A METHOD FOR THE DETERMINATION OF SUGAR IN NORMAL URINE

Abstract

Certain observations in this laboratory convinced the present writers some time ago that a quantitative study of the substances simulating reducing carbohydrate in the urine of normal animals would be of value, and the endeavor was made to construct a satisfactory quantitative method for this purpose.The procedure described below is the outcome of nearly 3 years of work on the subject and has been tested under widely varying conditions.We have not attempted to differentiate glucose qualitatively, but have provided for determination of "total sugar," and the sugar fermentable by yeast, by difference.The non-fermentable Ix&on, as determined in our method, is probably a true sugar, or is at lcast chiefly derived from carbohydrate in metabolism.These questions will be referred to again in subsequent papers.The procedure we have adopted is an adaptation of the colorimetric procedure of Lewis and Benedict for determination of sugar in the blood.Myers' has already reported an adaptation of the Lewis-Benedict method to urine, which provided for prcliminary precipitation of most of the creatininc as picrate according to the suggestion of Folin.Our work with the Myers' modification has convinced us that results by this method are too high (due partly to imperfect removal of creatinine).Indeed not infrequently the Myers method gives results at least, 100 per cent too great.This is not surprising when we remember that nitrogenous compounds other than creatinine may react with picric acid in alkaline solution (allantoin, for instance:, ,