.336 AN EMBRYOLOGIST'S VIEW 



ently normal hybrids, and yet they remain distinct even when 

 sympatric. In this situation there would be no adaptive advantage 

 to restricting hybridization, and hence the isolating mechanisms 

 that do prevent gene interchange could not have been perfected 

 by the natural selection of genes that prevent hybridization. 



Second, in areas of overlap where inferior hybrids are formed, 

 we must remember that hybridization must be frequent if natural 

 selection is to increase the frequencies of gene differences that 

 reduce crossing. Consequently, as isolating mechanisms become 

 more effective, the ability of selection to augment them will 

 steadilv decrease. 



Third, we must not assume that the wastage of gametes is 

 always a disadvantage to the species. There are some special 

 situations where the wastage of gametes might be a real advan- 

 tage to the population as a whole. As a hypothetical case let us 

 consider a species whose numbers are held in check by predators 

 and available food. If the source of predation is then removed, 

 there will be greater competition for the available food. In some 

 species severe competition may result in 100 per cent mortality. 

 In this, an unusual situation, some wastage of gametes would 

 be an advantage. 



There are some observations that have been cited in support 

 of Dobzhansky's thesis, such as those of Koopman ( 1950 ) . He 

 kept Drosophila pseudoobscura and pcrsimilis together in a 

 population cage and found that it was possible to augment the 

 existing isolating mechanisms by removing all the hybrids that 

 formed. In other words he was able to select for mating prefer- 

 ence. This is not an unexpected result. It is highly probable that 

 all genetic differences that serve as isolating mechanisms are con- 

 sequences of natural selection (though in most instances I feel 

 they are not selected for as isolating mechanisms per se). The 

 I act that one can select against hybrid formation does not prove 

 that an analogous situation will occur in nature. It is possible to 

 select in artificial situations for all sorts of things that are not 

 selected for under natural conditions. 



There are now a number of reports of species pairs which 

 overlap in their ranges and show greater genetic isolation from 



