150 BULLETIN 160, UNITED STATES NATIONAL MUSEUM 



found to be fairly abundant in the lowland swamps of Yucatan and 

 Qiiintana Roo. H. gracilipes is a very close relative of H. eximia, but 

 seems to be sufficiently distinct to be recognized as a full species, 

 rather than as a variety of the latter. The habitat of the tree frog 

 described by Brocchi as H. plicata is unknown, and, although the posi- 

 tion of the vomerine teeth is quite different from other Mexican 

 hylas, it is otherwise structurally quite similar to H. eximia. No 

 additional specimens of H. bistinda have been reported since it was 

 first described by Cope. The type of the peculiar little H. staujferi 

 remained unique for more than 60 years, though quite recently Dr. 

 E. R. Dunn discovered a second individual in the British Museum 

 that was collected at Motzorongo in Vera Cruz, and there are three 

 additional specimens in the United States National Museum from 

 Potrero, Vera Cruz. Some uncertainty still exists as to the validity 

 of H. taeniopus, a species based on two immature individuals. 



The genus Hyla, in spite of its nearlj^ world-wide distribution and 

 numerous species, seems to have retained a fairly uniform type of life 

 history. There is no evidence that any species of the genus skips the 

 tadpole stage, and with the exception of certain Jamaican species all 

 of them evidently lay their eggs in the water, where the tadpoles 

 follow the usual aquatic development. The eggs are deposited in 

 ponds, in flooded marshes, in stagnant pools of rain-water in forests, 

 along the edges of streams where there is little or no current, or in 

 little basins or pools on the edges of mountain brooks. In Jamaica, 

 Dunn found three species of Hyla that laid their eggs in water caught 

 between the leaves of bromeliads. Some of the large tree frogs of 

 South America build mud basins, in which the eggs are deposited and 

 in which the tadpoles pass through the larval stage. 



According to Noble " — 



In Hyla, as in many large genera of Salientia, the tadpoles do not conform to a 

 single type. Those species which lay their eggs in ponds have either broad-finned 

 tadpoles of the "sunfish type" or bottom-wriggling larvae of the "polUwog type." 

 The "sunfish type" has larger lungs, remains more frequently between surface 

 and bottom and is usually a graceful swimmer. The "polliwog type" has a nar- 

 row fin restricted to the tail or only to the tail and posterior part of the body. 

 This type is prevailingly a bottom-wriggler but may dart very rapidly in the 

 undulatory manner characteristic of our common Rana poUiwogs. 



KEY TO MEXICAN SPECIES OF HYLA 



1. Second to fourth fingers practically free, at most united at base 



by a vestigial web or entirely free from such 2 



Second to fourth fingers united at base by distinct web, which 



may extend for one-fourth to two-thirds of their length 8 



2. Vomerine teeth behind or at most at level of posterior edge of 



choanae 3 



Vomerine teeth between or near level of anterior edge of choanae 4 



" Noble, Q. K., The value of life history data in the study of the evolution of the Amphibia. Ann, 

 New York Acad. Sci., vol. 30, p. 95, Oct., 31, 1927. 



