It would appear that much of the divergence of opinion that has been formed has depended to a considerable extent upon the conditions under which cases have been observed. This is clearly brought out by contrasting the experience of Fantus[39] dealing with private cases in civilian practice, where pneumonia was relatively uncommon, with that of others dealing only with cases in large hospitals, where those admitted have been in large part selected seriously ill patients with a high incidence of pneumonia, the milder cases comprising from 60 to 90 per cent of those attacked by influenza never reaching the hospital. Variations in opinion with respect to the bacteriology of the epidemic, especially in regard to B. influenzæ, would appear to be due for the most part to differences in bacteriologic technic, in some degree to differences in interpretation. Accumulating evidence can leave little doubt that B. influenzæ was at least extraordinarily and universally prevalent throughout the period of the epidemic and thereafter, and that earlier reports of failure to find it were due to the use of methods unsuitable for its detection and isolation. The opportunity afforded the commission at Camp Pike to devote their full time to a systematic and correlated group study of the epidemic simultaneously from many aspects throughout its whole course made it apparent that influenza _per se_ is in the large majority of instances, in spite of the initial picture of profound prostration, a relatively mild disease which tends to rapid spontaneous recovery. This opinion is supported by the fact that the disease during the first waves of the epidemic in this country, which it is now recognized occurred pretty generally in the army camps during the spring of 1918, was so mild that it attracted only passing attention, since the disease at that time was not sufficiently virulent to predispose to any alarming amount of pneumonia. With the return of the epidemic in the late summer and early fall, however, the disease had attained such a high degree of virulence that it predisposed to an appalling amount of severe and often rapidly fatal pneumonia, which often detracted attention from the real nature of the preceding disease. Yet even during the fall epidemic from 60 to 90 per cent of the cases of influenza proceeded to rapid recovery without developing complications. On this ground alone it would seem only logical to regard pneumonia strictly as a complication of influenza rather than as an essential part of the disease, irrespective of whether the pneumonia may be caused by the primary cause of influenza or not. The complexity of the clinical features, the bacteriology and pathology of the pneumonias following influenza lend further support to this opinion. It seems better, therefore, to consider influenza first as a disease by itself and subsequently to take up the question of pneumonia and the relation of influenza to it.