80 THE ROYAL SOCIETY OF CANADA 



with 2 c.c. of glyoxylic acid reagent and 5 c.c. of concentrated sul- 

 phuric acid, mixed thoroughly, and allowed to stand for half-an-hour. 

 The colour developed in the globin solutions was then compared with 

 the series of standards. It was found that the globin solutions could 

 not in this way be distinguished the one from the other, and that 

 each gave a depth of colour about midway between that of the 0.02 

 and that of the 0.03% standard. This corresponds to a tryptophane 

 content for globin of between 2 and 3%. 



Since such a method as the above has no claim to accuracy we 

 applied also to our globin samples the quantitative colorimetric 

 method, based upon the Voisenet reaction, which has been proposed 

 by Fiirth and Nobel, ^ and by means of which it was that Furth and 

 Lieben reached their conclusion that globin yields no tryptophane at 

 all. The standard employed was a 0.1% solution of tryptophane. 

 The Voisenet reaction was developed, according to the technique of 

 Fiirth and Nobel, simultaneously in 2 c.c. of the standard and in 2 c.c. 

 of a 3% solution of globin. The colours produced were compared in 

 a Duboscq colorimeter. The three globin solutions gave absolutely 

 identical Vesults. With the standard set at 10 mm., each read 12.3 

 mm., corresponding to a tryptophane content (in the dry globin) 

 of 2.71 per cent. When, as a check upon the instrument, the globin 

 solution was set at 10 mm. and the standard itself adjusted to match, 

 the latter read in every instance 7.7 mm., indicating a tryptophane 

 content for the globin of 2.57 per cent. The average of the two 

 readings was therefore constant at 2.64 per cent. 



It is demonstrated, therefore, that repeated recrystallization of 

 oxyhemoglobin does not in the least affect its tryptophane content. 

 This can hardly be explained in any other way than by assuming 

 that tryptophane is an integral part of the hemoglobin molecule. The 

 proportion of tryptophane indicated is moreover quite considerable. 

 It is such as to suggest that the globin molecule contains two mole- 

 cules of tryptophane; for upon that assumption one may calculate 

 for globin a molecular weight of about 15,500, a figure in close agree- 

 ment with estimates attained by a great variety of independent 

 methods. 



It is not possible at present to explain the discrepancy between 

 our conclusion and that of Fiirth and Lieben. The only suggestion 

 we have to offer is that these observers applied the Voisenet test to 

 solutions of globin that were too dilute. They do not, for this parti- 

 cular protein, give any data upon which the concentration of their 



2Fùrth and Lieben: BioMhem. Zeitschr., cix, p. 124 (1920). 



