Snakes may be grouped, according to their mode of life, in five principal categories, gradually merging into each other, or two of them not infrequently found combined in one and the same species. These categories are:—Ground-snakes, Sand-snakes, Burrowing-snakes, Tree-snakes, and Water-snakes.

Ground-snakes may be defined as living above ground, and only occasionally climbing bushes or entering the water. Among European genera, Coronella and Vipera are perfect examples of this type, whilst Coluber and Zamenis approach the Tree-snakes in often ascending bushes, or even trees.

Sand-snakes are adapted for living on loose sand, in which they seek concealment. Such are Lytorhynchus and some Psammophis among the ColubridÆ, Cerastes among the ViperidÆ. Eryx connects this category with the next.

Burrowing-snakes live chiefly underground, and often have the visual organ atrophied in consequence, as in Typhlops; all the TyphlopidÆ, GlauconiidÆ, and UropeltidÆ, belong to this category; the Viperid Atractaspis is also a burrowing type.

Tree-snakes spend the greater part of their life on bushes or trees. Corallus among the BoidÆ, Dendrophis and Dendraspis among the ColubridÆ, Atheris and various species of Lachesis among the ViperidÆ, may be quoted as examples.

Of Water-snakes, some are exclusively aquatic, like the marine HydrophiinÆ and the typical AcrochordinÆ (Acrochordus, Chersydrus) and HomalopsinÆ (Hipistes, Herpeton). Chersydrus and Hipistes occur in the sea as well as in fresh water. Many species of Tropidonotus (T. tessellatus and T. viperinus in Europe), as well as the genera Helicops, Grayia, Boulengerina, etc., among the ColubridÆ, Eunectes among the BoidÆ, Ancistrodon piscivorus among the ViperidÆ, are chiefly but not exclusively aquatic.

Our Tropidonotus natrix stands between the Ground-snakes and the Water-snakes; Boas and Pythons are as much Water-snakes as Tree-snakes. As shown by these and many other examples which might be given, a division into categories cannot always be applied with precision, nor does it convey an expression of the natural relationships of the species, as was believed by many systematists of the last century, who appealed to such adaptations for the definition of families.

A vertical pupil denotes more or less nocturnal habits. Nevertheless our European Vipers, which are provided with such a contractile pupil, are far from exclusively nocturnal, delighting to bask in the sun, and pairing and feeding in the day-time. The BoidÆ appear to be more nocturnal, but no snake is known to be absolutely so, and the two species of Coluber which have been found living in perfect darkness in limestone caves in the Malay Peninsula and China, where they feed chiefly on bats, occur also outside the caves, and probably never breed in them.

It is often stated in books that the organs of locomotion for the exceedingly elongate body of snakes are the ribs, and these creatures have even been compared to Centipedes. This statement is no doubt true to a certain extent for slow locomotion on uneven ground, when the ribs and the corresponding ventral shields afford a point of support; but it does not account for the rapid movements, as when a snake darts like an arrow in pursuit of its prey or to escape from an enemy. Besides, the winding motions are not different from those of a Slow-worm or Glass-snake, in which, encased as they are in a bony armour, the ribs cannot come into play at all. The action of the muscles alone is quite sufficient to account for the reptation of snakes, without the ribs having to play an essential part.

Not only the Cobras, but several harmless snakes, are able to raise the anterior third of the body vertically, when taking up a threatening attitude in the presence of an enemy, at the same time widening or inflating the region behind the head.

Most snakes can climb, and in this case the ribs and ventral shields are of great assistance. The Tree-snakes, usually characterized by a very slender, sometimes compressed, body, or by a prehensile tail, are specially adapted for twining themselves round branches, and in several of them the presence of a keel on each side of the ventral and subcaudal shields, accompanied by a notch corresponding to the keel, affords an additional help for climbing on vertical uneven surfaces, such as the trunks of trees. This condition of the ventral shields has a bearing on the extraordinary mode of locomotion with which some Tree-snakes (Chrysopelea, and probably also Dendrophis) have long been credited by the Malays. We allude to the so-called Flying-snakes, remarkable for their habit of shooting down from trees and descending to the ground at an oblique angle, the body being kept rigid the whole time of the “flight.” It has been observed in Chrysopelea that the ventral surface between the lateral keels, which may be compared to hinges, can be drawn in and become deeply concave, whilst at the same time a slight dorso-ventral flattening of the body takes place. During this muscular contraction the snake is like a piece of bamboo bisected longitudinally, and is buoyed up in such a way as to explain its parachute-like descent.

