RICHARDS. — SOLID SOLUTIONS. ■. 379 



some easily identified substance such as sodic chloride. Mr. R. P. 

 Gushing kindly made a number of such experiments for me, and found 

 that not even from a saturated solution of salt did more than 0.1 percent 

 of the precipitate consist of sodic chloride. Of course this value fixes 

 the maximum, for some of the substance may well have been occluded 

 instead of included. 



Very different from the mechanical process of inclnsion is the behavior 

 of baric sulphate. Here the impurities may amount to several perceiit, 

 and the foreign material is disseminated equally throughout the mass. 

 The experiments of Schneider show that the amount of the occlusion is 

 greater as the concentration of the impurity in the solution increases, 

 although not in direct proportion. It is very clear that we are dealing 

 here with a special case of the distribution law ; but in this case the 

 distribution can take place only at the moment of precipitation, because 

 afterwards the rigidity of the solid prevents free motion. Kiister and 

 Thiel evidently thought of this possibility, but they did not amplify the 

 idea. In the present paper I have retained the name occlusion for this 

 phenomenon because no other term seems to be applicable. There is 

 indeed a certain analogy between the occlusion of hydrogen in palladium 

 and the dissemination of a solute out of a solution into a solid ; and since 

 the name gives rise to no misunderstanding it will answer its purpose. 



Obviously the study of the distribution law in other cases should throw 

 light upon this one by analogy. "When hydrochloric acid is distributed 

 between its aqueous solution and the vapor phase, the concentration of 

 the undissociated part alone is concerned in the distribution, because ions 

 can exist only in the solution. In the same way, when a moderately 

 strong acid is distributed between water and ether, only the undissociated 

 part of the acid comes into play, because ether does not cause important 

 dissociation. From these analogies one might infer that the o-roup which 

 is concerned in cases of concomitant precipitation are not ions, but rather 

 electrically neutral complexes. The inference is not a perfectly safe one, 

 for little is known about the possible degree of dissociation in solids ; 

 but experimental evidence is at hand to support this view. In 1894 

 Richards and Parker made the observation that the occlusion of baric 

 chloride by baric sulphate is much increased by the previous addition of 

 hydrochloric acid to the solution of baric chloride.* For example, the 

 addition of twenty cubic centimeters of strong hydrochloric acid caused 

 almost twice as much occlusion as the addition of ten cubic centimeters 



* Proc. Am. Acad,, 31, 74 (1894) ; Z. anorg. Chem., 8, 420 (1895). 



