All toxicity tests on mice were made with an exposure of ten minutes, while dogs were exposed for thirty minutes. In case death did not occur during exposure, the animals were kept under observation for several days. Toxicity and all other figures are expressed in milligrams per liter of air, though parts per million (p.p.m.) was frequently used during the early work. Another point of difficulty is the great individual variation in the susceptibility of animals. This is probably greater than when the poison is administered subcutaneously or intravenously. It necessitates the use of a large number of animals in making a determination of the toxicity of a gas. Again, the toxicity for different species may vary, and as the ultimate aim is a knowledge of the toxicity for man, a great many different species must be used. If the toxicity is widely different for different animal species, it is hard to arrive at a definite conclusion as to the toxicity for man. With longer exposures than thirty minutes the lethal concentration is usually less, there being a cumulative effect. This is not true for hydrocyanic acid. If the concentration is not enough to kill at once, an animal can stand it almost indefinitely. Whether the action is cumulative or not depends on the rate at which the system destroys or eliminates the poison. If the poison is being eliminated as fast as received the concentration in the tissues cannot increase. It is stated, for example, that the amount of nicotine in a cigar would kill a man if taken in one dose. If it is spread over twenty minutes, the destruction or elimination of the nicotine is so rapid that no obviously bad effects are noted. Another interesting thing about the work on poison gases is that in most cases a preliminary exposure to less than the lethal concentration does not seem to make the animal either more or less sensitive on a later exposure. This is quite unexpected, because we know that with irritating gases, especially lachrymators, men adapt themselves to much higher concentrations than they could stand at first. In view of the experiences of arsenic eaters, it is quite possible that the experiments, which showed no accustoming to toxic gases, were not continued long enough to give positive results. Not only does the susceptibility of different animals of the same species vary greatly for a particular gas, but the susceptibility of different species varies greatly with different gases. Thus while the effects of certain gases on mice are quite comparable to the effects on man, it is very far from being true with other gases. LACHRYMATORS While one cannot determine the lethal concentrations of poison gases for men, it is possible to determine the concentration that will produce lachrymation. The threshold value is that at which two-thirds of the observers experience irritation. The lachrymatory value is considerably higher than the threshold value.