Am.  Jour.  Pharm.  > 
Jan.  1, 1872.  > 
The  Goler  of  f  luorescent  Solutions. 
19 
general  conclusions,  nor  the  main  observations  on  which  they  were 
founded,  throws  out  one  of  the  corroborative  experiments  by  which  I 
thought  that  they  might  be  established  when  a  spectroscope  was  not 
at  hand. 
Obtaining  some  very  anomalous  results  of  late,  I  was  led  to  mis- 
trust the  action  of  the  Geissler  tubes  in  which  the  solutions  had  been 
examined. 
They  were  of  the  ordinary  kind  of  jacketed  spirals,  selected  as  be- 
ing nearly  identical  in  size  and  other  particulars. 
It  had  been  observed  from  the  first  that  the  internal  spiral  gave  a 
faint  blue  fluorescence  which  could  only  be  seen  on  close  inspection ; 
and  in  all  cases,  the  tube  being  but  partly  filled,  it  was  considered 
that  a  light  appearing  in  the  part  covered  by  the  fluid,  many  times 
more  bright  than  that  from  the  uncovered  part  of  the  spiral,  was  suf- 
ficient evidence  of  fluorescence  in  the  liquid. 
Late  experiments  have,  however,  proved  that  this  was  not  so.  Any 
liquid,  however  devoid  of  fluorescent  properties,  gives  all  the  appear- 
ance of  fluorescing  in  these  tubes,  and  on  a  little  thought  the  cause  of 
this  became  clear. 
The  only  fluorescent  light  that  can  be  seen  from  the  glass  of  the 
spiral  is  that  which  comes  off  tangentically  from  the  outer  surface, 
that  emitted  radially  being  marked  by  the  bright  electric  discharge 
behind. 
In  passing  from  the  glass  to  air,  most  of  the  light  will  suffer  total 
reflection  at  the  outer  surface  of  the  glass,  but  if  water  or  any  other 
liquid  is  substituted  for  the  air,  its  greater  refracting  power  (approach- 
ing that  of  glass)  will  diminish  the  above-named  action,  so  that  much 
more  of  the  light  will  reach  the  eye.  The  truth  of  this  explanation 
was  supported  by  the  observation  that  the  nearer  the  index  of  refrac- 
tion in  the  liquid  came  to  that  of  glass,  the  brighter  was  the  light 
seen  through  it,  while  a  liquid  of  higher  refraction,  like  carbon  bisul- 
phide, seemed  a  little  to  weaken  the  effect  by  diffusion. 
This  fact  renders  of  no  account  the  observations  before  made  on  fil- 
tered and  diluted  solutions  of  turmeric,  but  a  fresh  observation  with 
the  spectroscope  on  tubes  free  from  fluorescence  has  fully  confirmed 
my  former  conclusions  as  to  the  true  color  of  fluorescence  in  this 
liquid. 
No  correction  need  be  applied  to  the  description  already  published 
in  the  case  of  the  asphalt  solution,  but  I  may  add  to  what  was  there 
stated  another  striking  example. 
