Manchester Memoirs, Vol. xlviii. ( 1 904), No. 7. 3 



namely, some of the slips developed into cracks due it 

 was thought to grinding between the adjacent surfaces of 

 slip, for in these experiments — and it is important to 

 note it — the cycle of stress was from zero to a positive 

 maximum, then to an equal but negative maximum, and 

 then back to zero ; that is to say the loading was alter- 

 nating in direction. The same remarks also apply to 

 Professor Reynolds' experiments. 



Now when a bar of metal is subject to a cycle of 

 loading within the elastic limit (say from zero to a maxi- 

 mum tension and back to zero) at the end of the cycle it 

 is found that there is still a slight extension or lag or 

 hysteresis, which however disappears after a little time. 

 This phenomenon is marked when long wires are rapidly 

 unloaded. Fio-. i is intended to illustrate this. 



•^LAGie 



EXTENSION 



Fis. I. 



