1914} Alfred P. Lothrop 305 



Into a weighing flask put 15 c.c. of the copper sol. from a 

 graduated buret, and weigh the flask together with the liquid. An 

 ear lobe of the patient is pricked and 15 drops of blood are allowed 

 to fall directly into the copper sol. in the weighing flask. In experi- 

 menting with rabbits it is quite easy to insert a needle of large 

 caliber into an ear vein and thus draw off 15-20 drops of blood, the 

 needle acting like a canula. The flask is now weighed a second 

 time, and the amount of blood taken f or analysis is thus determined. 



The blood and the copper sol. in the flask are brought to a boil 

 and filtered through a Gooch filter with suction. The coagulum is 

 washed until the washings are perfectly colorless, 10-15 c.c. of 

 water being sufiicient for such washing. The washings are united 

 with the filtrate and titrated with the hydroxylamine sol., which 

 is added slowly from a graduated buret until all trace of blue or 

 green disappears from the filtrate. Since i c.c. of the hydroxylamine 

 sol. exactly decolorizes i c.c. of the copper sol., the amount 01 cop- 

 per present in excess is thus determined, and the ditference betwen 

 the original amount of copper and the excess represents the copper 

 reduced by the sugar in the blood. 



With this method a number of determinations of sugar in the 

 blood were carried out. In most instances the results obtained by 

 the above process were higher than the figures given by previous 

 observers, presumably due to more accurate manipulation because 

 of the greater simplicity of the method. 



It was found that normal human blood contained from 0.14- 

 0.19 percent of sugar. The blood of a diabetic patient who had a 

 glycosuria of 2.7 percent contained 0.37 percent. Two nephritic 

 patients had 0.21 percent (chronic interstitial nephritis) and o.ii 

 percent (acute parenchymatous nephritis) of sugar in the blood. 

 In a very large number of analyses of the blood of rabbits, it was 

 found that the normal sugar content varied between o.ii and 0.18 

 percent. Feeding sugar to rabbits raised the sugar content of the 

 blood to 0.24 percent. Subcutaneous administration of caffeine 

 (150 mg. per kilo of body weight) induced marked hyperglycemia, 

 the highest figure observed being 0.385 percent. While zinc causes 

 a marked glycosuria (by intravenous administration of the malate), 

 it does not produce hyperglycemia; the sugar content of the blood 

 in these cases seems to be normal or subnormal. 



