•248 
PJiOFP]RSOK J. A. EWING AND ML’. J. C. 4V. IIUMFREY ON THE 
It a 2 ) 2 )ears, tlien, that tliis material sutfers no damage fiom repeated reversals of a 
stress of 5 tons per S(p inch ; but tliat when tlie stress is raised to 7 tons per sq. inch 
signs of fatigue are apparent after many reversals. And further, that with a stress 
of no more than 9 tons per sq. inch, the damage caused by reversals is so considerable 
that cracks are formed and the piece breaks. In all 23robability fracture through the 
formation of cracks would occur with 7 tons also, thouoli all that is actuaUv 
demonstrated for this stress is that it causes slip-bands to appear and to become 
accentuated in the manner ^^Inch, with greater stresses, leads to the development of 
cracks. 
It is remarkaljle that these actions are Ijrought about by stresses much below what 
is ordinarily understood by the elastic limit of the material. A tensile test shows 
proportionality of strain to stress up to 12 or 13 tons per sq. inch, with no apparent 
defect of elasticity. The conditions under which tliese exj^eriments were made seem 
to exclude the possil^ility that vibration gave lise to local augmentation of the 
stresses. The manner in which the slip-lines appear shows that some crystals reach a 
limit of elasticity sooner than others. This is no doubt to be ascribed in part to their 
Ijeing so oriented that the gliding planes, on whicli slip occurs, are inclined in the 
direction wliich is most favourable to tire action of tangential stress. But besides 
tlris, it may be conjectured that in a complex structure built up of nrany crystals 
irregular in form the distribution of stress fi'om crystal to crystal is by no means 
honroo-eneous. 
o 
Whatever the selective action of the stress is due to, the experiments demonstrate 
that in repeated reversals of stress certain crystals are attacked and yield by slipping, 
as in other cases of non-elastic strain. Then, as the inversals proceed, the surfaces on 
which slijjping has occurred 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. Tlie efPect of this repeated sliding or grinding is seen at the 
polislied surface of the specimen ])y the production of a hiirr, or rough and jagged 
irregidar edge, broadening the slip-ljand, and suggesting the accumulation of debris. 
Within the crystal this repeated grinding tends to destroy the cohesion of the metal 
across the surface of slip, and in certain cases this develops into a crack. Once the 
crack is formed it (juickly grows in a well-known manner, by tearing at its edge, in 
consequence of the concentration of stress which results from lack of continuity. 
Engineers are familiar with the development of cracks, even in tlie most ductile 
materials, when these are initiated at air-bubbles or other flaws. The present 
experiments show liow a crack may be formed, without any flaw to serve as nucleus, 
the first breach of continuity being set up through repeated grinding on a plane of 
slip in ijerfectly sound metal. 
The experiments throw light on the known fact that fracture by repeated reversals 
or alternations of stress resembles fracture resulting from a “ creeping ” flaw in its 
abruptness and in the absence of local drawing-out or other deformation of shape. 
