483 



Station Record, vol. in, p. 22), and the New York Cornell Station, 

 reported in Bulletin No. 29 of that station (see Experiment Station 

 Record, vol. in, p. 231), are cited. These results showed practically no 

 loss due to delayed setting. The results with lots 3, 4, and 5 are in 

 accord with these, but those witli lots 1 and 2 are at variance with them. 



The difference in the creaming between the delayed and imiiifdiate setting is shown 

 in an entirely independent way by the quantity of cream obtained in the two 

 cases. It is a general rule that where the separatiin of cream is retarded in any 

 way it will occupy more space than where it is not rctarde<l. In the 99 trials made 

 in these tests 70 show more cream from the delayed portions, 16 have the same 

 amount upon both, and 13 show more upon immediate setting. 



In lots 1 and 2, where the most difference was found, every trial showed a lar<'-er 

 Volume of cream upon the delayed portion. In these two lots also every trial showed 

 more fat in the skim milk from the delayed portion. In the 99 trials 82 gave uujro 

 loss from the delayed, 7 gave the same, and 10 gave more in tlie immediate setting. 

 Lot 4 showed the least difference and lot 2 the most. 



Ih the use of ice profiiahle in deep setting f (p. 17). — Numerous trials 

 were made with herd milk set as nearly as i)ossible ujider the same con- 

 ditions, except that the temperature of the water in the tank varied from 

 35° to 58° F. The milk was skimmed after 11 to 12 hours. The losses 

 of fat in the skim milk per 100 pounds of milk set at the different temper- 

 atures of setting- were as follows : 



Ponnds. 



Ice, 350 to 45° F 0. 2.32 



No ice, 48'^ F 0. 297 



No ice, 540 to 56° F 0. 746 



No ice, .58° F 0. 949 



I am confident that in most cases [the loss of fat where no ice is used is] froiu it to 

 1 pound more per 100 pounds of milk than Avhere suthcieut ice is used to maintain the 

 temperature below 50° F. With a herd of ten average cows this makes a difference 

 in favor of ice of from 20 to 40 cents jier day, if butter sells for 20 cents per ])ound. 

 In most localities this would cover the necessary expense for ice and leave a ]ar"-e 

 margin for profit. 



Conclusions (p. 18).— The results indicated by the above experiments 

 are summed up by the author as follows : 



(1) In skimming by the Cooley system the siphon should be set so as to leave at 

 least 1 inch of skim milk with the cream. 



(2) The efficiency of creaming by deep setting is greatlj^ influenced by the charac- 

 ter of the herd, the milk from soiTie herds creaming very close, while that from other 

 herds, under the same conditions, creams poorly. This difference may amount to as 

 much as 1 pound of butter per 100 pounds of milk. 



(3) Delay in setting may cause a considerable loss with the milk from some herds 

 and scarcely any with that of others. To avoid the possibility of such loss it is rec- 

 ommended that milk be set as soon as possible after milking. 



(4) Deep setting without ice under the most favorable conditions results in con- 

 siderable loss, and where the temperature of the water used is not louver than .50° F. 

 the loss is excessive, reaching in some cases as much as 25 per cent of the total fat in 

 the milk. 



(5) The centrifugal system of separating cream overcomes all of these difficulties, 

 giving an efficient creaming with milk from all sources, either directly after milking 

 or after standing several hours. 



(6) The Baby Separator No. 2 may be used with advantage with herds of from ten 

 to twenty cows. 



