wash it in alcohol or ether, and dry. If the surface is clean no "rim" will be discernible when a drop of ether is allowed to evaporate on the surface.

Different workers prefer different etching fluids. Tincture of iodine, and dilute nitric acid are most frequently used.

I. Iodine.-Osmond employs tincture of iodine of two different strengths. The first contains

The second solution is made by diluting 25 c.c. of this up to 100 c.c. with alcohol. The stronger solution is used for mild steels, the more dilute solution for hard steels.

To apply the solution proceed as follows. Rub the solution over the polished specimen with the tip of the finger, or with a camel's-hair brush, and stop the etching when the brilliancy of the polished surface disappears, and the surface appears dull and grey. By rubbing the polished specimen in this manner, the etching is more uniform, and the constituents not acted upon by the etching fluid remain bright. Wash the specimen in alcohol, and finally dry in a blast of hot air, from, say, a Fletcher's hot-blast blowpipe.

If this treatment has not been sufficient to develop the structure, repeat the operation. Several mild etchings are better than one severe process. The etching which results from a prolonged exposure to the etching fluid often depends not so much upon the structure as upon mechanical imperfections of the surface, and other external causes. The lines of demarcation between the different constituents are

frequently indistinct and blurred. If on examination the etching is too deep, it will be necessary to repolish the surface. Deep etching is frequently misleading.

II. Nitric Acid.-Sorby used dilute nitric as an etching fluid. This is generally recognized as one of the best agents for steels containing but little phosphorus or arsenic. The strength may be—

Hold the specimen by means of a pair of crucible tongs, and immerse in the acid for about ten seconds. Wash in running water, then immerse in lime-water, then in water again, then in alcohol, and finally dry with hot air.

A. Sauveur1 adds: "The unreliableness and shortcomings of the usual treatment with dilute nitric acid are overcome by dipping the specimen in concentrated nitric acid (sp. gr. 1·42), which, on account of passivity, has little or no action on the polished surface. The specimen is then placed under an abundant stream of running water, and the acid is quickly and completely washed off. As soon as the layer of concentrated acid which covers the surface is diluted by the running water, it attacks the steel, at first vigorously, but for such a short time (since the water soon removes all traces of acid) that there is no danger of etching too deeply. This develops the structure clearly and sharply, the etching being of a uniform intensity all over the surface, and free from the objectionable coloured film, and from the unlike appearance of different parts of the field caused by local actions of

A. Sauveur, Metallographist, 3. 231, 1900.

varying intensity, which are so troublesome and misleading in etchings with dilute acid. It is sometimes necessary to repeat the treatment in order to develop structure to a proper depth, but more than two immersions are seldom required. The specimen is washed in water, alcohol, and quickly dried with a soft cloth, or under a hot jet of air."

An alternative1 way of conducting the etch with nitric acid is to employ the solution—

Arnold 2 treats a specimen with nitric acid (sp. gr. 1.20) immediately after leaving the No. 00 emery pad until bubbles of gas make their appearance (about one minute). The scratches are nearly all removed, and the structure is well enough defined for most commercial purposes. The preparation of a piece of steel for microscopic examination by this method only takes about five minutes.

III. Picric Acid.-A favourite etching fluid is prepared from

as recommended by Igevsky. It brings out the cell walls very distinctly. It does not colour ferrite.

IV. Hydrochloric Acid.-A solution is prepared by passing dry hydrochloric acid hydrogen chloride into absolute alcohol until the solution contains about 10

1 F. W. Spiller, Metallographist, 6. 264, 1903, replaces the alcohol by glycerine.

2 J. O. Arnold, Nature, 63. 613, 1901.

per cent. of acid. Then add one to 5 per cent. of anhydrous cupric chloride. Use solutions containing the least amount of cupric chloride for hard steels. Let one or two drops of this solution fall on each square centimetre of surface, and let the attack continue until the surface is covered with a light grey film. Wash and dry as before.

Other acids may be used-sulphuric, oxalic, chromic (Abel)-but all are inferior to nitric and picric acids.

V. Electrochemical Attack.-Very good results have been obtained by immersing the metal in a 10 per cent. solution of ammonium chloride, potassium sulphate, or warm sodium thiosulphate, while it is connected with the positive pole of a bichromate cell, the negative pole consisting of a piece of platinum foil, lead foil, or thin sheet iron. The current used varies from 0.001 to 0.01 ampere per sq. cm. of surface. The current should last a few minutes.

§ 44. Osmond's Polish Attack

Osmond recommends a lubricating solution of

in place of water in the last stage of the polishing with a parchment pad and rouge. The ammonium nitrate does not alone act upon the metal, but chemical action is induced by rubbing and friction.

An aqueous infusion of liquorice root was formerly used, but this has been abandoned in favour of ammonium nitrate.1

1 F. Osmond and G. Cartaud, Metallographist, 3. 1, 1900.