When absorbed by the _digestive_ tract, the venoms of COLUBRIDÆ often produce no ill-effects. It is otherwise with those of VIPERIDÆ. The venom of _Lachesis_, for example, if administered in sufficient doses, sets up acute inflammation of the gastric mucous membrane, and the animals speedily succumb with attacks of gastro-intestinal hæmorrhage, even before it has been possible for the toxic effects upon the nerve-cells to become apparent. These facts explain the contradictions that are to be found in the works of different investigators upon this subject. It is affirmed by some writers that venom can be swallowed without danger, and they even advise the sucking of venomous wounds in order to hinder its absorption. Others, including Sir Joseph Fayrer, Richards, and Weir Mitchell, have killed pigeons and fowls by making them ingest venom of _Vipera russellii_, or _Crotalus_. C. J. Martin, in experimenting upon rats with the venom of _Pseudechis_ (COLUBRIDÆ), has succeeded in keeping these animals alive for a whole week by providing them every day with a ration of bread and milk mixed with a dose of venom one hundred times greater than the lethal dose for a subcutaneous injection. This innocuousness of the venoms of COLUBRIDÆ, which I have frequently been able to establish by causing them to be ingested by different animals, is explained by the fact that the pancreatic juice and the ptyalin of the saliva very rapidly modify the proteic substances to which these venoms owe their toxicity, so that this disappears. No trace of them is found in the fæces. The glandular secretions of persons bitten by venomous snakes, and those of animals inoculated with doses of venom calculated to kill only after a few hours, are not infrequently found to be toxic. In the case of the urine in particular this has been shown to be so. Observations have also been recorded by C. Francis[18] and Sir Joseph Fayrer with reference to the passage of venom through the mammary gland. In the year 1893 a poor Mussulman woman died at Madras from the bite of a Cobra. She was nursing her child at the time, and the latter succumbed in its turn a few hours later, with all the symptoms of poisoning, although it had not itself been bitten, and had been suckled by its mother only once since the bite. The _histological lesions_ produced by snake poisoning have been particularly well studied by Hindale,[19] Karlinski,[20] Nowak,[21] Louis Vaillant-Hovius,[22] and Zeliony.[23] (1) ACTION UPON THE LIVER. Whether we are dealing with the venoms of VIPERIDÆ or COLUBRIDÆ, the anatomo-pathological processes are alike, and the changes produced are more or less profound, according to the degree or the slowness of the intoxication.