260 STATE BOARD OF AGRICULTURE. 



From an inspection of these tables one notes at once that with method 



A, extreme variations in the results arfe possible when waters are used 

 which differ among themselves as do those used in this work. Occa- 

 sionally there were also considerable variations between duplicate de- 

 terminations with the same water (see especially Table 6). It was 

 noted that a pink ring of alkaline precipitate formed around the periph- 

 ery of the solution and then the color gradually spread through the 

 entire mass becoming more intense the longer the solution stood. This 

 accounts for the variations noted but the reason for it is not clear. 



On the whole very little difference is to be noted between the results 

 of method B and C, except the greater ease and rapidity of the latter. In 

 general the results secured by method C are less variable than by method 



B. In a few instances, already mentioned, in which method B was un- 

 wittingly varied by adding the medium to the previously boiled water 

 while it was still hot instead of waiting until it became cold, the results 

 (which are marked in the tables with an asterisk) were decidedly lower 

 than the duplicates run in the regular way. 



It is hard to account for every change that takes place in a complex 

 system such as we have here. Bjerrum (1) has shown that it is im- 

 possible to establish precise relationships between the titre of hot solu- 

 tions and cold solutions in a complex system such as was used in this 

 work. Redfield (2) has shown that at different temperatures wide var- 

 iations in the titration of peptones may be procured. Tizard and Whis- 

 ton (3) have shown the effect of temperature on the color changes of 

 methyl-orange and on the accuracy of titrations. And finally, Clark 

 (4) has shown the fallacy of making titrations of hot culture media. 

 Our data would indicate that when it is necessary to titrate culture 

 media more uniform and accurate results may be obtained by neutraliz- 

 ing the excess of CO. with the alkali and proceeding with the titration 

 in the cold. On the whole, therefore, the writers are inclined to think 

 that method C is the most reliable especially where distilled water or 

 water of low salt content is not available. While the hydrogen ion 

 method is generally used in this laboratory, j^et we still teach titration 

 and method C is the one taught. 



REFERENCES. 



(1) Bjerrum. Die Theorie der alkalimetrischen und azidimetri- 

 schen Titrierungcn, 1914. Quoted from Clark (see ref. 4). 



(2) Redfield. A Study of Hydrogen Sulphide Production by Bac- 

 teria, Cornell Thesis, 1912. Quoted from Clark (see ref. 4). 



(3) Tizard, Henry Thos., and Whiston, John Reginald Harvey. The 

 Effect of a Change in Temperature on the Colour Changes of Methyl- 

 orange and on the Accuracy of Titrations. Jour, of The Chemical So- 

 ciety, Vols. 117-118, No. 688,' Feb. 1920, pp. 150-156. 



(4) Clark. The "Reaction" of Bacteriologic Culture Media. Jour, of 

 Infectious Diseases, Vol. 17, 1915, pp. 109-136. 



[G. L. A. Ruehle and F. W. Fabian.] 

 Problem No. 2. (a) To Determine the Number of Bacteria in Pasteur- 

 ized Ice Cream with a View to Determine a Bacteriological Standard, 

 (b) A Critical Bacteriological Study of the Process of Manufacturing 

 Ice Cream. 

 This problem is being worked out in conjunction with the ice cream 



