48 /. H. Powers 



most aquatic of all my adults, was assuming its great breadth and 

 length of tail. The posterior limbs of this same specimen, when 

 it had reached about 19 cm. in length and had lived about six 

 months in water, are shown in figure 7. The toes will be seen 

 to be very short and narrow. The posterior limbs of a stronger- 

 limbed and very typical aquatic adult are shown in figure 8. It 

 will be noticed that the toes have here become very narrow, 

 sharp, and cylindrical with unusually little webbing at their bases. 

 Permanent aquatic life, then, would, in all probability, reduce, 

 rather than develop the limbs of this species ; certainly it would 

 not develop a swimming foot. A comparison of the limbs and 

 feet of the more aquatic or the wholly aquatic forms among the 

 Urodela as a whole obviously substantiates this view. 



As I have watched the development and modification of the 

 limbs of this plastic species, the thought has occurred to me 

 again and again that the phenomena were suggestive of im- 

 portant factors which may have cooperated, not only in the 

 modification, degeneration, etc., of the vertebrate limb, but quite 

 possibly in its earlier stages of formation. Of course the improba- 

 bility of transmission reduces the importance of the otherwise 

 suggestive results of use and disuse. But not so with the equally 

 striking results of overgrowth. Excessive nutrition with these 

 larvae seems, as it were, to overflow into all peripheral parts, 

 quite regardless of function. It looks as if the tissues forming 

 the central axis of the body were incapable of growing at more 

 than a certain fixed rate, with the consequences that nutrition in 

 excess finds lodgment where it may, in hypertrophied gills or 

 limbs, regardless of any necessity laid upon them by function. 

 How easily might some preamphibian, Protopterus-like ances- 

 tor, overglutted with food in some fresh-water habitat, have thus 

 developed a fin-like appendage into the bulk if not the build of a 

 limb. And when we add to this the fact that these same condi- 

 tions of excessive feeding and overnutrition are the conditions 

 which make metamorphosis inevitable, and that metamorphosis 

 is essentially a random process, reducing weight and destroying 

 the less resistant tissues of the body while tending strongly to- 

 ward ossification of cartilaginous structures, it seems as if we had 



244 



