'254 Dr. Thomas Thomsons Observations 



A bare inspection of these tables shews how very close 

 approximations to the truth my atomic numbers would be. 

 This would be still further corroborated if we were to draw 

 up tables exhibiting the constitution of the compound 

 gases in volumes, and the degree of contraction which has 

 taken place. But this paper has already run out to such a 

 great length, and so much remains to be noticed, that I 

 must delay these tables, at least, for the present. 



There is another point, however, of so much consequence 

 that it would be unpardonable to pass it by ; because it 

 furnishes a method of determining which of two numbers 

 ought to be chosen for the atomic weight of a body, when 

 we have arguments in favour of two different numbers 

 almost equally balanced. Thus, we do not know for cer- 

 tain whether 0-125 or 0*0625 be the atom of hydrogen ; or 

 whether 1*75 or 0*875 be the atomic weight of azote. The 

 point to which I allude is an observation of Dulong and 

 Petit, that if the atomic weight of a body be multiplied into 

 its specific heat, the product is a constant quantity. This 

 subject has been lately taken up by M. Avogadro, who has 

 made further researches into the specific heats of bodies, in 

 order to put the law more completely to the test of experi- 

 ment. He has found the law to hold good in most of the 

 cases which have come under his review. 



The reader will observe, that if the specific heat of bodies 

 multiplied into the atomic weight be a constant quantity, 

 it will follow, that every simple atom is surrounded with 

 the same quantity of heat ; or, in other words, that the 

 specific heat of the atoms of all bodies is the same. This is 

 so important a proposition that it deserves a rigid investi- 

 gation. 



Our methods of determining the specific heat of bodies 

 are not yet so accurate that the numbers obtained can be 

 relied on as perfectly correct. But as they constitute at 

 least approximations to the truth, they will enable us to 

 perceive whether or not the law of Dulong and Petit holds 

 in general. Let us draw up tables of the atomic weights 

 and specific heats of all the simple substances so far as 

 known ; and let us multiply these two numbers into each 

 other, that we may see how far the product will be a con- 

 stant quantity. 



