294 



sprayed with the carbonate of copper solution, 28 pounds ; and those un- 

 treated, 22.5 pounds. The tubers from the Bordeaux-mixture rows were 

 much larger than the others." Lateness of planting, due to wet weather, 

 and the dryness of the growing season, made the test comparatively un- 

 satisfactory. In a case cited in the bulletin where the Bordeaux mixt- 

 ure was used on a large scale for this disease in 1890 on a farm in Ohio, 

 the results were unfavorable. 



"There is still much to be learned concerning this bacterial blight, 

 and further experiments are needed to determine whether it can be suc- 

 cessfully prevented on a large scale under ordinary commercial condi- 

 tions by the use of fungicides, and whether this can be done with suffi- 

 cient cheapness to pay for the additional outlay. One of the points that 

 especially needs to be determined is whether non-infected tubers should 

 not be planted to get the best results from the application of fungicides. 

 This would be the natural supposition, and experiments are now in 

 progress to determine the facts of the case." 



Pennsylvania Station, Bulletin No. 12, July, 1890 (pp. 30). 



Simple itrthods of determining milk fat, W. Freab, Ph. D., 

 AND G. L. Holter, B. S. (pp. 3-31, illustrated). — A detailed description 

 of the Short and Cochran methods for estimating the percentage of fat 

 in milk, together with a comparison of the results of these and the gravi- 

 metric (asbestos) method. Of the mean of duplicate analyses of twenty- 

 six samples of the milk of individual cows, 85 per cent by the Short 

 method were higher and 81 per cent by the Cochran method were lower 

 than those of the gravimetric. About one third of the determinations 

 by the Short method differed by over 0.2 percent from the gravimetric, 

 the difference in three cases being 0.4 to 0.49 per cent of fat. By the 

 Cochran method less than one quarter of the tests differed from the 

 gravimetric by 0.2 to 0.3 per cent, and in only one case was the differ- 

 ence as high as 0.3 per cent. The average error for the twenty-six 

 analyses was with the Short test +0.15, and with the Cochran —0.06 

 per cent. " In Short's method a somewhat closer agreement of dupli- 

 cates is obtained than in Cochran's method, as would be expected from 

 the finer subdivision of the apparatus and the greater quantity of milk 

 used in the former. In general, however, Cochran's method seems some- 

 what preferable, owing to the greater rapidity with which single deter- 

 minations can be made." 



Tests by the Short method in which the solution was boiled only one 

 half hour after the addition of the acid, gave i^ractically the same re- 

 sults as where the boiling was continued for one hour, as prescribed, 

 the mean difference of results between the prescribed and the modified 

 methods being only 0.08 per cent in eleven cases, " while by the orig- 

 inal method, comparing duplicates in forty-one analyses, the mean dif- 

 ference was 0.07 per cent." With the modified method " the fatty acids 



