INTERSPECIFIC STERILITY 79 



•[/«..] io, [1863.] 

 ' In plants the test of first cross seems as fair as test of 

 sterility of hybrids. And this latter test applies, I will 

 maintain to the death, to the crossing of varieties of 

 Verbascum, and varieties, selected varieties, of Zea. You 

 will say Go to the Devil and hold your tongue. No, 

 I will not hold my tongue ; for I must add that after 

 going, for my present book [ Variation under Domestica- 

 tioti\, all through domestic animals, I have come to the 

 conclusion that there are almost certainly several cases of 

 two or three or more species blended together and now 

 perfectly fertile together. Hence I conclude that there 

 must be something in domestication, — perhaps the less 

 stable conditions, the very cause which induces so much 

 variability, — which eliminates the natural sterility of 

 species when crossed. If so, we can see how unlikely 

 that sterility should arise between domestic races. Now 

 I will hold my tongue.' l 



Darwin made attempts to ' produce physiological 

 species by selection ', and thus meet his friend's criticism. 

 He thought out and suggested a plan of experiment 

 to W. B. Tegetmeier, 2 and gave a brief account of the 

 scheme to Huxley, December 28, [1862] : — ' I have . . . 

 given him [Tegetmeier] the result of my crosses of the 

 birds which he proposes to try, and have told him how 

 alone I think the experiment could be tried with the 

 faintest hope of success — namely, to get, if possible, 

 a case of two birds which when paired were unproductive, 

 yet neither impotent. For instance, I had this morning 

 a letter with a case of a Hereford heifer, which seemed 

 to be, after repeated trials, sterile with one particular and 

 far from impotent bull, but not with another bull. But it 

 is too long a story — it is to attempt to make two strains, 

 both fertile, and yet sterile when one of one strain is 

 crossed with one of the other strain. But the difficulty 

 ... would be beyond calculation.' 3 



The experiment was evidently unsuccessful, — perhaps 



1 More Letters, vol. i, pp. 231, 232, Letter 157. 



a Ibid. vol. i, pp. 223, 224, Letter 153, [1862, December] 27. 



s Ibid. vol. i, pp. 225, 226, Letter 154. 



