THE NATURE OF HEMOGLOBIN SOLUTION 79 



enough in the osmometer an equilibrium will never be established and 

 the final osmotic pressure will never be obtained. If, on the other 

 hand, you take sufficient time for the attainment of a final equilibrium 

 your solution may "go bad" and you may have quite a different 

 number of molecules from that with which you started. 



One way of meeting such difficulties — evidently not very satis- 

 factory — is frankly to give up the idea of obtaining equilibrium, to set a 

 time limit to the experiment, to take a series of readings of the osmotic 

 pressure at suitable intervals, and to trust to extrapolation for the 

 final result. Something, again, can be done to accelerate the attain- 

 ment of a final pressure in the osmometer by resort to the expedient 

 of prophecy. An initial pressure, adjudged about equal to what the 

 ultimate pressure will be, is established artificially by making the 

 osmometer work against a mercury column. This process is not 

 satisfactory either, because after all it does not hasten the processes 

 which are going to make for finality. 



Naturally the larger the surface relative to the volume of fluid 

 in the osmometer, the more rapidly will finaUty be reached, and of 

 course the more completely the fluids are stirred the better. When I 

 think of the dialyser which I used in order to obtain salt-free haemo- 

 globin (3) — a parchment sausage about 2 feet long and perhaps 

 3 or 4 inches in circumference placed in a jar containing half 

 a score of litres of distilled water, and agitated by being pulled up 

 and down by an engine — and when I compare that apparatus with the 

 beautifully simple and efficient osmometers of Adair (8), with collodion 

 membranes of texture so exquisite as just, and only just, to retain 

 the haemoglobin molecule, yet capable of standing high pressures, 

 membranes possessing a sufficient relative surface to allow of the 

 maximum rate of dialysis, then I reahse how far Adair has gone — I 

 appreciate a little of what lies behind the apparently naive state- 

 ment in one of his papers : "After a few years' practice the proportion 

 of failures was less than 10 per cent." (8) A few years' practice ! And 

 it was not that the first membrane which Adair made was a bad one. 

 I saw him make it. A biochemist, for whom I have a great respect, 

 came to our laboratory; he had had much experience as a manu- 

 facturer of membranes but he was making "heavy weather" of it; his 

 wife, for whom I also have a great respect, came along and "went 

 one better." Adair watched them both: at long last and in silence 

 he took some of the collodion, set a rod spinning, poured on the 

 fluid, and lo ! a membrane of no mean order. 



