considerable loss from abortion for several years past, and the presence of contagious abortion had been recognized by Professor Mumford, altho this diagnosis was disputed by other authorities. In order to settle the question it seemed best to undertake a bacteriological study of the disease. This seemed the more desirable because, so far as we have ascertained, there was no known microörganism generally recognized and accepted as the cause of the disease in this country, the bacillus described by Bang having been found only in Europe and his work having failed to be confirmed by American investigators.

Altogether eighteen parturient cows have been examined bacteriologically. Ten of these calved at term and the births appeared to be normal. The abortion bacillus was not found in any of these. Eight were cases of premature delivery, and of these, six appeared clinically to be cases of contagious abortion. Placental tissue from two of these cases was examined by the Nowak plate method and a bacillus isolated, apparently identical with that isolated by Bang in Denmark and by Nowak in Austria.

The organism is a very small short rod, usually oval in shape, from 0.8µ to 2.0µ long by 0.7µ wide, practically always single, rarely in short threads of two to four cells. It is not motile, and does not form spores. It stains with moderate rapidity with the ordinary anilin dyes, and is decolorized by Gram's method. The colonies on serum-agar are raised, with smooth circular borders, appearing almost like drops of dew. They are transparent and very clear, with a bluish gray color by transmitted light. Under the microscope a few coarse granules may be seen near the center of the colony but the greater part of it appears very homogenous and almost water-clear. The appearance of the colony is really a very characteristic feature of the organism and enables one to distinguish readily the colony of the abortion bacillus from other colonies on the serum-agar plates.

The behavior toward oxygen is another character upon which considerable reliance may be placed in the identification of strains recently isolated from the animal body. This is tested by transplanting the colonies from the serum-agar plates to two series of agar streak sub-cultures, of which one series is incubated in the atmospheric air and the other in the closed jar together with cultures of B. subtilis. Unless the growth under the latter condition is much better than the growth outside the jar, the culture may be discarded as one not belonging to this species.

A final important test in identification is that of pathogenicity. Nowak induced abortion in pregnant guinea pigs with great regularity by subcutaneous, intraperitoneal, and intravenous injection of pure cultures of the abortion bacillus. So far, four pregnant guinea pigs have been inoculated subcutaneously with the pure cultures isolated by us, and the inoculation has been followed by premature evacuation of the uterus with death of the fetuses in 3½, 8, 6, and 7 days respectively. In the first guinea pig the two fetuses were practically fully developed and covered with hair. In this instance the abortion bacillus was isolated only from the subcutaneous tissue of the mother at the point of inoculation, the cultures from the uterus, the placenta, and the fetuses remaining negative. In the other three cases the fetuses were undeveloped and the condition was that of a true abortion. In these instances the abortion bacillus was demonstrated by culture tests at the point of inoculation in pure culture in two, in mixed culture in the other one; in the interior of the uterus in pure culture in all three; in each of the four placentæ of two cases in mixed culture, as these placentæ had been passed some time before they were found, and in the three placentæ of the other case in pure culture; in the livers of all three fetuses of the one case in pure culture, but