358 Mr. Yngve Bjornstahl on Magnetic 



series A has been left standing a little longer than the 

 corresponding solution of series B. -It is known that colloidal 

 solutions often do not remain in complete stability: they 

 undergo a more or less rapid change of state — the sol 

 " grows old/'' The nuclear solution of A has had more time 

 to develop crystalline properties than B. 



The study of the phenomena of transition from the low 

 disperse systems to the molecular ones, the investigations 

 regarding the change of the properties of the colloids with 

 the degree of dispersity, have been of fundamental importance 

 for our understanding of solutions*. 



With regard to the magnetic double refraction, fig. 5 

 is a striking proof of the continuity when passing from low 

 disperse systems to high disperse ones. 



Coagulation experiments. — In order to study the influence 

 of electrolytes on the effect, I made some experiments with 

 the addition of sodium chloride. Even very small additions 

 caused a change in the effect. Between the limits of the 

 slow coagulation a quasi-stationary state seemed to be 

 attained. In the case of the rapid coagulation the conditions 

 are more definite. When the concentration of the sodium 

 chloride solution was c = 0'0167 normal, the progress of the 

 coagulation of a sol (2r= 70/-t/x), whose maximum of absorp- 

 tion was in the red, is illustrated by fig. 6 (PI. XIII.) 

 (Table VI.). In the beginning the effect decreases rapidly 

 with the time T, then more slowly. Another sol (2r*=40 /h/jl), 

 whose maximum of absorption was in the green, shows a 

 progress of coagulation represented by fig. 7 (PI. XIII.) 

 (Table VII.), when the sodium chloride concentration was 

 c = 0"0211 normal. The double refraction rises with the 

 time T; attains a maximum and then decreases slowly. A 

 quite evident difference of character seems to distinguish 

 the blue solutions from the red ones, the latter exhibiting 

 curves of coagulation (A = F(I)) with a maximum, while 

 the phase-difference of the low disperse sols decreases all the 

 time. I will return to this later on. With the aid of the 

 magnetic double refraction it is possible to record the 

 formation of the colloid. As an instance I adduce the form- 

 ation of gold colloid. Using a sol 2r=6±/jL/ju as nuclear 

 liquid, I prepared two solutions 2r=121 /jl/jl. the first with a 

 concentration r = 2'5 x 10 -4 normal, the other c = l*25 normal. 

 The effect during the formation is represented by fig. 8 

 (PI. XIII.) (Table VIII.). At first small, corresponding to 

 the rather insignificant quantity of nuclei, it rises rapidly. 



* Cf. The Svedberg, Die Existenz der Molekule l Leipzig, 1912. 



