^{page 571 note 1} Dulcit. This reaction is very valuable indeed. Its great importance as a differentiating characteristic was first pointed out by me at a laboratory meeting of the Pathological Society of London held on February 3rd, 1903.

^{page 571 note 2} Raffinose. These bacilli, just like all the members of the B. enteritidis Gaertner group, fermented raffinose with the production of acid always, and often also of a small amount of gas. But there seems some doubt as to the nature of this reaction.

L. Telesmin (Centralbl. f. Bakteriol. Abt. II., Bd. XII. 205—216, 1904) states that Kahlbaum's raffinose, after having been heated to 100°C., reduces Fehling's solution, and that on hydrolysis it yields glucose, galactose and fructose. In view of this statement a medium was made up composed of peptone 2% and gelatin 10% in tap-water, the whole being rendered slightly alkaline to litmus. Kahlbaum's raffinose was dissolved in cold distilled water and the solution sterilised by filtration through a porcelain filter. Enough of this raffinose solution was added to the gelatin to make the amount of raffinose equal to 1%. Neutral-red was then added as an indicator. The tubes were incubated to ascertain whether any contamination had occurred. As they were found to be sterile inoculations were made with about 15 different members of the group and all produced acid as evidenced by the change of colour of the neutral-red, a control tube remaining unchanged.

Messrs Merck, however, have informed me that there is no method of ascertaining whether a sample of raffinose is quite free from traces of fructose, etc. It would seem, therefore, that one cannot place much reliance upon this reaction.

^{page 572 note 1} Bacillus L. (Hume) differs from the rest in that it ferments adonit with the production of acid and gas.

^{page 572 note 2} Bacillus “Gl.” was isolated from the heart-blood of a guinea-pig which died after inoculation with the Bacillus mallei.