INNER STRUCTURE OF METALS: PROBLEMS OF ELASTIC FAILURE AND OF FATIGUE 72. Advances in experimental technique have made great changes in our outlook on the problems of elastic failure and of fatigue. In tests made for engineering purposes, we have seen that resultant effects, observed in a specimen of considerable size, are analysed as though due to continuous strains in a structureless material : even in the work of Bauschinger and others, on "yield," "recovery" and allied phenomena, no mental picture is suggested of the inner mechanism by which those effects are pro duced. That the older ideas suffice to explain the behaviour of materials from an engineering standpoint is shown by C. F. Jenkin's successful derivation of all the ordinary phenomena of low-speed tests from a conceptual "model" characterized merely by a resistance to distortion which is due in part to elasticity and in part to "solid" friction (Engineering, Nov. 17, 1922). But to explain the first occurrence of elastic failure, or the processes which result in fatigue and ultimate fracture, it has become clear that attention must be concentrated upon the inner structure of the crystalline aggregate.
The microscopical study of metals was initiated by H. C. Sorby in 1864 (Brit. Ass. Report, 1864). After a period of neglect, it was taken up and pursued energetically by many workers; at the hands of Ewing, Rosenhain, Humphrey, Beilby, Osmond and Arnold it has yielded results which are of fundamental importance in relation to the strength of materials. The technique employed may be studied in the article on METALLOGRAPHY : here it is sufficient to state that a polished surface, etched and examined in a microscope under oblique illumination, is seen to be made up of irregular areas hav ing well-defined boundaries. These areas are the sections of crys talline grains which constitute the mass of the metal; the irregular boundaries are the chance surfaces in which one grain has met another during the process of its crystalline growth. The size of
the grains depends very largely upon the heat treatment to which the metal has been subjected : sudden cooling from a high tempera ture tends to make the grains small, slow cooling to keep them large; and protracted exposure to moderate temperatures has been observed in some cases to favour the growth of very large grains.
When a metal is stretched beyond its limit of elasticity, the grains are found to have lengthened in the direction of that stretch. Subse quent exposure to fairly high temperature results in a recon struction of the grains : the original pattern is not reproduced, but the new pattern reveals no direction of predominating length. Researches by J. A. Ewing and W. Rosenhain (Phil. Trans. R.S., 190o) showed that "plastic" strain is the result of slips which occur in the cleavage or "gliding" planes of the individual crystals. These slips are observed, under the microscope, as sharply defined lines which appear on the surface of each grain. Seen under normal illumination the lines are dark; seen under oblique illumi nation they may be made to appear as bright lines on a dark ground : thus they may be recognized as narrow steps produced by the slipping of one part of the crystalline grain over another. Fig. 28 represents a section of two contiguous surface grains, hav ing "gliding" planes as indicated by the dotted lines; AB is a part of the polished surface. Under stress which exceeds the elastic limit (such as a pull in the direc tion of the arrows), yielding occurs by slip at a limited num ber of places, such as a, b, c, d, e. Many such lines appear as the process of straining goes on : in general three intersecting sys tems, and in many cases four, are seen. In this way severe deform ations may occur, which will not destroy the crystalline structure of any grain, although they will largely change its shape.