558 Mr. W. Kosenhain. Further Observations on [Feb. 9, 



problem is met with, and is satisfactorily overcome by embedding the 

 glass in pitch, supported by other pieces of glass. 



The author has adopted this principle of supporting the surface, which 

 in section becomes the edge, by means of an adherent layer of hard 

 material ; but the conditions which such a layer must satisfy for the 

 purposes of metallography are very stringent. The adherent layer should 

 be of approximately the same hardness as the metal under examination. 

 It must adhere to it with extreme closeness, and with strength enough 

 to resist the processes of cutting, grinding, and polishing ; there must 

 be no chemical action on the surface of the metal specimen, and the- 

 adherent layer must be free from grit, which would prevent the 

 polishing of the transverse section. Finally, it is desirable that the 

 protecting layer should be attached without heating the specimen, as- 

 even moderate elevations of temperature frequently affect the micro- 

 structure of strained metal. This last condition eliminates such sub- 

 stances as pitch or fusible alloys, which were, moreover, found unsatis- 

 factory in other respects also. The difficulty of cutting and polishing 

 eliminates the majority of cements and plasters. The author was there- 

 fore led to use a deposit of another metal obtained by electrolytic 

 means, and this method has proved satisfactory. 



The specimens used consisted of strips of the mildest steel, such as. 

 is used for transformer sheet, and after preparation an electro-deposit 

 of copper was applied to them. By first bending the strips into a 

 flat U shape, short portions of their length could be polished in the 

 usual manner for microscopic examination; subsequently the strips. 

 could be readily strained and even broken in tension^ by means of an 

 ordinary vice provided with clamping-plates attached to the jaws. 

 The slip-bands and other features of the specimens having been 

 satisfactorily observed, electro- deposition was proceeded with. The 

 specimens could not be immersed directly in the usual acid solution 

 of copper sulphate used in copper-plating, as the surface would have 

 been attacked and spoilt by the action of simple substitution ; they 

 were accordingly first coated with a thin film of copper in a bath of 

 copper cyanide under the action of a very weak current, and were 

 only then removed to the usual bath where deposition could be 

 carried on as rapidly as is consistent with the formation of a solid 

 and coherent deposit. The copper deposit was generally allowed 

 entirely to envelope the strips of iron, and deposition was carried 

 on until a thickness of 4 to 5 mm. was reached. This occupied from 

 8 to 14 days.* 



The specimens were then cut across, generally in a direction rough ly 

 parallel to the direction of the original tensile strain. In order to 



* The electro-deposition on the majority of the specimens here referred to Avaa 

 carried out by Mr. 8. Field, at the Northampton Institute, London, and the author 

 is greatly indebted to his skill and kindness. 