All snakes are able to swim, and the more aquatic kinds may spend a few hours under the water. A Python molurus is known to have remained alive in a basket sunk for thirty-six hours in a river. The best adapted for aquatic life are the HydrophiinÆ, or Sea-snakes, most of which never leave the water, and are quite helpless and soon die when brought on shore; their body is more or less compressed posteriorly, and the tail oar-shaped. Sea-weeds and barnacles sometimes settle on them. AlgÆ have also been observed growing on the fresh-water snake Herpeton tentaculatum.

As regards food, Burrowing-snakes, as well as a few small Ground-snakes, subsist mostly on worms, insects, and myriopods; Tree-snakes on lizards, frogs, birds and their eggs; Water-snakes on fishes and batrachians. Among the other types, some show a predilection for mammals, others for lizards or snakes, whilst not a few feed indiscriminately upon mammals, birds, reptiles and batrachians, even on slugs, insects, and worms, in addition. However surprising, it is a fact that spiny mammals are occasionally eaten, spines of the Madagascar Hedgehog (Ericulus) having been found in the excrements of a Boa madagascariensis. Even hard-shelled eggs and molluscs may constitute the principal or exclusive food of certain snakes.

Thus, Dasypeltis eats nothing but birds’ eggs, the shells of which are crushed in the gullet, by a special contrivance mentioned above (p. 80), and are soon after rejected through the mouth as a pellet. Other snakes, such as Coluber and Lioheterodon show themselves partial to eggs in addition to live prey, but their alimentary canal does not depart from the normal, the eggs being broken in the stomach and the remains of the shells passed with the excrements.

The AmblycephalidÆ subsist almost entirely on snails and slugs, the shells of the former being crushed in the anterior part of the intestine after their contents have been digested, and the dÉbris are rejected through the vent. A small land tortoise has been found in the stomach of a Cobra (Naia haie) from Algeria.

Snakes which take large prey secure it according to three methods: By catching it simply with the jaws, and immediately proceeding to swallow it, as in Tropidonotus and in some of the Constrictors when dealing with small animals; by constriction, after having seized it with the jaws, crushing it in the coils of their body and thus killing it previous to feeding, as in the BoidÆ and Coluber; or by poisoning, by a mere stroke with the fangs, the result being awaited before the meal is begun, as in most of the ViperidÆ. Other poisonous snakes proceed according to the first method, the use of the venom being to reduce the struggles of the victim and to relax its muscles. Such snakes as are in the habit of previously killing their prey show little reluctance to accept dead food in confinement, a thing which others usually refuse to do; they may, however, be deceived by the dead animal being agitated before them, and the system now adopted in our Zoological Gardens, of offering all snakes previously-killed animals, has been attended with comparative success.

Some species feed almost exclusively on other snakes, and often manage to swallow individuals as large as, or even a little larger than, themselves. Examples are known of harmless snakes showing a predilection for dangerous species, to whose poison they are immune (see p. 71).

As a rule snakes that eat fish will also eat batrachians, but nothing higher in the scale, although exceptions have been reported, such as the Anaconda feeding on mammals, birds, reptiles, and fish, and our Grass-snake having taken mice and birds. Some that feed chiefly on lizards and snakes will occasionally eat also mammals, and vice versa, but rarely frogs. On the other hand, European Vipers accommodate themselves to a more varied bill of fare, being known to feed on mammals, birds, reptiles, batrachians, insects, and slugs, and they have even been observed to eat voles showing signs of putrefaction.

The enormous prey which some snakes are able to swallow is quite astounding. Anacondas and Pythons, the largest snakes, have been known to swallow calves and good-sized antelopes with their horns, animals which, even after being somewhat crushed by constriction, very much exceed the calibre of the snake. A Python molurus 17 feet long is reported on good evidence to have swallowed a gravid Axis deer. A Grass-snake half an inch in diameter can manage a frog or toad three times that width, and a Dasypeltis of the same size a hen’s egg. Such feats are rendered possible by the mobility of the jaws and palato-pterygoid arch on the cranium, and the elasticity of the ligaments by which they are attached (see above, p. 42), as well as by the mobility of the ribs and the absence of sternal apparatus, together with the great distensibility of the skin. When a snake proceeds to dispose of a large prey, which, if it be a mammal or bird, is usually seized head-first, it pulls itself forward by alternate movements of the jaws, the maxillary and the mandibular ramus of the one side, and then of the other, being extended anteriorly and laterally, the snake at the same time producing an abundant salivation which renders the prey very slimy. Several repeated alternate movements of the jaws bring the head of the prey to the gullet, where the muscles and ribs come into play, and the two sides of the jaws work no longer alternately, but together. When once in the oesophagus, the prey progresses with much greater facility, and usually reaches the stomach in a few minutes, whilst the previous process of deglutition may have lasted half an hour. While this laborious operation is going on, the breathing of the snake is not impaired owing to a remarkable contrivance: the trachea can be protruded in such a manner as to bring its opening outside the mouth.

