It is manifestly desirable to obtain any required degree of hardness by a single process, if possible; hence by heating a known quality of steel to a definite temperature, and quenching it in water or other liquid or mixtures maintained at about an even temperature, the colour test is becoming, in some cases, dispensed with, the conditions of heating and cooling being varied to give any degree of hardness from the highest attainable down to normal softness. Another and very desirable method of hardening and tempering is to heat in a flue of some kind maintained at the required temperature over the fire, and cool either in water or a quenching or cooling liquid, and then, instead of employing the colour test, to provide a tempering bath composed of some substance that will heat, in the open air, to a temperature of about 450° F. (232° C). By placing the articles (after hardening them) in the tempering bath, and heating it to a temperature equal to the colour of temper (under the colour test) required, we have but to cease supplying heat to the tempering bath when a thermometer standing therein marks the required temperature, and a uniform degree of temper will be given to all the articles; and the operation will occupy much less time than would tempering either individually or collectively by the colour test, because a liquid is much more easily kept at an equal temperature throughout its mass than are the heated sand, hot pieces of iron, or iron tubes resorted to in tempering by the colour test.

Another method of tempering, which, if capable of reduction to uniformity, would be the quickest and hence most desirable of any, is to heat the steel to a definite temperature, and cool or quench it in a liquid having sufficient greasiness or other quality which acts to retard its retraction of the heat 'from the steel, and thus give a temper at - one operation. As an example of this kind of tempering, it may be mentioned that milk - and - water, mixed in proportions determined by experiment upon the steel for which it was employed, has been found to give an excellent spring temper. Nor is there any doubt that, carefully conducted, such tempering may be of the very best quality. A great deal, however, in this case depends upon the judgment of the operator, because very little variation in heating the steel or in the proportions of milk to water produces a wide variation in the degree of temper. If, on trial, the temper is too soft, the steel may be made hotter, or there may be more water added to the milk. If the steel was heated as hot as practicable without increasing the danger of burning it, more water must be added; while if the steel was made red - hot without being hot enough to cause the formation of clearly perceptible scale, the steel may be heated more.

It is desirable, in all cases, but especially with a high grade of steel, to. heat the steel not above a blood - red heat, although shear and spring steel may be, and often must be, made hotter, in order to cause it to harden when quenched in water. In former times the hardening by fire and water and tempering by the colour test were exclusively employed for hardening and tempering, except in cases where elasticity was the property sought to be imparted to the steel, and in this case a process termed blazing off (which will hereafter be treated of) was substituted for the colour test, and in those days the term "hardening" was understood to denote the process of heating to a cherry - red and cooling in cold water; this was sometimes further defined, if either of the terms "giving the steel all the water " or "hardened right out" was used to particularly specify that the steel was not to be extracted until reduced to the temperature of the water. The necessity for these terms arose from a practice that sometimes obtained, of withdrawing the steel from the water before it was quite cold, and many excellent hardeners and temperers there are, who at the present day withdraw the steel from the water when it has sufficient heat left in it to rapidly dry off the water adhering to it, the result being, it is claimed, to alter the degree of hardness to a practically imperceptible degree, but to add considerably to the strength of the hardened steel.

In those days tempering was understood to mean the second process, whose object was to modify, to a definite degree, by the colour test, the effects of the first hardening. Since the introduction of the other methods of hardening and tempering above referred to, the terms hardening and tempering have come to be used by many persons indiscriminately, and it is a fairly debateable question what process should be termed hardening and what tempering. First, then, any degree of hardness less than that obtainable in a given quality of steel, heated to the brightest degree without causing the change known to smiths as "burning the steel" to set in, must be a degree of temper, notwithstanding that it would have no representative colour under the colour test, because it is a degree of hardness less than the maximum.

