310 



THE CHEMISTRY OF BLOOD, I., 



washed in the centrifuge with normal saline 09°/ NaCl, until 

 the washings gave no precipitate with 2)otassium ferrocyanide 

 and acetic acid, or with nitric acid. The coi-puscles were then 

 dissolved in distilled water, the contents of each tube being 

 made up to 100, and the solutions filtered. Of these filtrates, 

 50cc., therefore, represent the haemoglobin from 25cc. of defibii- 

 nated ox-blood. Four lots of lOOcc. of the filtrate were taken 

 and precipitated as below. These experiments were carried out 

 during the progress of other work, and, owing to the fact that 

 about two months were taken up in the drying of the globin 

 precipitates, a somewhat unique preparation was obtained. The 

 details are as follows. Of the haemoglobin sulution, lOOcc. were 

 poured gradually into 1600cc. of a mixture of alcohol, ether, and 

 water in the proportions 5: 2: 1, the .solution having been pre- 

 viously acidified with 2cc. of sulphuric acid. The flocculent 

 precipitate was allowed to settle. The dark-coloured solution 

 was then removed by suction through cloth stretched over the 

 broac' end of a thistle funnel. This method is rapid, and permits 

 of the precipitate being repeatedly washed without incurring any 

 loss, as would happen ijsing filter paper. The washing was con- 

 tinued until the fluid was no longer coloured. The precipitates 

 were then transferred to weighed centrifuge-tubes, and the wash- 

 ing finished in the centrifuge with absolute alcohol, and finally 

 with ether. The precipitates when centrifuged, occupied about 

 70cc. in volume. The tubes were now placed in the incubator at 

 37°C., where they remained for six weeks. In order to assist in 

 the drying, ^'essels containing quick lime were also placed in the 

 warm chamber. At the end of this time, the tubes were trans- 

 ferred to an oven at 50°C., and remained there for two more 

 weeks. The tubes were now of constant weight. The material 

 occupied about 20cc., and was of the consistency of partially 

 baked porcelain. The preparation certainly appeared difl^'erent 

 from the material previously prepared in the laboratory, and the 

 conclusion seemed justifiable that we were dealing with pure 

 anhydrous globin, or, as it was subsequently shown to be, globin 

 sulphate. It is a difficult matter to remo\ e all the water from 

 the globin prepared in the ordinary way and dried quickly at 



