The experience of this laboratory has shown that results practically identical with those got as above, are obtained by conducting the digestions in hard glass beakers covered with watch glasses. The ease of manipulation in the modification of the process just mentioned is a sufficient justification for its use. =273. Separation of Cellulose.=—Hoppe-Seyler observed that cellulose, when melted with the alkalies at a temperature as high as 200°, was not sensibly attacked.[223] Lange has based a process for determining cellulose on this observation.[224] The process, as improved by him, is carried out as follows: From five to ten grams of the substance are moistened with water and placed in a porcelain dish with about three times their weight of caustic alkali free of nitrates and about twenty cubic centimeters of water. The porcelain dish should be deep and crucible shaped and should be placed in an oil-bath, the temperature of which is easily controlled. The contents of the dish are stirred with the thermometer bulb until all foaming ceases and the temperature of the mixture is then kept at from 175° to 180° for an hour. After the melt has cooled to 80° about seventy-five cubic centimeters of hot water are added to bring it into solution and it is then allowed to cool. The solution is acidified with sulfuric and placed in large centrifugal tubes. After being made slightly alkaline with soda lye, the tubes are subjected to continued energetic centrifugal action until the cellulose is separated. The supernatant liquid can be nearly all poured off and the separated cellulose is broken up, treated with hot water and again separated by centrifugal action. The cellulose is finally collected upon the asbestos felt, washed with hot water, alcohol and ether, dried and weighed. With a little practice it is possible to complete the separation of cellulose in two and one-half hours. =274. Solubility of Cellulose.=—Cellulose resembles starch in its general insolubility, but, unlike starch, it may be dissolved in some reagents and afterwards precipitated practically unchanged or in a state of hydration. One of the simplest solvents of cellulose is zinc chlorid in concentrated aqueous solution. The solution is accomplished with the aid of heat, adding one part by weight of cotton to six parts of zinc chlorid dissolved in ten parts of water. A homogeneous sirup is obtained by this process, which is used in the arts for making the carbon filaments of incandescent electric lamps. In preparing the thread of cellulose, the solution, obtained as described above, is allowed to flow, in a fine stream, into alcohol, whereby a cellulose hydrate is precipitated, which is freed from zinc hydroxid by digesting in hydrochloric acid. Hydrochloric acid may be substituted for water in preparing the reagent above noted, whereby a solvent is secured which acts upon cellulose readily in the cold.