B. influenzæ was found in the sputum coughed from the deeper air passages in the majority of cases, being present in 80, or 76.2 per cent, of the 105 cases. In the 58 cases of lobar pneumonia it was found 41 times, or 70.7 per cent, in the 47 cases of bronchopneumonia 39 times, or 82.9 per cent. The abundance of B. influenzæ in the sputum varied greatly in different cases. Microscopic examination of stained sputum films and direct culture of the sputum on blood agar plates showed that in general it was more abundant in the mucopurulent sputum from cases of bronchopneumonia than in the mucoid rusty sputum from cases of lobar pneumonia. This was by no means an invariable rule, however, since in the former the bacilli were sometimes very few in number, in the latter quite abundant. Whether B. influenzæ shared in the production of the actual pneumonia in these cases is difficult to decide and cannot be stated on the basis of the bacteriologic and clinical observations which have been made. =Clinical Features.=—One of the most striking aspects of pneumococcus pneumonia following influenza was the diversity of clinical pictures presented. These varied all the way from the classical picture of lobar pneumonia to that of bronchopneumonia of all grades of severity from the rapidly fatal coalescing type to that of very mild character with very slight signs of consolidation. For this reason it is questioned whether there is any real justification for speaking of a typical influenzal pneumonia, an opinion that seems well supported by the diversified picture found at the necropsy table. For purposes of presentation, pneumococcus pneumonia following influenza may be divided into three clinical groups: (1) Lobar pneumonia; (2) lobar pneumonia with purulent bronchitis; (3) bronchopneumonia. No accurate data are available as to the relative frequency with which these three types occurred at Camp Pike. In the group of 105 cases studied there were 58 cases of lobar pneumonia, 11 of which had purulent bronchitis, and 47 cases of bronchopneumonia. The majority of these cases, however, occurred during the early days of the epidemic of influenza and probably show a considerably higher proportion of lobar pneumonias than actually occurred in the total number of pneumonias throughout the epidemic. This is indicated by the fact that of 100 consecutive cases of influenza selected for observation at the height of the epidemic, 3 developed clinical evidence of lobar pneumonia and 12 of bronchopneumonia. (1) Lobar pneumonia presenting the typical clinical picture with sudden onset, tenacious rusty sputum, sustained temperature, and physical signs of complete consolidation of one or more lobes occurred in 47 cases; 36 cases in this group definitely followed influenza. In 11 cases no certain clinical evidence of a preceding influenza was obtained, and it is probable that some of these represent cases of pneumonia occurring independently of the epidemic of influenza.