10 METABOLISM IN SEVERE DIABETES. 



seconds under the tap in order to cool the mixture rapidly, but only to a few 

 degrees under the boiling-point, since the mixture is to be filtered hot without 

 delay. 



A Gooch crucible should have been previously prepared with a very heavy 

 filter, so that suction will draw water through it only slowly. Avery close and 

 satisfactory filter may be made by placing a layer of pure talcum powder 

 between two layers of asbestos in the Gooch crucible. A good filter may also 

 be made by substituting for the crucible a calcium-chloride drying-tube car- 

 ried by a one-hole rubber stopper. This is packed first with glass-wool and 

 upon this is made, in the bulb of the tube, a very heavy mat of asbestos, which 

 may also be made as close as desired by means of talcum. The same filter is 

 used many times, the precipitate being removed from it by nitric acid (original 

 specific gravity of 1.42) which has been diluted with, one-half of its volume of 

 distilled water. This requires less time than making a new filter for each deter- 

 mination. The suction-flask should have a capacity of about 200 c.c, so that 

 a titration on the filtrate can be performed in the suction-flask, thus avoiding a 

 transfer of the liquid. Not more than 15 to 20 c.c. of wash- water should be 

 used, as it all remains as part of the final volume at the end of titration. This 

 volume should not exceed 120 c.c, to which volume the concentration of the 

 reagents used for the determination of copper has been adjusted. For the 

 inexperienced it is recommended to make a rough measurement of the final 

 volume of the liquid obtained at the end of titration. 



The Determination of Reduced Copper. 



The amount of copper that has been reduced under the above conditions is 

 a quantitative index of the amount of sugar originally present. The reduced 

 copper may be determined directly upon the precipitated cuprous oxide or 

 residually by the determination of the amount of copper still remaining un- 

 reduced in the alkaline filtrate. The amount of copper in the filtrate may be 

 determined either electrolytically for control or volumetrically by the iodide 

 method. For rapidity combined with accuracy the iodide method is recom- 

 mended, but only when it has first been controlled for the conditions under 

 which it is to be used. For the alkaline tartrate copper solutions which are 

 here used (i. e., Allihn's alkaline tartrate, etc.) this control has been satisfac- 

 torily made. The residual method for the determination of the reduced copper 

 may therefore be regarded as certainly the most rapid as well as accurate when 

 the iodide method is applied as described in the previous references and under 

 the following provisions : 



(1) The thiosulphate solution should have a value between 12.5 and 13 

 mg. of copper per cubic centimeter. It should be standardized preferably 

 against the same stock solution of copper which is to be used in the sugar 

 analyses, and the true copper value of this solution should be determined elec- 

 trolytically. If , however, the stock solution is made of chemically pure copper 

 sulphate there will be no appreciable difference between the standard value of 

 the thiosulphate determined by this electrolytically controlled solution, and the 

 value obtained by the nitric-acid-talcum procedure above referred to, using 



