Atomic Volume and Specific Gravity, 523 



of the former oxide, when combined, to one of the latter. 

 We are now in a condition to consider the salts of ammonia. 

 It is quite unnecessary to remind chemists that there are 

 two rival theories regarding the constitution of ammoniacal 

 salts. One of them, proposed by the profound Berzelius, is 

 that the salts of ammonia contain a hypothetical radical 

 termed ammonium, consisting of one equivalent of nitrogen 

 and four equivalents of hydrogen. Sulphate of ammonia i^i 

 tp be viewed as sulphate of oxide of ammonium, the latter 

 hypothetical body being equivalent to potash ; and hence the 

 isomorphism between the salts of potash and ammonia. The 

 other view of the constitution of ammonia is that proposed 

 by Kane, and so elaborately supported by him in his paper 

 on subsalts and ammoniacal compounds*. Dr. Kane supr 

 poses that an ammoniacal salt is formed on the type of a 

 magnesian salt, carrying along with it constitutional water. 



Sulphate of copper . . . CuO, HO, SOg 

 Sulphate of ammonia . . HO, NHgH, SO3 



On this view, amide of hydrogen is equivalent to, and plays 

 the part of an atom of water. If this be the case, amidogene 

 must be analogous to oxygen, and ammonia and a magnesian 

 oxide must possess the same atomic volume. At present all 

 this is purely hypothetical, and must be subjected to the test 

 of experiment before wq can admit it as a safe foundation on 

 M'hich to rear a theory. The means of deciding this question 

 seemed to present itself in an examination of the amides of 

 mercury, and of the crystallized salts of copper and zinc, in 

 which the ammonia is present quasi ammonia; and such 

 compounds have been described in the beautiful researches of 

 Kane on this subject. Wohler's white precipitate, HgCl 

 + NHg seems to be constituted in the most simple manner, 

 and possesses a volume of 33'0, which, deducting the volume 

 22*0 for HgCl, leaves 11*0, or unity, as the volume of NHg. 

 But again, m hite precipitate, HgCl -f HgNH2, affects a volume 

 of 44'6 by experiment, which, deducting 22*0 for HgCl, leaves 

 HgNHg also equal to 22*0, and yet the latter compound 

 should correspond in volume to NHgH. The heavy yellow 

 powder obtained by boiling white precipitate with water has 

 a volume of 66"(), and is constituted according to the formula 

 (HgCl + HgAd) + 2HgO; so that deducting 44'0, the ascer- 

 tained volume of the double amide and chloride, 22 or 11 x 2 

 remains for two atoms of HgO, giving the same result as in 

 the former subsalts, viz. the equivalency of HgO to HO, but 

 not to HgNHg; and another proof of this is afforded in the 



* Transactions of the Royal Irish Academy, vol. xix. part 1. 

 2 M 2 