In cases where the victim is eaten alive, the snake has to contend with its struggles, but retrogression is rendered impossible by the backwardly-directed sharp teeth with which the jaws and palate are beset. A frog is usually caught by one of the hind limbs and swallowed back-first, the long hind limbs stretching forwards as they fold against the body; its struggles are often still apparent when it has reached the oesophagus. Snakes when caught immediately after a meal are in the habit of disgorging their food, and it sometimes happens that a frog or toad is thus vomited alive. An instance is known of a naturalist having captured a Grass-snake and put it in a linen bag. On opening it a short time after, great was his surprise to find the snake had escaped through a small hole in the bag, leaving instead a living toad too big to pass through the hole.

If not of too large a size, several animals will often be swallowed in rapid succession, after which the gorged snake will allow its digestive organs several days, or even weeks, of repose. A large Anaconda in the Paris Jardin des Plantes fed only thirty-six times in the course of seven years. Digestion is usually rapid in the small snakes, defecation taking place twenty-four to forty-eight hours after the feeding; it lasts much longer in the large Boas and Pythons. Thus, in the above-mentioned Anaconda it has been observed to take from nine to thirty-eight days. Even the hardest bones of birds are decomposed by the gastric juices, but hairs, feathers, and horny productions, are passed with the excrements, sometimes forming regular balls. It is in most cases possible to tell, from an inspection of the dried fÆces, what a snake has been feeding on, hairs, feathers, beaks, claws, epidermal horny shields, bits of tooth-enamel, being found mixed with the chalky matter which represents the decomposed bones. As a rule there is but one defecation after each meal, but there are in addition more frequent renal dejections, consisting chiefly of uric acid.

In captivity snakes show themselves capricious in the choice of food, one individual preferring mammals, whilst another, of the same species, will only take birds; and many, although to all appearances perfectly healthy, will persist in refusing all food, and allow themselves to die of starvation—a suicide which may require months, or even years, to accomplish. A Rattle-snake in the menagerie of the Jardin des Plantes in Paris has lived two years and two months without taking any food, a Python sebÆ nearly two years and a half, a Boa madagascariensis four years and a month. A Viper aspis was kept for three years without food and without losing its vicious temper. Specimens thus fasting do not, as a rule, renew their epidermis, or do so but very rarely. Our Common Adder can very seldom be induced to feed in captivity. Other snakes may rid themselves of all shyness to the extent of taking food from the hand, or show such appetite as to seize a prey immediately on being released from the small box or bag in which they have travelled for a considerable time.

Most snakes drink, and pretty often—not by lapping with the tongue, but by drawing in water from the mouth and immersing the anterior part of the head. Some are said to be fond of milk, but there is no foundation for the belief held by peasants, that they enter sheds with the object of sucking milk from the cows, which would be a material impossibility; their real purpose in visiting such places being a search for suitable dung-heaps in which to deposit their eggs.

Snakes cannot be credited with much intelligence or educability, nor do they display any very marked instincts. The least stupid and most easily tamed are the species of the genera Coluber and Coronella. There is, however, considerable difference in this respect between individuals of the same species. Most snakes, when freshly caught, defend themselves by biting, and some individuals retain their savage temper after months of captivity; others hardly ever bite, even if molested. The Common Grass-snake, for instance, hisses loudly and takes up a very threatening attitude, or even pretends to snap with open mouth, but very seldom bites; its principal defensive action when caught consists in voiding a most repulsive secretion from its anal glands, which it evidently controls, as it ceases doing so when accustomed to being handled. The same snake also produces, during the spring, an oily exudation from the skin which has the same repulsive smell. Mr. H. N. Ridley has observed a Malay snake allied to Tropidonotus, Macropisthodon rhodomelas, to exude drops of a white viscid liquid from the skin of its neck, which is flattened out like that of a Cobra when in an attitude of defence, and he noticed that his dog, seizing the snake to worry it, foamed at the mouth as if he had been biting a toad.

