1919] DAIRY FARMIlSrC DAIRYING. 277 



priate bran was daily given to each animal. During wheat bran feeding 1.2 

 per cent more milk and 1.3 per cent more butter fat on the average were pro- 

 duced. 



Proximate analyses of the feeding stuffs used are reported. The corn bran 

 was noticeably lower than the wheat bran in ash, protein, and fat. It is stated 

 that digestion trials made at the station with sheep show that the coefficient of 

 digestibility of dry matter of corn bran is 80 and of wheat bran G6. 



Milking machines: III, As a source of bacteria in milk. IV, Methods of 

 maintaining- in a bacteria-free condition, <!. L. A. Ruehle. R. S. Breed, and 

 G. A. Smith {New York State Sta. Bui. 450 (1918), pp. 113-181, flys. 11).— 

 These studies, of which a preliminary report has been noted (E. S. R., 38, p. 

 377), are grouped in a series \\ith two previous bulletins (E. S. R., 28, p. 472) 

 and include observations of the use of milking machines on the station herd and 

 at 10 commercial dairy farms. 



To stu^ly the milking apparatus as a source of bacteria, an artificial udder 

 was constructed from which sterile water was " milked " by the machine. To 

 determine at what stage bacteria gets into the water, means were devised for 

 collecting samples at two points in its passage into the pail. The chief source 

 of bacteria was found to be the teat-cups and rubber tubes, although the pail 

 in some cases Vv-as an important medium of infection. The suction trap placed 

 upon the recent types of one of the makes of milking machines used was found 

 to be efficient in preventing the return of contaminated condensation water 

 from the vacuum piping to the pails. 



The stable air which enters the machine during milking was analyzed and 

 found not to be responsible for increasing the bacterial content more than 

 34 per cubic centimeter under the worst conditions tested. The cotton filters for 

 which provision is made on one machine were found to remove more than two- 

 thirds of the bacteria from the air and also a small amount of dust. Since the 

 total amount of contamination from the air was insignificant, the improvement 

 due to the filters was too slight to be detected in ordinary analyses of machine- 

 drawn milk. 



Dropping the teat-cups to the floor in dirty bedding was found to cause rela- 

 tively large amounts of dirt to appear in the pails of the machines; but the 

 bacteriological analyses of the sterile water, milked through the machine at 

 the same time, did not reveal excessively high counts. These conditions failed 

 to give counts in excess of 2,5,000 per cubic centimeter. Here, as in other studies 

 of the amount of bacterial contamination derived from dust, the germ count 

 was found to be an unsatisfactory index of the amount of dirt and filth present. 



Although it was found possible to sterilize milking machines completely by 

 means of steam applied to the metal parts and by harmless antiseptics on the 

 rubber parts, the precautions necessary are considered impracticable where 

 certified milk is produced and quite impossible on the average farm. Satisfac- 

 tory bacterial results were secured where practicable methods were used for 

 keeping the machines cleaned, provided the teat cups and tubes were immersed 

 in any of the following antiseptic solutions: (1) Brines containing at least 

 10 per cent of salt; (2) solutions of chlorid of lime; (3) a combination of 

 brine and chlorid of lime; (4) lime water; (.5) cold running water; and (6) 

 a commercial germicide, which, however, is not recommended for general use, 

 because a trace of it gives milk a disagreeable taste. Chlorid of lime and 

 running water were ineffective if the temperature of the solution was as great 

 as 60° F. The solution recommended for general use is chlorid of lime dis- 

 solved in a saturated brine. Antiseptic solutions wei'e not found to be suc- 

 cessful unless the teat cups and tubes were maintained in a cleanly condition. 



