The gland which secretes the poison is a modification of the parotid salivary gland of other Vertebrates, and is usually situated on each side of the head below and behind the eye, invested in a muscular sheath. It is provided with large alveoli in which the venom is stored before being conveyed by a duct to the base of the channelled or tubular fang through which it is ejected.

In the Vipers, which furnish examples of the most highly developed poison apparatus, although inferior to some in its toxic effects, the poison gland is very large and in intimate relation with the masseter or temporal muscle, consisting of two bands, the superior arising from behind the eye, the inferior extending from the gland to the mandible. When the snake bites, the jaws close up, causing the gland to be powerfully wrung, and the poison pressed out into the duct. From the anterior extremity of the gland the duct passes, below the eye and above the maxillary bone, where it makes a bend, to the basal orifice of the poison fang, described above (p. 55), which is ensheathed in a thick fold of mucous membrane, the vagina dentis. By means of the movable maxillary bone (supra, p. 49) hinged to the prefrontal, and connected with the tranverse bone which is pushed forward by muscles set in action by the opening of the mouth, the tubular fang is erected and the poison discharged through the distal orifice in which it terminates.

In some of the Proteroglyphous Colubrids, as we have seen, the poison fangs are not tubular, but only channelled and open along the anterior surface; and as the maxillary bone in these snakes is more or less elongate, and not or but slightly movable vertically, the poison duct runs above the latter, making a bend only at its anterior extremity, and the tranverse bone has not the same action on the erection of the fangs. Otherwise the mechanism is the same.

In the Opisthoglyphous Colubrids, with grooved teeth situated at the posterior extremity of the maxilla, a small posterior portion of the upper labial or salivary gland is converted into a poison-secreting organ, distinguished by a light yellow colour, provided with a duct larger than any of those of the labial gland, and proceeding inward and downward to the base of the grooved fang; the duct is not in direct connexion with the groove, but the two communicate through the mediation of the cavity enclosed by the folds of mucous membrane surrounding the tooth, and united in front.

The reserve or successional teeth, which are always present just behind or on the side of the functional fang of all venomous snakes, are in no way connected with the duct until called upon to replace a fang that has been lost. It could not be otherwise, since the duct would require a new terminal portion for each new fang; and as the replacement takes place alternately from two parallel series, the new poison-conveying tooth does not occupy exactly the same position as its predecessor.

Two genera, Doliophis among the Elapine Colubrids, and Causus among the Viperids, are highly remarkable for having the poison gland and its duct of a great length, extending along each side of the body and terminating in front of the heart. Instead of the muscles of the temporal region serving to press out the poison into the duct, this action is performed by those of the side of the body.

When biting, a Viperid snake merely strikes, discharging the venom the moment the fangs penetrate the skin, and then immediately leaves go. A Proteroglyph or Opisthoglyph, on the contrary, closes its jaws like a dog on the part bitten, often holding on firmly for a considerable time.

The poison, which is mostly a clear limpid fluid of a pale straw or amber colour, more rarely greenish, sometimes with a certain amount of suspended matter, is exhausted after several bites, and the glands have to recuperate.

It must be added that the poison can be ejected otherwise than by a bite, as in the so-called Spitting Snakes of the genera Naia and Sepedon. The fact that some of these deadly snakes when irritated are in the habit of shooting poison from the mouth, at a distance of 4 to 8 feet, even apparently aiming at a man’s face, has been too often witnessed in India and Malaya, and especially in Africa, from the days of the ancient Egyptians, for any doubt to subsist as to their being endowed with this faculty, but the mechanism by which this action is produced has not been satisfactorily explained. In all probability, the poison escapes from the sheath of mucous membrane surrounding the base of the fangs, and is mixed with ordinary saliva, the membranes of the mouth perhaps acting as lips, in which case the term “spitting” would not be incorrect. The spitting, which may take place three or four times in succession, has been observed to be preceded by some chewing movements of the jaws. If reaching the eye, the poisonous fluid causes severe inflammation of the cornea and conjunctiva, but no more serious results if washed away at once.

