﻿524 
  Mr. 
  W. 
  Williams 
  on 
  the 
  Temperature 
  Variation 
  of 
  

  

  According 
  to 
  Fizeau, 
  zinc 
  behaves 
  in 
  an 
  anomalous 
  manner 
  

   as 
  regards 
  its 
  thermal 
  expansibility, 
  the 
  value 
  of 
  a. 
  being 
  

   greater 
  at 
  40° 
  than 
  at 
  50° 
  0. 
  This, 
  however, 
  is 
  not 
  in 
  accord 
  

   with 
  the 
  observations 
  of 
  Matthiessen, 
  according 
  to 
  whose 
  

   results 
  for 
  zinc, 
  Z=/ 
  [l 
  +-0000274U 
  + 
  '0000000234 
  * 
  2 
  ], 
  where 
  

   I 
  and 
  / 
  are 
  the 
  lengths 
  of 
  the 
  same 
  bar 
  at 
  f 
  and 
  0° 
  respec- 
  

   tively. 
  Hence 
  /3 
  = 
  '00172. 
  For 
  most 
  of 
  the 
  metals, 
  how- 
  

   ever, 
  for 
  which 
  both 
  Matthiessen 
  and 
  Fizeau 
  give 
  values, 
  

   Fizeau's 
  values 
  for 
  (3 
  are 
  about 
  half 
  those 
  deduced 
  from 
  

   Matthiessen' 
  s 
  observations. 
  Thus 
  : 
  

  

  Matthiessen. 
  

  

  Copper 
  -0025 
  

  

  Silver 
  -00149 
  

  

  Cadmium 
  *00354 
  

  

  Tin 
  -00259 
  

  

  Lead 
  -000543 
  

  

  Palladium 
  -00184 
  

  

  Gold 
  -00165 
  

  

  Platinum 
  -000823 
  

  

  Fizeau. 
  

  

  Ratio 
  

  

  •00125 
  

  

  20 
  

  

  •000806 
  

  

  1-85 
  

  

  •001124 
  

  

  3-15 
  

  

  ■001671 
  

  

  1*55 
  

  

  •000848 
  

  

  •64 
  

  

  •00088 
  

  

  2-09 
  

  

  •000567 
  

  

  2-91 
  

  

  •000923 
  

  

  •89 
  

  

  Mean 
  = 
  1-89 
  

  

  aluminium 
  are 
  confirmed 
  

   and 
  Mr. 
  Tutton. 
  Thus, 
  

   the 
  mean 
  value 
  of 
  a 
  for 
  

   00001718, 
  and 
  '00001883 
  

  

  Again, 
  according 
  

   the 
  value 
  

  

  Fizeau's 
  values 
  for 
  copper 
  and 
  

   by 
  those 
  of 
  Dulong 
  and 
  Petit, 
  

   according 
  to 
  Dulong 
  and 
  Petit, 
  

   copper 
  between 
  0° 
  and 
  100° 
  is 
  

   between 
  0° 
  and 
  300°. 
  Hence 
  £=-00101. 
  

   to 
  Mr. 
  Tutton 
  (Proc. 
  Royal 
  Society, 
  lxiii. 
  p. 
  208), 
  

   of 
  a 
  for 
  pure 
  aluminium 
  is 
  10" 
  8 
  [2204 
  + 
  2-12 
  t], 
  so 
  that 
  

   /3 
  = 
  '000962. 
  Hence, 
  perhaps, 
  until 
  more 
  decisive 
  deter- 
  

   minations 
  are 
  made 
  in 
  the 
  case 
  of 
  pure 
  zinc, 
  by 
  Mr. 
  Tutton' 
  s 
  

   improved 
  interferometer 
  method 
  for 
  instance, 
  and 
  as 
  the 
  

   molecular 
  constitution 
  of 
  the 
  metal 
  is 
  simple 
  like 
  that 
  of 
  

   copper 
  and 
  silver, 
  we 
  may 
  take 
  for 
  j3 
  half 
  the 
  value 
  deduced 
  

   from 
  Matthiessen's 
  observations 
  as 
  in 
  the 
  case 
  of 
  most 
  of 
  the 
  

   metals 
  tabulated 
  above. 
  

  

  From 
  the 
  values 
  of 
  /3 
  and 
  S 
  given 
  above, 
  we 
  get 
  the 
  

   following 
  results 
  : 
  — 
  

  

  . 
  

  

  P. 
  

  

  •00 
  

  

  8'. 
  

   •000 
  

  

  w-n 
  

  

  •000 
  

  

  a. 
  

   •00 
  

  

  1 
  

   b. 
  

  

  •ooooo 
  

  

  Copper 
  

  

  Silver 
  

  

  1509 
  

   0906 
  

   2163 
  

   1327 
  

   0959 
  

   0966 
  

   1206 
  

  

  590 
  

   393 
  

  

  700 
  

   468 
  

  

  488 
  

   533 
  

  

  478 
  

  

  460 
  

   245 
  

   731 
  

   431 
  

   236 
  

   217 
  

   364 
  

  

  4125 
  

   3910 
  

  

  4396 
  

   4096 
  

   3901 
  

  

  3882 
  

   4029 
  

  

  167 
  

  

  090 
  

   270 
  

   158 
  

   087 
  

   080 
  

   133 
  

  

  Tin 
  

  

  

  Lead 
  

  

  Zinc 
  

  

  

  

  