514 J. H. RICHARDS 



cultures for the present study were taken from patients who 

 were constantly afebrile excepting for the infrequent attacks of 

 fever, lasting for a short time only, during an exacerbation of 

 the arthritis, and the cultures were not taken at the time of the 

 fever. No case is included in this series that did not have at 

 least three months' history of arthritis. It was purposed to in- 

 clude no case that could possibly be considered as acute rheu- 

 matic fever. Likewise no case is included with a history of 

 syphilis or a positive Wassermann reaction, or a history of a 

 recent gonorrhoea or a positive gonorrhoeal fixation test. 

 The technique of blood culture was the same in all cases: 



1. North's medium was used. Heating this medium for steri- 

 lization caused different batches of the media to increase in 

 acidity to markedly different degrees. It was found necessary 

 for the development of the streptococcus obtained, to titrate 

 the medium so that after the final heating at the time of planting, 

 the reaction was 0.3 to 0.5 ™ acid, using phenolphthalein as an 

 indicator. 



2. Five to ten poured plates were used, and planting was done 

 immediately after taking the blood. 



3. The concentration of the blood in the plates was varied, so 

 that the first plate had in it only about 0.3 cc. of blood, and the 

 last 5 to 6 cc. This procedure showed that very often a growth 

 would be obtained in the dilute plates and none at all in the more 

 concentrated ones. It may be that the whole blood itself acts 

 to inhibit the development of the growth. It is well known that 

 whole blood makes every attempt to rid itself of bacteria, and 

 does not lend itself to use as a medium as well as other tissues of 

 the body. 



4. Incubation of the plates was continued for ten days. To 

 prevent drying, the plates were inverted as soon as they had set, 

 and were then placed over a basin of water in the incubator. 

 The development of the original culture did not appear in two 

 cases until the eighth day of incubation, and the average time was 

 five days. For this reason, some procedure was necessary to 

 prevent drying of the plates, for under the usual conditions the 

 plates dried on the third or fourth day. 



