Manchester Memoirs, Vol. Iv. (191 1), No. 24. 7 



structure well marked in the ingot section on the table. 

 (See Ftg-. 3, P/aU II.). 



Professor Howe, in America, tried to obtain side light 

 on this matter by pouring out the interior of freezing ice 

 ingots, but found the walls to be perfectly smooth. He also 

 experimented with blocks of slag in the same way, but 

 again found the sides of the emptied cavity smooth and 

 free from crystalline markings, although the solidified 

 portion had a strongly columnar structure. 



Referring to the very strongly marked columnar 

 structure of solidified steel at right angles to the cooling 

 surfaces, you will see, in the case of the ingot section 

 shown on the table, that the section is not wholly 

 columnar ; the section is nine inches square and the 

 largest columns about 3^" long, and an area of about 3" 

 square in the centre is confused or granular. The steel 

 has the following analysis : — 



Combined Carbon ... ... = "16 % 



Silicon ... ... ... = '046 „ 



Manganese ... ... ... = "55 



Phosphorus ... ... ... = "049 



Sulphur ... ... ... = "032 



Nickel = 5-35 



The composition of this steel indicates a high melting 

 point ; there is, therefore, less direct transition from the 

 liquid to the solid state, and the confused area would 

 represent the disturbance of the sluggish core sinking 

 more or less towards the lower part of the ingot. The 

 segregation in the centre of the section shows that the 

 same is only just clear of the bottom of the pipe, which, 

 as is usual, had formed in the top of the ingot. 



It is quite obvious that there is a similarity of cause 

 and effect between the phenomena of crystallization as 

 noted in the " pine-tree " crystal growth and that of the 



