Overstrain  and  Recovery  of  Aluminium,  etc.  381 
increasing  at  a  still  greater  rate  and  deviating  further  from 
the  law  of  proportionality  to  stress,  as  may  be  seen  most  readily 
by  plotting  load-extension  diagrams.  As  the  load  is  applied, 
the  material  exhibits  a  strain  which  is  increased  by  "creeping" 
more  or  less  slowly. 
Professor  Ewing  has  found  *  that  this  overstrained  state  in 
the  case  of  iron  and  steel  gradually  disappears,  and  that  the 
material  approaches  to  its  original  condition  after  a  sufficient 
interval  of  time.  Further,  Mr.  Muir  has  found  f  that  this 
process  of  recovery  is  greatly  accelerated  by  raising  these 
materials  to  such  moderate  temperatures  as  100°  C. 
After  consulting  Prof.  Ewing  and  receiving  a  favourable 
reply,  we  decided  to  find  to  what  extent  overstrained  aluminium 
recovered  its  original  properties  after  intervals  of  time  and 
after  subjection  to  moderate  heat,  and  to  investigate  other 
properties  bearing  on  this  plastic  condition.  The  general 
result  of  these  experiments  was  to  show  that,  as  in  the  case 
of  iron  and  steel,  recovery  of  aluminium  takes  place  with  rest, 
and  is  brought  about  more  quickly  by  the  application 
of  heat. 
The  recovery  with  lapse  of  time  is  nearly,  but  not  quite, 
complete  after  14  days,  while  Muir  J  found  that  steel  shows 
recovery  to  greater  hardness  than  the  original  in  17  days. 
The  heated  specimens,  although  they  recover  quickly,  do 
not  show  after  14  days  so  complete  a  recovery  as  the 
mi  heated  ones. 
Experiments  with  copper  gave  practically  the  same  results 
immediately  after  overstrain  as  before  it,  showing  that  if  the 
properties  are  altered  by  overstrain,  recovery  is  very  rapid. 
Specimens  of  aluminium-bronze  (Al  10  per  cent.,  Cu 
90  per  cent.)  were  also  examined  in  a  similar  way.  The  re- 
covery with  lapse  of  time  is  slow,  but  under  the  influence  of 
a  moderate  degree  of  heat  it  is  very  rapid. 
The  effect  of  mechanical  vibration  on  overstrained  alumi- 
nium appears  to  be  inappreciable,  but  in  the  case  of  aluminium- 
bronze  vibration  has  on  an  overstrained  specimen  the  effect 
of  augmenting  the  breakdown  of  its  original  properties  and 
retarding  its  recovery. 
The  effect  of  prolonged  loading  upon  tensile  strength  of 
aluminium  has  been  investigated  by  Prof.  E.  Wilson  and  by 
Mr.  J.  Gavey§,  who  pointed  out  that  a  load  of  two-thirds  of 
the   nominal    breaking   load    caused  rupture    when   applied 
*  Proc.  Hoy.  Soc.  vol.  lviii.  1895. 
t  Phil.  Trans.  Eoy.  Soc.  vol.  cxciii.  1899. 
t  Phil.  Trans.  Roy.  Soc.  vol.  cxciii.  1899. 
§  Journal  of  Inst.  Elec.  Engineers,  No.  154,  vol.  xxxi.  Feb.  1902. 
