It was interesting to endeavour to reproduce, with _Cobra_-venom, J. Bordet’s experiments upon alexins and anti-alexins. It was to be hoped that we had in this substance an anti-alexic body capable of being preserved for an indefinite time and constant in its activity, which would enable us easily to measure the dose of alexin contained in a small quantity of a serum, or other liquid of leucocytic origin. The experiment proved to Noc that, contrary to the ideas of Ehrlich and his pupils, and conformably to the results obtained by Bordet with serums and toxins, the neutralisation of venom takes place in a variable ratio. If a dose A of fresh serum is capable of neutralising exactly 5 milligrammes of _Cobra_-venom with regard to a sensitive microbe, on employing a dose of the strength of 2 A we ought to find a bactericidal dose, 1 A, in the excess of serum, according to the theory of definite proportions. No such bactericidal action is seen, however; the serum, on the other hand, acts in the contrary direction by means of its nutritive substances, and in the mixture 2 A + _venom_ we obtain a larger number of colonies of micro-organisms than in the mixture A + venom. We see, then, that the property of cells of fixing in excess the active substance in serums, discovered by Bordet for the hæmolysins (staining phenomena), is met with again in the case of extracts of organs, at least with regard to the bacteriolytic substance of _Cobra_-venom. It results, then, from the foregoing facts that _Cobra-venom contains a cytolysin, which acts upon micro-organisms and is capable of fixing the alexin of normal serums_. The application of these data to the living animal is evidently full of difficulties, by reason of the complexity of the substances that come into play. Let us see, however, to what extent they are capable of serving to explain the phenomena that are produced as the result of poisoning. It was observed by Kaufmann that the cadavers of animals which have died from snake-bite are very rapidly invaded by the bacteria of putrefaction. Welch and Ewing, referring to these phenomena of rapid putrefaction in cases of death from venom, explained them as being due to the loss of the bactericidal power of the serum. In hot countries, even when snake-bites are not fatal, they are frequently complicated by local suppuration or gangrene, occasioned by micro-organisms introduced at the time of the bite. The minute analysis of the phenomena of poisoning shows, in reality, that the organism undergoes different modifications according to the quantity of venom injected and its channel of penetration.