148 ATOMIC PRllfCIPIES OF CHEMISTRY. 



5th, 6th, and 287> 290. The 5th rule is "that a binary compound 

 I rues. gliould always be specifically. heavier than the mere mix- 

 ture of its two ingredients,*' The principle on which this 

 . rule is founded is recognised by chemists as general, if not 



universal; namely, that condiensation of volume is a neces- 

 sary consequence of the expulsion of heat by the exertion of 

 affinity. Thus, steam is specifically heavier than a mixture 

 of 2 parts hidrogen and 1 oxigen; ammoniacal gas is in like 

 manner heavier than 21 azote with 72 hidrogen. The 6th 

 rule is that **a ternary compound should be specifically hea- 

 vier than the mixture of a binary and a simple, which 

 would, if combined, constitute it ; and the 7th, that " the 

 above rules and observations equally apply when two bo- 

 dies, such as C and D, D and E, &c. are combined." 

 These rules are founded on the same principle as the f6r- 

 mer, which principle entirely precludes the notions of ni- 

 trous oxide and nitric acid being binary compounds, and 

 discountenances those of carbonic and sulphuric acid being 

 binary compounds. 



After making these observations on the general rules, 

 I shall now advert to more particular objects. I have 

 already remarked, that explanations and elucidations similar 

 to the above were what I thought unnecessary to enter upon 

 in the work alluded to : it is not improbable but I may have 

 been mistaken in this respect, especially if such inquiries 

 and observations as the following should be frequent. ** When 

 •* bodies unite only in one proportion, whence do we learn 

 *«that the combination must be binary ? Why is it not as 

 *• probable, that water is formed of two atoms of oxigen 

 «* and one of hidrogen, of two atoms of hidrogen and one 

 ** of oxigen, or in short of ani/ assignable number of atoms 

 •* of hidrogen and oxigen ? I do not perceive that Mr. 

 '* Dalton has given any reason in support of this binary 

 •* combination in preference to all the rest; and / am una* 

 '* hie to conjecture what reason can be urged in its favour,'^ 

 (page 283]. Ihopesuch remarks will be no more adduced; 

 and farther, that if any one should inquire, for instance, why 

 1 part of carbone, which takes 1'28 of oxigen, or 2-56, does 

 not also occasionally take 3*84 and 5' 12 parts of oxigen, 

 it will be understood, that the reason I should assign is, 



that 



