Classification of the pneumococci has been rendered difficult by the frequent discovery of atypical strains, many of which were eventually given individual type numbers. Avery (1) recorded such variants of Type II, and of these his IIA and IIB were later designated Types V and VI (2). Chemical relationships between the three serological types were clarified in a recent series of quantitative immunochemical studies (3, 4 a-4 c, 5). Similarly, an atypical Type III strain (6) soon became Type VIII (7) (9), and quantitative immunochemicat investigations (10 a-10 c).We have studied the chemistry of the capsular polysaccharides of Types VII and XVIII (11,12) and quantitative data are herewith given on the qualitatively long-known cross-reaction (2, 14 a) between these types. Types XVIII and VIII also cross-react (12), as do II and XIX (14 b), VI and XIV (4 b), and X and XIV, among others. Although the relationship of Types II and XX was discovered many years ago (13 a, 13 b, 14 a, 14 b), quantitative data on this instance of cross-reactivity are now given for the first time. In this connection, the X-XX cross-reaction, also long-known (13 a, 14 a, 14 b), was drawn into the study. The chemical basis of each cross-reaction is also discussed insofar as permitted by the present knowledge of the structures of the capsular polysaccharides, the principal determinants of pneumococcal type specificity. Perhaps the inadequacy of this present knowledge will stimulate further work in the field.