Atoms, Ions and Molecules. 413 



found to come together and to occupy certain distinct positions 

 in his series. This perhaps may be taken as additional proof 

 of the conformity of the system which I have proposed to the 

 essential characters of the elements. 



A few weeks after the first part of this paper was read before 

 the National Academy (April, 1895), M. Lecoq de Boisbaudran 

 brought before the French Academy a note on the subject of 

 the relations of the elements.* At page 1100 the elements to 

 which his system of nodes and decrements applies, are tabu- 

 lated. They are thirty-one in number, with three additional 

 elements somewhat widely separated from the rest. I observe 

 that all these thirty-one elements, without an exception, belong 

 to my first division and have atoms that are colorless at all 

 valencies. The three additional elements belong to the transi- 

 tional class. The elements of the third division, that is, having 

 atoms colored at all valencies, find no place in his classification. 

 The chances are enormously against this happening fortuitously. 

 The indications from both this and Thomsen's classification are, 

 therefore, confirmatory of the principle I have endeavored to 

 establish, that is, that the presence or absence of specific 

 absorption of a certain range of rays is a function of the atomic 

 weight and is closely related to the constitution of the elements. 



In considering the theory of ionic dissociation it is necessary 

 to bear in mind that no rigorous proof has ever been found for 

 it. It remains, therefore, a theory only, a fascinating theory 

 because it affords beautif ul explanations of phenomena which 

 otherwise have none. 



There is, however, an important difficulty connected with 

 this theory. The dissociated ions are often spoken of as " free 

 ions," which in the absence of exterior electrical agencies, 

 static or dynamic, they do not appear to be. It is certain that 

 after dissociation the ions continue in some way to influence 

 and control each other. This fact appears in a great number 

 of reactions, of which the following familiar one may be taken 

 as typical. 



If to a dilute solution of ferrous chloride we add dilute 

 hydrochloric acid in excess, completely excluding the air, there 

 will be a considerable dissociation and the ferrous ions will be 

 in presence of more than enough chlorine ions, if these were 

 free, to cause at least some portion of the ferrous ions to 

 acquire additional valency and become ferric ions. No change 

 of the sort can be detected. The chlorine ions which would 

 bring about this change are held in check by the hydrogen 



* C. R., exx, 20 (May 20th, 1895). 



