68 ANNUAL OF SCIENTIFIC DISCOVERY. 



. 



arranged in parallel groups or bundles, as before described, but 

 clustered together in several distinct crystalline layers, which 

 are not parallel to each other. The consequence is that the 

 needle-points, visible under the microscope, appear to cross each 

 other at certain places, or at least they point in such directions 

 that, if elongated, these lines would cross each other at a short 

 distance in front of the fractured surface. Wherever the cross- 

 ing actually takes place, a ridge or line is generally visible to the 

 naked eye, and the color of the two parts of the fractured surface 

 which contain the different groups is different, since the light 

 which falls upon one group, at the proper angle for reflection, 

 will be in such a position with regard to the other group as to 

 throw the points of the crystals into the shade. The one part 

 of the surface, therefore, will appear bright or silvery white, 

 while the other will look dark or gray in color. As usual, inferior 

 specimens are more instructive than the best qualities, because 

 there the peculiarities and faults come out most strikingly. We 

 have seen apiece of a Bessemer steel block from a spoiled charge, 

 in which the crystalline structure of the spiegeleisen was seen in 

 some spaces, particularly at the edges of the air-bubbles, perfectly 

 distinguished from the coarse-grained crystals of the mass of 

 steel all round. This mass, moreover, contained groups of very 

 different character within itself. In a specimen of steel or iron, 

 made by another process, we could discover clearly defined 

 crystals of p}Tites, indicating the existence of sulphur in an un- 

 expectedly tangible manner. Repeated melting, heating, or 

 hammering of steel has, in general, the effect of reducing the 

 sizes of crystals, and also of laying them more parallel. Still 

 there seems to be a difference between the treatment which gives 

 parallelism and that which causes the reduction of sizes in the 

 crystals. The former seems to be principally due to the action of 

 the heat, and repeated melting is the great panacea in this re- 

 spect. The small-sized crystals, or what is called fine-grain, can 

 be obtained by mere mechanical operations. In fact, hammering 

 at a dull red heat, or even quite cold, is known to produce the 

 effect of making the grain of steel extremely fine. This is a 

 property, however, which is lost by reheating, and at a suffi- 

 ciently elevated temperature steel seems to crystallize in large 

 grains, which remain, if it is allowed to cool slowly and un- 

 disturbed by mechanical action. Engineering. 



EXPANSION AND CONTRACTION OF STEEL. 



Expansion and contraction belonging to this subject is the en- 

 largement or increase, or decrease in the bulk of the steel, as the 

 case may be, in consequence of a change in the particles by the 

 process of hardening. It is pretty generally known to thos._> who 

 are employed in the process of hardening steel, and to those in 

 the habit of fitting up various kinds of work requiring great nicety, 

 that the hardening of steel often increases its dimensions ; so that 

 such pieces of work, fitted with nicety in their soft state, will not 



