Dr. Günther and Sir Joseph Fayrer both mention a peculiarity of this kind in some of the Eastern representatives of the _Crotalus_, viz. the _Trimeresuri_, Indian tree snakes. The former writes: ‘Some have prehensile tails, which, when not so occupied, vibrate rapidly, producing a rustling sound among the leaves.’[54] Others of the family have horny tails. Dr. Andrew Smith, in his _Zoology of South Africa_, mentions _Vipera caudalis_ especially, as having a ‘tail distinctly recognised, at the termination of his very thick body, and which is not often seen.’ In the vipers, however, more than others, tails are distinguishable, those of many of them being short as well as suddenly tapering to a point. The deadly Puff adder is called _Brachyura_ on this account, its tail being extremely short for the size of the snake. One exceedingly dangerous kind in St. Lucia is known as the ‘Rat-tailed snake.’ For climbing, and as a propelling power, this slender tail can be of little service. In St. Lucia is also a ‘Rat snake,’ _Crebo_ or _Cribo_ in vernacular (_Spilotes variabilis_), one of the active non-venomous kinds which, not content with rats and mice for food, wages war on its most venomous fellow-reptiles; as the ‘Racer’ and the ‘King snake’ do against the rattlesnake of the United States. This _Crebo_ is a graceful, elegant creature, and on account of its twofold virtues of mouser and ‘rat-tail’ catcher, is domesticated and petted in some of the islands.[55] In many of the Colubrine snakes it is almost impossible to distinguish where the ribs cease and the tail begins, except by the anus, so very gradually does the body taper. Nor does there appear to be any certain rule about the _length_ of tails, which in some snakes are even longer than their bodies, and in others not one-tenth the length. In giving the length of a few snakes (not in feet or inches, but in the number of their vertebræ), the reader will obtain a clear idea of this variation in tails. One species of rattlesnake has 194 vertebræ, of which 168 support each a pair of ribs, leaving 24 for its tail, or one-eighth. The python has 291 vertebræ, of which the 3d to the 251st support a pair of ribs, leaving 40 for its tail, or less than one-seventh of its length. Let me explain a seeming discrepancy of arithmetic. The spine of the boa constrictor consists of 304 vertebræ, of which 2 next the head support no ribs, and 252 support each a pair of ribs. Taking away the first two, which, having _no ribs_, may be said to form the neck of the snake, that leaves fifty joints for the tail, or about one-sixth of the entire length. Our little sums, therefore, are as follows, in reckoning the vertebræ:— RATTLESNAKE. BOA CONSTRICTOR. PYTHON.