THE ORIGIN OF SPECIES 



probably responsible for the inhibition of synapsis in these regions. He 

 concludes that the differentiation of Drosophila melanogaster and D. sim- 

 iilans from their common ancestor has involved as many as twenty-four 

 chromosomal rearrangements. 



CHROMOSOMAL REARRANGEMENTS AS 

 ISOLATING MECHANISMS 



This raises a very important question as to the extent to which such re- 

 arrangements, interfering as they do with the normal course of meiosis 

 in the hybrids, may constitute genetic barriers between related species. 

 It is evident, of course, that failure of synapsis cannot be imputed to ex- 

 tensive chromosomal rearrangements in every case, for simple lack of 

 homology will also prevent synapsis in a hybrid. But closely related spe- 

 cies must have a considerable amount of homology between their chro- 

 mosomes if there be any validity to the modern genetic approach to 

 problems in evolution. And there are many specific cases in which, like 

 that described above, chromosomal rearrangements produce considerable 

 disturbance of the mechanism of meiosis. 



Examples from Drosophila. The fundamental requirement for genetic 

 interspecific isolating mechanisms is that both of the homozygous types 

 (pure species) should be fully fertile, while the hybrid is largely or en- 

 tirely sterile. Generally speaking, chromosomal rearrangements studied in 

 the laboratory do not meet this requirement: while the heterozygous types 

 do show a reduced fertility, the homozygous rearrangements are com- 

 monly in viable in Drosophila (but often not in plants). Yet that this need 

 not be so is proven by the fact that fully fertile species are known in 

 nature, the chromosomes of which can be shown to be rearranged with 

 respect to one another. In the example above, if D. melanogaster be re- 

 garded as retaining the ancestral chromosomal pattern (an arbitrary 

 assumption ), then D. simulans must be homozygous for more than twenty 

 rearrangements, including certainly both inversions and translocations, 

 and possibly duplications and deletions as well. It is evident then, that 

 while the majority of rearrangements are deleterious, like the majority of 

 gene mutations, nonetheless there is always a possibility that a particular 

 rearrangement or combination of rearrangements may establish a new 

 harmonious genetic system which can be maintained in the homozygous 

 state and which is isolated from the parent type by the sterility of the 

 hybrids. Species pairs such as D. melanogasier and D. simulans prove that 

 this has actually happened in the formation of species now existing. 



Perhaps the most thoroughly studied case in which inversions differ- 

 entiate the chromosomes of species is that of the Drosophila psciido- 

 ohscura group, studied by Dobzhansky and his collaborators, as reported 

 in Chapter 14. For the present, the major fact may simply be restated, 

 that a series of overlapping inversions has made it possible to reconstruct 

 the exact phylogeny of three species, pseudoohscura, persimilis, and mi- 

 randa, together with the details of subspeciation in the first two species. 



282 



