Pe. Se Se 
Ce, ~ _w - 



New York AGRICULTURAL EXPERIMENT STATION. 495 
Statement of results.— The foregoing results appear to indicate : 
first. That the use of mercuric chloride in fresh milk largely 
prevents or retards the occurrence of those changes that effect 
the determination of casein. 
Second. That even when mercuric chloride is used in milk, it 
is generally desirable to use less than 1.5 cc. of dilute acetic acid 
in precipitating the casein. 
Third. That the use of mercuric chloride in the proportions 
stated did not precipitate in appreciable quantity any nitrogen 
compounds such as albumen. 
(7) General summary.— We may summarize the results of our 
- work as follows: 
frst. Precipitation of casein in cows’ milk gives essentially 
the same results whether made at forty degrees C. without car- 
bon dioxide or at ordinary temperatures with the aid of carbon 
dioxide, other conditions being uniform. 
Second. Between one and two cc. of a ten per cent. solution of 
acetic acid, generally about 1.5 cc., gave the best results. 
Third. The use of lactic acid, sulphuric acid, and hydrochloric 
acid gave results much less satisfactory than the use of acetic 
acid. 
Fourth. A variation of a few degrees from 40°C. and a 
variation of time of digestion after the addition of acetic acid 
exercised comparatively little influence on the results. 
fifth. The precipitation of casein in milk that had undergone 
noticeable change was generally found to give lower results than 
in case of the same milk when fresh, even when less than 1.5 ce. 
of acetic acid were used. Coagulated milk always gave low 
results compared with fresh milk. 
Stath. The use of mercuric chloride in fresh milk, in the pro- 
portion of one part of the former to 2,000 parts of milk, 
largely prevented or retarded the occurrence of those changes 
that affect the determination of casein in old milk. 
Seventh. In a general way, the character of the filtrate serves 
asa fairly reliable guide in regard to the completeness of the 
precipitation of casein from milk. When the filtrate is clear or 
only perceptibly cloudy, the precipitation is generally complete; 
but, when the filtrate is decidedly turbid or milky and two or 
_ three repeated filtrations do not remove the turbidity, the pre- 
