X CONTAMINATION OF MILK 183 



they believe that an accurate estimation of the number of B. 

 coli and B. lactis aerogenes in milk can be obtained. 



The writer has carried out a large number of milk examina- 

 tions to determine B. coli and allied forms, using at the same 

 time the L.B.B. tubes method and direct plating on aesculin 

 agar or L.B.A. He found the L.B.A. medium quite as useful 

 as aesculin agar, but on the whole with both media, it was 

 difficult to arrive at accurate estimations of the number of these 

 organisms from direct plating on solid media, and this pro- 

 cedure is far inferior to the use of L.B.B. tubes inoculated with 

 definite fractions of the milk. 



Identification tests. — The following tests should be carried 

 out for routine work : 



(a) Growth upon gelatine slope, for character of growth and 

 absence of liquefaction. 



(i) G-rowth in litmus milk at 37° C. for acid and clot. 



(c) Growth in lactose peptone litmus solution for lactose 

 fermentation. 



(d) Growth in peptone water for indol production. 



(e) Growth in saccharose peptone litmus solution for fermenta- 

 tion of saccharose. 



Of these, the saccharose fermentation test is omitted by 

 some workers. The others are the minimum number of tests 

 which should be employed. For research purposes, to accurately 

 study the distribution of the organisms included in one group 

 by the above tests, a greatly extended series of cultural tests 

 will be necessary. 



MacConkey has advocated a different series of tests for 

 members of the B. coli group. He at first suggested ^ the 

 omission of the fermentation of glucose, growth in milk, 

 character of the growth on gelatine, indol formation and 

 action on neutral red, and the substitution for these tests of 

 the following — fermentation of dulcite, adonite, and inulin, 

 and Voges and Proskauer's reaction. He retained the fer- 

 mentation of saccharose. In his opinion, " by so doing we 

 should gain a finer differentiation without increase of work, 

 and we should not be classing as B. coli organisms which may 

 have little in common with, and have a distribution entirely 

 different from that of the B. coli com.murds." 



' Journal of Byijienc, 1906, vi. p. 385. 



