1 1 2 Society for Experimental Biology and Medicine. 



band into the neck of the flask-shaped membrane. The latter, after 

 introduction of the colloidal solution, is immersed in a definite 

 volume of the pure solvent (water, or water plus electrolyte used) 

 contained in a battery jar ; the jar is covered by a glass plate to 

 prevent evaporation. The manometer tube is clamped in a ver- 

 tical position. The height to which the column of fluid rises is a 

 measure of the osmotic pressure ; a constant height is reached in 

 eighteen hours or less ; pressure readings thus obtained may easily 

 be translated into millimeters of mercury, if the specific gravity of 

 the solution within the membrane is known. 



The following general results have been gained. Non-electro- 

 lytes (sucrose, dextrose, glycerin, urea) have little or no influence 

 on the osmotic pressure of the above colloids. Electrolytes, on 

 the other hand, invariably produce a marked alteration. For 

 example, the osmotic pressure of gelatin is greatly increased by 

 the addition of small quantities of either acid or alkali. Thus in 

 one experiment a 1.5 per cent, solution of gelatin gave a pressure 

 of 8.4 mm. Hg ; the same solution with the addition of HC1 to 

 w/410 concentration gave a pressure of 41. 1 mm. Hg ; with 71/410 

 KOH it gave 26.3 mm. Hg. Egg albumin differs from gelatin in 

 showing a depression of osmotic pressure in presence of acid or 

 alkali. In all cases neutral salts depress osmotic pressure ; in 

 general there is seen a parallelism between the effectiveness of the 

 salts as precipitants and their action in lowering osmotic pressure. 

 The action is less pronounced — for equimolecular concentrations 

 — with alkali metal salts than with salts of alkali earths ; heavy metal 

 salts depress to a still greater degree. A typical series with 1.5 

 per cent, egg albumin and the following potassium salts gave this 

 result: (1) control: 22.6 mm. Hg ; (2) same solution -f- 7^/24 

 KC1 : 4.6 mm.; (3) w/24 KBr : 5.0 mm.; (4) KI : 5.4 mm. ; 

 (5) KNO, : 57 mm.; (6) KCNS : 6.0 mm.; (7) K 2 S0 4 : 4.0 

 mm. The action thus varies with the nature of the anion ; in 

 general the order of decreasing effectiveness of anions is some- 

 what as follows : SO, < CI < N0 3 < Br < I < CNS. This order 

 coincides with that found by Hofmeister and Pauli for the action 

 of anions in changing the aggregation-state of protcid solutions. 

 M;g6 CaCl 2 in a typical experiment depressed the osmotic pres- 

 sure of a 1.5 percent, albumin solution from 18.8 mm. Hg (control) 



