clency in alumina and silica to yield to the attraction of the electric current in 

 the direction of either pole, he inferred the probability of their partaking of the 

 .nature of nutro-saiine substances, and attempted their decomposition by pro- 

 cesses suggested by that supposition. Failing in these, and observing that 

 alumina and silica have both a strong affinity for potash and soda, and consid- 

 ering that such affinity could not proceed from the oxygen which might be one 

 of their constituents, he inferred that it must be a quality of their metallic bases, 

 and that it would, in that case, be probable that, if mixed with soda or potash, 

 and exposed to the electric current, the base might be made to separate, and to 

 attach itself to the base of the alkali. A mixture of silica and potash, in the 

 proportion of one to six, was accordingly put in a platinum crucible, and re- 

 duced to a fluid state over a charcoal fire. The crucible was put in connexion 

 with the positive pole of a battery of five hundred pairs, and a rod of platinum 

 connected with the negative pole was brought in contact with the alkaline 

 menstruum. The moment the end of the negative rod touched the liquid, glob- 

 ules rose through it to the surface, on which they swam about in a state of 

 brilliant combustion. When the mixture cooled, the platinum bar was removed, 

 and the alkali and salex which adhered to it detached ; there remained upon it 

 brilliant metallic scales, which, immediately on exposure, became covered with 

 a white crust, and some of which burnt spontaneously. Being plunged in wa- 

 ter, the end of the platinum produced effervescence, and an alkaline solution 

 was formed, which, upon examination, was proved to contain silica. The same 

 process applied to alumni gave a like result. 



It was now determined to try the effect of the Voltaic current upon the earths, 

 in contact with potassium itself. An amalgam of potassium, in contact with 

 silica, was negatively electrified under naphtha. After being acted on for an 

 hour, the amalgam was made to decompose water, and the alkali thus obtained 

 was neutralized by acetous acid. A white precipitate was obtained having all 

 the characters of silica. 



The same process was applied, with the same results, to alumina, glucinia, 

 and zirconia. It was inferred, therefore, that these earths were oxides of met- 

 als, to which respectively the names of SILICIUM, ALUMINIUM, GLUCINIUM, and 

 ZIRCONIUM, were given. 



Having established, by direct experiments, the fact that so many of the al- 

 kaline and earthy substances were oxides with metallic bases, it was consistent 

 with sound physical logic to assume, as a general law, that " the alkalies and 

 earths are oxides of metals." 



The question, how far the volatile alkali, ammonia, was to be regarded in 

 relation to such a law, naturally presented itself. Without reference to this> 

 analogy, or offering any hypothesis to explain the fact, Seebeck had already 

 shown that an amalgam could be obtained by the action of ammonia on mercu- 

 ry. This fact was reproduced by Berzelius and Pontin, and communicated by 

 mem, with various circumstances attending it, to Davy. Berzelius maintained 

 that ammonia came within the scope of the general law, and that an idea which 

 had been previously thrown out by Davy was justified by the phenomena which 

 showed that ammonia was a binary metallic base. This question was theft 

 taken up by Davy, and the experiments of Berzelius repeated, but without ar 

 riving at any certain or clear result. Gay-Lussac and Thenard opposed the 

 views of Davy and Berzelius ; and a contest arose, for which, as it has little 

 connexion with the progress of electrical science, we shall merely refer to the 

 scientific periodical works in which it was carried on.* 



It has been already observed, that the character of Davy's mind was to pass 



AnnalesdeC'iimie, torn. Ixxii., p. 193., Ixxv., 256-291.; Biblioth. Brit., June, 1809, p. 122, 



