EXPERIMENT STATION BULLETINS. 285 



tilis (9), in a closed chamber which has been partially exhausted of air 

 by means of a suction pump. It is generally believed that the initial 

 growth of this organism depends upon an atmosphere of lowered oxygen 

 tension. If this is true, we should expect newly isolated strains of the 

 organism to grow very rapidly when placed in a sealed tube or cham- 

 ber in which part of the oxygen has been absorbed by means of an 

 alkaline solution of pyrogallic acid. A series of experiments con- 

 ducted by the writer shows that the initial growth is not due to a 

 lowered oxygen tension, but is due to an increased carbon dioxide 

 tension of the atmosphere in which it is to be grown. The correct car- 

 bon dioxide tension may be derived by sealing the tubes containing the 

 inoculated medium ; by placing the inoculated medium in a closed 

 chamber (the growing organisms furnish CO, for the proper tension) ; 

 by placing the inoculated medium in a closed chamber in which 10 per 

 cent, of the air has been displaced by carbon dioxide gas. It was 

 found that when the latter method was employed colonies of Bad. 

 ahortus would appear on the surface of the medium in from 24 to 72 

 hours and often develop to 3 mm. in diameter. No visible growth ap- 

 peared until the fourth day or later when the sealed tube or closed cham- 

 ber method was used. When the growth became visible it was found that 

 a COo tension had developed in the sealed tube (measured by the H-ion 

 method) which corresponded to the CO, tension of a closed chamber 

 after 10 per cent, (by volume) of the contained air had been displaced 

 by COo gas. 



The methods which have just been discussed were used in studying 

 samples of milk from each of the four quarters of twelve cows. Cows 

 No. 1, 8, 9, IQ, 11, and 12 have never aborted. Cow No. 2 aborted a 

 six months fetus Dec. 18, 1910; Cow No. 3 aborted a five months fetus 

 Oct. 24, 1919; Cow No. 4 aborted a sLx months fetus Feb. 20, 1919; 

 Cow No. 5 aborted a seven months fetus Feb. 5, 1920; Cow No. 6 

 aborted a six months fetus Jan. 26, 1920; Cow No. 7 aborted a six 

 months fetus Jan. 1, 1917. The reactions of the milk to the agglutina- 

 tion test at the time the bacteriological examinations were made are 

 shown in Table I. , 



The samples of milk were collected in sterile test tubes after the 

 first milk had been discarded, and about 10 c. c. of each centrifugalized 

 for two hours at 2000 r.p.m. This length of time for centrifugalizing 

 is necessary owing to the fact that Ba0t. abortus is removed from sus- 

 pension with diflflculty. About 0.1 c. c. of the sediment was drawn from 

 the bottom of the tube by means of a small capillary pipette, then 

 placed on the surface of a solidified gentian violet agar plate and evenly 

 distributed by holding a sterile glass rod (bent at an angle of 90°) 

 against the surface of the medium at the same time rotating the plate 

 in a horizontal plane. 



The bacteriological examinations were in each instance controlled 

 by inoculating guinea pigs intra-abdominally with 5.0 c. c. of whole milk 

 from each quarter, and allowing a period of eight to ten weeks to 

 elapse before autopsying. The spleen and liver were then examined 

 culturally for the presence of Bad. abortus. The comparative results of 

 the two methods are shown in Table II. The milk from the four 

 quarters of cows No. 1, 5, 7, 8, 9, 10, 11, and 12 gave negative results to 

 each of the methods employed. The positive results correspond in all 



