158 IV. BILE PIGMENTS 



which gives a dye with different absorption bands, and is also easily dis- 

 tinguished from urobilinogen by its inability to pass into organic solvents. 

 The simple qualitative Ehrlich color test for urobilinogen in urine can, how- 

 ever, give misleading results, mainly due to indole {2010). Nitrite in urine 

 (0.2 milligram per cent) also interferes with the test {216). A green reaction 

 occasionally observed in such urines is due to the oxidation by nitrous acid 

 of bilirubin present in the urine, producing biliverdin. 



Watson's method of estimation, referred to above, requires a 48-hour 

 specimen of feces and a 24-hour specimen of urine; it is laborious and time 

 consuming. It is superior, however, to other simplified modifications {2914, 

 2598, cf. 2617) and to estimations on single samples of feces or urine {2598, 

 cf. also 2992,3067). It should be used for investigational purposes, but 

 whether it would give greater information for most clinical purposes than 

 repeated semiquantitative tests, is difficult to assess. Thus Watson himself 

 has recently {3003) suggested a simplified, though less exact, procedure for 

 single samples of urine or feces. 



Analysis of mixtures of mesohilene-{b) and tetrahydromesobilene-{b) . Fol- 

 lowing Lemberg's clarification of the relation between these two compounds, 

 Lemberg. Lockwood, and Wyndham {1713) investigated the relative amounts 

 of the two in a large number of normal and pathological urines. Their method 

 was only semiquantitative; it involved the determination of the position of 

 the main absorption band of the hydrochlorides in the Hartridge Reversion 

 Spectroscope, allowing a rough estimation of the relative amounts of each 

 from the position of the band. Since the difference in band position between 

 the two hydrochlorides is only 20 A, the method is not very exact. 



Legge {1667) developed a quantitative modification of the ferric chloride 

 oxidation method. Ferric chloride does not oxidize tetrahydromesobilene, 

 but transforms mesobilene into a mixture of mesobiliverdin and raeso- 

 bilipurpurin. By spectrophotometric measurement of the absorption of the 

 hydrochloride mixture before oxidation at 492 m/i, and after oxidation at 

 this and two other suitable wavelengths, the amounts of tetrahydromeso- 

 bilene and mesobilene in the mixture can be determined. 