In practice a toolsmith usually heats cast steel to what he terms a cherry - red; anybody, however, who has watched an ordinary blacksmith heating tools to harden will have observed that "cherry - red" practically includes all ranges of temperature between blood - red and a red verging upon deep yellow, the blacksmith being perfectly satisfied so long as the steel was not burned. The difference in the hardness obtainable by these two extremes of heating is not of practical importance in steels of fine grade; but in steels of inferior grade, as some spring and shear steel, it is so great that a blood - red will not appreciably harden, while a yellow - red will harden beyond the highest degree attainable under a colour test. The question arises then, shall a piece of steel possessing any of the degrees of hardness lying between that denoted by a yellow under the colour gauge and the highest attainable by giving the steel a maximum of heat (short of burning the steel) and " all the water," be termed hardened or tempered ? Now, of these degrees of hardness we have no clear conception, having no practical means of gauging it.

If we give to a machinist a tool of a particular shape, he has such a clear idea of the hardness and strength it will possess when tempered to a colour, that he can determine how hard it can be made to perform a given duty, or about to what colour it must be made to leave it strong enough to withstand the strain due to a given duty. Or if we give him a piece of steel soft at one end and hard at the other end, the graduation proceeding uniformly from end to end, he can take a file, and after testing the hardness, mark upon the steel with tolerable accuracy the sections corresponding in hardness to a blue, a purple, a brown, and a straw colour, and would know of what hardness to make a tool that would cut the steel at any particular section not too hard to entirely resist cutting. This knowledge he has obtained from manipulations performed upon steel of all degrees of colour temper; but if we were to give him a piece of steel harder than any degree denoted by a colour, and yet not of maximum hardness, he would be dealing with an utterly unknown quantity.

"Tempering," applied to define the degrees of hardness denoted by the shades of colour ranging between the palest yellow and the deepest blue, conveys (in connection with a colour as, say, tempered to a straw colour) a clear idea of a definite and recognizable degree of hardness; but made to include a greater degree of hardness than is deuotable under the colour test, it would cover in its meaning a number of unknown quantities which seriously impair, if not even destroy, its whole value. From this it appears that the word "tempered" (as applied to steel) should properly apply to all degrees of hardness denotable by colour in the colour test, and that " hardened " should include all degrees of hardness above that denoted by the palest yellow on the colour test. Under this interpretation, the meaning of the word will be extended to include all single processes which give degree of hardness denotable by colours; .hence the original meaning will be preserved to that extent. On the other hand, however, there is little doubt that the word was originally applied to the process because that process " tempered "or" modified" the degree of hardness of the steel, as indeed is proved by the fact that it was and is very often used interchangeably with the term "lowered"; thus, to lower to a blue was and is to temper to a blue; this is further attested by an old expression which is still in common use, viz., to "draw" the temper'; thus, drawn to a blue implies that a piece of steel, "hardened right out" or " given all the water," has had its hardness reduced, or drawn, to the degree denoted by a blue colour.

- We have, however, another consideration, in that steel given a definite degree of temper (corresponding to a colour under the colour test) by a single heating and quenching process has not been tempered in the sense that it has been lowered or suffered a reduction of hardness; on the contrary, it has received an increased degree of hardness. We may indeed stretch a point and claim that it has suffered a reduction of softness, but then temper is understood to mean, directly, a degree of hardness, because normal softness is in practice understood to be the condition as regards softness in which steel is supplied by the manufacturers. If from this condition it is rendered softer by softening processes, that is termed softening, while - all degrees of hardness above that must come under the head of either hardened or tempered. We have then to choose between including in the term "tempered" all processes which act to reduce the hardness of the steel to a degree recognizable under the colour test by a colour, or to confine its meaning to all processes by which the degree of hardness is modified, lowered, tempered, or lessened; and in view of the fact that the result reached and the object sought in either case is to obtain a definite degree of hardness comparable to that obtained under the colour test, the former interpretation is undoubtedly the better.

In any event it is an error to apply the term " tempered " to processes which, at one operation, leave the steel harder than any degree answerable to a colour in the colour test; while it has been shown that a process which reduces the hardness of the steel without bringing it down to a degree denoted by a colour in the colour test, should be termed hardening.