1814.] Chemical Proportions, S5 



Number of Weight of an 



atoms. iijlf^riii:t particle. 



Nitrous oxide 1 a -f- 1 o.... 2*803 



Nitrous gas 1 a -f 2 o . . . , 3*808 



Nitrous acid 1 a + 3 o . . . . 4*803 



Nitric acid \ a + 5 o. . . . 6*603 « 



Phosphorous acid 1 /> + 2 o.... 4*618 



Phosphoric acid I p + 3 o 5-618 h 



Ammonia 1 k + 1 a . . . . 1 '935 ■ 



Hydrophosphorous gas . . . . 4 // + 1 /> . . . . 3*146 k 



Weignt of an 

 atom. 



Magnesium 1 *577 l 



2SG Ammonium 1*149 m 



Alumina 2*136 n 



g The change introduced into the weight of an atom of azote 

 occasioned the alteration in the representation of these four com- 

 pounds of azote and oxygen, as explained in the Annals of Philo- 

 sophy, vol. iii. p. 135. 



h These changes are the consequence of the new weight adopted 

 for the atom of phosphorus, founded on the experiments of Rose 

 and Davy, as stated in a preceding note. 



' The reason of this alteration was explained in the Annals of 

 Philosophy, vol. iii. p. 135. 



k According to this statement, the gas is composed hy weight of 

 100 hydrogen and -:95 phosphorus. Now Sir Humphry Davy found 

 by analysis that it was composed of 100 hydrogen + 489*56 phos- 

 phorus. This differs little more than one percent, from the tabular 

 result. If we confide in the specific gravity of phosphuretcd 

 hydrogen gas taken hy Sir H. Davy, it is just half the weight of 

 hydrophosphorous gas, and ought therefore to be a compound of 8 

 atoms hydrogen 4- ' atom phosphorus. 



1 This is deduced from the analysis of sulphate of magnesia hy 

 Berzelius, which is probably more exact than the analysis of Dr. 

 Henry, on which my number in the original table was founded. 



I do not consider it as necessary to make any alteration in the 

 weights of potassium, sodium, barytium, calcium, strontium. Ber- 

 zelius' numbers deviate a little from mine ; but mine are founded 

 on the most accurate analyses hitherto made. 1 think no hypo- 

 thetical reasoning, however plausible, should lie allowed to influence 

 us in determining the weights of the atoms of bodies. 



'" 1 do not consider the reasoning of Gay-Lussac and Thenard 

 rejecting the amalgam of mercury hy ammonia as entitled to con- 

 fidence. 1 have explained the foundation on which this number 

 was <■ tablishcd iu the Mnnals of Philosophy, vol. iv. p, l i. 



" Pounded on Berzelius' analysis of alum. 



2 



