The substance is soluble in water and passes through a parchment membrane or a porcelain filter. Unlike the "B" it is apparently not adsorbed by fine precipitates such as fullers' earth or colloidal iron. Harden and Zilva showed that when a mixture of equal volumes of autolysed yeast and orange juice is treated with fuller's earth the "B" is removed and the "C" left unaltered. Eddy and La Mer have treated orange juice with fullers' earth and then tested the filtered off juice as cure and preventive of scurvy in guinea pigs. Their results showed that 6-2/3 cc. of the treated juice was curative, hence the loss due to adsorption must be less than 60 per cent to 70 per cent. Harden and Zilva were among the first to state that the vitamine is much more stable in acid than in alkali. They have shown, that even 1/50 N sodium hydrate at room temperature has a rapidly destructive effect. On the other hand Delf showed that when 0.5 gm. citric acid is added to the water in which germinated lentils are boiled, the loss of the antiscorbutic properties is, if anything, greater than when no addition of acid is made. She therefore concluded that in cooking vegetables there should be no addition of either acid or alkali to the cooking water if one wishes to conserve this vitamine. Sherman, La Mer, and Campbell have been engaged in experiments bearing on this point throughout the past two years. Some of their results have recently been published and their observations are worthy of special attention from their bearing on the character of reaction of the vitamine in general. They first proceeded to determine the amount of filtered tomato juice just necessary to produce scurvy in degrees extending from no protection to complete protection and they also constructed a basal diet which is apparently optimum in nutrients and all other factors except the "C" vitamine. They found that at the natural acidity of tomato juice (pH 4.2) boiling for one hour destroyed practically 50 per cent of the antiscorbutic power and by boiling for four hours they destroyed 70 per cent, which indicates that the curve of the destructive process tends to flatten more than that of a unimolecular reaction. This result was confirmed by heating experiments conducted at 60°, 80° and 100°. In all cases the temperature coefficients are low. (Q_10 equals 1.1-1.3) confirming Delf's results. When the natural acidity of the juice was first neutralized in whole or in part, the juice then boiled for an hour and immediately cooled and reacidified, it was found that at less than half neutralization (pH 5.1-4.9) the destructive effect of an hour's boiling was increased to 58 per cent. When alkali was added to an initial pH 11 (about N/40 titratable alkali to phenolphthalein) which fell to 9 during the hour's boiling the destructive effect was about 65 per cent. When reacidification was omitted and the neutralized boiled juice stored in a refrigerator for five days before