OF ARTS AND SCIENCES. 347 



when in presence of sesquioxide of iron, it is not at first sight easy to 

 decide which one should be preferred to the others for common use. 

 Bromine in presence of the alkalies appears to be the most powerful 

 of them, but since the acid solution of peroxide of hydrogen, which 

 is used in testing, would react upon any bromide which might have 

 been formed in the alkaline solution, bromine would be liberated, and, 

 by imparting its color to the ether, would obscure the reaction. A 

 similar objection applies of course to iodine. This difficulty is easily 

 overcome by neutralizing the alkali with nitric acid, and boiling for 

 a few minutes to expel the bromine : an objection to this process, how- 

 ever, is the fact that the solution is considerably diluted thereby. 



Chlorate of potash, with concentrated nitric acid, is in some respects 

 a very convenient agent : objections to it are, that some of the products 

 of the decomposition of the chlorate which remain in the solution, 

 appear to interfere with the formation of perchromic acid ; moreover, 

 if any nitrate of chromium remain unoxidized in the solution, it will 

 impart a bluish-purple tint to the aqueous solution of peroxide of hy- 

 drogen, which, though insoluble in ether, often interferes very materially 

 with the detection ©f the color of perchromic acid, if only traces of 

 the latter are present. This remark applies to any process of oxidation 

 in which nitric acid is employed. Both of these difficulties can gener- 

 ally be avoided by diluting the nitric acid solution with water, and 

 using a quantity of ether somewhat larger than is usually necessary 

 In any case where hyponitric acid is generated, it dissolves in the 

 ether, and may conceal the blue color of perchromic acid. 



Black oxide of manganese is not only a less energetic oxidizing 

 agent than pm'oxide of lead, but the chromate formed in its presence 

 does not color the solution so strongly. The latter remark applies also 

 to permanganate of potash, another objection to which is the necessity 

 of destroying its own color whenever it has been used in excess, be- 

 fore testinsf for chromic acid. 



On the other hand, the reagent of Chancel, peroxide of lead, is capa- . 

 ble, as I have shown, of oxidizing sesquioxide of chromium, even 

 when in presence of an enormous excess of peroxide of iron. It has 

 none of the disadvantages of the other substances to which allusion 

 has been made, and is especially to be preferred, since the chromate 

 formed in its presence imparts an intense yellow color to the alkaline 

 solution. 



