Annealing pure iron in hydrogen, followed by etching in a hydrogen atmosphere is shown to develop etch features at defects in the metal itself, rather than those of an oxide film. These features are contrasted with those reported for the same etchant on unreduced samples. The influence of an air-formed film in controlling etch morphology is discussed.The etch pitting of crystalline materials is now a well established method for pinpointing dislocations and following their movements u n d e r stress. Although much of the early work was confined to nonmetals, well authenticated examples do exist in the literature where researchers have produced etch pits at dislocation sites in metal single crystals; for example, the classical work of Livingston on copper (1) and Hibbard and D u n n on silicon iron (2). Later work included the etching of bismuth (3), tungsten (4), nickel (5), m o l y b d e n u m (6), zinc (7), cadmium (8), and niobium (9). A recent review of etching methods and etchants used has been made by Robinson (10).Subsequent research on polycrystalline metals, notably iron and iron alloys, indicated that an aging treatment was necessary in order to "decorate" the line defects with small quantities of an impurity or trace element. Suits and Low (11), for example, have shown that aging is necessary for etch pitting of dislocations in an Fe-3% Si alloy and that carbon is the principal segregating impurity. Pickering (12), again working with polycrystalline Fe-3% Si, found that a one-to-one correspondence between etch pits and dislocations is only possible when the dislocation density is below a critical value which cannot be displaced upwards by increasing the carbon content. He observed pitted areas in etched thin foils directly under the electron microscope, and saw that fresh dislocations introduced by plastic deformation were never pitted whereas aged dislocations were attacked to a degree depending on the carbon level. Pickering concluded that for a oneto-one correspondence to exist the amount of carbon in solution must be (a) equal to, or above, the level which is necessary to produce sufficiently large carbon atmospheres at every dislocation present in the alloy during aging; and (b) below that level which, by the formation of random clusters or precipitates during aging, is sufficient to cause two-dimensional nucleation of etch pits not associated with dislocations.More recently etchants have been developed which produce etch pits without an aging treatment. Liss13) developed etch pits in iron in what appear to be t Present address: Central Electricity Research Laboratories, Materials Division, Leatherhead, Surrey, England. K e y words: Surfaces, reactivity of solids, dissolution, defects in solids. glide bands. Pelessini and Alessandrini (14), also working on iron, demonstrated the increasing density of pits with work hardening, without intermediate aging treatment, on grains of all orientations.It is often difficult to interpret the results of etch pitting of polycrystalline metals because of the muc...