46 /. H. Pozvers 



cluced by wholly the same conditions. And as in the study of 

 variation the important thing is not the impressiveness of a char- 

 acter but the success with which we can trace it to its causes, I 

 shall devote a word to this delicate peculiarity separately. First, 

 I may say that the Vv^ebbing at the base of the toes is now and 

 then present, to a surprising degree, in old animals, and, like the 

 broad and depressed toes, it is, as I know by experiment, by no 

 means necessarily reduced by periods of burrowing as opposed 

 to aquatic life. Moreover, like previously described characters, 

 it is developed wholly during larval life. I find it decidedly more 

 resistant to metamorphosis, starvation, etc., than are most other 

 plastic and variable structures. Figure 8, plate IV, shows con- 

 siderable webs between bases of the toes, despite the narrowness 

 of these and the actual reduction which they have undergone. In 

 fact, I have found it possible to reduce the toes in some instances 

 until they were almost claw-like, with very delicate, cylindrical 

 and tapering tips, without detracting at all from a considerable 

 membrane at their bases. As to the cause of the development of 

 this feature during larval life, my observations are less complete 

 than I could wish. The character is more frequently determined 

 early in the larval life, at a time when I have had less chance for 

 observation. But such observations as I have made indicate a 

 purely Lamarckian origin. It develops, usually at least, about in 

 proportion as the young larvae use the feet to walk with. In one 

 instance, even in a large larva, I accidentally hit upon conditions 

 that very noticeably developed this "aquatic foot." A large, 

 slowly grown larva over 20 cm. in length, with slender toes, was 

 carefully slipped through its first summer without metamorphosis. 

 It was then placed in an ordinary fish globe kept on my table 

 throughout the winter. Here it grew slowly but continuously 

 until much too large for its quarters. Unable to expand fully, 

 and unwilling to lie coiled upon the disagreeable convex base of 

 the globe, it finally hit upon the fixed attitude of resting upon the 

 curved tail and the two outspread posterior feet which were ex- 

 panded to the uttermost. The body, raised thus to a semierect 

 position, threw much weight upon the posterior extremities, 

 which became very strong; while the feet, though the toes re- 



242 



