674 MR. Vr. T. BLANFORD ON PERSIAN REPTILES. [June 7, 



I now turn to the Central- Asiatic specimens. These are distin- 

 guished at the first glance from all the Persian examples by the 

 greater development of keels to the scales throughout, and by those 

 of the head, neck, and back being more spinose. This coincides 

 with Pallas's description' of Lacerta sanguinolenta — " L. squamis 

 carinatis imbricata, occipite muricato." The ventral scales are 

 described as " linea elevata carinata." 



Eichwald's description of Agama sanguinolenta is clearly appli- 

 cable to the same form. Agama aralensis is very imperfectly de- 

 scribed by Lichtenstein, but is stated to have fringed edges to the 

 toes ^ ; so I should have thought it a distinct species. Dr. Peters, 

 however, has probably examined the types, which should be in the 

 Berlin Museum. The toes are not fringed iu any of the Lizards 

 examined by me. 



C. Dumeril (/. c.) states that A. sanguinolenta is distinguished 

 from A. agilis by the ventral and lateral scales being keeled, and by 

 the diameter of the ventral being much larger compared with the 

 dorsal scales in A. sanguinolenta than in A. agilis. This latter dis- 

 tinction does not appear, judging from the specimens before me, to 

 be any more constant that the former. The ventral scales are rather 

 smaller in A. agilis; but the amount appears somewhat variable. 



I find that some of the largest males collected by myself in the 

 Indian desert between Sind and Jaisalmir have the strongly keeled 

 m' and mucronate scales of A. sanguinolenta. I see no reason for con- 



sidering these distinct from the other Sind and Persian specimens. 



The conclusion to which I have come is : — that, so far as I can see, 

 the form from Persia, Baluchistan, and Sind is the true Agama 

 agilis of Olivier ; that the identity of Trapelus fiavimaculatus of Rup- 

 pell is probable, but not satisfactorily proved ; and that the Agama 

 sanguinolenta of the countries north of Persia and east of the Caspian, 

 with which Agama aralensis is identified by Peters and ajjparently 

 by Strauch ^ is a variety of A. agilis with more keeled and spinose 

 scales. 



* Agama persica, sp. nov. (Plate LIX.) 



A. sine crista distinguenda nuchali vel dorsali, capite alto, trunco 

 depresso, squamis dorsalibus incequalibus carinatis atque in 

 maribus mucronatis, in medio dorso majoribus, ad latera mino- 

 ribus, aliis maximis, sed nunquam illas in medio dorso diametro 

 duplo excedentibus, singulatim intermixtis. 



Dehbid and Kazrun. 



Form moderately depressed, similar to that of A. agilis, except 



■ Pallas did not examine the species himself, but published Giildenstadt'a 

 notes on it. 



2 Eichwald describes those of Agama sanguinolc7ita as " subfimbriati." 

 ^ The list of reptiles in Severtzoif s ' Turkestanskie Jevotaie ' was drawn up 

 by Strauch, and includes Stellio aralensis. Specimens collected by Severtzoff, 

 and labelled by this name, are, as already stated, in the British Museum, and 

 are identical with Agama sanguinolenta. In the " Reptilia " of the Scientific 

 Results of the Second Yarkand Mission, p. 6, note, I suggested the possibility of 

 Lichtenstein's species not being a Stellio. 



