Chart 5.—Shows the time relation between the identification of hemolytic streptococci in the throats and the development of otitis media in 27 cases shown to be due to hemolytic streptococci. The onset of otitis media is represented by the ordinate marked ○. The number of days before or after the onset of the otitis, within which the throat culture which proved positive for hemolytic streptococci was taken, is marked off along abscissæ to the left and right of ordinate ○ respectively. On the curve plotted these symbols are used: A circle represents a throat culture positive for hemolytic streptococci in a case of otitis media without extension to mastoid. The plus sign represents a throat culture positive for hemolytic streptococci in a case of otitis media with mastoiditis and osteitis. ] In this series of cases (Chart 5) the appearance of S. hemolyticus in the throat and the onset of otitis media are very closely associated in those patients in whom further extensions of the streptococcus infection occurred. In instances in which appearance of streptococci and of otitis media are separated by an interval of more than seven days, no further extension occurred. In 8 cases in which this interval is seven days or less there has been no further extension of the infection. The Dissemination of Hemolytic Streptococci in Wards Beginning October 24 cultures for the identification of carriers of hemolytic streptococci were made from all patients in a ward and repeated at intervals of one week. Prior to this time individual patients had been examined at intervals of one week, so that an entire ward was never studied on any particular day. This system did not identify and remove all “carriers” in a ward at a given time and was abandoned because it failed to show the conditions present. Investigation of wards as units proved much more satisfactory. The studies made in four of the double wards used for the care of patients with measles are presented in Table LXV. During the time of this study hemolytic streptococci were more prevalent than at an earlier period. Cultures from the throats of all patients entering these wards were negative for S. hemolyticus on admission. The table showing the incidence of “carriers” of hemolytic streptococci each week in these wards demonstrates: 1. The separation of “carriers” and “noncarriers” by throat culture made on admission does not prevent the increase of streptococcus “carriers” in wards. 2. Removal of all “carriers” found by cultures on admission and at weekly intervals is inadequate. TABLE LXV WARD CONDITIONS WITH REFERENCE TO HEMOLYTIC STREPTOCOCCUS INFECTION