in its different States. 405 



potassium * about 40-5 ; and from an experiment which I 

 have detailed in the last Bakerian lecture, on the combus- 

 tion of potassium in muriatic acic! gas, oxymurialic acid will 

 be represented by 32*9, and muriatic acid gas, of course, 

 bv 33*9 ; and this estimation agrees with the specific gravity 

 of oxvmuriatic acid gas, and muriatic aeid gas. From my 

 experiments, 100 cubical inches or oxy muriatic acid gas 

 weigh, the reductions being made for the mean temperature 

 and pressure, 74*5 grains ; whereas by estimation they should 

 weigh 74'6*. Muriatic acid gas I find weighs, under like 

 circumstances, in the quantity of 100 cubic inches, 39 

 grains; by estimation it should weigh 38*4 grains. 



It is easy from these data, knowing the composition of 

 any dry muriate, to ascertain the quantity t - of oxide or of 

 acid it would furnish by the action of water, and consequently 

 the quantity of oxygen with which the inflammable matter 

 will combine f. 



In considering the dry muriates, as compounds of oxy- 

 muriatic acid and inflammable bodies; the argument that I 

 have used in the last Bakerian lecture, to show that potas- 

 sium does not form hydrate of potash by combustion, ig 

 considerably strengthened ; for from the quantity of oxy- 

 muriatic acid the metal requires to produce a muriate, it 

 seems to be shown that it is the simplest known form of the 

 alkaline matter. This I think approaches to an experi- 

 mentum crucis. Potash made by alcohol, and that has 



* Supposing potash to contain nearly 15'6 per cent, of oxygen. 



f I have stated in the last Bakerian lecture, that during the decomposition 

 of the amalgam from ammonia, one in volume of hydrogen to two of am- 

 monia is evolved: it is remarkable, that whatever theory of the nature of 

 this extraordinary compound he adopted, there will be a happy coincidence 

 as to definite proportions. If it be supposed that the hydrogen arises from 

 the decomposition of water; then the oxygen that must be assumed to exist 

 in ammonia, will be exactly bufllcient to neutralize the hydrogen, in an 

 equal volume of muriatic acid; or if it be said that ammonium is a com- 

 pound of two of ammonia and one of hydrogen in volume, then equal volumes 

 of muriatic acid gas aud ammonia will produce the same compound as oxv- 

 muriatic acid and ammonium, supposing they could be immediately com- 

 bined. I once thought that the phenomena of metallization might be ex- 

 plained according to a modified phlogistic theory, by supposing three dif- 

 ferent classes of metallic bodies: First, The metal of ammonia, in which 

 hydrogen was so loosely combinued as to be separable with great ease, and 

 whirl., in consequence of the small affinity of the basis for water, it had little 

 tendency to combine with oxygen. The second, the metals of the alkalies 

 and alkaline earths, in whicli the hydrogen was more firmly combined, but 

 in combustion, forming writer capable of being separated from the basis. 

 And, thirdly, the metals of the earths and common metals, in which the 

 hydrogen was more intimately combined-, producing by union with oxygen, 

 water not separable by any new attractions. '1 'lie phenomena of theaction 

 of potassium and sodium upon muriatic acid, referred to in the text, seem 

 however to overturn these speculation so far as they concern the metal* 

 from the fixed alkalies. 



2 C 3 been 



