1S97.J LIZAKDS OF THE GENUS SCELOPOBTJ?. 495 



2. S. darJcii differs from S. spinosus in having a greater number 

 of femoral pores (11 to 13 instead of 7 to 10). But as this is the 

 only difference I can perceive, and as I have no doubt an exami- 

 nation of a larger material would reveal greater variation than that 

 with which we are at present acquainted, resulting in an overlapping 

 of the two numbers, I consider this species merely as a form of 

 *S'. sjiinosiis, which " seems confined to South-eastern Arizona, 

 whence it is found southward into Mexico for an unknown distance, 

 probably confined to the western slope of the Sierra Madre." 



3. S. magister and *S^. Jloridanus (" or S. S2)ino»us Jloridanus if 

 the number of femoral pores should be found to intergrade ") are 

 both distinguished by longer ear-scales, as may be seen from the 

 figures annexed to Stejneger's paper, and the pores number 11 to 

 16. The character of the auricular scales is, however, much more 

 variable than one would infer on the authority of Stejneger. On 

 comparing his figures with mine (p. 494, d, e,f) and with Bocourt's 

 (pi. xviii. fig. 5), it will be seen that such a character is too incon- 

 stant for specific distinction \ in the absence of any others, which 

 I am unable to find and which Stejneger has, so far, not given. I 

 have carefully compared specimens from Texas (S. Jloridanus) with 

 some from Arizona (iS. mar/ister), and I am quite unable to separate 

 them. There is absolute identity between specimens from Waco 

 and from Tampico. I believe geographical considerations have 

 influenced Stejneger in separating the western from the eastern 

 form, — S. magister being found, according to him, in S.W. Utah, 

 the Grand Canon of Colorado, S. Nevada, the desert region of 

 Cahfornia, and Arizona ^ ; S. Jloridanus from Pensacola to Southern 

 Texas. But I would not allow such considerations to weigh in 

 the distinction of species or even varieties, as they must vitiate 

 any iiltimate deductions in the study of geographical distribution. 

 We know that gaps may occur in the distribution of a species. To 

 mention one example taken from the European fauna, the Eeptiles 

 of which are certainly better known than those of any part of 

 America, I will allude to the common Adder, Vipera herus, which 

 disappears from the plains of France south of the Loire, where it 

 becomes replaced, as well as in the Pyrenees, by V. aspis, and 

 reappears again in the hilly districts of North-western Spain and 

 North Portugal. 



On the whole, however, the ear-scales of S. Jloridanus (fig. 2, e) 

 and S. magister (fig. 2, /), which I unite as one form, are more 

 elongate than in S. clarJcii and S. spinosus, though they intergrade 

 in Texan specimens (fig. 2, d), and I therefore believe it best to 

 separate S. magister as a variety. 



The arrangement proposed is the following : — 



1. Var. horridus, Wiegm. Femoral pores 2-6 on each side; 

 auricular scales usually not or but slightly longer than broad. 



1 In S. undulatnsvre. meetwith a similar amount, of variation, andlam unable 

 to avail myself of the character for defining varieties, in spite of repeated efforts 

 to do so. 



^ Has since been recorded from New Mexico. 



33* 



