226 Lacertidx. 



of scales, as in E. ar(jvta. The scales on the sides of the toes some- 

 times form a very slight denticulatiou {E.fasciata, vermiculafa, argnta), 

 thus leading to the genus Scaptira, in which the character is more 

 pronounced. 



Eremias is certainly derived from Niicms, the two species placed at 

 the head of the series establishing the connexion with that genus. 

 Judging from the description, E. neumanni appears to agree with 

 Nucras in the nasal shields, and I have seen specimens of E. arcjus which, 

 owing to the wide separation of the subnasal (= lower postnasal) 

 from the rostral, answer to the definition of that genus so far as this 

 character is concerned, the essential difference between the two 

 genera residing in the presence or absence of keels on the subdigital 

 scales, in the same way as Latastia is separated from Lacerta. 



As in Nucras and in Acantliodactylus, the striation of the young of 

 many species (Sections Lampreremias, Mesalina, Eremias) is of a 

 very primitive type — 7 to 10 light streaks on the neck, 7 to 9 on the 

 body ; there is a frequent tendency for these streaks to break up into 

 ocellar spots, which become converted into transverse series, and 

 running together and losing the light centres, to black cross-bars, 

 exactly as in Nucras* 



34 species are here regarded as more or less well established, 6 of 

 which are not known to me from autopsy. Although I have examined 

 about 1000 specimens, the material at my disposal has not been so 

 exhaustive as for other genera, and I particularly regret a comparative 

 deficiency of Central Asian specimens, which is all the more to be 

 deplored on account of the difficulties with which the systematic 

 treatment of this group is still beset. Fortunately, I have been able to 

 avail myself of the detailed descriptions, based on over -500 specimens, 

 published liy Bedriaga in his monumental work on the Results of the 

 Przewalski Expeditions, but there are still some points that I cannot 

 settle to my satisfaction without direct comparison of the material 

 preserved in the Peti'ograd Museum. 



The genus is divided into 5 sections, which I am fully convinced 

 represent natural associations. The first establishes the passage from 

 Nucras and leads to the second, a terminus group. The third occupies 

 an isolated position and is probably derived from the first. The 

 fourth and fifth sections, although more remote from Nucras than the 

 first, have probably been independently derived from the same original 

 stock, and the species have been evolved on somewhat parallel lines, 

 as is evidenced by the similarity in the arrangement of the ventral 

 plates between some of the South African species and the Asiatic 

 * Cf. figures in Ann. S. Afr. Mns. xiii, 1917, pis. vi, vii. 



