1835.] Heat of Bodies to their Atomic Weights. 41 



stituent number is, \ -\- \-= -lb, the square root of which is, 

 •866 X -1875 = 1624 -r 3-5 = -464, the calculated specific 

 heat, which differs very slightly from -49 or '50, the results 

 of experiment. 



To obtain, therefore, the specific heat of oxides thus con- 

 stituted, it is only necessary to divide the weight of the 

 compound atom of each oxide by the number -1624. But 

 it is the double of this number or -3248 that it is necessary 

 to divide by the atom, resulting immediately from the 

 union of a whole atom with half an atom of oxygen; to ob- 

 tain the pame specific heat. 



By the application of this number we obtain for the cal- 

 culated specific heats of the oxides of mercury, zinc, and 

 copper, 0'048, 0*129, and 0*130, while the experiments of 

 Neumann gave *049, 0*132, 0*137. A similar calculation 

 affords for the protoxide of lead, protoxide of tin, and lime, 

 *047, *078, 0* 182, the results of experiment being *050, -094, 

 and -179. The calculated specific heat of magnesia is '251, 

 that of experiment *276. 



Neumann makes this constant number '697, the half of 

 which is *3485, constituting what he terms a stechiometrical 

 quantity. This is considerably above the estimate of 

 Avogrado. The difference is ascribed by the latter, to the 

 circumstance of Neumann taking as a basis of his calcu- 

 lation, the specific heat of magnesia, which has not been 

 properly ascertained. 



(2.) Those oxides which consist according to our present 

 ideas of 1^ atom base and 1 atom oxygen, are considered 

 by the views of Avogrado as compounds of | atom oxygen 

 and 1 atom base; to this class belong peroxide of iron, 

 deutoxide of lead, arsenious acid, and perhaps alumina. 

 The specific heat of alumina by experiment was found to be 

 *2. Now supposing the solid atom of alumina to consist of J 

 atom aluminum and f atom oxygen, the calculated specific 

 heat would be *2185. New experiments are required upon 

 pure specimens of peroxide of iron and arsenious acid, 

 before any conclusion can be come to with regard to the 

 constitution of these atoms. But in reference to magnetic 

 iron which consists of 1 atom protoxide, and 2 atoms per- 

 oxide of iron, important data have been obtained. If we 

 consider the atom of this substance divided into 8 portions 

 so that the solid atom may be f F + -JO, the weight of 



