512 



THE AMERICAN NATURALIST [Vol. XLIII 



ordinary hypothesis. I can not help suspecting that the parent 

 slug had really paired earlier with a specimen of ater, the 

 progeny in consequence not actually having the origin stated. 

 In that case, supposing ater to be dominant, the results would 

 not be so anomalous. It is not so easy to explain the results 

 of the first experiment by the hypothesis of previous pairing, 

 as in that, except for the original pair, the slugs were under 

 observation from their birth. 



Seeking a possible explanation of the albolateralis case, I con- 

 sulted Dr. C. B. Davenport, who practically concurs with my 

 view, writing: ' 'The result of Collinge's mating is inexplicable 

 to me except upon one or the other of two hypothesis; either 

 that the parents were heterozygous or else, as you suggest, the 

 supposed parents were not the actual ones, and one had previ- 

 ously paired with a black slug. Of course if albolateralis is a 

 heterozygote, then striping is dominant as is usually the case 

 •and uniformity recessive. Uniform black would then be active 

 in one quarter of the offspring, but the great proportion of pure 

 black speaks against this hypothesis" (litt, March 22, 1909). 



T. D. A. COCKERELL. 



