles. 



Specific differences in breeding season or seasonal occurrence 

 are extremely common in insects and are not rare in other 

 groups, though complete isolation by this means is probably 

 rather rare. We can only mention here a few typical examples. 

 One of the most striking instances is seen in the Seventeen-year 

 Cicada (Tibicen septemdecim) of the United States (Marlatt, 

 1907). To begin with there are two races, the 17-year race 

 (mainly northern) and the 13-year race (mainly southern). 

 The number of years refers to the time spent as a subterranean 

 nymph. These two main races scarcely differ in structure, 

 but do not appear to interbreed where they meet. Almost 

 every year a brood of each race emerges in some part of the 



