282 PHYSIOLOGICAL CHEMISTRY 



To secure accurate results the method of Bang must be rigidly con- 

 trolled, all new solutions and absorbent papers being checked up 

 against pure 0.2 per cent glucose solutions. Taylor and Hulton 1 

 also suggest the following precautions. A blank check must be made 

 on the reagents each day an estimation is made. 0.10-0.15 gram of 

 blood should be taken and must spread smoothly on the paper. The 

 proteins are best coagulated by heating of the blood-impregnated 

 papers in the hot air oven at 100 (as recommended by Gardner and 

 McLean) 2 for five minutes with corks of flasks inverted. The solu- 

 tion should be boiled four minutes for complete reduction. The 

 iodine solution must be fresh each day and checked each day. Deter- 

 minations should be made in triplicate. Results cannot be depended 

 upon to be more accurate than to 0.005 gram glucose in 100 c.c. blood. 

 Other authors have recommended that an hour instead of half an hour 

 be allowed for the diffusion of the blood sugar, the fluid being brought 

 to the boiling-point twice during this period or kept in a bath at 40C. 



Method of Epstein. 3 Principle. This method is a modification of the Lewis 

 and Benedict procedure, being based on the same principle but making possible the 

 determination of reducing sugar in finger blood (0.1-0.2 c.c.) with a sufficient 

 degree of accuracy for clinical purposes, and with little expenditure of time. In- 

 stead of a Duboscq colorimeter the less expensive Sahli-Gower hemoglobin color- 

 imeter is recommended. 



Procedure. The apparatus 4 shown in the illustration (Fig. 82) and the fol- 

 lowing reagents are necessary: 



1. Picric acid, saturated solution. 



2. Sodium carbonate, 10 per cent solution. 



3. Sodium fluorid or potassium oxalate, 2 per cent solution. 



Put one or two drops of the fluorid or oxalate solution into the graduated test- 

 tube (see illustration). By means of the blood pipette, 3, 0.2 c.c. of blood is 

 obtained from the tip of the finger or the lobe of the ear and is discharged into the 

 tube containing the fluorid solution. The pipette is rinsed two or three times 

 with distilled water and the washings added to the blood in the tube. Distilled 

 water is then added to the i.o c.c. mark. After laking of the blood has taken place, 

 picric acid is added to this (a few drops at a time) up to the 2.5 c.c. mark, shaking 

 the tube gently with each addition of the acid. Precipitation of the blood-proteins 

 takes place; the sugar, together with an excess of picric acid sufficient for the 



1 Taylor and Hulton: Jour. BioL Chem., 22, 63, 1915. 



2 Gardner and McLean: Biochem. J., 8, 391, 1914. 



3 Epstein: /. Am. Med. Assn., 63, 1667, 1914. 



4 The tubes belonging to this hemoglobinometer are not all equally calibrated. With 

 some the 50 per cent mark represents a volume of i.o c.c.; with others, i.o c.c. of fluid 

 reaches up to the 43, 45, 46 or 47 per cent mark. The error in the calibration is generally 

 below the 10 per cent mark; the graduations above this mark are usually correct. By 

 means of the standard i.o c.c. pipette one can readily determine whether or not a given 

 tube is properly calibrated. In order to facilitate a direct reading of the percentage 

 of sugar on these hemoglobinometer tubes, it is essential to have i.o c.c. of fluid stand at 

 mark 50. To overcome a discrepancy (if any exists) in the calibration of a given tube, 

 one may put one, two or three small glass beads in the bottom of the tube, of such size 

 asjio raise the meniscus of i.o c.c. of fluid up to the 50 per cent mark. 



