106 U. S. NATIONAL MUSEUM BULLETIN 206 



separated teeth between the choanae, two examples (97579 and 97582) 

 have a set of teeth on one side of the mouth and none on the other, 

 while one (97581) is entirely toothless. 



In other structural characters there is not much variation. The 

 webbing of hands and feet is quite constant, and the proportions of 

 the limbs do not vary much. The tongue is slightly concave at its 

 posterior margin in some of the specimens. The color pattern varies 

 greatly in intensity, and in a very few examples the back is immaculate, 

 with only the dark lateral stripe persisting. In some of the frogs, the 

 brilliant golden f^-shaped mark surrounding the deep chestnut mid- 

 dorsal area suggests somewhat the color pattern of H. leucophyllata. 

 Many have small dark spots over the back, particularly noticeable 

 on the anterior half. 



Remarks. — This frog might be taken for the young of Hyla leucophyl- 

 lata, were it not for its shorter head and foot. In alcohol, the dorsal 

 patterns of H. decipiens and Brazilian examples of leucophyllata appear 

 similar at first glance, consisting of a light head and light dorsolateral 

 lines, with rectangular middorsal brown area between the light borders. 

 In life, the light areas on the back of decipiens are less metallic appear- 

 ing. Upon close examination, other differences in pattern are ap- 

 parent. The white areas of leucophyllata are much more distinct 

 posteriorly than they are in decipiens, since the dorsolateral white 

 stripes scarcely reach to the groin in decipiens, while the apparent 

 stripes in leucophyllata are really the dorsal color showing beside the 

 dorsal rectangular brown mark, hence the entire sacrum is white 

 posteriorly behind the brown mark, this white area tapering to a 

 point just above the anus. 



On February 14, 1935, a collecting trip to Sacco Sao Francisco was 

 made by Dr. Adolpho Lutz, Joaquim Venancio, and myself. In a 

 boggy spot where an inch or two of standing water was sometimes 

 found, and where small shrubs and vines of several kinds were plenti- 

 ful, some egg-masses had been deposited on leaves hanging over the 

 water. It is well known that Phyllomedusa deposits its eggs in such 

 a manner on leaves overhanging the water, but this locality is not 

 one at which Phyllomedusa is known to occur, and since a full-grown 

 Hyla decipiens was taken at the spot, the conclusion was that these 

 were Hyla eggs, uniquely placed over the water instead of in it. 

 Collections of 30 or 40 different egg-masses were made in a few minutes, 

 and the next day some of these were photographed in the laboratory. 

 The remainder of the leaves and their adhering egg-clusters were put 

 on the sides of glass funnels where the already hatching tadpoles slid 

 down into the water in the aquaria beneath them. From this point, 

 watercolor sketches were made at each developmental stage. The 



