208 itEroRTS ox the state of science:. 



probably a pr&ponderatiug influence. Organic sols which have received 

 definite charges usually behave with electrolytes like the inorganic. 



Spring has pointed out that ' coagulation ' (flocculation and precipi- 

 tation) takes place in two stages, though both may proceed simulta- 

 neously. In the first, the particles flocculate or adhere in larger 

 aggregates ; and in the second, these precipitate or settle under the 

 influence of gravity. The rapidity of this second stage is much in- 

 fluenced by the size and density of the aggregates and the viscosity of the 

 liquid, while the first stage is more clearly marked by the beginning of 

 turbidity or opalescence, and, though not usually instantaneous, is better 

 suited to observation. 



Numerous expei-iments have shown that in order to produce floc- 

 culation, the concentration of the electrolyte must reach a certain 

 minimum value ('Schwellenwert'), which varies with the electrolyte and 

 the sol from a mere trace to a considerable amount. 



Below this value, lapse of time produces no change, but, once it is 

 passed, the sol is rapidly or slowly precipitated. In each case that ion of 

 the electrolyte which has the opposite charge to the particles of the sol, 

 has the determining eflect, which increases rapidly with its valency. It 

 has been stated that with divalent and trivalent ions it is proportional to 

 the square and cube of that of monovalent, and Wetham ' has shown 

 that such a relation can be accounted for on the theory of probabilities 

 with certain assumptions as to the electric charge required for coagu- 

 lation ; but the agreement of the observed numbers with each other 

 and with those required by theory is not sufficiently close to make such 

 speculation of much importance. It is also evident that the proportion 

 will vary with the numerical value assumed for the monovalent ion. 

 Perhaps more important is the observation of Hardy ^ that the specific 

 conductivities (ionic dissociation) of acid solutions exerting equivalent 

 coagulating powers on negative and of alkaline on positive colloids are 

 appi'oximate constants : while acid solutions on positive and alkaline 

 solutions on negative colloids have very varied values. This probably 

 simply means that free H and free OH ions each have constant precipi- 

 tating values for the colloids of opposite sign, while the acid and basic 

 ions act according to their nature and valency. 



It has been stated by Spring^ and others that the superior influence 

 of polyvalent ions was not due directly to their valency, but to the usually 

 greater hydrolytic dissociation of their salts, and, though this has been 

 contradicted, it is not probable that the difl'erent ionic concentration 

 produced by hydrolysis is without influence. Jordis '' considers that the 

 colloid state and its phenomena are due to complex chemical compounds. 



When a colloid sol is precipitated by an electrolyte the resultant 

 gel always contains a small and apparently definite quantity of the 

 precipitating ion of opposite sign to the particles, which cannot be removed 

 by washing without restoring the gel to the sol condition, or destroying 

 its colloidal character. This quantity is proportional to the equivalent 

 Aveight of the precipitating ion, and may, in many cases, be substituted 



' Phil. Mag., 1899 (5), 48, 47-1-77. 



•■^ Proc. Roy. Soc, 1899, 66, 110-125. 



" Pec. Trav. chim. Pays-Pas, 1900, 19, 201-236. 



* Z.f. Plel'tr, 1904; 10, 509-518 ; Situinf/sbcr d. phys.-med. Soc. Erlangen, 1904, 

 36, 47-U7 ; Z. f. Elekirotech., 1904, 10, 500-518 ; and Kaiiter, Z. aiwra. Ch., 1903, 36, 

 lG-22. 



