750 



SCIENCE 



[N. S. Vol. XLIV. No. 1143 



on an agar plate develops from a single bac- 

 terium. 



The development of microscopical methods 

 of counting bacteria in milk have now made 

 it possible to check up this matter. Studies 

 at the N. Y. Agricultural Experiment Station 

 by J. D. Brew, as well as cooperative analyses 

 carried out by the dairy husbandry depart- 

 ment of the N. Y. State College of Agricul- 

 ture at Ithaca and the bacteriological depart- 

 ment of the Agricultural Experiment Station 

 at Geneva have shown that the number of 

 bacteria in market milk is rarely less than 

 twice the number of colonies developing on 

 agar plates even after prolonged incubation at 

 two different temperatures; and that the num- 

 ber of bacteria is usually from three to six 

 times the number of colonies. In those 

 fairly common market milk samples where the 

 predominant bacterial flora consist of long 

 chain streptococci, the actual number of bac- 

 teria present may be fifteen to twenty-five 

 times the number of colonies on agar plates. 



With these facts established, there seems to 

 be no justification for continuing the present 

 unscientific custom of referring to agar- 

 plate counts as showing the number of bac- 

 teria in milk. As a matter of fact they show 

 the number of colonies developing on nutrient 

 agar (or other culture medium) under the 

 conditions of incubation used, and nothing 

 more. In the earlier literature the latter form 

 of expression was common and is still used by 

 some investigators. Americans, however, gen- 

 erally use the inaccurate form of expression 

 especially when discussing sanitary milk prob- 

 lems. 



It does not require a vivid imagination to 

 picture the dismay of the layman, whether 

 consumer, milk dealer or farmer, when he dis- 

 covers that what he has been told about the 

 number of bacteria in milk is all based on a 

 fallacy and that the real numbers are from 

 one and a half to twenty-five or more times 

 the figures which have been given to him. 

 Neither does it require a vivid imagination to 

 predict that those forces which find it to their 

 advantage to resist the efforts which are being 

 made to control our milk supplies will be quick 



to seize upon the seeming inconsistencies of 

 bacteriologists as a means of discrediting the 

 use of bacterial counts for controlling milk 

 supplies. 



So long as there was no available method 

 by which the actual number of bacteria in 

 milk could be counted, the use of the short 

 form of expression had some excuse because 

 of its convenience. Now that the real facts 

 are known, its continued use will increase the 

 present confusion. This confusion does not 

 trouble bacteriologists, nor will it do so, for 

 the majority of them have understood all of 

 the time that they were probably not telling 

 the truth about the matter; but it does be- 

 wilder the uninitiated. 



Eobert S. Breed 



N. T. Agricultural Experiment Station 



OSTWALD'S HANDBOOK OF COLLOIDAL 

 CHEMISTRY 



In a criticism 1 of my review 2 of Professor 

 Fisher's translation of Wo. Ostwald's " Hand- 

 book of Colloidal Chemistry" Professor 

 Richard C. Tolman disagrees with my state- 

 ments concerning negative surface tension, and 

 submits certain thermodynamic considerations 

 and experiments as evidence of the existence 

 of negative surface tension. The question is 

 one over which two people may disagree inas- 

 much as it depends solely on their point of 

 view. Professor Tolman relies principally 

 upon thermodynamic considerations, while I 

 refuse to consider energetics as infallible in 

 the present case, but base my reasoning on 

 ordinary atomistics. In fact I regard the 

 application of thermodynamics to disperse 

 systems as decidedly hazardous. 



In the first place it is well known (Max- 

 well) (Smoluchowski) that the second law is 

 no longer valid when applied to particles ap- 

 proaching molecular dimensions. Secondly the 

 first characteristic of all colloidal solutions is 

 unstahility. I have yet to experience an abso- 

 lutely stable permanent colloidal solution. 

 Once we admit the absence of true thermo- 



i Science, 44, 565, 1916. 

 2 Science, 43, 747, 1916. 



