I'HESIDENTS ADDRESS SECTION H. 619 



article^! in " Eiig-iiieering:,"' 1897-1898, gives the results oi the 

 microscopic examination of the fractures of a number of steel shafts 

 and axles that had broken at ordinaiy work, and found that the 

 breakage began at some spot where, owing to the presence of im- 

 purities of some kind, the ciystals were not in perfect union. Now, 

 a small flaw in an otherwise liomogeneous material produces a con- 

 centration of stress in its neighbourhood, and thus under repeated 

 stresses it tends to continually extend, the rupture in this case being 

 pi'imarily due to a separation of the crystals along their boundaries. 

 Andrews propounded the view that this is the real explanation of 

 fatigue in metals, and that the giving way under repeated loads is 

 thus primarily due to imperfections, generally in the nature of 

 microscopic flaws in the material, of which he says — 

 '■ Imperfections abound, without, Avithin, 

 In toughest metals as metallic sin." 

 According to this view, if it were possible to produce a perfect metal, 

 free from such microscopic imperfections, there should be no such 

 thino' as deterioration by fatigue. This view has been adopted by 

 Professor Jolmson, in his work on '" Materials of Construction," who 

 j)refers in consequence to speak of " The gradual fracture of metals 

 under repeated loads " as being a more truly descriptive phrase than 

 the term Fatigue, which seeins to imply that the metal itself has 

 somehow deteriorated. It is, however, hard to understand how the 

 remarkable general conformit}^ of the numerous fatigue experiments 

 by Wohler, Spangenberg, Bauschinger, Baker, and the rest can be 

 explained on the supposition of the existence of chance flaws of this 

 nature. More recent work of Ewing and J. C. W. Hunifrey (Phil. 

 Trans., A., vol. 200, 1903) has shown, that in a series of fatigue ex- 

 periments undertaken by them failure took place not by the separa- 

 tion of the crystals from one another along their boundaries, but the 

 line of fracture generally went through the crystals themselves, and 

 a mici'oscopic examination apparently showed the eft'ect to be inde- 

 pendent of small flaws in the material and enables something of the 

 mechanism of fatigue for the first time to be traced out. They 

 experimented upon rotating square bars, subjected to a bending 

 moment after the manner of Wohler, the sides being polished so that 

 they could be submitted to microscopic examination at intervals. 

 They found that after a certain number of reversals of load here and 

 there isolated grains began to show slip bands or gliding planes 

 similar to those thit make their appearance in metal under direct 

 tension or compression. "As the reversals proceed, the surfaces on 

 which slipping has occun-ed continue to be surfaces of weakness. 

 The parts of the crystal lying on the two sides of each such surface 

 continue to slide back and forth over one another. The effect of 

 this repeated sliding or grinding is seen at the polished surface of 

 the specimen by the production of a hurr, or rough and jagged 

 irregular edjje, broadening the slip band and suggesting the accumu- 

 lation of debris. Within the crystal this repeated grindino- tends to 

 destroy the cohesion of the metal across the surt'ace of sliii, and in 

 certain cases this develops into a crack. Once the crack is formed 

 it quickly grows in a well-known manner, by tearing at tlie edge in 

 conse(|uence of the concentration of stress which results from lack of 

 continuity. The experiments show how a crack mav be fonned 



