640 



REPORT OF NATIONAL MUSEUM, 1902. 



Tho, female and young differ chiefly in the absence of dewlap and fin 

 to the tail, the female also by entire absence of postanal plates. 



Variation. — The greatest variation is found in the head scales, the 

 most important being in the relation of the supraorbital semicircles to 

 each other and to the occipital plate. In the majority of cases the 

 semicircles are broadly in contact across the interorbital space, but a 

 small scale is often developed at the intersection of the sutures in this 

 region, and it is very rare that not at least one pair of semicirculars 

 touch. The number of scale rows separating the occipital varies 

 between 2 and -i, the latter number but rarely, however. The 

 enlargement of the two median rows of scales or granules on the back 

 is variable, forming an appreciably raised line in many individuals, 

 but almost absent in others. The height of the caudal tin is also 

 somewhat variable, as it is not always the highest in the largest males, 



I 95 l 97 



Figs. 95-97.— Anolis cristatellus. Culebra Island. 95, top of head; 96, side head. 2| x natural 

 size. 97, side of tail at level of fifth verticil. 5i x natural size. No. 25777, U.S.N.M. 



but I believe it to be present in all full-grown males and do not at all 

 understand the remark by Reinhardt and Luetken to the effect that the 

 caudal crest is absent in specimens from Porto Rico and Vieques. In 

 some specimens the ventral scales are more or less distinctly keeled. 



One of the individuals, a halfgrown male (No. 25777), from the island 

 of Culebra, collected by the Fishhawh part}% differs at first sight so 

 much from the other specimens of the same locality collected by Mr. 

 Baker that at one time I believed it to belong to a different species. 

 The scales separating the occipital and supraocular semicircles, as well 

 as the enlarged scales forming the supraocular disk, are rather numer- 

 ous, but these peculiarities are matched by various Porto Rican species, 

 and a careful comparison of other points fail to discover an}^ tangible 

 differences. It is very light colored, with scarcely any dark mark- 



