FROGS OF SOUTHEASTERN BRAZIL — COCHRAN 309 



line over the ear), resembling the pattern figured by Miiller for L. 

 nanus (see Lutz, 1926b, pi. 32, figs, 10, 11). 



Remarks. — This confusing species has been described under several 

 different names and perhaps further study may add others, as Parker 

 (1935, p. 508) has already suggested. It is one of the species that 

 tends to bridge the gap between Leptodactyhis and Eleutherodactylus, 

 for the terminal phalanges of the toes are distinctly T-shaped, while 

 the terminal disks show a corresponding enlargement, unhke most 

 of the other species in the genus Leptodactylus. 



The pattern is more variable than is usual in the other leptodactyli 

 of Brazil, some examples having three pale red stripes on the back, 

 a fact that led Dr. Lutz to consider them as a species distinct from 

 L. marmoratus {=nanus Miiller). Since this variation has been 

 found in practically every locality from which marmoratus itself is 

 known, as well as intergrading patterns wdth the stripes greatly re- 

 duced, it seems proper to add the name irivittatus to the synonymy 

 of marmoratus. 



It is quite apparent that L. marmoratus is the southern representa- 

 tion of L. hylaedadylus. These little frogs are so similar in size, 

 color, and general bodily proportions that they might easily be con- 

 fused were it not for certain fixed differences that separate them. 

 L. hylaedadylus is much more rugose dorsally than its southern ally; 

 it has dorsolateral glands which often appear as true folds; there are 

 many short glands on the center of the back as well as on the sides; 

 and the head is altogether more pointed than that of marmoratus. 

 Viewed from above this is not so apparent, but from below the chin 

 the outline is oval-pointed in the former, nearly semicircular in the 

 latter. Thus, marmoratus appears to have a wider head and a shorter 

 snout than hylaedadylus. The interorbital width seems shghtly 

 greater in marmoratus, and the eyelids are usually much narrower 

 and the eyes less projecting, although sometimes, as in USNM 97475, 

 the width of the lids approaches that of hylaedadylus. Color 

 patterns in both species are extremely similar. Sometimes an example 

 of hylaedadylus, as in AMNH 44782, will have a light dorsolateral 

 stripe (as in some of the marmoratus specimens called irivittatus b}^ Dr. 

 Lutz for the same reason). The body in hylaedadylus seems longer 

 in proportion to its total length, perhaps because of its pointed head, 

 but the proportions of hand and foot are almost identical in the two 

 species. The upper tibial and femoral surfaces of hylaedadylus are 

 rough and tubercular, while those of marmoratus are practically 

 smooth. 



I saw a nest of L. marmoratus on February 7, 1935, buried over a 

 foot deep in a clay bank about 10 feet above a shallow pool near 

 Covanca, Jacarepagud. The eggs were discovered by the collector 



