_Morbid poisons._ Hoffa found in anthrax cultures a ptomain which killed with anthrax symptoms. Hankin obtained a deadly albumose which in small doses procured immunity. Brieger and Fraënkel separated a toxalbumin, and Martin too, a protalbumose and a deuteralbumose together with a ptomain. Marmier separated a toxin which did not give the reactions of the albuminoids, albumoses, peptones nor alkaloids. This was not poisonous in small doses, to animals possessing immunity—natural or artificial. It was weakened but not destroyed by 110° C., and was rendered harmless by treatment with alkaline hypochlorites. Immunity could be induced by its use in small non-fatal doses. This is present in the bacilli and being soluble in water can be secured from these by diffusion in watery fluids and especially so if aided by heat. _Incubation._ The implanted bacillus begins at once to multiply in the tissues, but the encrease is at first slow and the resulting morbid phenomena slight, so that there appears to be a period of incubation. In experimental inoculations in which this can be certainly noted it extends from one to two and even three days in the rabbit and Guinea pig, from two to four days in sheep, and from three to six days in horses and cattle. It may be shortened by giving an overdose and especially if this is introduced intravenously, the chemical poisons apparently acting at once. In young animals, too, the period is shortened, unless they have been rendered refractory by a milk diet or otherwise.