BY R. GREIG SMITH. 453 



but did not darken upon boiling the mixture in a test-tube. 

 Sulphuric and phosphotungstic acids produced a curdy precipitate. 

 No reactions were obtained with neutral lead acetate, barium 

 hydrate, silver nitrate, mercuric chloride, copper sulphate, milk 

 of lime, borax, iodine, Fehling's solution or Schweitzer's reagent. 

 With the exception of the phosphotungstic acid precipitate 

 which is given with most bacterial gums, these reactions are 

 identical with those given by arabin. But I did not, from the 

 nature of the slime, expect the gum to be arabin, and this 

 expectation was confirmed by the examination of the products of 

 hydrolysis. After the gum had been freed from reducing sugars 

 it was boiled for 5 hours with 5 % sulphuric acid, when a portion 

 was found to contain no gum precipitable by alcohol and to con- 

 tain reducing substances. Fehling's solution was rather slowly 

 reduced for a sugar, and as the whole portion gave but a trace 

 of osazone it was evident that the gum was rather difficult to 

 hydrolyse. In this respect it approached the pararabin gums. 

 In two subsequent tests the gum was attacked by evaporating 

 the solution in the water-bath with sulphuric acid until it began 

 to char; then the solution was diluted with water to form a 

 10 % solution which was boiled for two hours. A portion 

 reduced Fehling's solution quickly, and upon the osazone being- 

 prepared and purified it was found to be galactosazone. Arabin- 

 osazone was carefully sought for at all stages of the purification, 

 but it could not be detected. The gum was therefore a galactan, 

 and was peculiar in giving the chemical reactions of arabin. In 

 view of this behaviour, I propose to call the organism Bacillus 



2?seudarahi7ius. 



Bacillus pseudarabinus, n.sp. 



Shape, etc. — The organism is a coccobacterium or short thick 

 rod with rounded ends. In nutrient agar culture the rods 

 measure 0*6 : 0-6-Q-9/i, and in saccharose-potato-agar they are 

 0-7 : 0'8-1/i. In bouillon cultures the cells are actively motile. 

 They stain readily, but are decolorised by the Gram method. 

 The flagella are numerous and peritrichous; up to nine have been 

 seen. No spores were observed. yfC^AV i^ 



W ^ t- ■::: -~"^ / 



^ ^^^'^ V 



LIBRARY 



» v> 



