iSIr. D. Muslict on the Crystallization of Iron. 85 



foiiiul united base to base, but more often obliquely, taking 

 the most slender filament as the foundation of the next cry- 

 stal. 



Investigating masses of crystallized cast-iron, I have found 

 the slender I'amifications present dots or burnished points 

 impressed with the parent tbrm of the crystal ; and in cells 

 where the glass can only reach, the most minute but most per- 

 fect forms may be discovered. 



The most perfect specimens of crystallized cast-iron are 

 to be found at foundries where cannon, mortars, lai-ge rollers, 

 or such heavy iron machinery is made, as it is necessary to 

 cast with large feeders or heads, to supply the diminution of 

 volume or subsidence that takes place by the gravitation of 

 the fluid iron. 



At one time I possessed a very large and beautiful collec- 

 tion made at the Clyde Iron- Works, where a considerable 

 manufactory of Government and other guns was carried on 

 during the early part of the late war. These crystals were 

 obtained in the greatest abundance, when by accident the 

 head or feeder of the gun had not been filled sufficiently high 

 with fluid iron to overcome the quantum of subsidence that 

 afterwards took place in cooling. When the gun was carried 

 to the cutting carriage, and the knife applied to separate the 

 head close by the termination or muzzle of the gun, the de- 

 ficiency was soon discovered by the iron cutting open and 

 spongy, and an imperfect castuig indicated. A blow from a 

 heavy sledge, under such circumstances, was suflicient to se- 

 parate the head from the body of the gun. The fracture 

 occasioned by such separation was generally convex to the 

 muzzle end of the gun, and concave to the head or feeder; the 

 latter of which presented a fine arrangement of crystals, in- 

 serted in the solid iron, while the convex surface presented 

 the more i:)eriect and detached forms grouped base to base, or 

 obliquely, as circumstances had determined. 



There can be no doubt that every gun or large castinrr, 

 made with a head or feeder, contains a portion of crystallized 

 iron, though it becomes only obvious and most easily got, 

 when the casting is imperfect at the point where the knife or 

 chisel is applied ; and probably these opportunities are at 

 all foundries sufficiently numerous to furnish an ample supply 

 of crystallized iron, though a supply might be obtained from 

 a more perfect casting, particularly a mortar or cannon, by 

 first measuring the depth of the shrinkage on the head or 

 feeder, and applying the cutting knife from four to six inches 

 lower than the ascertained point. As soon as the iron begins 

 to cut open and spoiigy, the operation is to be stopped : and 



the 