Snake poisons is a subject which has always attracted much attention, and which has made great progress within the last quarter of a century, especially as regards the defensive reaction by which the blood may be rendered proof against their effect by processes similar to vaccination—antipoisonous serotherapy. The studies to which we allude have not only conduced to a method of treatment against snake-bites, but have thrown a new light on the great problem of immunity. They have shown that the antitoxic serums do not act as chemical antidotes in destroying the venom, but as physiological antidotes; that, in addition to the poison glands, snakes possess other glands supplying their blood with substances antagonistic to the poison, such as also exist in various animals refractory to snake poison, the hedgehog and the mungoose for instance. Unfortunately, the specificity of the different snake poisons is such that, even when the physiological action appears identical, serum injections or graduated direct inoculations confer immunity towards one species or a few allied species only. Thus, a European in Australia who had become immune to the poison of the deadly Notechis scutatus, manipulating these snakes with impunity, and was under the impression that his immunity extended also to other species, when bitten by a Denisonia superba, an allied Elapine, died the following day. In India, the serum prepared with the venom of Naia tripudians has been found to be without effect on the poison of Naia bungarus, the two species of Bungarus, and the Vipers Vipera russelli, Echis carinatus, and Lachesis gramineus. Vipera russelli serum is without effect on Colubrine venoms, and on those of Echis and Lachesis. In Brazil, serum prepared with the venom of Lachesis lanceolatus has proved to be without action on Crotalus poison. These examples, and others which could be given, show that the hopes which were at first entertained as to the benefits to be conferred on mankind by the serum treatment were somewhat over-sanguine—at least as regards countries like India, where, different kinds of poisonous snakes occurring together, it is sometimes impossible to know by which the bite has been inflicted.

Chemistry teaches that snake venoms consist for the most part of solutions of modified proteids, and all attempts to separate the toxic principles from such proteids have hitherto been unsuccessful. Accordingly, at the present time we must regard such toxic principles as residing in some special grouping of a portion of the atoms in the complex venom proteid molecule. The analysis of their physiological actions has proved them to be made up of a great many more constituents than would be imagined from their chemical composition.

The effect of the poison of Proteroglyphous Colubrids (Hydrophids, Cobras, Bungarus, Elaps, Pseudechis, Notechis, Acanthophis) is mainly on the nervous system, respiratory paralysis being quickly produced by bringing the poison into contact with the central nervous mechanism which controls respiration; the pain and local swelling which follow a bite are not usually severe.

Viper poison (Vipera, Echis, Lachesis, Crotalus) acts more on the vascular system, bringing about coagulation of the blood and clotting of the pulmonary arteries; its action on the nervous system is not great, no individual group of nerve cells appears to be picked out, and the effect upon respiration is not so direct; the influence upon the circulation explains the great depression which is a symptom of Viperine poisoning. The pain of the wound is severe, and is speedily followed by swelling and discoloration. The symptoms produced by the bite of the European Vipers are thus described by the best authorities on snake poison (Martin and Lamb): The bite is immediately followed by local pain of a burning character; the limb soon swells and becomes discoloured, and within one to three hours great prostration, accompanied by vomiting, and often diarrhoea, sets in. Cold, clammy perspiration is usual. The pulse becomes extremely feeble, and slight dyspnoea and restlessness may be seen. In severe cases, which occur mostly in children, the pulse may become imperceptible and the extremities cold; the patient may pass into coma. In from twelve to twenty-four hours these severe constitutional symptoms usually pass off; but in the meantime the swelling and discoloration have spread enormously. The limb becomes phlegmonous, and occasionally suppurates. Within a few days recovery usually occurs somewhat suddenly, but death may result from the severe depression or from the secondary effects of suppuration. That cases of death, in adults as well as in children, are not infrequent in some parts of the Continent is mentioned in the last chapter of this Introduction.

