21G REPORTS ON THE STATE OF SCIENCE. 



Rodewald ' has done this in the case of starch, and ha? calculated irom 

 the results its internal pressure, which was found to vary from 2073 

 kilos, per sq. cm. in dry starch to 561 kilos, in starch fully saturated with 

 aqueous vapour. He also calculated from its vapour- pressure its mole- 

 cular weight in the solid form as 4370, closely corresponding to twenty - 

 seven times the empirical formula CgHidO.^. [Phis is interesting as 

 derived in a different way from the molecular weight determinations in 

 liquid^. The method is obviously applicable to other solid colloids. 



The pressures exerted in the early stages of colloid swelling are very 

 large, though they become almost inappreciable as the maximum is 

 approached. The ancient use of dry wood wedges subsequently moistened 

 is an instance of this, and it may be mentioned that stones of the 

 trilithon at Baalbec in Syria, weighing over 1,000 tons, have obviously 

 been split in this way. The effect has usually been ascribed to capillary 

 contraction, but a little consideration will show that capillarity cannot 

 produce expansive effects. A pile of thin plates (e.g., cover-glasses on the 

 microscope stage) will be compressed and not expanded if water is allowed 

 to be drawn between them by capilhirity. The effect is necessarily mole- 

 cular and osmotic, the liquid getting vytthin the sphere of molecular 

 attractions, and so sustaining and liberating for expansion a portion of the 

 enormous internal pressures which have just been mentioned (in the case 

 of'starch some 1,500 atmospheres). 



The point of maximum swelling, v/hether in liquid or vapour, is 

 obviously a definite equilibrium which is reached when the attraction of 

 the colloid for the liquid or vapour is balanced by the internal attractions 

 of the colloid and the liquid in themselves. The liquid and the colloid 

 are in complete osmotic equilibrium, and consequently, according to a 

 well-known law, are both in equilibrium with the vapour. P. von 

 Schroeder ^ draws attention to an apparent exception to this law, which 

 is of considerable importance, and for which he was unable to offer a 

 satisfactory explanation. He found that agar and gelatine jellies swelled 

 to a considerably larger extent when immei-sed in water than they did in 

 saturated aqueous vapour, and that when the jelly, saturated by im- 

 mei'sion, was suspended in saturated vapour it dried till it was again 

 in equilibrium with the vapour ; thus evidencing a greater vapour- 

 pressure than the liquid water with which it was in osmotic equilibrium. 

 He found, however, that when the jelly^ was swollen with an N/ 100000 

 solution of a sulphate instead of water, no contraction, but slight further 

 swelling, occurred ; while an N/ 1000000 solution diminished but did not 

 prevent the evaporation. As the N/IO"' solution may be assumed to be 

 completely ionised it will contain three gram-ions in 2 x 10* c.c, and at 

 0' C. will exert an osmotic pressure of 340 dynes per cm.-. This would 

 involve work of 340 ergs per gram of water removed, and would cor- 

 i-espond to raising a gram of water only 0-344 cm. in height, but Von 

 Schroeder took special precautions to avoid difference in level. If, 

 however, a portion of water is removed from the mass, the surface is 

 extended and work is done (on the assumption of spherical form) of 

 370 ergs per gram. The .surface-tension is less, and the osmotic pressure 

 greater, at laboratory temperature (about 365 dynes at 20°), so that the 



» E. f. Phys. Ch., 1897. 24, 193 ; and UnUrsucMngen iibfr Hie Quellung der 

 Starke, Lepsius and Fischer, Leipzig, 1896, 

 » Ibid., 1903,^5, 75-U7. 



