36 MR. G. A. BOULEXGER 0^' LIZARDS 



superciliaries aud the supraoculars is not a constant character for the separation of 

 .L. viridis from I/, agilis, and he alluded to specimens from Turin and Persia as being 

 devoid of such granules. I may add that out of 37 specimens from France and the 

 Channel Islands, 12 lack the granules altogether, whilst in 9 others they are reduced 

 to one, two, or three. 



Among other characters, not clearly formulated before for the identification of 

 perplexing individuals of the two species, I would mention the number of subdigital 

 lamellaB under the fourth toe (16 to 23, usually 18 to 21, in L. agilis; 21 to 31, 

 usually 23 to 27, in L. viridis) and the number of shields bordering the nostril 

 (3 or 4, rarely 2, in L. agilis; 5 or G, rarely 4, in L. viridis). The more elongate 

 head of L. viridis, though pretty striking in most cases *, is not a cliaracter to be 

 absolutely relied upon, as I can lay side by side individuals of the two species which 

 are absolutely identical in this respect. The same is true of the length of the 

 tail, which, although usually at least twice as long as head and body in tlie adult 

 L. viridis, varies between If and 2§ times, and between Ij and 1| times in L. agilis ; 

 there is, therefore, a considerable overlap which must be borne in mind when making 

 use of this character for the identification of critical specimens. 



Although it is undeniable that a nearly complete passage exists between L. agilis 

 and Ij. viridis, a sufficient number of characters are available, taken in combination, 

 to remove all doubts as to the correct naming of any individual, however aberrant, out 

 of the very large number that have hitherto passed through my hands. But these 

 characters are not those usually given in books, and 1 would, on this occasion, allude 

 to one in particular — Schreiber's new edition of the ' Herpetologia Europa;a,' — to warn 

 beginners from placing too much reliance on descriptions which contain an undue 

 proportion of errors, often the result of uncritical compilation, errors which must mis- 

 lead the student unable to distinguish Avhat is original from what is second-hand. I 

 regard the treatment of the genus Lacerta in tliat work as particularly unfortunate, 

 the key intended to facilitate the distinction of the species being most misleading. 



Laceeta parva. 



Lacerta parra Bouleng. Cat. Liz. iii. p. 22, pi. i. fig. 1 (1<S87) ; "Werner, Sitzb. Ak. Wien, cxi. i. 

 1902, p. 1080, pi. i. fig. 4, & pi. ii. ; Nilvolsky, Mittli. Kauk. Mus. iv. 1909, p. 306 ; Nesterov, 

 Ann. Mus. Zool. St. Petersb. xvii. 1912, p. 74. 



Head and body feebly depressed. Head small. If to If times as long as broad, 

 4^ to 4^ times in length to vent in males, 4f to 5 times in females, its depth equal to 

 the distance between the anterior corner of the eye and the tympanum ; snout 

 obtusely pointed, as long as the distance between the eye and the tympanum, with 

 feebly concave loreal region ; cheeks not much swollen ; length of pileus twice its 



* Fatio (Vert. Suisse), usuallj' so careful in Ids descriptions, is certainly mistaKeu in stating " la jilua 

 graude largeur egalant, eu general, la moitie dc la longueur." According to my measurements, the width is 

 1^' to 1| times in the length. 



