; 
SECTIONAL TRANSACTIONS.—B. 299 
copper and palladium, and conclusions are drawn about the mechanism of 
evaporation of hydrogen from a state of adsorption in atomic form. 
More recently a discussion of dissociation and recombination of molecules 
at a crystal surface is being developed. 
AFTERNOON. 
Visit to Macaulay Institute for Soil Research, Craigiebuckler,. 
Tuesday, September 11. 
Jotnt Discussion with Section M (Agriculture) on The chemistry of 
milk (Section B room, 10.0) :— 
Prof. H. D. Kay.—Introduction. 
Dr. J. F. Tocurer.—(a) The composition of milk and the present 
regulations (10.15). 
The proportions of the constituents of milk are known to vary widely 
from sample to sample even in the case of bulked milk. In 1925 the author 
described the form of variation for each constituent. In the case of fat and 
solids-not-fat percentages, it was shown that many cases occurred where 
the values fell below the prescribed presumptive limits under the regula- 
tions. ‘These regulations were made at a time when no accurate knowledge 
existed of the observed minimum limits in the case of herds. Legal enact- 
ments, however, should follow scientific knowledge, not precede it. One 
of the difficulties encountered is a method of detecting ‘ watering.’ Many 
cases have occurred where genuine milk has been held to be watered. An 
equation has been found from which it is possible to detect watering within 
certain limits, which can be used in conjunction with the observed freezing 
point. The results will be published at an early date together with the 
freezing point results obtained from the same samples. 
(b) Variations in the freezing point of milk. 
It has been found by various workers that there is very little variation in 
the freezing point of milk, even if samples are taken from individual cows. 
_ It is the least variable of all the physical characters, the coefficient of variation 
being approximately 1-5 as against 4°5 for refractive index and 5 for specific 
gravity. On account of its low variability the freezing point of milk has 
been frequently used as a criterion of ‘ watering.’ No general agreement 
has, however, been reached as to the actual range of values in genuine 
samples. Different results have been obtained from the same sample, due 
chiefly to the practice of placing alcohol between the freezing tube and the 
ether flask. In one case the alcohol is removed after cooling, while in the 
other it is retained. Much more constant and accurate results are obtained 
by removing the alcohol, which is useful only to promote rapid cooling. 
The author, in 1925, showed that, in the absence of alcohol, the values 
varied from — 050° C. to — 0°56° C. in fresh samples from individual 
cows. Certain workers hold that if the freezing point of a sample is greater 
than — 0°52° C., water has been added. Before, however, one can estimate 
whether watering has taken place one must know the number of cows whose 
milk has been bulked. Variation in bulked samples is naturally greater 
than in samples from one cow. Values of — 0:50° C. have been obtained 
