THE GROWTH OF GROUPS 133 



deduction that the chances against one kind mating with 

 the other are seven to one. The experiments certainly 

 show that the sexual attraction, decemlineata to pallida, 

 is weak, and that the attraction, decemlineata to decem- 

 lineata, is strong. But they also show that the attraction 

 pallida to pallida is strong, and this last will be an im- 

 portant factor in causing the establishment of a group, 

 when there are two like sports of opposite sexes on the 

 same plant or on neighbouring plants. Tower relates 

 that he found two or three sports together at various 

 times. 



Let us consider also the case of the single sport alone 

 among an unfriendly multitude of decemlineata, what 

 chance has it of finding a mate ? It is said that the chances 

 are seven to one against it. If we express this statement 

 in terms of events we shall see more clearly that out of 

 every eight isolated mutants, occurring in separate places, 

 one will unite with a decemlineata and seven will die un- 

 mated. But if this is true it follows that some hundreds 

 of isolated sports will find a mate in every year in the 

 United States for the mutant occurs in the proportion 

 of at least two hundred to a million of the normal 

 kind, and we also know that a quarter of the number 

 of their grand-offspring from a hybrid union will resemble 

 them in body and germ. 



According to Tower the three factors which combine 

 to make it impossible for mutants to give rise to a species 

 in nature are as follows : firstly, the rarity of the mutants ; 

 secondly, the difficulty they would experience in finding 

 mates ; and thirdly, their delicacy. We have discussed 

 the first two factors, and the third is not valid in the 

 particular case of pallida, since when this mutant was 



