148 THE AMERICAN NATURALIST [ Vol. LII 



for about the same period of time ; i. e., all life histories 

 involve about the same period of activity and rest. 



An inspection of the table shows that the time of reach- 

 ing the adult stage varies for the different species, so that 

 there is a general change of spiders in the adult stage as 

 the season progresses. This is all that seasonal succes- 

 sion can mean under any conditions ; the fact that the eggs 

 or other young stages can not be identified or their loca- 

 tion is unknown does not change the character of phenom- 

 enon in any locality where the species are resident. 



The causes of the succession of species may be roughly 

 summarized as follows : Species differ in the time in the 

 annual climatic rhythm at which development begins, in 

 the time of occurrence of dormancy and in the conditions 

 necessary to break it up, in threshold of development rela- 

 tive to several climatic factors, in velocity of development 

 relative to several climatic factors, food, etc., and in size 

 and total energy expended. These may be taken up one 

 at a time. 



Considering differences in the time in the annual cycle 

 at which development begins, as a factor in seasonal suc- 

 cession, we must notice first that this can be a controlling 

 factor only where there is no dormancy in the life history 

 or where the available total of temperature, moisture, 

 light, etc., above the thresholds is just enough to produce 

 one generation per year and not to permit of a gradual 

 moving of the time of appearance to an earlier date each 

 season, during several successive long seasons. The test 

 of this would come in the migration of agricultural pests 

 which are arrivals in localities where the growing season 

 is longer. There appear to be no easily available facts, 

 and for the present this type of maintaining a definite 

 time of appearance is to be regarded as a theoretical pos- 

 sibility. The fact that the life histories of various ani- 

 mals which have been known to migrate extensively into 

 new territory appear not to be accelerated indicates that 

 dormancy may control appearance and thus time of be- 

 ginning development may be a secondary consideration. 



