AND DIABETES 171 



ones, because, as in this instance, a trace may be 

 found normally. The presence of diacetic acid is of 

 more clinical importance. 



Tests for acetone in the urine : — To 3 c.c. of urine 

 add a few drops of fresh sodium nitroprusside (a 

 crystal in 5 c.c. of water). Cover with strong ammonia. 

 A magenta ring appears at the hne of junction, and 

 spreads upwards (Jackson-Taylor). Or, to 5 c.c. of 

 urine add \ c.c. of 5 per cent sodium nitroprusside ; 

 make just alkaline with caustic soda, and acidify 

 with acetic acid. A reddish-violet colour develops. 

 Diacetic acid also gives this test. 



Acetone is excreted in the breath as well as in the 

 urine, and the sweet odour is perfectly apparent to 

 many medical men, more so to some than others ; 

 some can smell a diabetic excreting acetone at a great 

 distance. It is remarkable how the flies may con- 

 gregate about a diabetic in a ward. 



Test for diacetic acid in the urine : — To 3 c.c. of 

 urine add a few drops of Hq. ferri perchlor. A 

 deep red colour which disappears on heating is 

 positive. The test is often performed on a white 

 slab as a contact test. One must not be deceived 

 by the frequent reddish precipitate of iron phosphate 

 from a normal urine. 



Turning now to the prevention and treatment of 

 acidosis, we may clear the ground by reserving 

 diabetic coma for a special word later on in the 

 chapter, and delayed chloroform poisoning for con- 

 sideration in Chapter XL 



We saw that the cause of this peculiar perversion 

 of metabohsm is inabihty on the part of the tissues 



