6o MANUAL OF BACTERIOLOGY. 



in the steam sterilizer for fifteen minutes or longer on each of 

 three consecutive days. When precipitates form, they are 

 usually caused by a too alkaline reaction. That may be cor- 

 rected by the addition of a little weak hydrochloric acid, drop 

 by drop, testing frequently with litmus-paper. 



A more accurate way of obtaining the proper reaction is Schultz's method. 

 Take of the bouillon lo c.c; add a few drops of phenolphthalein* (alcoholic 

 solution, J per cent.); with a burette add, drop by drop, a solution of caustic 

 soda, 0.4 per cent., until a faint red color appears, which indicates the beginning 

 of the alkaline reaction. This procedure is followed with three samples. The 

 amount of soda solution required in each case is noted and the average taken. 

 If now, on the average, for each 10 c.c. of bouillon i c.c. of soda solution needs 

 to be added, for 1000 c.c. of bouillon 100 c.c. of the soda solution must be 

 added; only, instead of adding a weak soda solution, one-tenth as much is 

 taken of a solution ten times as strong. 



Another method of making bouillon is to use, instead of beef- 

 extract, 500 grams (one pound) of finely chopped, lean beef, 

 which is placed in one Uter of water and kept on ice for twenty- 

 four hours. It is strained, thoroughly cooked to coagulate the 

 albumen in it, filtered and a Kter of fluid obtained, adding wafer 

 if necessary. The peptone and salt are then added and the 

 medium heated to dissolve them. It is then neutraHzed, filtered 

 and sterilized. Although bouillon made with solid beef-extract 

 is convenient and serviceable for most purposes, it is advisable 

 to use meat when the bouillon is to be employed for the develop- 

 ment of bacterial toxins. Meat should also be used in the 

 preparation of either bouillon, gelatin or agar-agar when new 

 species of bacteria are being studied for publication. 



In both of these cases the recommendations of the American Public Health 

 Association should be foUowed.f 



* In neutralizing an acid culture-medium it has been found that when the 

 medium appears to be neutral or slightly alkaline to litmus, it may still be 

 acid if phenolphthalein be employed as an indicator. Fuller. Journal American 

 Public Health Association. 1895. 



t See the Report of the Committee of the American Public Health Asso- 

 ciation entitled Procedures Recommended for the Study of Bacteria. 1898. 

 Rumford Press, Concord, N. H. 



