104 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser., 



of the others, and the dorsal, or posterior, longest of all. 

 Later, however, this discrepancy in relative length of the 

 anterior one becomes much more pronounced. 



There are at hatching usually one or two filaments on the 

 anterior gill, and about three or four on the posterior. 



The mouth is not yet broken through when the larva 

 escapes from the jelly mass, but the anterior nares are con- 

 spicuous pits on the ventral side of the head quite remote 

 from their ultimate terminal position. 



The so-called balancers are prominent during the early 

 larval life (fig. 9), but their atrophy begins in a few weeks 

 after hatching, and no trace of them remains by the time 

 the larva has reached its full size. 



At their full development blood corpuscles can be seen 

 to enter these structures. This suggests that they may be 

 useful as accessory gills; it is, however, true that the blood 

 stream entering them is much less in volume than that which 

 passes through the gill filaments. I have seen no indication 

 that these appendages are a compensation for the absence 

 of limbs, as they are stated to be in some urodele larvas 

 (Baird ['51], Clark ['80]), or are used as clinging organs, 

 as is said by Hay ('89) to be the case in Amblystomea micro- 

 stomum. In D. torosus they are cylindrical and of uniform 

 diameter from end to end. The epithelial cells at their dis- 

 tal ends are produced at their outer ends into blunt though 

 distinct processes, so that when viewed with a low power 

 of the microscope the end of the balancer reminds one of 

 the tentacles of various hydroids armed with their lasso cells. 

 Hay ('89) mentions the same condition in Amblystoma micro- 

 stomum. 



The extreme transparency not only of the gills but also 

 of the entire animal, makes these larvas especially favorable 

 for studying the movements of the blood, the character of 

 the living blood cells, the development of capillaries, and 

 other points in the circulatory system; but this subject does 

 not come within the scope of the present paper. 



As is the general rule with the larva? of the Urodela, the 

 fore limbs are the first to appear. They are mere buds at 



