﻿178 Canadian Record of Science. 



outside of the matte pot. The second consists of ferro- 

 nickel crystals/ generally found near the centre or bottom, 

 and having the form of squares or rectangular triangles 

 about \ to yV i^^ch in diameter. These are tin white, very 

 thin, flexible and highly magnetic, and have the formula 

 FcgNig. Wliile comparatively rare, yet close examination 

 will discover the presence of ferro-nickel in every pot 

 of matte. 



It has been known for several years that matte is 

 not homogeneous throughout each casting, nor is it 

 surprising that in such a fluid mixture of different 

 sulphides the elements should, during the time of cooling, 

 attempt to arrange themselves with regard to their 

 respective affinities. A long series of experiments to 

 determine what these tendencies were may be thus briefly 

 summarized. 



Numerous analyses showed' that in one and the same 

 matte casting a sample broken from the top will be, as a 

 rule, higher in copper and lower in nickel than a sample 

 from the bottom. Eleven pots thus examined gave an 



average as follows : — 



Cu. m 



11 top samples 23.26 20.15 



1 1 bottom samples 21.14 20.32 



2.12 0.17 



Gain Cu at top, 2.12 per cent. 

 " " bottom, .17 per cent. 



The copper seems to vary more rapidly than tlie nickel 

 from top to bottom. 



Further analyses showed that nickel was higher at the 

 centre than at the bottom of the casting. A few examples 

 will illustrate this tendency. A pot casting broken into 

 quarters was sampled at the points shown in the sketch, 

 and analyzed as follows : 



1 Jour. Anal. Chem., March, 1892. 



