The only articles considering this important question from the experimental side are those of Jackson and Moody, and of Moore, who conclude tentatively that scurvy may be a bacterial infection. Jackson and Moody cultivated a diplococcus from the tissues of scorbutic animals after death, reproduced hemorrhages by inoculating cultures of these microörganisms into the circulation, and recovered the bacteria from the tissues some weeks later. Their results are open to the criticism that bacteria were found only after death, and that all blood cultures during life proved negative. An article by Moore, however, which has just appeared from this same laboratory, states that "an organism of the streptococcus viridans type was isolated from the blood" in a case of adult scurvy. In one instance we recovered an organism of this type from the blood of an infant suffering from scurvy. It is highly important that more blood cultures should be carried out in the course of human or animal scurvy, and that particular note should be made of the stage of the disorder when they are taken. There is no doubt that invasion of the blood-stream does occur readily in the course of scurvy, but this takes place generally after the disease has developed and must be regarded as a secondary phenomenon and therefore unessential from an etiologic standpoint. Indeed one of the striking and important symptoms of scurvy is the marked susceptibility to infection (furunculosis, nasal diphtheria, "grippe," etc.), which comes about as the result of the nutritional disturbance. An excellent example of this interrelationship is the "epidemic" of hemorrhagic scurvy described in the chapter on symptomatology. Hemorrhages coming about in this way should be regarded as focal complications rather than as truly scorbutic. It should be realized that, at the present time, it is not possible to distinguish between local symptoms which are truly nutritional or scorbutic in nature, and those which are bacterial and of secondary origin.