68 Scientific Intelligence. 



vent the formation of graphite. Hence Swedish iron was heated 

 in a carbon crucible for three minutes by means of an arc given 

 by 900 amperes and 600 volts, and the fused mass poured into 

 water. A very hard, brittle metal resulted, frequently highly 

 crystalline, containing 3 or 4 per cent of combined carbon but 

 very little graphite. To isolate any carbide present, the author 

 used (a) an electrolytic method and (b) the action of very dilute 

 acids, out of contact with air. Normal nitric acid gives a car- 

 bide mixed with carbon, but a half normal acid gives practically 

 pure carbide. After separating adhering impurities, iron carbide 

 is obtained in the form of brilliant white crystals having the com- 

 position Fe 3 C and a specific gravity of 7*07 at 16°. It is appar- 

 ently identical with the carbide present in steel and is not affected 

 by oxygen at the ordinary temperature, though it is acted on by 

 moist air containing carbon dioxide. In the state of very fine 

 division it ignites in the air below 150°. It becomes incandescent 

 in vapor of sulphur at 500°, in chlorine below 100°, in bromine 

 vapor at about 100°, and is decomposed by iodine at a red heat. 

 H) drogen chloride decomposes it at 600° with evolution of hydro- 

 gen containing traces of hydrocarbons ; but heated in a sealed 

 tube, the products of the reaction are hydrogen and methane. 

 Neither water nor solutions of sodium or magnesium chloride 

 have any action on the carbide at 150°. — C. JR., cxxiv, 716-722,. 

 April, 1897. G. r. b. 



4. On the Production of Peroxides in Slow Oxidation. — In 

 order to test the question whether slow oxidation is always 

 accompanied by the formation of peroxides, Bach has made nu- 

 merous experiments, the results showing that nascent hydrogen, 

 phosphorus, sodium, potassium, zinc, iron, lead, methyl alcohol, 

 ethyl alcohol, isopropyl alcohol, glycerin, formaldehyde, acetal- 

 dehyde, benzaldehyde, glucose, acetic, oxalic and tartaric acids, 

 ethyl ether, acetic anhydride, phenol, resorcinol, catechol, tannin, 

 pyrogallol, dimethylaniline, diethylaniline, phenylhydrazine, for- 

 mamide, acetamide, terebenthene, benzene, petroleum, quinine sul- 

 phate, morphine acetate, brucine and strychnine all give peroxides 

 on slow oxidation in the air either in the light or in the dark. 

 Since these compounds belong to very varied groups and since 

 the oxidation involves the splitting up of the oxygen molecule 

 O : O, it is reasonable to suppose that in the earlier stages of the 

 oxidation, the substances combine with the grouping • O • O • to 

 form peroxides, the splitting up of O : O into • O • O • requiring 

 less energy than into * O * and • O \ Peroxides are also pro- 

 duced by energetic oxidation, and they may be detected in water 

 on which a flame of hydrogen or carbon monoxide is allowed to 

 play ; percarbonic acid being formed in the latter case. More- 

 over these peroxides, when formed, facilitate further oxidation. 

 When air is passed through a solution of indigo mixed with tere- 

 benthene or benzaldehyde the indigo is rapidly oxidized to isatin. 

 Nascent hydrogen liberated from palladium produces a similar 

 effect when oxidized by air, and in this case the indigo is com- 



