The isolation of D-glycero-D-galactoheptose and other sugar components from the specific polysaccharide of Chromobacterium violaceum (BN).

Although ketoheptose sugars are known to occur in Nature both free (La Forge, 1916-17; La Forge & Hudson, 1917) and in phosphorylated form (Robison, Macfarlane & Tazelaar, 1938; Benson, Bassham & Calvin, 1951; Horecker & Smyrniotis, 1952; Buchanan, 1953), it is only in very recent years that aldoheptose sugars have been detected in natural products. The aldoheptoses have been found, up to the present time, only as constituents of polysaccharides extracted from Gram-negative bacteria. From a strain of Shigella sonnei, Jesaitis & Goebel (1952) isolated a polysaccharide which they showed to contain glucose, galactose, glucosamine and an aldoheptose. This heptose was not identified but its behaviour on paper chromatograms was similar to that of D-gala-L-mnannoheptose (D-glycero-L-mqannoheptose). The specific polysaccharide of a strain of Shigel1a flexneri was described by Slein & Schnell (1953); this contained glucose, rhanmose, glucosamine and phosphorylated sugars, one of which was an aldoheptose. Derivatives of this heptose had very similar properties to those of D-gala-L-mannoheptose but the melting points of mercaptals and hexa-acetylmercaptals were lowered in the presence of the corresponding products of authentic D-gala-Lmannoheptose. It was considered that the sugar