On the Non- existence of Sugar in thi Blood. 21 



half to the ounce, tlie cryslallization is not prevented ; but 

 even this small quantity is perceptible by a degree of black- 

 ness that appears after evaporation ; occasioned, as I sup- 

 pose, by the action of a small excess of acid on the sugar. 



If ^i'lQ. grains have been added, the crystallization is very 

 imperfect, and soon disappears in a moist air by delitjues- 

 cence of the sugar. 7'he blackness is also deeper than in 

 the former case. 



By addition often grains to the ounce, ihe crystallization 

 of the salts is entirely prevented, and the degree of black- 

 ness and disposition to deliquesce are of course more mani- 

 fest than with smaller quantities. 



As I was aware that the sugar obtained from diabetic 

 urine is a different substance from common sugar (ap- 

 proaching more nearly to the suo;ar of iigs), I had the pre- 

 caution to repeat the same series of experiments upon se- 

 rum, to which I made corresponding additions of dry sugar, 

 that I had formerly extracted irom the lu'ine of a person 

 who voided it in ci)nsiderable quantity; and I found the ef- 

 iccts to be perfectly similar in every respect. 



As a further test of the absence or presence of sugar, I 

 found it convenient to add a little nitric acid to the salts 

 that remained alter crystallization of the drop. If the serum 

 has been successfully coagulated without any addition of 

 sugar, the addition of nitric acid merely converts the mu- 

 riatic salts into nitrates, and nitrate of soda is seen to cry- 

 stallize without foam or blackness. But when sugar has 

 been added, a white foam rises round the margin of the 

 drop; and if further heat be applied, it becomes black in 

 proportion to the quantity of sugar present. 



Such are the appearances when the proportions have been 

 duly adjusted, and the proper heat for coagulation applied. 

 I must own, however, that I could not always succeed to 

 my satisfaction at the time when these experiments were 

 conducted, and F am inclined to ascril)e occasional failures 

 to having used more muriatic acid than was really necessary, 

 which by excess of beat miolit redissolve a part of the co- 

 agulated albumen, and thence occasion appearances which, 

 without careful discrimination, might be ascribed to sugar. 



After having, bv this course of experiment, satisfied my- 

 self as to the phaeiionicna exhibited by serum in its natural 

 state, and the eflccts of any small additions of sugar, I then 

 proceeded to the examination of such specimens of diabetic 

 blood or of strum, as i was able to procure. 



The first whicli I examined was a portion of blood that 



had been taken froixi a person whose urine had been ana- 



B 3 lysedj 



