24 BULLETIN" 184, UNITED STATE'S NATIONIAL MUSEUM 



phate, slightly acidulated. (The action of Stead's, Le Chatelier's, 

 and other special reagents used in microexamination is too slow.) 

 On application of the solution a red film of copper quickly flashes 

 over the surface of the iron. If cohenite is present it is covered 

 with the film of copper, but schreibersite inclusions are not. 



To identify cohenite on a microetched sample, the author has 

 found the most satisfactory tests to be electrolytic etching with 

 alkaline sodium picrate and the use of Murikami's reagent. These 

 methods, which darken cohenite but not schreibersite, are described 

 in chapter XVII. Electrolytic etching with chromic acid, a stand- 

 ard test for carbide in artificial irons, proved less satisfactory. 



Finally, a test that should infallibly distinguish the two is spec- 

 troscopic analysis, which reveals as little as 0.01 percent of phos- 

 phorus. If the spectrum of a doubtful inclusion shows no phos- 

 phorus lines it cannot be schreibersite. 



Oxides. — Iron oxides are often present in meteoric irons, not as a 

 primary constituent but as a product of oxidation, either during 

 atmospheric flight or by weathering. 



The commonest type is the hydroxide (limonite), which is the 

 result of weathering. Few iron meteorites are free from it, unless 

 freshly fallen. It forms a brown-black crust and penetrates more or 

 less into the mass. In some cases the kamacite is oxidized, but the 

 taenite lamellae remam unchanged, preserving the outlines of the 

 octahedral pattern. In other cases the iron is wholly changed to 

 limonite with only faint traces of its original structure. The hy- 

 droxide tends to penetrate octahedrites along the planes of the 

 Widmanstatten pattern and also along grain boundaries, and some- 

 times along the planes of Neumann lines. See plate 67. 



Magnetic or ferrosic oxide (Fe304 or FeO.FeaOa), which is miner- 

 alogically identical with magnetite, is formed by the superficial 

 fusion and oxidation of the iron during its flight through the atmos- 

 phere, which produces the fusion crust often observed on iron me- 

 teorites. The oxide is driven into the cracks that often are formed 

 in the mass by stresses resulting from extreme atmospheric pressure 

 during flight, and it occasionally appears as small rounded intrusions 

 invading the surface slightly. It probably never occurs as pure 

 magnetic oxide, as it naturally would carry in solution more or less 

 ferrous oxide, iron, troilite, and schreibersite. This oxide, called 

 by the earlier German writers Eisenglas, has a black, somewhat 

 glassy appearance. 



Oxides are easily recognized, being visible on an unetched surface. 

 The magnetic oxide, however, sometimes is not easily distinguished 

 from the hydroxide, except by the manner of its occurrence. Inva- 

 sions of the former have rounded outlines and sharply defined edges, 

 are generally independent of octahedral structures, and never show 



