REDUCTION OF VARIABILITY. 121 



uals which produce off-spring to all the individuals is estimated, 

 it is found to be small. Moreover, the niunber of individuals of 

 a given group nearly always vaiies within wide limits. In 

 extreme cases, such as that of parasistes, it is the rule that one 

 or a few individuals found a colony which may attain to enor- 

 mous niunbers, before it dies out. It is easy to see, how such 

 organisms like malaria plasmodia must form genetically pure 

 groups in very short time. When conditions are favourable, any 

 group of organisms expands in numbers. The proportion of in- 

 dividuals having of f-spring is large. When a group gets into 

 adverse conditions, it may die out altogether and leave only 

 daughter-colonies, or a few individuals may survive. In the lat- 

 ter case, the potential variability of the new population must 

 be a fraction of the old one, no matter if the organism is self- 

 fertilizing, asexually propagating, or aUogamous. 



In nearly every wild animal, which is for some reason or 

 other subject to careful observation, it is found that the num- 

 bers in which it exists vary considerably from generation to 

 generation. In not a lew instances a family of rapidly increas- 

 ing animals, e.g. moths, periodically fluctuates in numbers, 

 because of the influence of a parasite. In a few years the moths 

 will multiply amazingly, imtil the parasites have caught up 

 with them, so that in the next generation hardly any moths are 

 left, and the parasites almost die out because of the dearth of 

 hosts. The moths left are relatively free to multiply for a few 

 generations, imtil the parasistes catch up with them again, 

 and so on. In both animals the periodical catastrophe leaves 

 only a very small number of individuals, a very small fraction 

 of the number in the preceding generation. After each catas- 

 trophe the total potential variabihty of both animals must be 

 considerably reduced. 



If crossing be excluded, any group of animals or plants 

 gradually becomes pure for its genot3^e, and consequently for 

 its phenotype, its characters, even without any selection. We 

 have tried to show in the chapter on Variation, how the varia- 

 tion we see in groups of organisms is due almost exclusively to 



