NO. 1443. CAVE SALAMANDERS— BAXTA AND Me A TEE. 77 



areas. The pigment forming the lower border of the first series 

 and the upper margin of the second now collects into the most con- 

 spicuous series of spots on the lateral .surface. The pigment on the 

 sides of the tail is also beginning to form spots. 



It is possible that stages of the various components of the color 

 pattern presented by this specimen ma\' he omitted b}^ some at trans- 

 formation, and on the other hand they may be prolonged, in other 

 cases for some time afterward. Spelerpes hilineatus seems especially 

 lialjle to such vagaries at transformation, and this suggests the possi- 

 bilit}' of a parallel case in the present species. However, the larva is 

 undoubtedly intermediate in age between the preceding and succeed- 

 ing examples making the color stage valid for our series if it is not 

 for every one. 



The next stage (55.5 mm. fig. 7, Plates VIII and IX) is a typical 

 newly-ti'ansformed cave salamander. The ground color is lemon yel- 

 low, the spots are distinct with but few traces of the primitive pig- 

 ment reticulation. The pigment on the legs is now for the first time 

 collected in spots. The head is more plainl}^ spotted, approaching 

 nearly to the adult condition. The spotting on the sides of the tail 

 is definite though not 3'et perfect, and of the series of immaculate lar- 

 val areas onl}" one, the second, is traceable. On each side of this 

 series of clear areas the bands of pigment cells are collected into dense 

 bars, which now break up to form two rows of spots, which are very 

 conspicuous in the adult salamander. The third and lowest series of 

 larval areas is no longer present. The nebulous pigment surround- 

 ing them forms irregularly distributed blotches on the lower 

 sides of the salamander and contributes many cells to the ventral area, 

 which now becomes entireh", if sparsely, pigmented. In a salamander 

 of this age, with the second series of larval areas almost intact, the 

 original position of the three series of clear areas and the fate of their 

 accompanying pigment masses is not difficult to trace. But in the 

 fully grown adult, at first blush, it does not appear that the grouping 

 of the blotches is the result of even a fairh^ definite system. The 

 influence of their origin upon their arrangement becomes apparent at 

 once, however, when compared with newly transformed specimens, 

 such as the one just described. 



The scattered pigment cells at this stage are more abundant than in 

 the full grown adult, and the pigment blotches less distinct. As the 

 animal grows the scattered pigment collects more and more in the 

 blotches, which grow larger and at the same time more dense. Stages 

 in this aggregative process may be seen about the edges of the blotches 

 on an}' adult. Approaching and newl}^ united pigment cells form a 

 fringe about most of them. 



Few other changes occur in the pigment beyond the stage just 

 described (55.5 mm.). In many adults the chaining of blotches along 



