THE HERPETOLOGY OF HISPANIOLA 



277 



almost invariably be relied upon, I found to be in the size relation of 

 the interparietal to the scales adjacent to it on either side; in chryso- 

 laema, the interparietal scale is smaller, never larger, than the adjacent 

 scales, while in affinis the interparietal scale is larger, never smaller. 

 Rarely a specimen of either species may have the interparietal and 

 its adjacent scales approximately equal in area; this is due to a little 

 more splitting than usual at the border of a scale. The difference in 

 ratio is due to the amount of splitting up of these large scales, indeed 

 in the case of chrysolaema, the prominent area for splitting up seems 



Figure 76. — Ameiva chrysolaema chrysolaeyna: a. Top of head; b, side of head; c, chin and 

 throat; d, forearm showing antebrachials; e, anal and femoral regions. U.S.N. M. No. 

 12140, cotype, from near Port-au-Prince, Haiti. One and one-fourth times natural size. 



to be that of the median occipital region, while in affinis the splitting 

 has avoided the central region and displays its activity chiefly at the 

 sides of the occiput. 



The other differentiating characters are more or less unstable, as I 

 have found through studying carefully a large series of examples. 

 Coloration is usually of some importance, the intensely spotted style 

 being that of chrysolaema, while a wide, black, lateral band practically 

 uninvaded by light dots ordinarily marks a specimen of affinis. 

 In addition, the brachials are usually quite conspicuously enlarged in 

 chrysolaema, but very small and scarcely differentiated from the sur- 

 rounding granules in affinis. Sometimes, however, an example of 

 chrysolaema will be found without the considerably enlarged scales. 



