On Kreatinins, 



497 



at a glance. In this table all the reductions are expressed in terms 

 of glucose. 



Summing up all the evidence, there is no doubt that a reducing 

 agent is present as a normal constituent of the urine of healthy men, 

 and that it confers upon normal urine the property of reducing cupric 

 oxide to the same extent as if it held in solution (on the average) 

 6 grains of glucose in every 10 fluid ounces of urine, or 1*34 grams 

 per litre. But, considering the impossibility of causing alcoholic fer- 

 mentation to take place in solutions of this reducing agent, and its 

 property of reducing picric acid to some extent in presence of 

 potassium hydrate at the ordinary temperature, its identity with 

 glucose appears to be very doubtful. Further doubt is thrown upon 

 this identity by the fact that if a solution of mercuric chloride be added 

 to normal urine and afterwards potassium hydrate, the yellowish pre- 

 cipitate which forms, becomes grey in a few minutes by reduction at 

 the ordinary temperature, whereas glucose does not effect reduction 

 of mercuric oxide in presence of potassium hydrate without applica- 

 tion of heat in less than one hour. 



Endeavours to Isolate the Reducing Agent of Normal Urine. 



Having examined qualitatively the reactions of the reducing agent 

 of normal urine, and having found reasons to doubt its identity with 

 diabetic sugar, I next proceeded to employ various precipitants with 

 a view to its removal from the complex fluid in which it is dissolved. 



Normal lead acetate, basic lead acetate, baryta-water, solution of 

 ammonia, alcohol, were all employed in turn, but the filtrates in each 

 case reduced picric acid as strongly as before precipitation, showing 

 that the precipitants employed had failed to remove the reducing 

 agent from solution. 



Finally 1 succeeded in removing the whole of the normal reducing 

 agent from urine by complete precipitation with strong aqueous solu- 

 tion of mercuric chloride. 



Previous Researches on the Action of Mercuric Chloride upon Normal 



Human Urine. 



It has long been known that an aqueous solution of mercuric 

 chloride produces in all specimens of normal urine a flocculent pre- 

 cipitate, when added in sufficient excess. But the state of our 

 knowledge as to the nature of the substance or substances thus 

 precipitated is very unsatisfactory. 



Dr. John Greene, of Birmingham, published in the 'British 

 Medical Journal ' (May 10th, 1879) a research in which he describes 

 the isolation from this precipitate, produced by mercuric chloride in 

 normal urine, of a white flocculent albumen-like substance, precipi- 

 table by lead acetate as well as by mercuric chloride, and refusing to 



