98 BULILETIN 18 4, UNITED STATES NIATIONIAL MUSEfUM 



acid in alcohol) are most commonly used, of which the former is 

 preferred by the author for all ordinary work. It develops the 

 structure the same as nital, but its action is slower, and therefore 

 the operation is more controllable. It also does not cause stains to 

 run over in very fine structures, which sometimes occurs with nital 

 etching. 



Nital, however, occasionally works better on individual irons or 

 on particular structures. A mixture of picral with 10 percent to 

 50 percent of nital also has been found to be the most satisfactory 

 etchant for certain details of structure. 



Nitric-acetic acid (equal parts of concentrated nitric acid and 

 glacial acetic acid) is used on high-nickel steels and has proved useful 

 on a few nickel-rich ataxites that revealed little or no structure with 

 ordinary etchants. 



Ferric chloride (FeCla 5 grams, HCl 50 cc, alcohol to make 1,000 cc.) 

 is considered effective in developing etching pits and Neumann lines 

 in artificial irons. Its usefulness with meteoric irons is minor. 



Stannous chloride, used on artificial irons to show oxide, is also 

 of minor value because such inclusions are visible on an unetched 

 surface. 



A number of reagents used with artificial irons (such as Stead's, 

 Le Chatelier's, Rosenhain and Haughton's) depend for their effect 

 upon the selective deposition of a film of copper. The author has 

 not found them useful. They are employed chiefly for revealing 

 the presence and variations of phosphorus content, which in mete- 

 oric irons is more satisfactorily shown by etching with neutral 

 sodium picrate. 



Ammonium persulphate (10 percent by weight in water, freshly 

 prepared) is recommended strongly by some technicians for bringing 

 out grain structures and other details in artificial irons, with a lighter 

 attack than is needed with other etchants. The author has used it 

 with fair results in most cases, but the sm'face is apt to appear 

 clouded with minute specks and fine structural details are less clear 

 than they are with picral or nital etching. 



Observations on etching. — The structural details and inclusions in 

 artificial irons are very much finer than in meteoric irons, and there- 

 fore some etching methods that are necessary with the former are 

 not so with the latter. The same reagent is often effective on mete- 

 oric irons with a much shorter application than is required with 

 artificial irons — e.g., etching for carbide with boiling alkaline sodium 

 picrate. 



As the attack of any solvent etchant is strong along the interface 

 of an inclusion, the appearance of very minute inclusions in artificial 



