Abstract
The action of B. subtilis on a sucrose nutrient medium gave a 2:6 anhydro fructofuranose polysaccharide, levan, identical in structure with that synthesized by the action of B. mesentericus. The yield of levan (calculated on the fructose portion of the sucrose taken) amounted to 60–65%. The pure, ash-free, snow-white, powdery levan on hydrolysis with oxalic acid, was converted into fructose, the yield of the pure, crystalline hexose amounting to 99%. On methylation the levan was converted into trimethyl levan, the yield of partially purified product amounting to 88.5%, that of the highly purified material being 75%. The trimethyl levan on hydrolysis was converted into 1:3:4 trimethyl fructofuranose identical with that obtained from the levan synthesized by B. mesentericus from the same sucrose nutrient medium. The pure, snow-white, crystalline 1:3:4 trimethyl fructofuranose was obtained in a yield of 98.5%.Attention is drawn to the marked influence exerted on the course of the methylation by small amounts of inorganic impurities present in the levan and to the necessity for using only highly purified, ash-free products for physico-chemical investigations on the properties and structure of polysaccharides. It would appear that insufficient purification of many of the products, such as starch and inulin, used previously by different investigators, renders somewhat uncertain various conclusions drawn regarding the structure of these substances.