Morphological Variation and Its Causes in A. tigrinum 47 



mained of the type shown in figure 9, plate II, became more fully 

 webbed than in any specimen which I had previously seen. I 

 have repeated this experiment with more or less success upon 

 several larvae, wintering one specimen, which became over 20 cm. 

 long, in a one-quart Mason jar. This individual adopted a sim- 

 ilar attitude and with similar results in the posterior feet. But 

 the development was eventually checked when the captive 

 adopted, very successfully, the reverse habit of standing, for the 

 most part, upon its head. I am able, then, to substantiate Cope's 

 assumptions as to this character, in principle if not in fact. The 

 difference is that the plasticity, assumed by Cope for the adult, 

 exists only in the larva, and, in case of this character, mainly 

 (aside from exceptional stimulus) at an early age. Then, too, 

 the stimulus is not from swimming, which Cope, though he does 

 not state as much, doubtless has in mind, but from walking under 

 water. 



I may now make a final statement which will, I hope, seem less 

 controversial at this point, about the actual effect of aquatic life 

 upon the feet of adults. As before stated, I have kept a number 

 of adults in water of varying depths for over two years after 

 metamorphosis, and many others for lesser periods of time. 

 These animals have been kept in glass jars or other small aquaria 

 and have been fed upon cut liver. They have, therefore, not been 

 so situated as to use the limbs to a great extent in moving about 

 in search of food or the like. To some extent the results may be 

 due to this. Yet it is significant, none the less, that these aquatic 

 amblystomas showed noticeable 'degeneration of, or at least dis- 

 proportionately weak development of the limbs and feet, the toes 

 becoming rounder, more pointed at tips, etc. Even the webbing 

 at their bases seemed to be slowly reduced under these circum- 

 stances. An examination of figures on plate V will show this, 

 although the toes are not well shown upon all of the figures. 

 Figure i still shows a very broad toe, though reduction had be- 

 gun near the tips. The breadth of toe had been previously ac- 

 quired by the larva, a very rapidly grown anabolic specimen. 

 Figure 2 shows toes that were almost awl-shaped at the tips, this 

 form having been slowly assumed at the same time that this, the 



243 



