Overstrain  and  Recover//  of  Aluminium,  etc. 
387 
Table  VI. — Specimen  No.  8  (Third  Overstrain). 
Time 
Load 
in 
in 
Minutes. 
Tons. 
0 
0 
1 
0-5 
2 
1-0 
3 
1-5 
4 
20 
5 
25 
(') 
30 
7 
35 
8 
40 
9 
4-5 
10 
5-0 
11 
5-5 
12 
60 
13 
6-5        : 
14 
7-0 
15 
7'5 
10 
8-0 
17 
8-5 
19 
1 
110 
1st  Loading. 
Dec.  1.  1904. 
2nd  Loading. 
After  3  mts.  at 
99°  C.  and 
20  mts.  after 
overstrain. 
3rd  Loading. 
After  3  more 
mts.  at  99°  C. 
and  1  hour 
after  overstrain. 
4th  Loading. 
5  days  after 
overstrain. 
Reading.    Diff.    !  Reading. I   Diff.  j  Reading. 
200 
236 
273 
310 
345 
384 
424 
465 
504 
544 
585 
626 
673 
722 
768 
822 
884 
956 
36 
37 
37 
35 
39 
40 
41 
39 
40 
41 
41 
47 
49 
46 
54 
62 
200 
236 
270 
307 
343 
380 
419 
459 
500 
543 
588 
636 
36 
34 
37 
36 
37 
39 
40 
41 
43 
45 
43 
200 
236 
272 
310 
349 
387 
426 
465 
505 
546 
589 
633 
Diff.     Reading. 
36 
j  36 
38 
39 
38 
39 
39 
40 
41 
43 
44 
200 
237 
273 
311 
348 
385 
422 
460 
500 
540 
580 
621 
Diff. 
37 
36 
38 
I    37 
37 
37 
I    38 
i     40 
40 
40 
41 
and  III.) *  and  the  broken  lines  represent  that  of  specimen 
No.  8  recovering  under  the  application  of  heat,  as  in 
Tables  IV.,  V.,  and  VI.  Fig.  3  is  plotted  from  Tables  I. 
and  IV.,  and  represents  the  first  overstrain  of  the  two 
specimens  and  their  recovery.  Fig.  4  is  from  Tables  II.  and 
V.  for  the  second  series  of  tests,  and  fig.  5  is  from  Tables  III. 
and  VI.  for  the  third  series.  The  order  of  letters  A,  B, 
C,  D  gives  the  order  of  tests  in  every  case. 
Rate  of  Time  Recovery. 
The  rate  at  which  a  newly  overstrained  piece  of  aluminium 
recovers  its  original  properties  with  lapse  of  time,  during 
comparatively  short  intervals,  was  investigated  by  experiments 
with  a  piece  (No.  14)  of  the  same  bar  from  which  specimens 
Nos.  8  and  9  were  cut.  It  was  overstrained  in  exactly  the 
same  manner  as  before,  and  then  a  load  of  four  tons  was 
applied  at  intervals  and  quickly  removed.  The  extensometer 
was  read  at  each  application  of  the  I-ton  load,  the  reading 
for  no  load  being  as  before  200. 
The  manner  in  which  these  readings  decreased  is  shown  in 
