22,5 Taylor: Herpetological Fauna, HI 525 



(which is invariably absent in R. magna and R. woodworthi), 

 in the absence of angular canthi rostrales, and in the arrange- 

 ment of the dorsal skin folds. The leg is longer than in R. 

 magna, and the dorsolateral fold of R. woodworthi is absent. 

 The webbing of the toes is much the same as in R. woodworthi 

 but less than in R. magna. 



The specimens were collected about a small sluggish stream. 

 Many that were seen escaped capture by diving in the water and 

 remaining under for a considerable time. 

 Rana leytensis Boettger. 



Rana leytensis Boettger, Zool. Anz. (1893) 365; Boulengeb, Proc. 

 Zool. Soc. London (1897) 228; Taylor, Philip. Journ. Sci. 16 (1920) 

 246, pi. 2, fig. 1. 



The type of this species from Leyte measured 28 millimeters, 

 and the largest one of the specimens reported by Boulenger, 

 from Celebes, measured only 38 millimeters. Philippine spec- 

 imens for the most part appear to attain a larger size. Most 

 of the adults in my collection from Zamboanga and Basilan 

 average about 50 millimeters while the largest specimen (from 

 Abung-abung, Basilan) measures 56 millimeters The speci- 

 mens show the same color variations as recorded by Boulenger. 

 Some have a more or less continuous dorsolateral fold, but m 

 by far the greater number the folds are broken and more or 

 less irregularly arranged. The V-shaped mark on the shoulders 



^TnTspLmens from Jolo were collected for the most part 

 on the top of Bud Daho, an old volcanic crater with an eleva- 

 tion of about 600 meters. These are very much smaller than 

 the Zamboanga and Basilan specimens, are lighter m color ^ ana 

 the pads on the toes are appreciably larger. The largest of 

 thirty-four specimens measures only 39 millimeters 



I have called attention to the presence of the en arged eeh 

 in the lower jaw of this species,* which are plainly visible in 

 specimens measuring between 40 and 50 milhmetersin length, 

 smaller male specimens show the teeth to some degree. 



I have examined two specimens from Sibuyan Is ^££ 

 I have doubtfully referred to this species. They are ™J£*£ 

 and one of them has the digit of the egs *^™£™ 

 More material must be obtained before this point can 1* deadeo. 

 Specimens of the Polilto frog which I referred to this species 

 belong to the species Rana woodworthi sp. nov. 