The hissing is produced by the rapid expulsion of air from the lungs through the trachea and the notch at the end of the mouth, which is kept shut at the time. Snakes provided with an epiglottis (see p. 79) produce a much louder hissing. Other sounds are produced by some snakes. Thus, the Indian and African Vipers of the genera Echis and Cerastes make a curious, prolonged, rustling noise, by rubbing the folds of the sides of the body against one another. This sound is produced by friction between the serrated keels of the lateral scales, which are disposed obliquely with their tips directed downwards and backwards; the noise can even be repeated after the death of the animal, by twisting the body and thus rubbing or rasping these little saws against one another. The same thing probably takes place in the African genus Dasypeltis, in which we find a similar arrangement of the scales, though to a less degree.

The best known sounding apparatus is that of the Rattlesnakes, described on p. 20. When alarmed, these snakes gather the body in a few coils or roll themselves up in a spiral, with the tail erect in the centre, and vibrating with great rapidity, whilst the head is ready for attack. Other snakes, such as the Ancistrodon and some species of Coluber and Zamenis, when excited, vibrate the tail in the same manner; but, being deprived of the sound-producing apparatus, this expression of their anger does not attract the same attention. It is from such a habit, however, that the rattle must have been evolved and perfected, not necessarily in a Lamarckian sense, but through the different steps by which evolution or creation has proceeded; Natura non fecit saltus, as LinnÆus well said. Many suggestions have been made as to the use of the rattle. One of them is that the rattling resembles the sound made by locusts, and serves to decoy insect-eating birds; another, that it serves to call the sexes together. Probably it is useful to the snake as a warning to keep off disturbers which cannot serve as food, and thus prevents useless expenditure of venom, or even the breaking of the fangs. At any rate, it gives expression to the snake’s excitement, as does the voice in the case of many other animals, and it seems reasonable to suppose that it may be applied to different purposes. With the advent of man, this means of attracting attention must tend to the more rapid extermination of the snakes which possess it.

Another curious behaviour is that of feigning death, as observed in a harmless but vicious-looking snake, Heterodon, often called Puff-adder in America. It looks more like a Viper than a harmless snake, and when disturbed hisses loudly and flattens out the anterior part of the body, much as does a Cobra, and pretends to strike, although it is one of the few snakes that never bite man. If, however, this display proves of no avail in frightening away the intruder, the snake rolls on its back and opens its mouth, and then lies for a time, which may exceed a quarter of an hour, absolutely motionless, as if dead. As soon as it thinks the danger over, it awakens from its spasm and rapidly moves off. It is the opinion of those who have most experience of this snake that this extraordinary behaviour is not to be explained as a convulsion or faint due to fright, but constitutes a deliberate trick to save its life. Individuals of the South African Ringhals (Sepedon hÆmachates) and of the Common Grass-snake have also been observed to feign death.

The notion that snakes fascinate their prey, attracting it or reducing it to immobility by a mysterious power in their glittering eyes, is pure fable. Animals placed in a cage with a snake evince no particular fright, and fly away when pursued, if not actually turning round to defend themselves. It is even dangerous to offer a good-sized snake a wild rat for food, as all keepers of menageries know.

In cold and temperate climates snakes hibernate, lying more or less torpid in holes or hollow trees, sometimes assembled in numbers and coiled together in a mass. The first thing they do in awakening in the spring is to cast the outer coating of the epidermis, as described above (p. 20). Several exuviations take place during the period of activity, sometimes pretty regularly every month, sometimes at very irregular intervals. A few days previous to this operation the snake is languid and abstains from feeding; its skin is dull and the sight impaired by the opaque condition of the lid; a day or two before moulting, the outer stratum of the epidermis becomes again transparent and the eye clear, through this stratum becoming detached from the subjacent tissue, until it is pulled off in one piece, by the snake rubbing itself against stones or bushes. The first exuviation takes place very shortly after birth.

Snakes are long-lived, although the limit of duration of life is not known in any of them. They grow slowly, and do not appear ever to reach sexual maturity until the fourth year, when they continue increasing in size for a long period. A Python reticulatus and an Ancistrodon piscivorus are reported to have lived twenty-one years in captivity in Paris. The young of many snakes are very secretive, and are not often found in the open, those that are met with being as a rule either new-born or approaching sexual maturity.

Snakes are tenacious of life, and remarkable for the reflex movements which take place after they have been cut to pieces, the severed parts of the body and tail wriggling for a considerable time, and the head endeavouring to bite. Accounts of decapitated Rattlesnakes turning round and striking with their bloody stumps are probably not snake stories.