Mr. R. W. Atkinson on the Atomic Theory. 119 



accrue by concealing from oneself tlie fact that this is only a 

 clumsy expedient for getting rid of the word d^iom. ? the essential 

 idea conveyed is that of an indivisible portion of matter^ i. e. an 

 atom. Thus^ then^ the generalization being bound up with the 

 hypothesis, it follows that the convention is also bound up with 

 them, the symbols and suffixes merely being used as an abbre- 

 viation for such an expression as " one atom of chlorine weighing 

 35'5 tinges an equal volume of hydrogen combined with one 

 atom of hydrogen weighing 1/'"' indicated by the formula HCl. 

 Hence Dr. Wright has not shown that symbols can be used to 

 represent the quantitative composition of bodies without invol- 

 ving the notion of atoms. If, as is done by Dr. Wright, sym- 

 bols be used to represent elements, and suffixes to indicate the 

 multiple of some fixed number mentally associated with that 

 symbol, it is merely putting the notion of atoms in a form of 

 speech which is calculated to produce the erroneous impression 

 that that idea has been dispensed with. 



Having thus shown, first, that the atomic theory now in use 

 is only a development of Dalton^s original conception and is 

 perfectly consistent with it, and, secondly, that the notion of 

 atoms is involved in Dr. Wright^s account of Dalton^s conven- 

 tion, his objections to my remarks can be easily disposed of. 



The first objection has already been noticed. The generaliza- 

 tion and convention cannot be separated from the theory which 

 Dalton proposed, they are so completely interwoven. But, from 

 Dr. Wright^s remark^ it would seem useless to discuss this point 

 further. 



The next objection is that, if my line of argument were correct, 

 it would be impossible to assign any formula to any thing ; Dr. 

 Wright omits to add the condition which was attached, viz. that 

 the notion of atoms be not held. Without that belief one has 

 no right to take the numbers corrected by the atomic theory in 

 preference to those obtained by experiment. In speaking of the 

 law of multiple proportions. Dr. Wright must mean either mul- 

 tiples of equivalent weights or of atomic weights. If the latter, 

 as it would appear, how much more certain are they than the 

 atomic weights of which they are the multiples ? 



Respecting the "combining number ^^ of aluminium, the se- 

 lection of which Dr. Wright thinks was very unfortunate, the 

 determination of the vapour-density of aluminic chloride by De- 

 ville must be well known, which shows that its formula is Al^ Cl^". 

 With regard to Buckton and Odling^s determination of the 

 vapour-densities of aluminic methide and ethide, the results are 

 totally at variance with those obtained in the case of other similar 

 compounds, and, further, have never been accepted by chemists. 



If Dr. Wright had read my paper carefully, he would have seen 



