EXPERIMENT STATION BULLETINS. 631 



at the end of four days. If these colonies are picked off with a platinum 

 needle and agar shakes made it will be found that a zone of growth will 

 appear at a depth below the surface at which the proper amount of 

 oxygen is present. This is usually about 1 cm. with a second faint zone 

 sometimes appearing at a greater depth. While the writer has isolated 

 Bad. abortus from milk sediment by this method, it is too tedious a 

 process to apply to any number of samples. Plates are likely to be over- 

 grown with colonies of fast-growing organisms, and the method has the 

 further disadvantage of requiring several weeks to isolate and identify 

 the cultures. 



Alkaline pyrogallol has been used by several investigators to obtain 

 the desired amount of oxygen in a closed jar, but without any great 

 success. 



Others claim that if an inoculated agar slant is sealed and incubated 

 there will be a period at which the proper amount of oxygen is present 

 due to partial absorption of oxygen by the agar. 



Miss Evans succeeded in isolating from milk, organisms which she 

 believes to be Bad. al)ortus, by plating on ordinary lactose agar to 

 which 10 per cent of sterile blood serum Avas added just before plating. 

 After incubating for four days, the colonies which developed were trans- 

 ferred to nutrient broth containing 1 per cent of glycerine and to tubes 

 of whole milk containing litmus. 



Dr. Giltner has devised an H tube which has been fo^md of great value 

 in growing newly isolated cultures of Bad. ahortus. One arm of the 

 tube contains ordinary agar upon which B. suMilis is grown. The other 

 arm of the tube contains a serum agar slant upon which the culture of 

 Bad. ahortus is inoculated. The cotton plugs are pushed down into 

 the tubes and tight rubber stoppers inserted. This method is more con- 

 venient than the ordinary jar method as it is possible to open one tube 

 without disturbing others as in the jar method. It has the further ad- 

 vantage that the culture can easily be watched and handled. 



Another method of study, the inoculation of guinea-pigs with the 

 milk, while more reliable, is far from satisfactory, owing to the fact that 

 it takes 8 to 10 weeks for the lesions to develop, and it is probable that 

 the organisms must be present in large numbers to cause the character- 

 istic lesions with the 5 c.c. of milk used for inoculation. 



In the investigation of the effect on milk of the diseases of the cow, 

 with special reference to infectious abortion, it was found desirable to 

 examine a large number of samples to determine whether or not Bad. 

 ahortus was being passed with the milk. The cultural and animal in- 

 oculation methods were the onlj^ ones found available for this work. 

 As these methods are unsatisfactory because of the length of time re- 

 quired, it has been found necessary to develop new technic in order to 

 study a large number of samples. Knowing that this organism is some- 

 times present in considerable numbers in millv as it comes from the 

 cow's udder, it was thought that this might indicate an infection of the 

 udder and a consequent local production of antibodies. Zinsser says 

 in this connection, that "it is more than probable that antibodies may 

 be formed anywhere in the body and that the locality of their produc- 

 tion is largely dependent upon the locality in which the antigen is con- 

 centrated." 



This led the writer to believe that the agglutination and complement 



