508 REPORT OF NATIONAL MUSEUM, 1898. 



cariiiuted. The trausverse series on each side vaiy from thirty-nine to 

 fifty-two. The head plates are quite symmetrical; their general char- 

 acter will be found detailed under G. muUicarinatiis. The variations are 

 in the size of the large median frontal and of the outer nasals; the 

 number of the postnasals, whether two, three, or four; and the outer 

 row of supraorbitals, whether three or two. 



Young individuals are much more brilliantly colored than adults and 

 have a more distinct color pattern of alternating light and dark cross- 

 bars. On young individuals were proposed the species G. cceruleus and 

 G. webbii. The colors fade out with maturity in most of the species, 

 becoming shades of olive or brown with lateral bars, or if brilliant 

 colors are retained as in G. gramineus and G. auritus, the crossbars 

 disappear. 



There is considerable irregularity in the scales of the head in some 

 of the species, which renders it necessary to analyze closely their 

 homologies. Between the prefrontals and the internasals on each side 

 of the median or interfrontonasal plate there may be one, two, or three 

 plates. When there is one, it generally meets its mate of the opposite 

 side in front of the interfrontonasal. This plate is the prefrontonasal. 

 Sometimes, as G. auritus or G. liocephalus, there may be another pair of 

 prefrontonasals, which mayor may not meet in front of the interfronto- 

 nasal. In addition to these there may be another, generally smaller, 

 plate, which lies between the interfrontonasal and loreal on each side, 

 which I call the lateral prefrontonasal. This never meets its mate of 

 the other side, but is sometimes fused on one or the other side with the 

 prefrontonasal. On the side of the muzzle there may be one or two 

 scales between the nasal and internasal, the supranasals. Posterior to 

 the nasal there may be one or two plates in contact with it, one above 

 the other, the postnasals. Immediately in front of the eyes there are 

 two lilates, one above and one below, the preoculars, which may rarely 

 be fused. Between these and the postn;isals^or nasals when postnasals 

 are absent, is one or two plates, the anterior and posterior loreals. 

 When there is but one plate between the nasal and the postloreal, it is 

 sometimes difficult to determine whether it is the preloreal or the post- 

 nasal. In the case of G. burnettii I do not know which it is. It is 

 probably preloreal, since in the allied G. multicarinatus the inferior post- 

 nasal is sometimes wanting. 



The typical characters of the head ijlates are important as definitive 

 of the species, but their variations render it necessary to rely on other 

 characters. The Sonoran and Western forms are more difficult to dis- 

 tinguish than the Toltecan. I append the following key, which will 

 aid in their determination : 



I. Dorsal scales laij^er than veutrals. 



A. Two pairs of prefrontonasal plates. 

 a Two loreals. 



No lateral perfrontonasals ; elongate supraauricular processes ; green, 

 scales reticulated Avith black; below yellowish.. G. auritus Cope. 



