268 Dr. Thomas Thomsons Observations 



sent state of chemistry, I am not aware of any advantage 

 which attends this doubling. 



The fourth table exhibits the atomic weights and specific 

 heats of 11 ternary compounds. The products differ a good 

 deal from each other, shewing that the specific heats have 

 not been determined with accuracy. But, if we leave out 

 olefiant gas which deviates too far from the rest, the mean 

 product of all the others gives 1*071. Now, 0*375 x 3 = 

 1 * 125 approaching pretty near the mean. Hence, we may con- 

 clude, that each of these ternary compound atoms contains 

 thrice as much heat as any of the atoms entering into its com- 

 position. So that in ternary compounds each simple atom 

 retains all the heat which was united with it when in an in- 

 sulated or uncombined state. 



I am afraid that it would hardly be safe to carry our 

 computations much farther. 



Table fifth contains one substance, chrysolite which is a 

 compound of four atoms. Now, the product of its atomic 

 weight by its specific heat is 1*332, which approximates to 

 1-500 = 0*375 x 4. 



Sulphuric acid, the other substance in this table, is a 

 compound of one particle of real sulphuric acid containing 

 four atoms, and one particle of water containing two or 

 three atoms. So that it is a compound of six or seven 

 atoms according to the view which we take of the constitu- 

 tion of water. Now, the number which it gives, namely, 

 2*144, which is not far from 2*250 = 0*375 x 6. So that, 

 if we reckon water a binary, and sulphuric acid a quarter- 

 nary compound, all the constituent atoms enter into com- 

 bination without parting with any of their heat. 



Table sixth contains the only two examples of quarter- 

 nary compounds whose specific heats have been determined. 

 Now, 0*375 x 5 = 1*878. A number approaching to that 

 which alcohol yields. 



If water be a binary compound, the hydrate of lime is 

 only a quaternary, and its number 1*387 approaches 1*500 

 = 0*375 x 4. 



The seventh table exhibits the atomic weights and speci- 

 fic heats of 12 sextenary compounds. Now, the mean pro- 

 duct furnished by all these bodies is 1*805. But 0*375 x 

 5 = 1*865. It M'ould seem from this that the binary atom 



