482 



Estimating the fat in the buttermilk iu the manner described above, 

 the average total loss of fat iu buttermilk and skim milk per lUO i>ounds 

 of milk set is calculated for the two systems as follows: 



Comparative losses of fai per 100 pounds of milk creamed by centrifuge and deep setting. 



When the Baby Separator was first tried by us the cream screw was so adjusted 

 that the cream contained over 30 per cent of fat. This cream, when ripened, was 

 too thick to churn well, the buttermilk containing from 0.5 to 1 per cent of fat. By 

 diluting this cream with about one fifth its volume of water, the fat in the butter- 

 milk was reduced to about 0.2 per cent and in some cases as low as 0.15 per cent. 

 This improvement I believe to be due entirely to having reduced the consistency of 

 the cream to a jioint where it was uniformly churned. This is closer work than is 

 usual with cream from deep setting, and warrants the eonclnsiou that centrifugal 

 cream, containing about 25 per cent of fat, will churn more etUciently than that from 

 deep setting, which contains less than 20 per cent. 



Delayed setting (pp. 11-17). 



The following trials were undertaken to determine whether the creaming of differ- 

 ent samples of milk would be similarly affected by delay in setting when deep setting 

 is used. The tests were made with milk from the same groups of cows used in the 

 preceding experiment. In each trial the milk from each group of cows was mixed 

 and divided, one half of it being set immediately in ice water, the other half being set 

 in the same tank after standing in the open air for periods ranging from 15 min- 

 utes to 3 hours. The delayed milk was mixed just before being put into the tank. 

 The samples were skimnu'd in tlie order in Avhich they wore placed iu the tank, 

 both the delayed and immediate setting being allowed the same time — usually 12 

 hours — for creaming. In most cises the milk was drawn off to within 2 inches of 

 the cream line, although some milk was skimmed to within 1 inch. All compara- 

 tive trials were skimmed in the same way. [Eighteen comparative tests were 

 made with milk from each lot. The results are tabulated.] • • ♦ 



Although on the whole these trials show considerable loss by delayed setting, their 

 most interesting feature is the influence which delay has upon the creaming with 

 different kinds of milk. 



The loss by delay with milk from lot 3 was practically nothing; with lots 4 and 5 

 the loss was also very small, while with lots 1 and 2 it was very large, being largest 

 of all with lot 2. The loss of fat from delay appears to be independent of the 

 amount of fat in the milk, although as a rule it is somewhat larger in the rich milk 

 than in the poor milk. * * « 



If the loss of fat in the buttermilk be considered, the difference in favor of imme- 

 diate setting would be still more marked on account of the increased quantity ol 

 buttermilk from the delayed portions. * * * 



Test« in which the milk from each lot of cows was divided as before, one half 

 being set immediately after milking, the other half after 30 minutes* delay, the 

 milk being kept warm until put into the tank, [indicated] that the efficiency of 

 the creaming by delayed setting is not materially imjiroved by keeping the milk 

 warm. 



The results of similar experiments at tlio Miiinc Stiition. r('i)ortrd in 

 the Annual lieport of that station for 181)0, part ii (see Experiment 



