Experiments by Harden and Zilva showed that exposure of lemon juice to _ultra-violet rays_ for eight hours does not influence its antiscorbutic activity, that exposure of autolyzed yeast for the same length of time likewise does not impair its "antineuritic" potency, but that under identical conditions the fat-soluble factor in butter becomes inactivated. Similar tests carried out by the author with orange juice led to the same result. It thus appears that one of the vitamines--not the most unstable--manifests a peculiar sensitiveness to a certain form of physical or chemical action. In view of the fact that _shaking_ partly destroys pepsin and rennin, as shown by Shaklee and Meltzer, it would be of interest to ascertain whether this process brings about any impairment of the antiscorbutic vitamine. Particular interest is attached to this question because it is generally recognized that milk loses some of its potency in the course of handling--whatever may be comprised by this term. In experiments on the fat-soluble vitamine Steenbock and his co-workers found that "somewhere in the course of the manipulation to which the butter fat had been subjected, factors had been introduced which were responsible for a vitamine destruction." Before discussing the question of the manner in which the antiscorbutic vitamine functionates, it may be well to state briefly the type of the disturbance which its deficiency occasions. The chief manifestation is damage to the integrity of the endothelium of the vessels, resulting in hemorrhage--whether from diapedesis or from rhexis or both of these conditions, the microscope does not inform us. Nor can it be stated positively that the endothelium has not been injured by a secondary toxic or bacterial factor. However this may be, the end result of the deficiency is endothelial damage, a pathologic condition which may be demonstrated clinically in scurvy by the "capillary resistance test" (chapter VII). The other marked functional alteration in scurvy is increased susceptibility to infection; but how a vitamine deficiency induces this vulnerability cannot be explained. Clinical tests show that the blood contains sufficient antitoxin (diphtheria) to afford protection. Harden and Zilva found that "guinea-pigs fed on an unrestricted mixed diet, on a quantitatively restricted mixed diet, and a scorbutic diet showed no differentiation in amboceptor and agglutinin titres, and in the complement activity of the blood." If this is to be interpreted as indicating that the protective substances of the body undergo but little alteration, we must consider whether susceptibility to infection, as well as tendency to hemorrhage, is not due largely to alteration in the cement substance of the endothelial and epithelial membranes.