DISCONTINUOUS VARIATION. 67 



thor, cannot stand, at least for members of the Animal 

 Kingdom in which there is no promiscuous union of the 

 sexes. For instance, if 10 per cent, of the members of 

 a species are thus physiologically isolated, so as to be 

 fertile inter se, but sterile when crossed with any of the 

 other members of the species, and if the interbreeding 

 take place purely according to the laws of chance, then 

 on an average only one-tenth of these 10 per cent, 

 will happen to pair with individuals with which they are 

 fertile, and the remaining nine-tenths will form abso- 

 lutely sterile unions. Thus this physiologically isolated 

 section will never be able to increase in numbers and 

 establish itself. In the case of flowering plants which 

 are fertilised by insects, each of which perhaps visits 

 ten or more flowers in a journey, Fletcher Moulton has 

 shown * that Wallace's objection does not hold, as the 

 dimunition in fertility in such a case is practically neg- 

 ligeable. In his more recent discussion of the theory 

 Romanes somewhat modified his views, and laid more 

 stress on the fact that the mutual sterility may have 

 been slight at first, and have been subject to a gradual 

 development, it acting as a segregating cause in a de- 

 gree proportional to its completeness. However, he 

 makes no suggestion as to why and how such physio- 

 logical incompatibility should arise, other than general 

 ones such as the influence of food and climate, and 

 spontaneous variability of the reproductive system. 



It seems to me that the origin of this physio- 

 logical barrier which so generally exists between 

 species can be most readily accounted for by 

 assuming that in some cases at least it is, as Romanes 

 * " Darwin and after Darwin," vol. iii. p. 165. 



