4 JOURNAL, BOM HAY NATURAL HIST. SOCIETY. Vol. XXI. 



pupil is vertically elliptical, and its iris beautifully speckled with 

 gold. The nostril is slitlike, and placed high on the snout. The 

 tongue is pale at the base, but blackish at the tips. The 

 tail is short, and tapers very rapidly so that it is conical in 

 shape. It is even rougher above than the hinder part of the 

 body. 



Colouration. — The under pails are butt', uniform, or with but little 

 trace of mottling. In the flanks there is a mottling of brown, 

 sometimes of a light shade, sometimes as deep as chocolate. At 

 flrsi very fine this mottling becomes coarser as it ascends the flanks, 

 and then vertical bars of the ground colour pass up to the spine. 

 These bars are much narrower than the intervals. When they meet 

 over the back large somewhat irregularly squarish blotches are 

 formed which proceed from the nape to the tail tip. More often the 

 bars of the two sides alternate, and an irregular dark patchy 

 confluent pattern results. The head is light above with sometimes 

 dark speckling especially about the lips, and a dark irregular stripe 

 passes from the eye to the gape. Dr. Annandale*, who captured a 

 mother and young, saj^s the latter are more brilliantly coloured. 



Identification. — Rtissell's earth snake is very like Linrie's earth 

 snake (jaculus), so much so that I have no doubt the two have 

 been confused repeatedly in the Punjab where they arc associated. 

 It was only in 1909 in this journal that I reported the occurrence 

 of jaculus for the first time within Indian limits, the specimen 

 being captured at Jhelum. Whether it is as rare as this single 

 record might lead one to suppose, remains to be seen. 



The dual association of small head scales, with ventrals so nar- 

 row that they are only twice or little more than twice the breadth 

 of the last costal row, suffices to pronounce the snake an Wryx. 

 ' Onions differs from the other two Indian species (johni and jaculus) 

 in having no groove beneath the chin, and no angular transverse 

 ridge on the rostral shield, so that the identification is extremely 

 easy. A similar specimen with a conical tail, mental groove, and 

 angular ridge on the rostral would prove to bejaculus. 



Dimensions. — The largest specimen I know of was a gravid $ 



* Mem. A> Soc, Bengal, Vol. 1. 1". |>. 193. 



