The quantitative estimation of hemicelluloses by direct isolation.

The various methods described in the Hterature for the quantitative estimation of hemicelluloses were reviewed in an earher paper (5) ^, It was pointed out that the analytical methods, based either on the determination of the quantity of reducing sugars produced when these carbohydrate complexes are subjected to acid hydrolysis or on the determination of the percentage of furfural, afford results that are difficult to evaluate in terms of actual quantity of bemicellulose in the material under examination. Freeze (ö) has introduced a method for the quantitative estimation of hemicelluloses based on the isolation and weighing of the hemicelluloses. According to the Preece procedure, the sample (20-25 gm.) is given a preliminary extraction first with a hot 0.5-percent aqueous ammonium oxalate solution, then with a 1-percent boiling sodium hydroxide solution in 50-percent ethanol, and finally with a 50-percent ethanol solution.. The residual material is then repeatedly extracted, at room temperature (25°-30^ C), with a 4-percent aqueous sodium hydroxide solution, until a portion of the extract no longer gives a precipitate when neutralized with acetic acid and diluted with an equal volume of acetone. The combined alkaline extract is then rendered, acid by the addition first of glacial acetic acid and then of acetone equal in volume to the acidified solution. The precipitate is filtered, dried in vacuo and the weight of the moistureand ash-free product is determined. The result thus obtained is calculated as percentage of free bemicellulose. The material remaining from the extraction with the alkali solution is further extracted with 4-percent boiling aqueous sodium hydroxide solution, and the product is isolated as previously described. The result is reported as percentage of combined bemicellulose. So far as the writers are aware, no attempt has yet been made to examine the Preece method critically. In a slightly modified form, the method was used by Bus ton (2, 3) and by Bennett {!) in their studies of the cell-wall constituents of plants. However, when the plant material is extracted according to the Preece procedure, not only are the hemicelluloses removed but some of the lignin also, with the result that both the free and combined bemicellulose fractions of Preece are contaminated with lignin. Moreover, when the extraction