576 REPORT— 1881. 



deserve and ac-quire the confident trust of chemists. But, as often happens in such 

 •cases, this confidence in the result carried many of them too far. It led them to 

 assume that atomic values in all other chemical compounds must be alvraj's the 

 same as in the compounds under consideration. They sav? that they had got hold 

 ■of the truth, and they thought it was the vs^hole truth. For instance, one most 

 distinguished chemist assumed that each elementary atom has only one value in its 

 compounds; that the atom of nitrogen has alv^ays the value three, as in ammonia 

 and its products of substitution, and that in sal ammoniac the atom of nitrogen is 

 chemically combined only with three atoms of hydrogen, whilst the molecule of 

 ammonia is in a state of molecular combination with hydric chloride. Another 

 most distinguished chemist admitted that nitrogen and phosphorus have two atomic 

 Talues, but not more than two. He held that the respective combining powers are 

 always satisfied by the same number of atoms, no matter what the character of the 

 uniting atoms may be. 



With respect to these views it may be noticed that the assumption of combina- 

 tion between molecules as due to some other force than that which binds together 

 the constituents of each molecule — in fact the assumption of molecular combination 

 as an unknown something different from chemical combination, is open to even 

 more grave objections tlian those which led us to abandon the dualistic system. 



To represent a molecule of sal ammoniac as a compound containing two mole- 

 cules, each one biult up by the chemical combination of the constituent atoms, and 

 the two united together by some other force called molecular, was hardly a step in 

 advance of the view which represented it as containing two molecules united 

 together by the same kind of force as that which holds together the atoms in each 

 of the constituent molecules. 



The other form of the theor)' of atomicity as an inherent property of each atom 

 enabling it to combine with an equal number of other atoms, whatever the cha- 

 racter of those other atoms may be, seems difficult to reconcile with such facts aa 

 the following: An atom of nitrogen is not known to combine with more than 

 three atoms of hydrogen alone, or of substances like hydrogen, but it forms 

 stable compounds with five atoms (as in the ammonia salts), when four of them 

 are basylous and one of them is chlorous. An atom of sulphur is not known to 

 combine with more than two atoms of hydrogen alone, but it forms stable com- 

 pounds with four atoms, if three of them are like hydrogen, while the fourth is 

 chlorous. Instances like these are plentiful, and they lead us to look to the 

 -chemical characters of the atoms bound together in one molecule as a fundamental 

 condition of the atomic value of the element which binds them together. 



Theoretical limitations of natural forces are very difficult of proof, and it is 

 ■well to be slow and cautious in adopting any such limitation. 



A careful consideration of the facts of the case has led me not only to dotibt the 

 Talidity of the supposed limits of atomic value, but to doubt whether we have 

 grounds for assigning any limits whatever to such values. 



Atomic values- appear to me to be in their very nature variable quantities, and 

 I venture to think that chemistry will be greatly advanced by a full and careful 

 study of the conditions of variation of atomic values. 



Two conditions of change of atomic value are particularly worthy of notice : — 



I. Temperature. 



II. The chemical character of the uniting atoms. 



Atomic values increase with fall of temperature, and diminish with rise of 

 temperature. An atom which is combined with as many basylous monads as it can 

 take up by themselves will take up chlorous monads, or both chlorous and basylous, 

 and reciprocally. 



In illustration of the diminution of atomic values with rise of temperature, 

 I may adduce the following well-known reactions : Sal ammoniac containing 

 nitrogen combined with five monads breaks up at a high temperature into am- 

 monia and hydric chloride ; and in like manner other ammonia salts decompose by 

 heat, forming ammonia or an amide, with trivalent nitrogen. The highest chlorides 

 of phosphorus and of antimony are decomposed by heat into free chlorine and the 

 lower chloride. Potassic fluosilicate is decomposed by heat into silicic and potassic 



