FROGS OF SOUTHEASTERN BRAZIL — COCHRAN 89 



the color patterns of the two species are notably different in most 

 features, details of structure definitely lead to the conclusion that the 

 two are allied. The very small tympanum of the type of clepsydra, 

 one-third the eye diameter, is often paralleled in the large series of 

 mtdtilineata, in which the tympanum may be as great as one-half 

 the eye diameter but is sometimes much smaller, so that the finger 

 disk practically covers it as has already been noted in variation in. 

 multilineata. In the contour of the head, the shape of the snout, and 

 the manner in which the prominent eyes are set upon the head, the 

 similarities become quite apparent when actual specimens are com- 

 pared. The vomerine teeth are also much the same. An important 

 characteristic not found often, except in larger frogs, is the prominent 

 rudiment of a poUex, noticeable equally in clepsydra and in all male 

 examples of multilineata. This at once places the two species in a 

 more than accidental juxtaposition. Furthermore, the limb propor- 

 tions coincide. The similarities in color pattern, as seen in the water 

 color sketch by Sandig of the living frog, seem to be the dark patches 

 behind anus and on heel, the dark loreal region, and the narrow brown 

 bars across the upper surface of the femur. A significant fact, 

 however, is that the anterior and posterior femoral surfaces and top 

 of foot of both species are pink or red, a condition not too frequently 

 found in species of Hyla. The hoiu"glass-shaped dorsal marking 

 which characterizes clepsydra is not approached by any individuals 

 in the series of multilineata, and it seems well to keep the two species 

 separate until a larger series of clepsydra can be brought together for 

 full variational studies. 



Specimens examined 

 BRAZIL: 



Rio de Janeiro: Bonito, Serra da Bocaina, IOC (type; from bromeliad) ; 

 IOC (1), A. Lutz, December 1928. 



Hyla polytaenia Cope 



Plate 7, Figures c-g 



1869. Hyla rubicundula (not of Reinhardt and Liitken) GtJNTHER, p. 489, pi. 40, 



fig. 3. 



1870. Hyla polytaenia Cope, 1870, p. 164 (type locality, Brazil); 1871a, p. 554. — 



BouLENGER, 1882a, p. 394. — Goeldi, 1895, p. 93. — Gadow, 1901, p. 

 198.— Baumann, 1912, p. 107.— Nieden, 1923, p. 293.— Miranda- 

 RiBEiRO, 1926, p. 83. — L. Muller, 1927, p. 266. — Barbour and Love- 

 ridge, 1929, p. 280.— deWitte, 1930a, p. 228.— Myers, 1946, pp. 13, 31. 



1872. Hyla striata Peters, 1872a, p. 681 (type locality. Nova Friburgo, Rio de 

 Janeiro) . 



1925. Hyla semiguttata A. Lutz, 1925b, p. 212 (type locality, SSo Bento, Santa 

 Catarina) ; 1926a, pp. 7, 14. 



