1906.] IN MEXICAN LIZARDS. 299 



new varieties, races, &c. are formed, or at least initiated. 

 Whether they can hold their own, can become common, or 

 predominant, depends upon the test of life and time. In any 

 case, when we speak of them as new varieties or races &c., we 

 thereby but register the fact that the characters of these lizards 

 in certain localities average differently from what we are pleased 

 to consider the normal, more universal stock. 



On the other hand, where environmental conditions are stable, 

 or when the new homes imply no bionomic change, there should 

 be no reason for shaking up the organism ; it should arrive at a 

 settled condition, and the only changes, if any, should be very 

 gradual and orthogenetic, following the law of improvement. 



It follows from these considerations, that the evolution of new 

 species should be most active, most obvious in varied, not in 

 monotonous districts. It also follows that change of environmental 

 conditions need not imply migration, or spreading, but change of 

 conditions in loco. 



It is a kaleidoscopic process — a stirring up, and there are new 

 combinations, some of transient existence, others are obvious 

 failures, others are lucky hits which should be the most successful 

 according to all canons. They seem to fall into the category of 

 " mutations," but to a rather mild extent, since the game is played 

 fairly with all the pieces or characters ujDon the board, none more 

 and none less. The game has been played incessantly and in 

 many places by these lizards, hence the possibility of the occurrence 

 of the same combinations at different times and places ; and such 

 coincidences become probabilities when the performers are of a 

 kindred stock and play under similar conditions those subtle rules 

 and influences and traditions which Nature is able to " corriger 

 la fortune." 



The distressingly laborious examination of half a thousand 

 specimens, of Cnemidophorus has not been in vain, since it has 

 revealed not a few instances which are favourable to the inter- 

 pretations and to the general conclusions given in this paper. 

 None amounts to proof, but even an occasional glimpse may lead 

 to a path into and eventually out of a tangle which hides quagmires 

 and precipices with its luxui'iant and often thorny growths, but 

 which above all attracts us by its glorious and mysterious beauty. 



Definition of the genus Cnemidophorus Wagler, pt. ; Boulenger, 

 Cat. Lizards, ii, 1885, p. 360. — Swift- I'unning, strictly humivagous 

 Tejidse, with a long, narrow arrow-headed tongue, not retractile into 

 the base ; lateral teeth compressed, bi- or tricuspid ; head covered 

 with lai"ge regular shields ; anterior nasals in contact with each 

 other; dorsal scales granular ; ventrals large, forming regular rows; 

 with a double collar-fold ; with femoral pores ; tail round ; eyelids 

 and ear-opening well developed. 