The bite of all the Proteroglyphous Colubrids, even of the smallest and gentlest, such as the Elaps or Coral-snakes, is, so far as known, deadly to man. The ViperidÆ differ much among themselves in the toxicity of their venom. Some, such as the Indian Vipera russelli and Echis carinatus, the American Ancistrodon, Crotalus, Lachesis mutus and lanceolatus, the African Causus, Bitis, and Cerastes, cause fatal results unless a remedy be speedily applied. On the other hand, the Indian and Malay Lachesis seldom cause the death of man, their bite in some instances being no worse than the sting of a hornet. The bite of the larger European Vipers may be very dangerous, and followed by fatal results, especially in children, at least in the hotter parts of the Continent; whilst the small Vipera ursinii, which hardly ever bites unless roughly handled, does not seem to be possessed of a very virulent poison, and, although very common in some parts of Austria-Hungary, is not known to have ever caused a serious accident.

It is noteworthy that the size of the poison fangs is in no relation to the virulence of the venom. The comparatively innocent Indo-Malay Lachesis alluded to above have enormous fangs, whilst the smallest fangs are found in the most justly dreaded of all snakes, the Hydrophids.

Little is known of the physiology of the poison of the Opisthoglyphous Colubrids, except that in most cases it approximates to that of the Proteroglyphs. Experiments on Coelopeltis, Psammophis, Trimerorhinus, Dipsadomorphus, Trimorphodon, Dryophis, Tarbophis, Hypsirhina, and Cerberus, have shown these snakes to be possessed of a specific poison, small mammals, lizards, or fish, being rapidly paralyzed and succumbing in a very short time, whilst others (Eteirodipsas, Ithycyphus) do not seem to be appreciably venomous. Man, it is true, is not easily affected by the bite of these snakes, since, at least in most of those which have a long maxillary bone, the grooved fangs are placed too far back to inflict a wound under ordinary circumstances. There are, however, exceptions. A case was reported a few years ago of a man in South Africa nearly dying as a result of the bite of the Boomslang, Dispholidus typus, the symptoms, carefully recorded, being those characteristic of Viperine poisoning, an important fact to oppose to the conclusions, based on the physiological experiments on Coelopeltis, which appeared to disprove the theory that the ViperidÆ may have been derived from Opisthoglyphous Colubrids.

Experiments made with the secretion of the parotid gland of Tropidonotus and Zamenis have shown that even Aglyphous snakes are not entirely devoid of venom, and point to the conclusion that the physiological difference between so-called harmless and poisonous snakes is only one of degree, just as there are various steps in the transformation of an ordinary parotid gland into a poison gland or of a solid tooth into a tubular fang.

The question whether all snakes are immune to their own poison is not yet definitely settled. Most snakes certainly are, and it is a remarkable fact that certain harmless species, such as the North American Coronella getula and the Brazilian Rhachidelus brazili, are proof against the poison of the Crotalines which frequent the same districts, and which they are able to overpower and feed upon. The Cribo, Spilotes variabilis, is the enemy of the Fer-de-lance in St. Lucia, and it is said that in their encounters the Cribo is invariably the victor. Repeated experiments have shown our Common Snake, Tropidonotus natrix, not to be affected by the bite of Vipera berus and V. aspis, this being due to the presence, in the blood of the harmless snake, of toxic principles secreted by the parotid and labial glands, and analogous to those of the venom of these Vipers.

The Hedgehog, the Mungoose, the Secretary Bird, and a few other birds feeding on snakes, are known to be immune to an ordinary dose of snake poison; whether the pig may be considered so is still uncertain, although it is well known that, owing to its subcutaneous layer of fat, it is often bitten with impunity. The Garden Dormouse (Myoxus quercinus) has recently been added to the list of animals refractory to Viper poison.