One end might have been finished so cold as to crush the grain, and the other end so hot as to cause incipient disintegration, but a competent inspector would discover either condition at once and reject the bar.
There is, then, a specific structure due to temperature; it is modified by carbon and by treatment under the hammer or in the rolls. If a bar of steel be heated up to the highest plastic limit, just so that it will not fall to pieces, and then cooled slowly without disturbance, and a fracture be taken, it will be found to be coarse and with an exceedingly brilliant lustre. Now let it be heated again to a bright lemon color, but still plastic, and cooled as before; it will be found to be coarse, with bright lustre, but neither so coarse nor so bright as the first piece. Then let it be treated in this way to lemon color, light orange, medium orange, dark orange, and orange red; as the heats go down the grain will be finer and the lustre will be less, until at about medium orange the lustre will be absent.
If any number of bars of even composition be heated in this way, the fractures will all be alike for each temperature.
If a series of bars of the different full tempers, about seven in all, be treated in this way, the structures due to a given temperature will all be similar, but there will be no two exactly alike, because high steel is much more profoundly affected by heat than low steel.
Seven tempers are mentioned here, because that is the number of full tempers in common use. Steel is graded out into fifteen tempers ordinarily by the interpolation of half numbers; this is easy and sure in the ingot inspection. In the above experiment the differences due to carbon are not quite so delicate, and the work is hampered in the heating by the personal equation, so that the use of seven full tempers is refinement enough. There is a difference due to every separable quantity of carbon, which could be shown if all of the operations of the experiment were exact.
If when a bar is broken cold the fracture is uneven, with coarse grain in one part and fine grain in another, it shows that there has been uneven heating. If one side has large grain and the other side is fine, the bar has been a great deal hotter on the side having coarse grain than on the other: the heater has let the bar lie in the furnace with one side exposed to a hot flame and the other protected from the flame in some way; he has neglected to turn the bar over and heat it evenly.
If the outside of the bar is fine and the centre is coarse, the bar has been very hot all through and has been finished by light blows of the hammer or by light passes in the rolls; it has been worked superficially and not thoroughly.
If the outside of the bar is coarse and the centre is fine, the steel has been heated on the surface too hot and too quickly; it has not had time to get hot through, and it has had too little work in the finishing.
If the grain is dark, with the appearance of a rather heavy india-ink tint, the steel has been finished too cold, and it will be found to be brittle.
If the grain is very dark, especially about the middle, looking almost black, then it has been finished altogether too cold: the grain is disintegrated, and the bar is fit only for the scrap heap.