328 Tie/larches refpetl'utg 



alkaline bale which is added docs not afTume its fhare of the 

 acid, but the metallic oxide, on the contrary, divides the 

 precipitate \vjth the acid ; as when we form fulminating gold, 

 or the orate of ammonia. Sometimes alio the precipitant, 

 the acid, and the metallic oxide, form two complex com- 

 binations; one of which is infoluble and the other remains 

 liquid, as we have feen in the decompofition of the corrofive 

 mercurial muriate by ammonia, and in the experiments that 

 I have defcribed (firlt feries, Nos. X. and XI,), In general, 

 but particularly with refpecf to the metallic folutions, we 

 muft not feparate in our reafonings the fubftance employed 

 by the name of precipitant from the liquid in which the pre- 

 cipitation is effected. Attention ought to be. paid equally to 

 all the fubftances prefent which can poflibly form new com- 

 binations. 



3. Coloured oxides may produce colourlefa combinations ; 

 but on giving out a part of the acid with which they have 

 been combined, their colour will again appear in proportion 

 to the quantity of acid they have parted with 3 to that this 

 colour is an indication of the constitution thus eftablilhed, 

 provided the ftate of oxidation has not been changed by any 

 particular circumftance. 



4. Metallic oxides cannot be compared together, unlefs 

 they are taken in a determinate (late of oxidation. All the 

 combinations they arc capable of forming vary, not only 

 from this caufe, but alfo from the proportion of acid they 

 retain, when this proportion is not determined by a cryftaU 

 lization. With refpe£t to oxidation there are onlv two ex- 

 tremes ; that of the leaft and that of the greateft oxygenation, 

 which can be confidered as conftant ; and hence it follows, 

 that the nomenclature can only indicate vaguely, and with 

 great latitude, any metallic combinations in which the oxida- 

 tion and the proportion of the acid are not determined. 



5. The acids do not follow the fame order in their affinity 

 relative to the degrees of oxidation. There are fome in which 

 the affinity diminimes with the oxidation, fuch are the nitric 

 and fulphuric acids; in others again it appears to increafe, 

 fuch as the muriatic acid. Hence we may perceive, inde-. 

 pendently of all the confiderations 1 have mentioned, how little 

 reafon there was fof clafling the affinities of metals for the 

 different acids by confider'mg them as conftant forces. 



6". The remits of the complex affinities of metallic folutions, 



mixed with other falls, may alfo vary from the proportion of 



thofe falts, conformably to what has been obferved (on the 



influence of proportion in the complex affinities):, fo that then 



proportion of oxygen in the metallic oxule, thai of the 



6 oxide 



