Fracture surfaces have, no doubt, been studied throughout the history of mankind, probably starting with observations on stone-age tools. In the 16–18
centuries, the macroscopic appearance of fracture surfaces was used to assess the quality of metallic materials, with studies by Réaumur in 1722 [
] being the most notable (Fig. 1(a)). However, it was not until 1943 that fracture surfaces were first examined at high-magnifications (using optical microscopy up to 1,000x) (Fig. 1(b)) (Zapffe and Moore [
]), and that the first attempts were made to examine replicas of fracture surfaces using transmission-electron microscopy (TEM) (Barrett and Derge [
]). Early replicas had poor fidelity and resolution, and it was not until 1956 that Crussard
] pioneered high-resolution TEM fractography using shadowed, direct-carbon replicas (Fig. 2). This technique (and its subsequent variations) revolutionised fractography and led to a plethora of studies in the 1960’s and 70’s. It therefore seems appropriate to commemorate the 50
anniversary of high-resolution electron fractography with a review of how it, and subsequent scanning electron microscopy (SEM) and other techniques, have led to a better understanding of fracture processes. Such understanding has been invaluable in failure analysis and in developing improved materials. Milestone observations for a number of important modes of fracture in inert environments including cleavage, brittle intergranular fracture, dimpled overload fractures, and fatigue fractures, are described first, followed by examples of key observations for fractures produced in embrittling environments (Fig. 3). Fractographic features and aspects of fracture modes that are not well understood are also discussed.