In 1895 Fraser confirmed these results,[98] and on May 15 in that year exhibited before the Medico-Chirurgical Society of Edinburgh a rabbit vaccinated against a dose of cobra-venom fifty times lethal. At once considering the possibility of obtaining serums highly antitoxic against snake-venoms, and of practical utility in the therapeutics of snake-bites, I prepared to vaccinate a certain number of large animals, horses and donkeys, in order to procure great quantities of active serum. I at first experienced some difficulties in providing myself with a sufficient store of venom. But thanks on the one hand to the obliging collaboration of some of my old pupils or colleagues, and on the other to the valuable co-operation of the Colonial Governments of Indo-China, the French Settlements in India, and Martinique, I soon received poisonous snakes and dried venom in abundance. [Illustration: FIG. 93.--VACCINATING A HORSE AGAINST VENOM AT THE PASTEUR INSTITUTE AT LILLE.] After this I was not long in pushing the vaccination of a few horses until I made them resist, in a single injection, 2 _grammes_ of dry cobra-venom, a dose about _eighty times lethal_; for I was able to satisfy myself that about 0·025 gramme of cobra-venom was sufficient to kill fresh horses in from twelve to twenty-four hours. The immunisation of horses to this very high degree of tolerance of venom is not obtained without difficulties; many animals succumb in course of treatment from endocarditis or acute nephritis; in the case of others, each injection of venom leads to the formation of enormous aseptic abscesses, which have to be opened and drained. It may be said that on an average an interval of _sixteen months_ is necessary in order to obtain a serum sufficiently antitoxic. [Illustration: FIG. 94.--ASEPTICALLY BLEEDING A HORSE VACCINATED AGAINST VENOM IN ORDER TO OBTAIN ANTIVENOMOUS SERUM AT THE PASTEUR INSTITUTE AT LILLE.] When a horse is well vaccinated and tolerates without a reaction 2 _grammes_ of dry cobra-venom in a single subcutaneous injection, it may be bled on three consecutive occasions in the space of ten days, and in this way 20 litres of blood may be drawn from it (fig. 94). The bleeding is arranged in the following manner: _Twelve days_ after the last injection of venom the horse is bled for the first time to the extent of 8 litres; five days later it is bled for the second time to the extent of 6 litres; five days later still the third bleeding takes place, when 6 litres are again withdrawn. The animal is then allowed to rest for three months and supplied with strengthening food, and during this period 2 _grammes_ of venom are again injected on two occasions at the end of a month, followed, a month and a half later, by the injection of 2 more _grammes_. The antitoxic power of the serum is thus maintained approximately at the same standard.