326 ZOOLOGICAL GEOGRAPHY. [PART III. 
problem admittedly depends. It will, perhaps, be said that a 
great many of the 88 genera above given are very scarce and very 
local; but this is certainly not the case with the majority of them ; 
and even where it is so, that does not in any. degree affect their 
value as indicating zoo-geographical affinities. It is the pre- 
sence of a type ina region, not its abundance or scarcity, that is 
the important fact ; and when we have to do, as we have here, 
with many groups whose habits and mode of life necessarily 
seclude them from observation, their supposed scarcity may not 
even be a fact. 
Reptiles and Amphibia.—Reptiles entirely agree with Mam- 
malia and Birds in the main features of their distribution. 
Out of 17 families of snakes inhabiting Hindostan, 16 range 
over the greater part of the entire region, and only two can 
be supposed to show any Ethiopian affinity. These are the 
Psammophide and Erycide, both desert-haunting groups, and 
almost as much South Palearctic as African. The genus 7’7o- 
pidococcyx is peculiar to the sub-region, and Aspidura, Passerita 
and Cynophis to the peninsula and Ceylon; while a large number 
of the most characteristic genera, as Dipsas, Simotes, Bungarus, 
Naja, Trimeresurus, Lycodon and Python, are characteristically 
Oriental. 
Of the six families of lizards all have a wide range The 
genera Humeces, Pentadactylus, Gecko, Hublepharis, and Draco, are 
characteristically or wholly Oriental; Ophiops and Uromastix 
are Palearctic; while Chamelcon is the solitary case of decided 
Ethiopian affinity. 
Of the Amphibia not a single family exhibits special Ethiopian 
affinities. 
IT. Sub-region of Ceylon and South-India. 
The Island of Ceylon is characterised by such striking pecu- 
liarities in its animal productions, as to render necessary its 
separation from the peninsula of India as a sub-region ; but it 
is found that most of these special features extend to the Neil- 
gherries and the whole southern mountainous portion of India, 
and that the two must be united in any zoo-geographical pro- 
