320 A CENTURY OF SCIENCE 



For example a molecule of sodium chloride was supposed 

 to give the two ions Na+ and Cl~, thus exerting twice as 

 much osmotic pressure as a single molecule. 



Determinations of osmotic pressure or related values, 

 such as depression of the freezing point and of electric 

 conductivity, indicated that ionization could not be 

 regarded as complete in any case except in exceedingly 

 dilute solutions, and that the extent of ionization varied 

 with different substances. The fact that osmotic pres- 

 sures and electric conductivities gave closely agreeing 

 results in regard to the extent of ionization in various 

 cases, is the strongest evidence in support of the theory. 



It was difficult at first for many chemists to believe 

 that atoms, such as those of sodium and chlorine, and 

 groups such as NH 4 and S0 4 could exist independently 

 in solution, even though electrically charged. However, 

 the theory rapidly gained ground and is now accepted 

 by nearly every chemist as a satisfactory explanation of 

 many facts. 



During recent years, many investigations relating to 

 osmotic pressure and ionization have been carried out in 

 the United States, but only the work of Morse, A. A. 

 Noyes, and the late H. C. Jones can be merely alluded to 

 here. It should be mentioned that the eminent author 

 of the ionic hypothesis gave the Silliman Memorial course 

 of lectures at Yale in 1911 on Theories of Solution. 



Colloidal Solutions. Graham, an English chemist, in 

 1861 was the first to make a distinction between sub- 

 stances forming true solutions, which he called crystal- 

 loids, and those of a gummy nature resembling glue, 

 which in solution do not diffuse readily through parch- 

 ment membranes, as crystalloids do, and which he called 

 colloids. The separation of colloids by means of parch- 

 ment was called dialysis, and this process has come into 

 extensive use in preparing pure colloidal solutions. 

 Slow diffusion is now regarded as characteristic of col- 

 loids rather than their gummy condition. 



Colloidal solutions occupy an intermediate position 

 between true solutions and suspensions, resembling one 

 or the other according to the kind of colloi(J and the fine- 

 ness of division. By preparing filters with pores of 

 varying degrees of fineness, Bechold has been able to 



