leucocytosis had been present the more serious was the subsequent leucopenia. We regard this form of secondary leucopenia, if one may use such a term, as a prognostic sign of some value. As in lobar pneumonia, a high leucocyte count has been, as a rule, a favorable feature. The second group, or those showing a leucopenia throughout their course, was by no means an unusual thing. This is a cardinal point—in fact, one of the most striking clinical features of the epidemic. The leucopenia here does not have the prognostic value that it seems to have in the group just referred to previously. We have observed cases go through a pneumonia with 4,000–5,000 white cells in a relatively easy manner. When, however, the leucocytes fall to 3,000 or under, one may be reasonably sure that the outcome is doubtful, even with the general condition of the patient at the time favorable. In the pneumonias of this group which died the leucocytes have always fallen to about 2,000 cells. We have a number of observations taken from one-half to four hours before death showing counts in the immediate neighborhood of 2,000, but never below this number. Where recovery has taken place the cells go forward to the normal, more or less keeping pace with the general clinical picture. Of group three there is not much to say, except that on one hand it tends toward a leucocytosis, and on the other to a leucopenia. This group comprises a considerable number of the pneumonias. We are not in a position to say anything regarding the relative mortality of this group. The development of a leucopenia from these cases after a period of some stability in the leucocytic curve is of bad prognostic import. Not infrequently we have noticed rather wild abrupt rises to 20,000 in the leucocytes toward the late half of the disease. This curve was nearly always sustained until the end, which, as a rule, was recovery. We do not need to consider at any length the effect on the leucocyte count of complications not of lung origin. Acute sinuses in head, otitis media and meningitis always produced a variable moderate leucocytosis. The change was not so marked in meningitis, as our cases were all preceded by a pneumonia which had independently invoked a slight leucocytic response. As a complication of the pneumonia we have noted an abrupt rise following an acute pleuritis with effusion, and similarly after the onset of an empyema. These complications seemed to be able to induce a leucocytosis with more certainty and ease than the more serious pneumonic condition. Possibly, as they occurred toward the end of the infection, the toxic factor of the epidemic influenza was more or less spent, and the secondary invader had a freer hand to act in its normal way.