SEPARATION BY AGAR MEDIA 61 



here, as the agar does not adhere well to the sides. If to the 

 agar 2 per cent, of a strong watery solution of pure gum arabic 

 is added, Esmarch's tubes may, however, be used. 



2. Scjmration I>y Stroking Mixture on Surface of Agar 

 Mulia. The bacterial mixture, instead of being mixed in the 

 medium, i* spread out on its surface. The method may be used 

 both when the bacteria to be separated are in a fluid, and when 

 contained in a fairly solid tissue or substance, such as a piece 

 of diphtheritic membrane. In the case of a tissue, for example, 

 a small portion entangled in the loop of a platinum needle is 

 stroked in successive parallel longitudinal strokes on sloped 

 agar, the same asj)ect being brought in contact with the agar in 

 all the strokes. Three strokes may be made in each tube, and 

 three tubes are usually sufficient. In this process the organisms 

 on the surface of the tissue are gradually rubbed off, and when 

 growth has taken place it will be found that in the later strokes 

 the colonies are less numerous than in the earlier, and sufficiently 

 far apart to enable parts of them to be picked off without the 

 needle touching any but one colony. When, as in the case of 

 diphtheritic membrane, putrefactive organisms may be present 

 on the surface of the tissue, these can be in great part removed 

 by washing it well in cold water previously sterilised (vide 

 Diphtheria). In the case of liquids, the loop is charged and 

 similarly stroked. Tubes thus inoculated must be put in the 

 incubator in the upright position and must be handled carefully, 

 so that the condensation water, which is always present in 

 incubated agar tubes, may not run over the surface. Agar, 

 poured out in a Petri's capsule and allowed to stand till firm, 

 may be used instead of successive tubes. Here a sufficient 

 number of strokes can be made in one capsule. Sloped blood- 

 serum tubes may IK- used instead of agar. The method is rapid 

 and easy, and gives good results. 



Separation of Pathogenic Bacteria by Inoculation of 

 Animals. It is found difficult and often impossible to separate 

 ly ordinary plate methods certain pathogenic organisms, such 

 as b. tuberculosis, 1>. mallei, and the pneumococcus, when such 

 occur in conjunction with other bacteria. These grow best on 

 special media, and the first two (especially the tubercle bacillus) 

 grow so slowly that the other organisms present outgrow them, 

 cover the whole plates, and make separation impossible. The 

 method adopted in such cases is to inoculate an animal with 

 the mixture of bacilli, wait until the particular disease develops, 

 kill the animal, and with all aseptic precautions (vide p. 1 !">) 

 inoculate tubes of suitable media from characteristic lesion- 



