336 Analysis of Scientific Books. 



Davy has found that lOO parts of platinum combine with 11.86 

 parts of oxygen ; and on the other hand, M. Berzelius has de- 

 termined that 100 parts of this metal combine with 16.494 parts 

 of oxygen. The mean of the results of Edmund Davy and Ber- 

 zelius, would be 14.177, for the oxygen in the peroxide; now 

 this quantity does not differ much from 13.269, which repre- 

 sents the quantity of oxygen which would be necessary to form 

 a tritoxide.' " Thismethodof treating the Science of Chemistry," 

 adds Mr. Berzelius, " is peculiar to M. Thomson *." 



Well may the Swedish Philosopher say so. This travesty of ex- 

 perimental results is of a piece with that account of thechloro- 

 cyanic acid of M. Gay Lussac, which has excited so much ridicule 

 among the French Chemists. Our Doctor actually describes as 

 a liquid, what M. G. Lussac analyzed in the gaseous form ; and 

 he enjoins as the mode of making it, a method of verifying its 

 nature after it is made. No man could prepare it by Dr. 

 Thomson's misdirections. 



The Swedish Chemist, in his remarks on the prussiates pub- 

 lished in the 15th volume of the Annales de Chirnie et de Phy- 

 sique, has the following passage: " This is the first example," 

 says Dr. Thomson, " which does not agree with chemical 

 theory ; I therefore invite chemists to seek for its explanation, 

 because nothing contributes so much to the advancement of 

 science, as the development of what may appear to contradict 

 opinions already received." To which quotation Berzelius sub- 

 joins, " he ought to have added, with the exception of those 

 cases where the contradiction is the result of an experiment ill- 

 made and inexact, whereby the Science makes a very negative 

 gain't." 



As the public have it in their power to compare at leisure our 

 Review with the Regius Professor's Answer, we had resolved at 

 first to leave the matter entirely in their hands, without 

 further comment, perfectly tranquil as to the final decision. 



• " Les experiences analytiques de M Cooper, ainsi que les conclusions 

 qu'il en a tiroes, me pariirent ne rn^riter aucune attention. Cepeudant 

 M. Thomson, dans la septieme Edition de son Systeme de Chitnie, adopta le 

 protoxide de M. Cooper, rejeta celui dont j'avais fait I'analyse, et dont 

 j'avais d^crit les combinaisons avec les acides et les alcalis, et forma sur 

 les donn^es analytiques de M. Cooper, un tritoxide deplatine de la nianiere 

 suivante ; M. Edniond Davy, a trouv6 que lOO parties deplatine se com- 

 biuent avec u.86 p. d'oxigene,et d'un autre cot^ M. Berzelius a determine 

 que loo parties de ce ni^tal, se combineut avec 16,494 part, d'oxigfene. Le 

 terme moyen des r^sultats d'Edmund Davy et de Berzelius serait 14.177 

 pour I'oxigene dans le peroxide ; or ce terme raoyen ne differe pas beaucoup 

 de 13 Q69, qni repr^sente la quantity d'oxig&ne qni serait n^cessaire pour 

 former un tritoxide. Cette m^thode de trailer la science est particuliire a 

 jyi. Thomson" — Ann. de Chirnie et de Phys. Oct. 1821, p. 148. 



•f- A quoi, ce me semble il auroit faller ajouter ; a I'exception des cas oil 

 la contradiction est le resultat d'line experience mal faite et iuexacte oti la 

 •cience fait un gain bien negatif. Ann. x v. 147. 



