The Report of the 1875 Commission said:—

"It is not possible for us to recommend that the Indian Government should be prohibited from pursuing its endeavours to discover an antidote for snake-bites; or that, without such an effort, your Majesty's Indian subjects should be left to perish in large numbers annually from the effects of these poisons."

Certainly it was not possible; and the numbers are large indeed. During 1897, 4227 persons were killed by wild animals in India, and 20,959 by snakes. (British Medical Journal, 5th November 1898.)

Sir Joseph Fayrer's name must be put in the highest place of all those who have studied the venomous snakes of India.

Sewell, in 1887, showed that animals could be rendered immune, by repeated inoculation with minute quantities of rattlesnake-venom, to a dose seven times as large as would kill an unprotected animal. Kanthack, in 1891, immunised animals in the same way against cobra-venom. He also made experiments to ascertain whether the blood-serum of these animals acted as an antidote to the venom. Then came the work of Calmette, Fraser, Phisalix, Bertrand, Martin (Australia), Stephens, and Meyers. Professor Fraser's observations on the antidotal properties of the bile are, of course, of the utmost importance; not only in preventive medicine, but also in physiology. The results obtained by Calmette are a good instance of the fineness and accuracy of the experimental method. It is to be noted that the animals were inoculated with a fine needle, not thrust into cages with snakes, as at zoological gardens; and that an animal thus poisoned has a painless death. The different venoms were measured in decimal milligrammes, and their potency was estimated according to the body-weight of the animal inoculated. As with tetanus, so with snake-venom, there must be a standard, or "unit of toxicity."

"The following table gives the relative toxicity, for 1 kilogr. of rabbit, of the different venoms that I have tested. To denote this toxicity I use terms such as Behring, Roux, and Vaillard used for the toxin of tetanus, taking the number of grammes of animal killed by one gramme of toxin:—

1.Venom of naja0.25 mgr. per kilogr.
of rabbit.
One gramme of this venom kills 4000 kilogrammes
of rabbit; it has, therefore, an activity of		 4,000,000
2.Venom of hoplocephalus0.29 mgr.3,450,000
3.Venom of pseudechis1.25 mgr800,000
4.Venom of pelias berus4.00 mgr250,000

"Of course, this estimation of virulence is not absolute; it varies considerably according to the species of animal tested. Thus the guinea-pig, and still more the rat, are extremely sensitive. For instance, 0.15 mgr. of viper-venom is enough to kill, in less than 12 hours, 500 grammes of guinea-pig; so that the activity of this venom with a guinea-pig is 3,333,000, but with a rabbit is not more than 650,000. With more resistant animals, the opposite result is obtained; about 10 mgr. of cobra-venom are necessary to kill a dog of 6.50 kilogrm. weight; but to kill the same weight of rabbit 1.65 mgr. is enough. Thus the virulence of this venom with the rabbit is 4,000,000; but with the dog not more than 650,000."

By experiments in test-tubes, Calmette studied these venoms under the influences of heat and various chemical agents. He found how to attenuate their virulence, and how to diminish the local inflammation round the point of inoculation; and it was in the course of these test-tube experiments and inoculations that he discovered the value of calcium hypochlorite as a local application. Working, by various methods, with attenuated venoms, he was able to immunise animals:—

"I have come to immunise rabbits against quantities of venom that are truly colossal. I have got several, vaccinated more than a year ago, which take, without the least discomfort, so much as 40 mgr. of venom of naja tripudians at a single injection; that is to say, enough to kill 80 rabbits of 2 kilogr. weight, or 5 dogs.