376 EXPERIMENT STATION RECORD. [Vol. 40 



time. The cost of 100 lbs. of milk is computed on the assumption that the 

 cow is dependent solely on pasture for five months, beginning May 15, and is 

 on full feed for the seven months beginning October 15. 



From the data at hand for these Ohio herds, it was found that during winter 

 the items not offset by credits required to produce 100 lbs. of milk are 44 lbs. 

 of grain, 159 lbs. of silage, 43 lbs. of hay, 31 lbs. of Stover, 2'_» lbs. of bedding, 

 and 4.7 hours of labor and management. During pasture season the charges 

 are 6.1 days of pasture and 3.1 hours of labor. 



A scale is suggested for monthly deviations from average prices which gives 

 a uniform percentage from November to April of 128. The results are com- 

 pared with the formulas and sliding scales proposed by i". a. Pearson and 

 G. F. Warren. 



Concerning milk costs and prices. V. W. RADKB [Washington Stu., V 

 Wash, st, i. i/o. Bui., 6 (1918), No. .''. i>i>. i :' \ discussion of the author's 



figures for cost of milk production in western Washington (E. s. i:.. 89, p. 

 784) in the light of criticisms from distributors and condensing interests sum- 

 marized by W. A. Linklater (B. s. R., 10, p. 87). 



The examination of milk for public health purposes. J. RAC1 I V< If) Fork: 

 John Wiley & Sons, 1918, pp. VI+&4, figs. ♦).- This handbook, by the city 

 bacteriologist and food examiner \'<<v < Ottawa, * tatario, Includes considerably more 

 than the routine methods of chemical and bacteriological examination required 



of officials charged with enforcing pure food laws, and aims to i f general 



use to students. In the section on chemistry, in addition (<> the official methods 

 for the determination of fats and total solids and the detection of preservatives, 

 alternative procedures are outlined. Selected methods are given for the deter- 

 mination of lactose, total proteins, caselnogen, albumin, mineral constituents, 

 total acidity, refractive Index, etc., to aid In the correct diagnosis of sophisti- 

 cation. The section on bacteriological examination includes a general Burvey 

 of sources of infection, an outline of methods of enumeration, and a discussion 

 of the possibility of detecting excremental and pathogenic forms. Details of 

 the standard methods of the American Public Healtb Association are given in 

 some instances. Many of the illustrative tables are summaries of the author's 

 unpublished Investigations. 



It is pointed old that averag unta may be misleading If deviations from 



the mean are large or the number of variants is small. In these cases the 

 median would be a better indication of the quality of the supply, but it is best 

 to group the results into classes having a comparatively small range of bac- 

 terial counts. 



An appendix gives the composition of certain culture media ami tables for 

 computing specific gravit\. 



The significance of the colon count in raw milk. S 11. ATKBS and F. W. 

 Ci.emmer (U. 8. Dept. Agr. Bid. 739 [1918), pp. •>■>. pis. .1 fig. /).— This is a 

 study of the significance of the colon count as an index of cleanliness in milk 

 production, based upon an examination of the literature on the subject and a 

 series of bacteriological studies under controlled conditions. 



The colon count aj* at present determined by plating methods does not dis- 

 criminate between the Bacillus c<>u ami />. orogenes types of organist 

 Since the first was found to be mostly of fecal and the latter of nonfecal origin, 

 the colon count Is not a direct measure of manorial contamination. Using the 

 Clark-Lubs method (E. S. R.. 34, p. 136), the authors found only 4 cultures o{ 

 B. wrogenes among 1,100 cultures Isolated from 20 samples of fresh cow fe. i 

 but found this type common in the air of barns and very generally Introduced 

 into the milk by dirty utensils. 



