96 ANIMAL LIFE AND SOCIAL GROWTH 



lence such that in the case of the most recent 

 rabbit cycle in Minnesota, the "crash" when the 

 animals suddenly died off in great numbers, was 

 due to the appearance of a particularly virulent 

 form of bacteria, which came when the number of 

 immune rabbits was low. 



Among the climatic factors, variations in 

 temperature, rainfall and humidity are obviously 

 important. The amount of sunlight received, 

 including the amount of ultra-violet, is also im- 

 portant and this is subject to cyclic changes. 

 With mammals, knowledge of the presence of 

 cycles of abundance is least in regard to those of 

 the tropics and greatest concerning the sub-arctic 

 regions. In the former, the climatic conditions 

 are most constant; in the latter, the most varied. 

 There are strong indications that cycles are more 

 likely to occur on the margins of available habitats 

 rather than in more favorable regions; this holds 

 whether the margin is due to coldness or to drought. 

 Under such conditions a small fluctuation in cli- 

 mate may produce a catastrophe for the animals 

 present. 



There is much good evidence that in fishes at 

 least, the numbers of animals present depend on 

 the length of time since there has been a good 

 spawning year. Fishes frequently produce un- 

 believable numbers of eggs; a single female cod- 



