Reviewers of our knowledge of the mode of formation of endospores in bacilli usually present the evidence under one or more of three theses: (1) the endospore is formed by the growth of a specialized granule in the bacterial protoplasm; (2) the endospore is formed by the fusion of a number of granules, some of which may be nuclear material; (3) the endospore is formed by gradual condensation of cell-substance, possibly due to dehydration. These notions have been derived from the study of both stained and unstained bacilli. Although the origins of these conceptions are obscure, they can be associated with names of distinguished bacteriologists. The first of these opinions was expressed by Koch (1876) in his description of the growth of the spore from a granule in Bacillus anthracis. de Bary (1887) noted the growth of the spore from a granule in Bacillus megatherium and Ward (1895) described the same process in Bacillus ramosus. Ernst (1888-1889) and Babes (1889) are the sponsors of the second theory of spore-formation and their view of the origin of the spore in the fusion of granules, composed chiefly of nuclear material, has been corrected and extended in modified form by Schaudinn (1902) and Dobell (1908, 1909, 1911). The third conception of spore-formation by protoplasmic condensation has been in the minds of many investigators. Matzuschita (1902) mentioned it definitely and lately von DarAnyi (1927, 1930) has emphasized it as a consequence of his observations on the effects of col