552 



SCIENCE 



[N. S. Vol. XXXI. No. 797 



raflBnose and mannite), and determined by titra- 

 tion the amount of acidity produced by each 

 strain in each medium. An examination of the 

 results obtained confirms and harmonizes the work 

 of the English observers in all particulars. The 

 commonest streptococci in human faeces are 

 8. mitis (acidifying dextrose and lactose), 8. 

 fwcalis (dextrose, lactose and mannit) and 8. 

 equinus (dextrose alone). In the faeces of the 

 cow 8. equinus and 8. mitis are present; but 

 8. fwcalis is absent and a form rare in human 

 faeces, 8. salivarius (dextrose, lactose and raf- 

 finose), is fairly abundant. In the faeces of the 

 horse practically all the streptococci present are 

 of the S. equinus type. (Full paper. Journal of 

 Infectious Diseases, VII., 1.) 

 The Determination of the Number of Leucocytes 



in Milk by a Direct Method: S. C. Peescott 



and R. S. Breed, Boston, Mass. 



The methods in general use for determining the 

 number of leucocytes present in milk are all based 

 on the use of the centrifuge. The assiunption is 

 that all but a small fraction of the leucocytes are 

 precipitated and also that this fraction is a fairly 

 constant proportion of the whole and can safely 

 be neglected. An investigation carried on in the 

 Boston Biochemical Laboratory during the past 

 summer has shown both of these assumptions to 

 be incorrect. By the use of a new method, it has 

 been found that the distribution of the leucocytes 

 in a given sample of milk after centrifuging varies 

 greatly in dififerent samples of milk, although 

 their distribution is approximately the same in 

 different samples of the same milk. Usually more 

 than half are present in the cream, one fourth or 

 less in the precipitated slime, and the remainder 

 in the skim milk. 



The variation in position of leucocytes in dif- 

 ferent samples is apparently due to the variable 

 percentages of cream present. The distribution 

 of the leucocytes in a centrifuged sample corre- 

 sponds closely to the previously known distribu- 

 tion of bacteria in similar samples. 



The new method by which these facts have been 

 ascertained is as follows: a measured drop 

 (.01 c.c.) of milk to be examined is spread evenly 

 over a measured area (1 sq. cm.) on a glass slide, 

 dried with gentle heat, the fat dissolved out with 

 xylol, fixed with alcohol for a few minutes, the 

 slide again dried and over-stained with methylene 

 blue and partially decolored with alcohol. The 

 number of leucocytes present is then determined 

 by examination with the microscope. Results 

 done in duplicate show a small percentage varia- 



tion proving that the practical error is not a 

 large one. 



A series of tests of milk show that much larger 

 numbers of leucocytes are normally present in 

 milk than has been supposed. The average num- 

 ber of leucocytes present in the samples examined 

 is approximately 1,500,000 per cubic centimeter, 

 while numbers less than 100,000 per c.c. are 

 uncommon. 



The Bacteriology of Condensed and Evaporated 



Milks: S. C. Pbescott and R. N. Hott, Massa- 

 • chusetts Institute of Technology. 

 Some Problems of Sanitary Milk Production: 



P. G. Heinemann, a. B. Luckhaedt and A. C. 



Hicks, The University of Chicago. 



A series of experiments was made during the 

 month of September at a sanitary dairy to throw 

 light on the following points: (1) the bacterial 

 content of separator milk and cream, (2) the 

 value of narrow top pails with and without 

 strainers, (3) the bacterial content of milk after 

 straining through layers of absorbent cotton, (4) 

 a study of body cells in separator slime and an 

 attempt at classification. 



It was found that the bacterial content of the 

 separator cream was very small, the average of 

 48 tests being 132 bacteria per cubic centimeter 

 of 40 per cent, separator cream. The separator 

 milk in the same number of tests contained 2,130 

 bacteria per cubic centimeter and the original 

 milk contained 738 bacteria per cubic centimeter. 

 We conclude from these experiments that the ac- 

 tion of the separator tends to break up clumps, 

 chains and imperfectly divided forms so as to 

 increase the colony count. 



The experiments with the narrow top pail with 

 and without strainer showed that the count was 

 620 bacteria per cubic centimeter with the 

 strainer and 674 without the strainer. These 

 figures are the averages of 108 tests. The small 

 difference in favor of the strainer may possibly 

 be due to experimental error and of little signifi- 

 cance. Still we think that the strainer should not 

 be omitted, since the milk needs straining at some 

 point or other of production to remove foreign 

 material which is bound to gain access even in 

 the most carefully managed dairies. 



As a result of 240 consecutive tests we con- 

 clude that straining milk through thick layers of 

 absorbent cotton, as is customary in many dairies, 

 is decidedly disadvantageous. The force of the 

 milk being poured on top of the strainer seems to 

 break up bacterial aggregates so as to increase 

 the colony count in the strained milk. 



