674 REPORT OF NATIONAL MUSEUM, 1902. 



SO many of the head shields develop into horns and tubercules, the 

 boundaries between the scales and their shapes become oliscured. 



The adults which reach a length of at least 1 meter are character- 

 ized b}^ the enormous development of some of the head shields into 

 pointed horns or cones, and the increase in the height of the spines 

 which form the crests on neck, back, and tail. 



Thus, in No. 29366, an adult male about 1.06 m. long, the frontal 

 horn is 9 mm. high and the lower preauricular shield has grown out 

 to form a stout horn not less than 12 mm. high; the posterior pair of 

 the prefrontal shields are also raised, forming hornlike protuberances; 

 many other scales have their keels form sharp ridges or knobs; thus 

 several at the posterior end of the superciliary ridge, the median pre- 

 auricular, a smaller shield in front of the big lower preauricular at 

 the angle of the mouth, one or two of the posterior suboculars and 

 the posterior scales of the mandibidar or malar series. The nuchal 

 crest is very low, scarcely more than 2 nun. high, but the scales of 

 the median dorsal series have developed into flattened, more or less 

 falcate spines, the longest of which are 17 nmi., while the spines of the 

 caudal crest, which are stronger and broader at the base, measure no 

 less than 19 mm. In this specimen the vertical diameter of the tym- 

 panum is 17 mm. and equals seven of the larger keeled scales on the 

 forearm and ten dorsal scale rows. The femoral pores are in two rows, 

 with a few additional forming a third row near the upper end. Length 

 from tip of snout to ear, 101 mm. ; width of head, exclusive of pre- 

 auricular spines, 76 mm. 



No. 29365, another adult male, differs but little from the above, the 

 main dift'erence being that the preauricular spines are hardly devel- 

 oped. The femoral pores are arranged in three unequal rows, 18-19 

 in the anterior row, which is the longest. 



The adult female (No. 2!»642) does not difl'er much, except that the 

 spines are somewhat smaller; thus the lower preauricular spine is 

 onl}^ 9 mm., the highest dorsal also 9, and the highest caudal 13 mm. 

 The femoral pores are not large and are arranged in one complete 

 series, with a second incomplete in the intervals between the first. 



The number of dorsal spines seems to be quite constant, var3"ing as 

 it does in our four specimens only between 50 and 55, as follows: No. 

 29365, 50; No. 29366, 50; No. 29367, 51; No. 29642, 55. 



llahltat. — If the present species is correctly referred to Cydura 

 cormda, the habitat embraces the whole island of Haiti as well as 

 Mona. In the latter island Mr. Bowdish found it among the rocks. 



