250 REPORT OF NATIONAL MUSEUM, J89S. 



Cat. No. 2732, froui Cliihuabua, has a series of large rounded dark 

 spots ou each side behind tlie fore leg, and the supraorbital plates are 

 flatter than usual. Cat. No. 2768, from the Colorado Eiver, also shows 

 the spots to less extent, but there is not the same difference iu the 

 orbital plates. The specimens are not in sufficiently good condition to 

 furnish satisfactory indications on other points. 



I)r. Stejneger has pointed out ' that specimens of this species from 

 the Sonorau region always possess two series of frontal scales, while 

 those from the Central region and the Texan district present generally 

 but one row. He states, also, that the specimens which enter the former 

 category have smaller supraocular scales, a narrower head, and a longer 

 muzzle. He regards these as belonging to another species, which he 

 calls G. haileyi. He observes, further, that "there can be no doubt that 

 both forms are subsi^ecies of the same species, but whether the form 

 now named for the first time should receive a trinomial appellation or 

 not is quite another thing, depending, according to the code of zoolog- 

 ical nomenclature adopted by the American Ornithologists' Union, 

 which I adhere to in all my writings, on whether the two forms are now 

 known to intergrade or not." Stejneger then goes on to observe that, 

 " In the collection before me there are a few specimens which present 

 features which at first sight might seem to indicate intergradation," 

 and he points out three specimens of an intermediate character so far 

 as regards the character of the frontal scales. 



My examination of the series in the national collection has convinced 

 me of two things — first, that the differences observed by J>r. Stejneger 

 exist, and are mainly characteristic of distinct geographical areas, and, 

 second, that the transitions are so numerous that a distinct specific 

 appellation is not practicable, and that even a distinct subspecific 

 name is of doubtful utility. In arranging the specimens I set apart 

 three lots — first, those presenting two rows of frontal scales ((7. haileyi) ; 

 second, those with one row, and, third, those bottles which contain 

 specimens, mostly from the same localities, of both kinds. Of the first 

 lot there are thirty-six specimens, in the second twenty-two, and in the 

 third twenty-two. The proportion of specimens which do not show geo- 

 graphical limitation of range, it will be observed, is equal to those of the 

 typical C. collar is, which do so.' Let us now analyze the characters of 

 these two lots. In lot second, six specimens exhibit two rows ; eleven 

 have one scale only which crosses the frontal region; four have two 

 scales which cross, and one has three scales. In lot thiid (C collaris 

 typical), four have one scale across the entire frontal bone; seventeen 

 have two scales, and one has three scales so extending. It seems, then, 

 of specimens in which the series of two scales is interrupted, that in 

 fifteen one scale crosses the frontal space; in twenty-one two scales 

 cross, and two have three scales crossing. Thus we find that the char- 

 acter of the frontal scales is variable, since there are fifteen specimens 



' North American Fauna, No. 3, 1890, p. 103. 



