1917.] NATURAL SCIENCES OF PHILADELPHIA. 17 



completely disappeared in S. longicaudus while they are still quite 

 distinct in S. bislineatus. The dorsal region of S. longicaudus has 

 only a faint stripe down the median line. The larvae of *S. bislineatus 

 have the dorsal region either uniform or darker towards the sides. 

 In *S. ruber the dorsal series is formed of a few very small areas, and 

 is only distinct in smaller larvae (40 mm.). This forms no part in 

 the development of the color-pattern. The lateral series is of many 

 small areas which form a dotted line down the side of the larva. 

 It disappears before transformation, and also plays no part in the 

 formation of the adult pattern, or rather the early disappearance 

 of the larval series is responsible for the lack of pattern in S. ruber. 

 The same may be said of G. porphyriticus and D.- quadrimaculata . 

 It is interesting to note in this connection that the pattern of D. 

 ochrophea is certainly the result of the development of the dorsal 

 areas from their insignificance in D. quadrimaculata, their late 

 coalescence in D. monticola, their early coalescence in D. fusca, to a 

 coalescence in D. ochrophcea, which takes place so early that even in 

 the smallest examples seen no trace of it has been observed. This 

 suggests the absence of an aquatic larval stage, and indeed D. 

 ochrophcEa is the most terrestrial of the species of Desfnognathus. 

 The color pattern of the "red-back" stage of the entirely terrestrial 

 Plethodon erythronotus is almost exactly like that of D. ochrophcea. 

 The "cinereus" phase is certainly not to be explained in the same 

 way, and the coloration of the species of Plethodon presents a different 

 problem. It has been stated that the dark pigment underlies the 

 red in P. erythronotus. Now the bleachment and the color pigment 

 (usually red or yellow), in Spelerpes and Desmognathus are distinct, 

 the black pattern being the result of the change in the larval areas 

 already referred to, while the color pigment appears at transformation 

 and usually is not localized. A possible suggestion of the coloration 

 in the genus Plethodon is that the dark color is uniformly over the 

 animal and the color pigment is localized or in scattered spots. 

 Thus there would seem to be no trace in Plethodon of the influence 

 of any larval areas on the coloration, which would be correlated 

 with the apparent absence of any aquatic larval stage. 



In order to make clear the relations of several recently described 

 species, and a new species of Plethodon described later on in these 

 pages, the analysis below is given. Species marked f are restricted 

 to the southern AUeghanies and those marked * are northeastern 

 in distribution. 



