1918.1 DAIRY FARMING DAIRYING. 579 



tained by exposing glass petri dishes containing a sterile nutrient medium in 

 the dairy barn of the station and the adjoining milk room. 



The results indicate that in a well-cleaned dairy barn very few bacteria are 

 present in the air, but when dust is present the number is very large. " When 

 the cows are in the barn the number of bacteria in the air is materially in- 

 creased. The presence of bedding in the stable greatly increases the bacteria 

 of the stable air. Dusty mill feeds and hay are fruitful sources of bacterial 

 contamination. The act of brushing the cows adds myriads of bacteria to the 

 surrounding atmosphere. Washing the cows' udders, even when apparently 

 clean, greatly reduces^the number of bacteria falling from the udder. Infection 

 from washed udders is less than one-seventh of that from unwashed udders. The 

 air in a pasture upon which is growing a heavy sod of grass is nearly sterile. 

 In a dairy room having smooth walls and a concrete floor which are regularly 

 and carefully washed the air should be practically free from bacteria." 



It is suggested that " to keep down the bacterial contaminatfon of the air 

 in the stable where the cows are milked the room must be kept well cleaned 

 and every effort must be made to keep down all dust. Dusty bedding must be 

 avoided, and if bedding is used dust should be laid by sprinkling. Since cow 

 hairs are laden with bacteria, all loose hairs should be removed from the cows 

 by currying, in order to prevent them from falling into the milk. However, 

 sufficient time should elapse after currying to permit the dust and bacteria to 

 settle to the floor before milking begins. The cows' udders should be carefully 

 washed with tepid water and a clean cloth before milking begins. Wiping the 

 udder with a clean, damp cloth greatly reduces bacterial contamination of the 

 milk. In feeding mill feeds and hay care should be taken to raise as little dust 

 as possible." 



A preliminary report on a series of cooperative bacterial analyses of milk, 

 R. S. Breed, W. A. Stocking, et al. (Jour. Dairy ScL, 1 {1911), No. 1, pp. 19- 

 SJf). — In this paper, which was read before the Laboratory Section of the 

 American Public Health Association, at Cincinnati, October 24, 1916, a pre- 

 liminary report is submitted of bacteriological analyses of milk made to deter- 

 mine whether the results secured in laboratories by university men trained in 

 i-esearch methods were as irregular as those secured in the commercial and 

 control laboratories in New York City. Bacterial analyses were made of four 

 sets of five samples each by seven men, four working in the Cornell University 

 laboratory at Ithaca, and three at the New York State Experiment Station. 

 Each man used the technique he thought would give accurate results. The 

 results are compared with those secured in New York City and already 

 reported (E. S. R., 33, p. 767). 



A second series of analyses was made on samples of three lots of high-grade 

 milk inoculated with a culture of the colon bacillus, for the purpose of com- 

 paring results secured by the plate method, by the dilution method, and group 

 and individual counts by the microscopic method. These counts were made 

 by one person only. 



It is concluded that " research men using technique which differs much in 

 details may be depended upon to secure much more consistent agar plate counts 

 from ordinary samples of market milk than laboratory assistants working 

 rapidly and using the routine methods of analysis recommended for the 

 purpose. Inexperienced workers are apt to make gross errors in count when 

 using the direct microscopic method as a means of making exact counts. 

 Experienced workers, however, secure results which compare favorably with 

 those secured by workers who have had experience with the plating technique. 

 The labor and time necessary in order to make relatively accurate counts by 



