on the Atomic Weights of Bodies. 

 Of these 17 simple bodies, there are 8, namely, 



259 



that agree so closely in the product of their specific heat 

 multiplied by their atomic weight, that no doubt can be 

 entertained that the atomic weights for these bodies given 

 in the tables are the true ones. The atomic weights of 

 Berzelius for these bodies differ but little from mine ; yet, 

 if we were to adopt his numbers, with the exception of 

 nickel, his number for which I have approximated to in 

 the table, the results would deviate farther than mine from 

 0-375, as may be seen by the following table in which his 

 numbers are adopted. 



Atoms of Specific PlodncU 

 Berzelius. heat. 



ex 



1° 



Mean 

 377. 



Sulphur, 2-01165 0-188 0-378 



Antimony, 8-06452 0-047 0-379 



Nickel, 3-69675 0-1035 0383 



Lead, 12-94498 0*029 0-375 



Gold, 12-43013 0-0298 0-370 



Copper, 3-95695 0-095 0-376 



Tin, 7-35294 0-051 0-375 



Platinum, 12-33499 0-031 0-382 



The deviation is undoubtedly but trifling, because Berze- 

 lius's atomic numbers for these bodies are almost identical 

 with mine. But the reader will perceive, at a glance, how 

 little is gained by the long string of decimals appended to 

 each of his atomic numbers. 



There are three of these simple bodies, namely, 



Tellurium, . . 0-364 



Bismuth, . . . 0-360 



Mercury, . . . 03625 



Mean, 



0-3623 



b2 



