704 PROCEEDINGS OF THE AMERICAN ACADEMY. 



shapes its signals by an iron magnet. Even the telephone speaks to 

 us through a steel magnet and an iron diaphragm. 



The earth itself is probably mostly composed of iron, for we have 

 reason to believe that we are living on a layer of cooled and modified 

 slag surrounding a still hot spherical iron ingot five thousand miles or 

 more in diameter, the product of the celestial furnaces. We get our 

 usable iron from the surface flecks of rust here and there. 



The art of steel making, as first practiced, was merely an art. There 

 was no science to assist, as it long antedated metallurgical science. 

 The processes were the result of accident or empirical trials. Only 

 within the half century just passed has steel production been brought 

 under control of scientific methods, and only within the past twenty- 

 five years has there been reached a just understanding of the nature of 

 the changes involved. That singularly valuable property of true steel 

 which enables its hardness, elasticity and toughness to be adjusted by 

 a simple heat treatment, has received its explanation. There is no 

 need to emphasize the enormous importance of the property of harden- 

 ing and tempering which is the characteristic most prized in tool steel. 

 Chemical analysis and microscopic examination of etched faces or sec- 

 tions have revealed in large measure the actual structure of the steel 

 in its various states. Such microscopic examination constitutes the 

 comparatively new science of metallography, which has in recent years 

 contributed so much to our knowledge of metals and alloys, but more 

 particularly the structure of steel. 



We now know that not only is steel a complex product, but that its 

 properties, as varied by hardening and tempering, or heat treatment, 

 depend upon the greater or less predominance and upon the dis- 

 tribution of certain chemical constituents, the relative amounts of 

 which are not the same even in the same bar when heated to differ- 

 ent temperatures. Names are given to these constituents ; ferrite, 

 cementite, pearlite, martensite or hardenite, graphite ; and they are, 

 except the last, composed of iron associated with certain proportions of 

 carbon in combination. By sudden cooling of red hot steel to harden 

 it, we catch and fix, as it were, the components in their relation as ex- 

 isting in the hot metal, and before they can adjust their proportions to 

 correspond with that normal to cold metal. In the hot metal so fixed 

 by sudden cooling, there is a preponderance of a hard constituent, mar- 

 tensite or hardenite. If the hot steel had been slowly cooled, then 

 there is time for a reproportioning in which the hard martensite disap- 

 pears and gives place to a mass composed largely of softer constituents. 



Still more valuable is that property of tempering, without which 

 steel tools would be either too hard and brittle, or too soft and flexible. 



