igii] Ernest D. Clark 201 



Each of the dialysates from the soluble starch Solutions prepared 

 at both temperatures was then saturated with magnesium sulfate. 

 A precipitate was produced in each instance. It was filtered off, 

 washed with a saturated Solution of magnesium sulphate, redis- 

 solved in pure water and treated with dry sodium carbonate in 

 quantities just sufficient to precipitate the magnesium as carbonate. 

 This precipitate was removed by filtration. The filtrate yielded a 

 deep hlue color with iodine Solution and slightly reduced Fehling- 

 Benedict Solution. 



The filtrates from the carbohydrate precipitated by magnesium 

 sulfate were treated with sodium carbonate for the precipitation 

 of the magnesium as carbonate. After removal of the latter by 

 filtration, each filtrate yielded a red color with iodine Solution and 

 reduced Fehling-Benedict Solution. Saturation of these filtrates 

 with ammonium sulphate failed, in each case, to produce further 

 precipitation. 



These experiments with typical diffusates involving fractional 

 precipitation with magnesium sulfate were repeated several times 

 and all the results confirmed. This process has shown that Lintner 

 soluble starch, when dissolved in water and subjected to dialysis, 

 yields to the diffusate a mixture of substances including one or 

 more erythrodextrins. 



Diffusion residues. The Solutions remaining in the bags at 

 the conclusion of dialysis were tested in each instance with iodine 

 Solution and also with Fehling Solution. The former reagent pro- 

 duced the same colorations, respectively, as those indicated on 

 page 200, while the Fehling Solution was reduced in each case. 



5. Fractional precipitation with alcohol. Alcoholic pre- 

 ciPiTATES FROM DIFFUSION RESIDUES. It was obviously desir- 

 able to subject the Solutions in the bags to further investigation, 

 Each of these Solutions was poured into 400 c.c. of 95 per cent. 

 alcohol and, after thoroughly shaking them, the mixtures were 

 allowed to stand undisturbed for a month. Each fluid was, from 

 the beginning, an opalescent colloidal Solution which did not show 

 any tendency toward agglutination even during this long period. li 

 was found, however, that a single drop of 10 per cent. sodium Chlo- 

 ride Solution sufficed, in each case, to induce instantaneous precipita- 



