December 14, 1900.] 



SCIENCE. 



911 



through a tube containing small gravel 

 stones, with a current of air drawn in the 

 opposite direction. Further study of this 

 entire subject will place it on a more sub- 

 stantial basis. 



Dissolved Oxygen. — There appear to be sev- 

 eral methods which can be used success- 

 fully for this determination. The method 

 in most general use, however, is that of 

 Winkler as described in the Special Report 

 of the Massachusetts State Board of Health, 

 1890, Part I. 



By way of general comment it may be 

 added that recent developments in sewage 

 purification have indicated the desirability 

 of special attention to several matters. 

 Among them is the advantage coming from 

 a more general use of the determination of 

 suspended matters in sewage, with the loss 

 on ignition. Another point is the desir- 

 ability of determining the organic nitro- 

 gen in unpurified sewage by the Kjeldahl 

 method, in view of the varying percentage 

 of this constituent which is afforded by the 

 albuminoid ammonia. The so-called ' in- 

 cubation test ' to show the relation to putre- 

 faction of sewage after purification seems 

 to have much practical value, under certain 

 conditions, although the details cannot be 

 considered at this time. 



Quantitative Bacterial Examinations. — With 

 reference to this subject, there has recently 

 been a marked improvement in the general 

 results obtained in this country. It is true, 

 however, that methods of diiferent workers 

 are still variable to a degree which seems 

 unnecessary, and which is certainly not de- 

 sirable, when we consider that the value of 

 this class of data relates largely to purposes 

 of comparison. 



The culture medium now in general use 

 is nutrient gelatine, prepared substantially 

 as recommended by the Bacteriological 

 Committee in their Eeport of 1897. Meat 

 extract, however, is still used by a number 

 of workers, in place of meat infusion. Data 



are lacking to justify this as a general pro- 

 cedure. For some special lines of work 

 nutrient agar is used with apparent ad- 

 vantage. These conditions refer to analyses 

 of decomposed or stale sewage, where the 

 number of bacteria capable of rapid lique- 

 faction of gelatine is very large ; and also 

 to certain lines of field work. Several in- 

 vestigators have tried media of modified 

 composition, containing new ingredients in 

 some instances, but the present evidence is 

 altogether too inconsistent and indefinite 

 to permit of any recommendations along 

 this line. 



Concerning the reaction of the nutrient 

 gelatine, the optimum varies under differ- 

 ent conditions. Speaking in general terms, 

 the majority of waters now studied appear 

 to require ordinarily about 1.0 per cent, 

 acid. There are some waters for which 

 this reaction is too acid, and the sewage of 

 some manufacturing cities evidently re- 

 quires an alkaline medium. For important 

 continuous work the reaction to be used 

 should be carefully worked out with refer- 

 ence to the local conditions. 



The amount of agitation which the 

 sample of water should receive before 

 plating, in order to insure mixing and a 

 separation to a reasonable degree of groups 

 of bacteria, is afforded by 25 vigorous 

 shakes of the partially filled sample bottle. 



Most workers arrange, so far as practi- 

 cable, to have not more than about 200 col- 

 onies on the ordinary plate, such as Petri 

 dishes having a diameter of about 4 inches. 

 For those waters in which such numbers of 

 Bacteria are contained in small fractions of 

 one cubic centimeter, it is the general prac- 

 tice to dilute them with sterilized water, 

 rather than to use pipettes delivering small 

 fractions of one cubic centimeter. 



The amount of nutrient gelatine used for 

 each plate ranges at difierent laboratories 

 from 5 to 10 cubic centimeters. Most 

 workers use more than 7, while in some of 



