710 PHYSICAL CHEMISTRY OF THE PROTEINS. I 



Our practise has been to pipette an aliquot part of a preparation 

 into a Pyrex volumetric flask of from 25 cc. to 100 cc. capacity. When 

 this was at first attempted it was found that the error due to sampling 

 the suspension of a flocculent protein precipitate was frequently as 



Fig. 1. 



great as 3, or even 5 per cent. Accordingly an apparatus was devised 

 by means of which the protein precipitate could be maintained in a 

 state of fine subdivision during pipetting (Fig. 1). 



This was accomplished by placing the bottle (A) containing the protein in a 

 carriage (B) hung, as is a compass, in concentric rings (C) and {D). As a result, 



