Whilst Calmette has found that the venom of different kinds of reptiles exhibits marked differences in its toxic character, he has also discovered that the venom secreted by one and the same serpent varies considerably, according to the length of time the animal has fasted. He describes how he kept a _naja haje_ (Cleopatra's asp) in his laboratory, which during the whole eight months that it lived never took any food whatever, although it was offered the most diverse dainties. On its arrival it was made to bite on a watch-glass, this being one method adopted for collecting the venom; the liquid was at once dried, and 0·7 milligramme was found to kill a rabbit weighing nearly four pounds in four hours. Two months later on, when the venom was again collected, 0·25 milligramme proved a fatal dose. On the death of the animal, at the end of eight months, the venom extracted from the glands was so toxic that it only required 0·1 milligramme to kill a rabbit of about the same weight as the previous one. The same curious fact was noted in the case of a cobra's venom. Another circumstance which appears to control the degree of toxicity inherent in serpent venom is the interval of time which elapses between two successive bites. The longer the interval the more virulent is the venom; and Calmette points out that these observations are in accordance with what has for a long time been known in France with respect to indigenous vipers--that their bites are far more dangerous and far more fatal in the spring, after the winter period of torpor is over, than in the autumn. Until quite recently it was thought that the only creatures which could resist the fatal action of this poison were serpents, both poisonous and non-poisonous. Calmette was led to this conclusion because, although he inoculated large doses, as much as ten drops, into cobras, they suffered absolutely no inconvenience, and the same results were obtained with harmless snakes. On repeating these experiments, however, and using much larger quantities of venom, Calmette has found that they do ultimately succumb. That their susceptibility in comparison with other animals is very slight, may be gathered from the fact that a lethal dose of venom for reptiles is roughly estimated to amount to as much as three times the quantity of venom normally present in their respective poison glands. These animals, therefore, although very refractory, are not absolutely immune from the action of venom-toxin.