176 BULLETIN 15 4, UNITED STATES NATIONAL MUSEUM 



The variation in the frontoparietals, .sni)raociilars, and supraocular 

 granules has been studied on a general series of 496 specimens, 

 consisting of the examples used for the above tables with the excep- 

 tion of those listed from San Esteban Island. Since geographical 

 variation is not evident, it does not seem necessary to present these 

 data by localities. 



The frontoparietal was found to be single in 2 specimens, double 

 in 418 specimens, in 3 pieces in 33 specimens, in 4 pieces in 20 speci- 

 mens, in 5 pieces in 17 specimens, in C pieces in 5 specimens, and in 

 7 pieces in 1 specimen. 



The supraoculars were found to be 3 in 2 specimens, 4 in 437 

 specimens, 5 in 56 specimens, and 6 in 1 specimen; and the supra- 

 ocular granules extended to the anterior border of the fourth supra- 

 ocular in 2 specimens, to the middle of the third supraocular in 437 

 specimens, to the anterior border of the third supraocular in 43 

 specimens, to the middle of the second supraocular in 6 specimens, 

 and to the anterior border of the second supraocular in 8 specimens. 

 The forward extent of the sujjraocular granules is usually to the 

 middle of the third supraocular as indicated here, but it is usually 

 at the posterior border of the third supraocular in the United States 

 specimens of gularis and 2)erflexus. 



The preanals are highly variable in number, so much so in fact 

 that little attempt has been made to take data concerning them in any 

 species of C nemidopliovus. Fourteen specimens of tcssellatas from 

 Cerros Island were examined for this character and the number was 

 found to be 3 in 1 specimen, 4 in 7 specimens, 5 in 1 specimen, 6 in 

 2 specimens, 7 in 1 specimen, 8 in 1 specimen, and 9 in 1 specimen. 



A large number of abnormalities in the arrangement of the head 

 plates have been noted, but only some of the most conspicuous and 

 probably most important are given here. A specimen (C.A.S. No. 

 40248) was found in Avhich the frontal was divided by a transverse 

 suture so as to leave a posterior frontal as large as the area occupied 

 by the frontoparietals. (This variation occurs regularly in certain 

 other teiids, such as Ameiva hifrontata hifrontata, Ameiva hifron- 

 tata concolor, and Ameiva hifrontata divisa of northern and north- 

 western South America.) In C.A.S. No. 17368 the right fronto- 

 parietal is united Avith the frontal. The same variation is seen in 

 C.A.S. No. 37683, which also has a small median plate at the base 

 of the frontoparietals. This latter abnormality is of rather fre- 

 quent occurrence in Cnem'/dophorus, but in spite of this Barbour 

 (1928) has mentioned it as a diagnostic feature of his C. lemniscatus 

 ruatanus, which is represented by only the type specimen. Many 

 specimens have been examined which show the progressive union 

 of the two frontoparietals into a single scute. These are not con- 



