THE CARBOHYDRATES. 299 



Isolation. The collected urine of twenty-four hours is precipi- 

 tated with lead subacetate and filtered. After washing with water 

 the filtrate and washings are mixed and treated with ammonia. 

 The resulting precipitate is filtered oif and the filtrate again pre- 

 cipitated with lead subacetate and ammonia, and so on until the 

 final filtrate is optically inactive. The precipitates, with the excep- 

 tion of the first, are then mixed, washed with water, decomposed 

 with hydrogen sulphide, and filtered. In the filtrate the excess of 

 hydrogen sulphide is removed by a current of air, and freed from 

 any acids that have been liberated by shaking with argentic oxide. 

 The mixture is filtered, freed from soluble silver with hydrogen sul- 

 phide, treated with barium carbonate, and concentrated to a small 

 volume ; 90 per cent, alcohol is then added, which causes the 

 formation of a flocculent precipitate. This is filtered oif. The 

 filtrate is placed in the desiccator, when on standing crystals of 

 lactose gradually separate out. These may be purified by recrys- 

 tallization, decolorization with animal charcoal, and extraction with 

 60-70 per cent, alcohol. 



Lsevulose. The occurrence of a laevorotatory sugar has been at 

 times, though rarely, observed in the urine of diabetic patients, 

 where it was present either alone or in association with glucose. 

 Like dextrose, the substance reduced Fehling's and Nylander's solu- 

 tion, and formed an osazon with phenylhydrazin, with a melting- 

 point of 205 C. It was fermentable, but, unlike true lasvulose, 

 could be precipitated with basic lead acetate. 



Leo's laiose, which was also obtained from a diabetic urine on 

 one occasion, was very similar to the body just described, but, unlike 

 this, could not be fermented. With phenylhydrazin, moreover, it 

 formed a yellowish-brown non-crystallizable oil. 



Of the nature of these bodies nothing further is known. 



The presence of a Isevorotatory sugar can, of course, readily be 

 established by the common tests, supplemented by a polarimetric 

 examination, if it is present alone. If glucose, however, also is con- 

 tained in the urine in amounts sufficient to counteract the Isevorota- 

 tion, the matter is more difficult. In such an event, however, it will 

 be observed that higher values are obtained in estimating the sugar 

 by titration, or according to the differential density method, than 

 with the polarimeter, for reasons which are self-evident. 



According to Neuberg and Strauss, the presence of laBvulose can 

 be definitely established in the urine, the blood-serum, ascitic fluid, 

 or in pleural effusions by its transformation into the corresponding 

 methyl-phenyl-osazon, which can be obtained in crystalline form. 1 



Lsevulose also gives SeliwanofFs reaction, viz., it gives rise to the 

 formation of a red pigment on boiling with resorcin and hydro- 

 chloric acid, which is soluble in alcohol. The reaction in question 

 is common to all the ketoses of the hexose series (fructose and 



l The method is described in my Clinical Diagnosis, 6th edition. 



