INTRODUCTION. 9 



carries a thermometer. The thermometer should not have a greater error at 

 the boiling-point of water than 0.5 C. when the barometric pressure is taken 

 into consideration. The lower end of the thermometer should be placed at a 

 distance of about 2 mm. from the bottom of the flask and the region of the stem 

 above 35 C. should extend above the upper surface of the stopper so as to be 

 plainly visible. This flask should be placed in the center of a plate of unper- 

 forated asbestos gauze carried by a ring-stand so that it can be raised or low- 

 ered. The distance between the burner and the asbestos should always fall 

 within the limits of 3 to 5 cm., thus avoiding a long flame which would be easily 

 affected by air-currents. Room-temperature and freedom from drafts should 

 prevail. If the latter condition can not be provided the heating apparatus 

 may be wholly or partially protected by a cylindrical hood of sheet metal or 

 asbestos. The flask as above described containing 60 c.c. of distilled water 

 is placed on the asbestos and the time observed that is required for the mer- 

 cury to rise from 35 to 95 C. The rough adjustment to 120 seconds is made 

 by regulating the flow of gas so as to change the power of the flame with the 

 position of the asbestos constant at 4 cm. The finer adjustment is made by 

 moving the asbestos plate toward or away from the burner within the limits of 

 3 to 5 cm. The conditions are not properly adjusted until repeated determina- 

 tions of the time give constant results within limits of error of only a few 

 seconds. At 120 seconds the standard heating power here adopted has been 

 established. 



The Process of Reduction. 



The reduction mixtures should be placed in the same kind of flasks, with 

 the same stopper, thermometer, etc., and with the same conditions of heating 

 power as were used for the standardization of the latter. The volume of the 

 reduction mixture is always 60 c.c. It consists of 20 c.c. of Allihn's alkaline 

 tartrate solution plus 20 c.c. of Fehling's copper sulphate solution plus 20 c.c. 

 of distilled water for a blank determination, or, in the case of sugar analysis, of 

 an aqueous sugar solution with enough distilled water added to make a final 

 volume of 60 c.c. 



The Fehling-Soxhlet solution contains 34.639 grams of pure crystallized 

 copper sulphate in a volume of 500 c.c, but the purity of the copper sulphate 

 is more important than its exact weight. Since its correct copper content can 

 not with reliability be deduced from the formula CuS0 4 .5H 2 0, this value is 

 found by means of standardized thiosulphate as described on page 11 for the 

 "basis titration." The Allihn alkaline tartrate solution contains 173 grams of 

 sodium-potassium tartrate (Rochelle salt) and 125 grams of potassium hydrox- 

 ide in a volume of 500 c.c. 



For a determination of sugar the flask containing the reduction mixture is 

 placed on the heating apparatus and the time is observed, but in this case the 

 reckoning begins only when the thermometer reaches 95 C. The flask should 

 stand undisturbed in the center of the asbestos gauze until the expiration of 20 

 seconds after the temperature of 95 C. has been reached. It is then promptly 

 removed, with the stopper and thermometer still in place, and twirled for a few 



