METAMORPHIC CHANGES, R. PIPIENS 43 
plete. Larvae in which metamorphic changes were obviously 
initiated, but whose fins and gills were still of approximately 
maximum size when they were brought into the laboratory, 
emerged as young adults and resumed feeding nine days 
later. This is probably a fair estimate of the interval required 
for the completion of a metamorphosis in the animals under 
observation. Ambystoma tigrinuwm is an exceedingly variable 
species. While the Ambystoma population of any given ket- 
tle-hole was relatively uniform, specimens taken from various 
kettle-holes in the vicinity showed wide variations in size. 
While it is quite probable that metamorphosis requires no long- 
er interval in the larger than in the small varieties, the data 
at hand do not justify the conclusion that the interval of ap- 
proximately nine days here indicated is the average interval for 
the species. Powers’ (’03) observations indicate an interval 
considerably shorter than nine days. 
The morphological changes involved are far less extensive in 
Ambystoma than in Rana. The difference in the extent of re- 
duction in the length of the digestive tube and the changes in 
the size of the liver in these two types of Amphibia, doubtless 
are correlated with their habits of feeding. The larvae of Rana 
are primarily herbivorous while those of Ambystoma are es- 
sentially carnivorous. The relatively enormous length of the 
digestive tube in the larvae of Rana is an adaptation correlat- 
ed with their herbivorous diet. The great reduction in the 
length of the digestive tube during metamorphosis is an essen- 
tial part of the adjustment of the little frog to a carnivorous 
diet. The lesser reduction in the length of the digestive tube 
during metamorphosis in Ambystoma, doubtless must be ex- 
plained on the same basis. While the larvae of Ambystoma are 
essentially carnivorous they are not strictly limited in their di- 
et. Those used in this study were taken in a habitat in which 
there was very little vegetation. Their food consisted primar- 
ily of aquatic insects and insect larvae. However, the ccn- 
tents of the digestive tube included some mud. Larvae of 
Ambystoma taken in another habitat in which algae were grow- 
ing in abundance were found to be feeding on these plants ex- 
tensively. Their stomach contents consisted almost exclusive- 
ly of Entomostraca and minute algae. Doubtless, the diet of 
