570 JOURNAL, BOMBAY NATURAL HI/ST. SOCIETY, Vol. XXVI. 



The Hill Tree Snake — ^ Dipsadomorphus ceylonensis (Gunther). 



In the paper referred to in the footnote I gave in tabular form a series 

 of 21 speeimens characterised by costals in midbody 19, ventrals varying 

 from 214 to 235 and subcaudals 98 to 109. Between 1909 and 1917 I 

 received from Ceylon and the Hills of South India 11 more specimens with 

 costals 19, ventrals 209 to 240, and subcaudals 94 to 107. This year I have 

 acquired 40 more specimens with costals 19, ventrals 214 to 239, and sub- 

 caudals 94 to 110. In the aggregate then, I have had 72 specimens with 

 costals 19, ventrals 209 to 240, and subcaudals 94 to 110. 



Habitat. — All except seven of these were captured in the Wynaad. 



Sexes.— Oi 33 sexed, 19 were $ and 14 2 . 



Food. — Frogs had been swallowed by five examples, once a Rana limno- 

 charis, and twice a species of Lvalus. Seven other specimens had eaten 

 lizards, three Salea horsfieldi, and one a Calotes versicolor. 



Breeding. — In August I got two females in each of which 7 follicles were 

 impregnated. In September four other females were egg-bound, the 

 numbers of eggs varying from 5 to 8. These, very elongate in shape, 

 measured one inch long in a specimen killed during the last ten days of 

 that month. The prospective dams measured from 321- to 34 inches, 

 (length of two not recorded), a length it would appear they attain at the 

 end of the second year of life. 



Grorvth. — Three examples in August and September measured from 12^ 

 to 15^ inches, four others from 20 to 26\ inches, 10 others from 31| to 39 

 inches, and six others from 44|^ to 50^ inches. These various ranges 

 appear to denote the broods of successive years. 



My largest c5' was 50^, and $ 34 inches. 



Lepidosis. — In three specimens the scale rows were found in places to be 

 21 for a brief interval or intervals, but when critically examined it was 

 found that at these spots the costals remained the same. One or more 

 shields in succession in the vertebral row were divided into three and 

 accounted for the increase to 21, and I find this tendency in certain in- 

 dividuals of all the species of this genus of which I have examined a large 

 series. Posteriorly the costal rows reduce to 15 with great consistency. 



Dentition. — The maxilla in Nilgiri specimens supports from 18 to 20 teeth 

 followed after a short gap by 2 large, grooved, and obliquely-placed 

 pseudo fangs. (In Ceylon specimens there are only 14 to 15 teeth ante- 

 riorly). The palatine has 9 teeth. (In Ceylon specimens 7 to 8). The 

 pterygoid has 19 ? to 21 teeth (Ceylon specimens 18 to 20). The mandible 

 holds 25 to 28 teeth (Ceylon specimens 21 to 24). 



* In 1909 I published a note in the Records of the Indian Museum (pag-es 

 1.51 et seq.) upon certain "forms" of Dipsadomorplms. I expressed the view 

 that ccyZoncjisis (Gunther) as described in Boulenger's Catalog'ue (Vol. iii, i5. (JO) 

 includes four distinct " forms," and ^ave in tabular form the shield characters 

 for each upon which reliance is mainly placed in the separation of the species of 

 this genus. I sufrg'ested that they should each receive recognition as distinct 

 species. Dr. Annandale in a later issue of the same Journal (Vol. iii, part 111. p. 

 281) dissented from my views. The conclusions drawn by me from the rather 

 small series of specimens of each " form "' that I had examined when I wrote the 

 paper referred to, are completely confirmed by the very much larger material now 

 available vvith regard to t>vo of those "forms". The others (one from Ceylon, 

 and one from the Andamans) do not concern us here. The fact that there is some 

 slight overlapping in the ranges of the ventrals shields does not I think invalidate 

 the recognition of two distinct " forms " which I still choose to regard as 

 " species " though I exiject others will not accord to them so exalted a rank. 



