70 



THE BRITISH NATURALIST. [April 



Euclidia mi.— Plentiful on low coast lands. Occasionally at Onchan. 

 Euclidia glyphica. — Plentiful on Bank's Howe. 



Mollusca. 



Notes on Varieties. The observations on variation recorded by 

 Mr. Gain, on p. 35 are of the greatest interest. I do not agree with 

 all his opinions on this subject, but the careful breeding of the several 

 varieties is just the thing we want, and will go further towards 

 solving difficult problems than any amount of discussion based on 

 cabinet specimens. If it can be proved that the different variations of 

 Helix nemoralis and hortensis, though living in the same hedge, rarely 

 mate and produce fertile offspring ; and that individuals of the same 

 variation give rise to others like themselves, it is a matter of the 

 highest importance from an evolutionist's point of view. Of course, 

 much has been written on this particular question, notably by Messrs. 

 Gulick and Romanes, and I need not here go into the details of what 

 is known as " Physiological Selection," but I think the warmest 

 advocates of that theory would be the first to admit the desirability of 

 fresh and numerous experiments, such as those of Mr. Gain, to test 

 the point. Early in i8go I received a circular from Prof. Romanes, 

 urging botanists to make observations testing this theory, and as it 

 explains what is wanted briefly and lucidly, and applies equally well to 

 animals, I Vv^ill quote portions of it: — " Several years ago I published 

 " An Additional Suggestion on the Origin of Species," which, briefly 

 stated, consists in the hypothesis that in many cases, species owe 

 their origin to a sexual variation, such that w^hile fertility continues 

 unimpaired within the limits of the physiologically affected group, 

 some degree of infertility supervenes between members of this group 



and the unaffected portion of the species Thus, the 



occurrence of some degree of sterility between allied species, which is 

 so frequently observable, may be explained by the supposition that 



the sexual change was the prior change " " In order to 



test this theory, constant varieties of plants growing on common areas 

 (or areas not geographically separated) ought to be tested as to the 

 comparative fertility of Ax A, BxB, AxB, BxA, where A is one 

 variety and B the other. I believe you would find that in most cases 

 where there is a marked difference of fertility even as between such 

 first crosses, — A x A and BxB being more fertile than A x B or 

 BxA, while in many cases I believe you would find AxB and 

 B X A absolutely sterile. But if there should be no difference 

 observable in the case of first crosses, the experiment ought to be 



