316  Heating  Effects  produced  by  Rontgen  Rags  in  Metals. 
meter,  for  one  reason  or  another,  had  to  be  taken  down  and 
re-adjusted. 
Columns  VI.  and  VII.  contain  the  deflexions  reduced  to 
the  steady  state,  by  dividing  by  (1  — <?~7*).  The  values  of  7 
used  were  :  for  lead  1*5,  for  zinc  0*67.  Column  VIII.  con- 
tains the  ratios  of  VI.  to  VII. ;  it  is  the  ratio  o£  the  heat 
generated  per  second  in  the  lead  strip  to  that  generated  in 
the  zinc  strip. 
The  first  three  experiments  were  made  with  the  vane? 
between  the  window  and  the  strips  ;  the  last  six  with  the 
strips  between  the  window  and  the  vanes, 
E\p. 
I. 
II. 
III. 
IV. 
v.     ! 
VI. 
VII. 
VIII. 
1 
1-5 
-•&4:5 
4-1-30 

-3-85 
+  206 
1-87 
2 
1-5 
-  6-84 
+2-26 
— 7'<55 
4-3-58 
2-13 
3 
1-5 
4-5-20 
-190 

+5-82 
-3-01 
1-93 
4 
1-5 
-5-51 
+243 
+2-16 
-6-18 
+  3-41 
1-81 
5 
1-5 
-6-67 
4-2-64 
+  2-25 
-7-40 
+  3\u) 
210 
6 
1-5 
-7-07 
4-2-90 
1 
-790 
4-3-96 
2-00 
7 
1-5 
+  7-09 
-2-30 
4-0-39 
+  7-12 
—  0  bo 
1-96 
8 
1-5 
4-9'76 
-3„ 
+8-80 
+9-85 
-5-07 
1-94 
9 
30 
-13-75 
4-64 
-13-95 
,4-7-4 
i 
J    1-88 
The  average  of  the  ratios  in  Column  VIII.  is  1*96  ;  re- 
ducing to  equal  absorptions,  we  get,  as  the  result  of  these 
measurements,  that  when  Rontgen  rays  are  equally  absorbed 
in  lead  and  in  zinc,  the  quantity  of  heat  generated  in  the 
lead  is  1*93  times  the  quantity  generated  in  the  zinc. 
The  necessity  of  the  writer's  returning  to  America  has 
temporarily  interrupted  this  investigation.  Further  experi- 
ments are,  however,  now  under  way  in  which  other  metals 
will  be  compared,  and  the  experimental  conditions  varied  as 
much  as  possible.  An  attempt  will  be  made  to  detect  the 
effect  with  a  thermopile  as  well  as  with  the  radiometer ;  and 
to  ascertain  whether  cathode  rays  .produce  similar  effects,  as 
might  be  expected.  If  the  present  results  are  due  to  some 
hitherto  unsuspected  source  of  error,  it  is  hoped  that  the 
error  may  reveal  itself  as  the  conditions  are  varied. 
