1. Feeding the purified soluble specific substance of Type I pneumococcus protects rats against an intraperitoneal injection of the virulent organism.
2. This increased resistance resembles that obtained when the intact (dead) or dissolved bacteria are fed, as follows: (a) one feeding is sufficient, (b) the interval between the feeding and the appearance of the immunity is the same, (c) the duration is approximately the same, (d) when the immunity is exhausted it can be renewed by a new feeding, (e) the immunizing action is type-specific.
3. The differences between the effects of feeding the purified specific substance and the intact or dissolved organism to rats, appear to be quantitative rather than qualitative, the proportion of animals protected and the height of the immunity being generally, though not always, less in the case of the former.
4. In contrast to the immunizing action which the soluble specific substance possesses when administered to rats, feeding it to mice failed to protect them. Neither were mice definitely immunized by parenteral administration.
5. A sodium glycocholate solution of Pneumococcus Type I lost part of its immunizing activity on standing for 1 year.
6. The failure to immunize mice and the loss of activity of the bile salt solution of pneumococcus, on standing, are discussed in terms of (a) the possible presence of a second cell constituent which is active by mouth, and (b) a possible intramolecular change in the type-specific polysaccharide associated with a loss of immunizing action while retaining the precipitin reaction.