Zinc. Oxygen. 1766. Cavendish 100 + 23.3 1785. Lavoisier -- + 19.6 1790-1800. Wenzel and Proust -- + 25 1801. Desorme and Clement -- + 21.7 Davy -- + 21.95 Berzelius -- + 24.4 Gay Lussac -- + 24.4 Thomson -- + 24.42 My own -- + 24 Now if 24 oxy. ∶ 100 zinc ∷ 7 oxy. ∶ 29 zinc, nearly, which is therefore the weight of an atom of this metal, on the supposition that the oxide is 1 oxygen and 1 metal; and the atom of oxide = 36. I formerly estimated the atom of zinc at 56 (VOL. 1, page 260). This was occasioned by taking the above as the _deutoxide_ instead of the _protoxide_. By violently heating the oxide of zinc in a close vessel, Desorme and Clement reduced the oxygen nearly one half, so as to afford a presumption that an oxide with half the oxygen of the common one subsisted. Since that time some observations of Berzelius seem to shew that a suboxide of zinc exists. It does not appear however, that such oxide is ever found in combination with acids; and, granting the accuracy of the observations, it is rather to be presumed to be the semi-oxide, or 1 atom of oxygen and 2 of metal, than the protoxide. No higher oxidation of zinc than the above has yet been obtained, and probably does not exist. 15. _Oxides of potassium._ Since writing the articles “potassium and sodium,” in the former volume, a very important essay relating chiefly to these subjects has been written by Gay Lussac and Thenard (a copy of which they were so good as to send me), entitled “Recherches Physico-chimiques, &c.” in 2 VOL.--Many of the most interesting experiments of Davy have been repeated on a larger scale, and a great number of original ones added; these ingenious authors endeavour to sum up the evidences for and against the two hypotheses concerning potassium and sodium, namely, as to their being metals or hydrurets, and upon the whole incline to the former, allowing however, that the facts afford great plausibility to both. One thing they seem to have discovered and established, that the new bodies or metals admit of various degrees of oxidation, and of course these products have a claim to be classed amongst oxides in general though the nature of their bases may still be an object of dispute.