HERPETOLOGY OF PORTO RICO. 587 



a similarly colored spot on the occiput and cross-bars on the hind legs, 

 all the pale marks more or less plainly edged with dusky; the ground 

 color varies from dark to light gra}dsh brown overlaid to a varying 

 deerree with yellowish or reddish. This style is evidently only a 

 modification of the one described above, the dark ground color having 

 encroached upon the light flanks. Some of these specimens are so 

 dark that the pattern can be made out only with difficulty. One of 

 the large adult specimens shows distinct traces of the dorso-lateral 

 pale semilunar marks. 



The next style, represented by a few specimens only, is very different. 

 These are nearly uniform dark brownish gray with a very narrow, 

 sharply defined pale line from tip of snout down the middle of the 

 back to the vent, where it bifurcates and extends along the posterior 

 edge of each femur; another line, less pale, crosses this one on the 

 interorbital space. Several specimens combine the common pattern 

 as described above with this narrow median line. 



A single young specimen in the lot is light olive gray with a pale 

 longitudinal band on the middle line from snout to vent so broad as to 

 nearly fill the interorbital space; the post-femoral streak is faintly indi- 

 cated; the pale lines are indistinctly outlined with dusky and there 

 are similar cross marks visible on the hind legs. 



A curious feature is also the variability in the coloration of the iris. 

 In the younger specimens it is brassy, more or less overlaid with 

 dusky. In the adults, however, it is golden with a vertical blackish 

 streak in the middle of the lower half. However, if the sides of the 

 face are much darker than the region above, the iris is also darker in 

 the lower half below the narrow horizontal pupil which, as indicated 

 above, forms the continuation of the canthal black line. 



That the coqui or tree toad in other localities assumes a coloration 

 closel} T agreeing with the surroundings is proven by several specimens 

 collected in the western part of the island. A large adult specimen 

 caught by Dr. Richmond near the summit of the pass between Adjuntas 

 and Ponce, was of a reddish brown ground color, closely resem- 

 bling the clay upon which it was found (U.S.N.M. No. 27280; L. S. 

 No. 9067). The four specimens, on the other hand, which we took 

 under stones in the white clay hills east of Ponce on April 16, were 

 nearly uniform pale clay colored, with no markings, not even a trace 

 of a pale dorsolateral band. 



Habitat. — The "coqui" is common all over Porto Rico, from sea 

 level to the top of the highest mountains; in fact we found it most 

 abundant in the excessively damp woods near the top of El Yunque. 

 It does not occur on Vieques, nor has it been found on Mona Island, 

 a distribution quite unexpected in view of the fact that the species also 

 occurs in Santo Domingo and eastern Cuba. 



Habits. — Although by no means confined to living on or among the 



