THE SNAKES 2539 



ing to do with real affinity. Certain groups of snakes, such as the members of the 

 viper family, may, however, be wholly poisonous; while in other groups, such as 

 the typical snakes, some species may be venomous and others innocuous. Many 

 attempts have been made to draw up a list of characteristics by means of which the 

 harmless members of the suborder can be distinguished at a glance from those 

 which are hurtful. On this point Mr. Boulenger writes " that there is no sure 

 method of distinguishing the two kinds of external characteristics; except, of 

 course, by a knowledge of the various forms. And even then, a cursory examina- 

 tion is not always sufficient, since there is, in some cases, a striking resemblance be- 

 tween snakes of totally different affinities, by which even specialists may at first 

 be deceived. In short, nothing but an examination of the dentition can afford 

 positive information as to the poisonous or nonpoisonous nature of an unknown 

 snake. ' ' 



_. .. . Geologically speaking, snakes are a comparatively-modern group, 



being scarcely known below the lowest portion of the Eocene division 

 of the Tertiary period, although one or two forms have been described from the 

 underlying Cretaceous rocks, and one has recently been recorded from the Gault of 

 Portugal a formation underlying the Chalk. It is noteworthy that one of the 

 North-American lower Eocene snakes has the additional articular facets of the ver- 

 tebrae but very imperfectly developed; and there can be little or no doubt but that 

 the whole group is an offshoot from the lizards. From the commencement of the 

 Tertiary period, the group seems to have gone on steadily increasing in numbers; 

 and it is now represented by some fifteen hundred species, ranging all over the 

 world except New Zealand. Snakes are, however, much more abundant in the 

 moist tropical regions of the globe than in colder regions, and it is there only that 

 they attain their maximum development in point of size. India and the Ma- 

 layan countries, where there are representatives of the whole of the nine families 

 into which the suborder is divided, are the home of a greater number of both gen- 

 era and species of snakes than any other part of the world. Tree snakes are very 

 common in this region, while the gigantic pythons are shared by it in common with 

 Africa. The proportion of poisonous to innocuous species is likewise very high in 

 the Oriental region, and has been estimated at about one in ten. Africa has 

 scarcely half the number of snakes found in the Oriental region; and it is note- 

 worthy that the forms inhabiting Madagascar have but little in common with those 

 of the mainland; the so-called lycodonts, which are so common in Africa, being 

 unknown in Madagascar, while some of the forms from that island are closely al- 

 lied to South-American types. Whereas pit vipers are absent, an especial feature 

 of Africa is the number of typical vipers which inhabit that country; and Australia, 

 which differs so remarkably from India in its tortoises, possesses snakes (and like- 

 wise lizards) closely allied to Oriental forms. Next to the Oriental region, tropical 

 America is richest in ophidians, although the number of generic types is not so 

 great. The proportion of poisonous species is, however, high, and has been esti- 

 mated at as much as one in eight. In Southern Argentina and Patagonia snakes 

 become scarce. Unlike its Chelonians, the snakes of North America present a re- 

 semblance to those of Central America. Indeed, a feature of the whole of America 



