280 EXPERIMENT STATION RECORD. 



or 3 per cent, and within a few days they contained 5 or 6 per cent. Tlie milk from 

 cows near the end of the lactation period showed the least variation from day to day, 

 and was, as a rule, of higher quality than that from cows yielding a large amount. 

 The difference between the morning's and evening's milk was mostly in the percent- 

 age of fat, the remainder of the solids showing little variation. A cow yielding large 

 quantities of milk gave a product lower in solids-not-fat than those near the end of 

 lactation. The average composition of the morning's milk was fat 3.2 per cent, 

 solids-not-fat 9.2 per cent; evening's milk, fat 4.5 per cent, solids-not-fat, 8.9 per 

 cent. 



Variation in milk tests, F. W. Bouska {Hoard's Dairyman, S2 {1901), No. 24, 

 pp. 512, 513). — A discussion of the causes of the changes in the milk tests with the 

 same animal from day to day. 



Variations in milk tests, T. L. Haeckee {Hoard's Dairyman, 32 {1901), No. 22, 

 pp. 470, 471). — An address before the Lincoln, Nebr., convention, outlining results 

 at the Minnesota Experiment Station. 



The physical state in which fat exists in cream, H. D. and S. 0. Richmond 

 {Analyst, 26 {1901), No. 302, pp. 117-123). — It was found that there was a distinct 

 change in the density of cream at about the melting point of fat, while no such change 

 was observed with separated milk. This led the authors to study the question whether 

 fat in milk was always liquid, as it was expected that owing to the different heat 

 capacities of solid and liquid fat, there would be a change in the rate of heating at 

 the melting point of the fat. The plan adopted in studying this question was to heat 

 20 cc. of cream in a tube with a stirrer bearing a thermometer reading to 0.1° C, 

 the readings being made every 15 seconds. The estimations were made with cream 

 that had been kept in the refrigerator 18 hours, and with cream that had l)een 

 heated to 50° C. and rapidly cooled down to 15°. The results are given in tables and 

 also platted. With well cooled cream it appears that there was a change in the spe- 

 cific heat capacity, and a slight though distint;t absorption of heat at about 34°. 

 With heated cream rapidly cooled, a very slight change occurred at this point. The 

 only possible change that could produce these results is the melting of the fat. The 

 authors conclude that in well cooled cream the fat is solid, while with rapidly cooled 

 cream it is mostly liquid, though there is a distinct indication that it had become 

 slightly sohdified in a quarter of an hour. The melting is practically instantaneous. 



Detection and. estimation of preservatives in milk, M. W. Blyth {Analyst, 

 26 {1901), No. 303, pp. 148-151). — The author proposes a method for detecting pre- 

 servatives in milk which permits the examination of a large number of samples in a 

 short time. Ten cc. of th^ suspected milk and of a control sample known to be free 

 from preservatives are measured into test tubes. To each there is added a strong, 

 slightly alkaline solution of litmus. If the tubes are not of the same color, a half 

 normal solution of sodium hydrate is added drop by drop to the suspected sample 

 until it is of the same color as the check. The tubes are then plugged with cotton, 

 and both heated in the water bath for 10 min. at 80° C. They are afterwards 

 inoculated with J cc. of a solution of ^ cc. of sour milk in 100 cc. of water, shaken, 

 and kept 24 hours at a temperature between 15 and 22° C. Tubes containing pre- 

 servatives will remain blue or pink, while those without will become white in the 

 same way as the check. The length of time it takes for the blue or pink color to 

 become white depends upon the quantity of the preservative present. Having found 

 the samples which contain preservatives, the nature of them must be determined by 

 the ordinary methods. 



This method is also recommended for the determination of the amount of formic 

 aldehyde in the sample. A series of samples is made up with known amounts of 

 this preservative, and the quantity in the sample under investigation is estimated by 

 comparison. 



