Ajpeil 23, 1915] 



SCIENCE 



619 



duced by them in the animal body. Their physio- 

 logical activity is measured by the end products 

 of their metabolism. Physiologically, bacteria, 

 generally, have evolved in two main directions — 

 one group possessing marked carbohydrate split- 

 ting properties, the other having developed the 

 property of digesting various protein substances. 

 The streptococci belong to the former division, 

 showing but little tendency to proteolysis. 



It appears natural enough to assume that the 

 biologic activities of a cell would correspond with 

 the chemical nature of its protoplasm. Yet such 

 a correlation has not been worked out except in a 

 few isolated eases. Among the streptococci such 

 a correlation, if it exists, would be especially sig- 

 nificant in that it would help to differentiate the 

 various members of a genus that has puzzled 

 many investigators. 



The agglutination, fermentation and hemolytic 

 properties of sixty strains of streptococci obtained 

 from various pathological conditions, were stud- 

 ied, using four agglutinating sera having a titer 

 of SOO-1,000, and six carbohydrates and other 

 fermentable substances as follows: 



Disaccharides{^^=t°^^^^^g 



Trisaccharide Eaffinose 



Alcohol Mannite 



Glucoside Salicin 



Starch Inulin 



Only a limited number of the strains were ag- 

 glutinated by the sera used. A definite correla- 

 tion was, however, obtained between the aggluti- 

 native and fermentative characters. The serum 

 produced by a strain of one fermentative group 

 (the group that fermented salicin, for instance) 

 agglutinated only cultures of its particular di- 

 vision and failed to agglutinate members of any 

 of the other groups. No such correlation was ob- 

 tained with the hemolytic property, members of 

 one hemolytic group being agglutinated by the 

 sera produced by strains from other hemolytic 

 groups. 



The results obtained indicate that a separation 

 of the streptococci obtained from various patho- 

 logical conditions into three fermentative types 

 would coincide most closely with their natural re- 

 lationship. 



The groups suggested are: 



1. Salicin fermenters only, generally hemolytic 

 — Str. pyogenes. 



2. Raffinose fermenters, salicin usually fer- 

 mented, mannite always negative, generally pro- 

 duces a green colony on blood agar — Str. sali- 

 varius. 



3. Mannite fermenters, generally ferment sa- 

 licin, rarely ferment raffinose, variable in their 

 reaction to blood — Str. fecalis. 



The Filterability of B. ironchiseptieus : with an 



Argument for a Uniform Method of Filtration: 



N. S. Feeby. 



The purpose of the paper was to place on rec- 

 ord a series of filtration experiments with Bacillus 

 ironchisepticus, described as. the cause of canine 

 distemper by Ferry in 1910, McGowan in 1911 

 and Torry in 1913. 



The experiments were conducted as follows: 

 The organism was grown twenty-four hours both 

 on agar and in bouillon. The bouillon growth 

 was filtered, undiluted, while the agar growth was 

 taken off in bouillon and made into a suspension 

 of about the same density as the bouillon culture. 

 The method of testing the integrity of the filters 

 was that described by Bulloch and Craw in the 

 Journal of Hygiene in 1909, which depends upon 

 the measure of the pressure of air as it is allowed 

 to pass through the pores of the candles while 

 immersed in water. The filtration was conducted 

 at room temperature, one hour taken as the 

 length of time for filtration, and three pressures 

 were used; gravity, 15 lbs. (negative) and 225 lbs. 

 (positive). 



The experiments proved conclusively that the 

 Bacillus hronchisepticus is a filterable microor- 

 ganism. The work also corroborates the results 

 of previous investigators with regard to the fact 

 that the less pressure used the more easily will 

 some organisms pass through the filters. 



Some interesting possibilities were suggested 

 by the outcome of this work. Since 1905, when 

 Carre claimed he had produced typical symptoms 

 of distemper in susceptible dogs from the filtered 

 discharges of diseased dogs, the majority of 

 writers have classified the etiology of canine dis- 

 temper as a filterable, invisible or ultramicro- 

 scopic virus, and it is so described in many text- 

 books. The work of Ferry, McGowan and Torry 

 with the Bacillus ironchisepticus tends to refute 

 the statements of Carre. 



The results of the present filtration experiments 

 puts an entirely new light on the subject. If the 

 Bacillus ironchisepticus is the cause of canine 

 distemper, then the experiments corroborate the 

 work of Carre. If the work of Carre was correct, 

 and if the causative agent of canine distemper is 

 a filterable virus, then the experiments point very 

 conclusively to Baoillus hronchisepticus as the 

 etiological factor and confirms the findings of the 

 three previously mentioned investigators. 



