CMiAP. XIV. RECAPITULATION. 413 



graclutited .steps. There are, it must be admitted, cases of 

 special diiriculty oppo.se(l to tlie theory of natural selection ; 

 and one of the most curious of tlicse is the existence of two or 

 three defined castes of workers or sterile female ants in the 

 same community ; but I have attempted to show how these 

 dilKculties can be mastered. 



With respect to the almost universal sterility of Species 

 when first crossed, which forms so remarkable a contrast with 

 the almost universal fertility of varieties when crossed, I must 

 refer the reader to the recapitulation of tlie facts given at the 

 end of the eighth chapter, which seem to me conclusively to 

 show that this sterility is no more a special endowment than 

 is the incapacity of two distinct trees to be grafted together; 

 but that it is incidental on differences confined to the repro- 

 ductive systems of the intercrossed species. We see the truth 

 of this conclusion in the vast difference in the results of cross- 

 ing the same two species reciprocall}- — that is, when one 

 species is first used as the father and then as the mother : 

 analogy from the consideration of dimorphic and triraorphic 

 plants clearly leads to tlic same conclusion, for when the forms 

 are illegitimately united, they yield few or no seed, and their 

 offspring arc more or less sterile ; and these forms of the same 

 undoulited species dlfler in no respect from each other except 

 in their reproductive organs and functions. 



Although the fertility of varieties when intercrossed and of 

 their mongrel offspring has been asserted by so many authors 

 U) 1)0 universal, this cannot be considered correct after the fiicts 

 given on the authority of Gartner and Kijlreuter. Nor is the 

 very general fertility of varieties, when crossed, surprising, 

 when we remember that it is not likely that their reproductive 

 sj-stems should have been profoundly modified. Moreover, 

 most of the varieties which have been experimente(l on have 

 been produced under domestication ; and as domestication (I 

 do not mean mere confinement) almost certainly tends to elim- 

 inate sterility, we ought not to expect it also to produce 

 sterility. 



The sterility of hyl)rids is a different case from that of a 

 first cross, for tli(» reproductive organs of hybrids arc more or 

 less functionally impotent ; whereas in first crosses, the organs 

 of both species arc of course in a perfect condition. As we 

 continually see that organisms of all kinds are rendered in 

 some degree sterile from being exposed to slightly-changed 

 conditions, wc need not feel surprise at hybrids being in some 



