io8 Journal of Agricultural Research voi. x, No. 3 



purple are sterilized. That this is due to change in hydrogen-ion' con- 

 centration is shown by the following experiment : Two equal portions of 

 the same sample of fresh-skimmed milk were autoclaved, one sample with 

 bromcresol purple present, the other with the indicator absent. After 

 sterilization the uncolored sample received the same quantity of indicator 

 that had been added to the first sample. The two samples of heated 

 milk then had the same color. Both showed in like degree a change 

 from the color of an unheated control containing the same concentration 

 of indicator. The change in Pg which occurred during sterilization was 

 measured by means of the hydrogen electrode. The value of the un- 

 heated sample was Ph = 6.6o, while that heated in the autoclave at 17 

 pounds for 15 minutes was 6.36. When the unheated milk was acidified 

 until its Pg value was close to that of the heated milk, the two samples 

 matched almost perfectly in their color with bromcresol purple. 



These observations indicate quite conclusively that the color change 

 which occurs when milk containing bromcresol purple is heated is due 

 to a change in the hydrogen-ion concentration of the milk, and that it is 

 not due to an alteration of the indicator itself. The same conclusion 

 holds for the permanent color change in litmus milk. This is supported 

 by the fact that the different degrees of color change which occur when 

 milks are heated for a longer or shorter period or at higher or lower 

 temperatures are proportional to the changes in hydrogen-ion concen- 

 tration accompanying these treatments. It should be noted that in the 

 more severely treated milks there is a coloration of the milk itself, which 

 is superimposed upon the color of the indicator. 



It is also important to note that the changes in t e reaction of milks 

 which are brought about by sterilization may be considerable. It 

 has been stated above that a sample of milk with a Pg value of 6.60 was 

 changed to Ph = 6.36 during 15 minutes' sterilization at 17 pounds' steam 

 pressure. A duplicate sample, when held for 15 minutes longer at that 

 pressure, had a Pg value of 6.13. . What the variation may have been 

 in milks of diverse qualities, treated with litmus solutions of various 

 degrees of impurity, adjusted with different additions of alkali or acid, 

 and heated at various pressures for different lengths of time, it is quite 

 impossible to say. The variations may not have been important for 

 crude cultural tests of most organisms, but in the study of certain ones 

 it is entirely possible that at times the reaction was brought to the 

 border of or placed outside the optimum range for growth. 



In the utilization of milk as a culture medium it may be foun_ advan- 

 tageous in particular instances to adjust its initial P^ to some point other 

 than that obtained in sterilized fresh milk. The writers have found, for 

 instance, that if milk is to be brought to Pg 7.0 when sterilized, brom- 

 thymol blue is more serviceable than bromcresol purple. For most 

 purposes it will be found advantageous to adhere to the use of fresh, 

 unadjusted milk containing some definite quantity of bromcresol purple. 

 If, then, the temperature and time of sterilization are kept constant, the 



