The numerous attempts made in the last twenty-five years to reproduce lobar pneumonia in animals practically failed in all instances. By intrabronchial insufflation of pure cultures of pneumococcus in dogs, we produced experimental pneumonia successively in forty-two cases, with a mortality of at least 16 per cent. The fatal cases resembled closely lobar pneumonia in man. In the non-fatal cases, the pathological and bacteriological findings were again in accord with the findings in man. Clinically, however, the cases of nonfatal experimental pneumonia run a milder and shorter course than in man. In a few instances, lobar pneumonia has been produced experimentally also with the Pneumococcus mucosus and with Friedländer's pneumobacillus. The anatomical findings in these experiments have shown some characteristics agreeing with the findings in the pneumonias of man produced by these organisms. The quantity of the injected culture seemed to have a definite influence upon the outcome of the disease; in the fatal cases larger quantities of the culture had been injected. The animals were neither selected nor prepared in any manner. The experimental success did not, therefore, appear to depend upon the degree of resistance of the individual hosts. It is suggested that the uniformly successful results of the experiments were due to the obliteration of a large number of bronchi by the injected culture, through which mechanical effect a favorable opportunity was provided the pneumococci to develop and display their pathogenic activities consisting in the calling forth of a characteristic local, more or less effective, widespread, inflammatory reaction of the lung tissue.