Manchester Memoirs, Vol. h. (191 1), No. %A. 5 



coincident with or parallel to the "terraces" on the 

 exterior so plainly to be seen in Plate I. These parallel 

 lines are intersected again at right angles by evenly spaced 

 short lines, which can be clearly counted in the photograph. 



To the metallographist the photograph suggests the 

 appearance of rectangular lines of lighter coloured ferrite 

 upon a darker pearlite background, but this is not the 

 case. The bulk of the ground, it is true, is pearlite, but 

 the pattern, as shown, is produced by lighter coloured 

 lines of pearlite on the dark ground, apparently brought 

 out by slight overetching. On turning the section so as 

 to allow the light to catch the bright uncoloured ferrite, 

 it will be seen that this has arranged itself with almost as 

 obvious a general symmetry along the boundaries of the 

 network pattern as shown by the photograph. Under a 

 low power of magnification the ferrite appears as discon- 

 nected uneven lines and rectangles, but the pattern made 

 by the ferrite as a whole is, as before said, unmistakeably 

 in perfect agreement and relation with the pattern in the 

 pearlite itself. 



Under a high power there is little to distinguish the 

 structure, as revealed in so limited a field, from that of an 

 ordinary unannealed but slowly cooled casting. 



The structure also showed a great deal of the dove- 

 coloured sulphide of manganese, which was not surprising 

 in view of the analysis. This sulphide of manganese 

 mostly occurred, as is commonly the case, in the centre of 

 the ferrite lines and patches. 



There is not the smallest doubt that the pattern 

 revealed by the etching stands in intimate relationship to, 

 and is indicative of, the interior crystalline structure of 

 the steel "pine-tree " crystal, and affords valuable evidence 

 of crystallization in a regular system. 



