of Notation in Chemistry. 439 



tation, do not sufficiently interpret themselves ; and even such 

 can only be constructed for a few detached instances. 



When the constitution is at all less simple than in the above 

 examples, the expression to describe it becomes still more diffi- 

 cult to construct. If we have 3 atoms of lime and 4 of silica, 

 there is no very compendious chemical name for the compound. 



But if the usual phraseology be defective and inconvenient 

 in compounds of two ingredients, it becomes unmanageable 

 and almost impossible when there are more. Stilbite is one 

 atom oftrisilicate of lime combined with four atoms oftrisiUcafe 

 of alumina 9 and six atoms of water. Such a mode of descrip- 

 tion conveys no distinct impression, except by being considered 

 as a mathematical form ; and, if it be so considered, can be 

 expressed much more simply and briefly by means of mathe- 

 matical symbols. 



Moreover, such a mode of description is, in some de- 

 gree, hypothetical ; for the direct and certain result of the 

 analysisls only this: that the mineral just mentioned contains 

 certain quantities of silica, alumina, lime, and water, respec- 

 tively ; and the process of connecting one certain portion of 

 the silica with the lime, and another portion with the alumina, 

 depends upon the assumption of the doctrine of atomic combi- 

 nations. It is also, in some measure, arbitrary, even granting 

 that doctrine ; for there are generally, in such cases, more 

 possible ways than one of making such combinations, though 

 one way may often be simpler than the others. Thus, one 

 atom of silicate of lime combined with one atom of bisilicate 

 of alumina might just as well be considered to be one atom 

 of bisilicate of lime with one of silicate of alumina ; and as 

 this choice of the mode in which the combination is to be 

 taken is often a matter of great doubt, it is not desirable to 

 adopt a notation which necessarily affirms one particular mode. 

 The notation of Berzelius, however, does restrict us, in this 

 manner, to one selected and frequently arbitrary view of the 

 body's constitution. 



It appears to me, that the objects of notation in such cases 

 will be gained, and all the inconveniences avoided, if we adopt, 

 the most simple and natural method of representing chemical 

 combinations by symbols. Let S represent the weight of an 



2G2 



