SNAKE POISON. 57 
venom. By appropriate experiments they excluded micro-organisms, 
ferments, alkaloids, ptomaines, and crystalline acids. 1 They next showed 
that there are three protoids in the secretion; one, an albumin, is not 
irulent; but the other two, which are proteoses (pn it .0- and hotero-pro- 
teose), are extremely poisonous. Their action is the same as that of the 
venom itself. They, like the venom, can he momentarily boiled without 
impairing their activity, but prolonged boiling for days destroys then' 
virulence. 
The action of the poison is local and general. The most marked 
local effect is oedema ; the general symptoms in non-lethal doses consist 
of twitching and convulsions. A fatal dose kills within a few seconds 
or minutes. There is also a peculiar effect on the blood. More than a 
century ago, the Abl >e Fontana 2 noticed that the blood of animals 
killed by viper bite remained fluid. Brainard, 3 writing more than forty 
years ago, states that when death occurs immediately, in animals bitten by 
rattlesnakes, the blood is found, at the post-mortem examination to be 
clotted : but if some time elapses before the animal succumbs, the blood 
remains fluid in the vessels. The continued fluidity of the blood has 
since then been noted by numerous observers in the case of various 
snakes. These observations are explained by C. J. Martin's researches. 
He found that different doses produce different results. Immediately after 
the introduction of the venom, the coagulability of the blood increases, 
and this increase in the case of moderate or large doses (more than 
- 0001 grin, per kilog. of body weight) culminates in intravascular 
clotting of greater or less extent. The injection of smaller doses 
produces a transient phase of increased coagulability, but after two 
minutes this is succeeded by a negative phase ; the blood when drawn 
either fails to clot at all, or does so only after the lapse of several hours. 
The thrombosis occurs more readily in venous than arterial blood, and is 
frequently confined to the portal area. These results show a great 
resemblance between the action of the venom and that of nucleo-proteid. 
The effect of diminished coagulability is not unexpected, seeing that the 
principal substance in the venom is proteose, but the minuteness of the 
di ise necessary is very striking and distinctive. The smallness of the dose 
suggests that the injected material does not itself contribute to fibrin- 
formation. It probably acts by producing disintegration of the cells of 
the endothelium of the blood vessels, or, according to Martin's later 
observations, of the red corpuscles ; in either case the result would be 
liberation of nucleo-proteid material. 
With regard to the question of how these poisonous proteoses are 
formed, Martin puts forward the following hypothesis : the cells of 
the venom gland exercise a hydrolysing agency on the albumins supplied 
them by the blood, the results of which influence are the poisonous 
proteoses found in the venom. A difference between the process and 
digestion by pepsin, or by anthrax bacilli, is that the hydration stops 
short at the proteose stage, and is not continued so as to form peptone, 
or simpler nitrogenous materials, like leucine, tyrosine, or alkaloids. 
Gland epithelium is certainly capable of exercising such a hydrolysing 
influence ; the conversion of glycogen into sugar in the liver cells is 
one of the best known examples. 
1 A questionable trace of organic acid found did not possess toxic properties. 
2 Fontana, "Poisons," Trans, by J. Skinner, London, 1787. 
8 Rep. Smithstm. Inst., Washington, 1854. 
