88  Detection  and  Determination  of  Alcohol.  { Xm'J™\l%!m- 
stances  (e.g.  aldehyd,  acetone,  methyl  alcohol,  propyl  alcohol,  volatile 
oils,  gum,  sugar,  lactic  acid,  etc.).  Another  general  test  is  Hardy's 
(abs.  Y.  B.  P.,  1872,  161),  depending  upon  the  production  of  a  blue 
color  when  guaiacum  resin,  hydrocyanic  acid  and  sulphate  of  copper 
are  added  to  an  aqueous  solution  of  alcohol,  but  it  is  incapable  of 
detecting  less  than  1  part  in  150.  In  the  same  Year-Book  (page  16a) 
is  an  abstract  of  a  method  proposed  by  M.  Berthelot.  Benzoic 
chloride  is  mixed  with  the  suspected  solution,  the  mixture  warmed, 
and  a  little  caustic  potash  added,  when  a  characteristic  odor  of  benzoic 
ether  is  evolved  if  alcohol  be  present.  This  test  is  very  sensitive 
with  a  4  or  5  per  cent,  aqueous  solution,  but  when  smaller  proportions 
are  present,  or  the  fluid  contains  other  odorous  substances,  it  is  quite 
useless. 
The  old  chromic  acid  test  as  hitherto  applied  is  perhaps  the  most 
fallible  of  all,  since  innumerable  substances  give  the  same  reaction,  in 
virtue  of  their  greed  for  oxygen.  A  more  modern  test  is  that  proposed 
by  Davy  (abs.  Y.  B.  P.,  1877,  I09)>  ana  depends  upon  the  production 
of  a  blue  coloration  when  molybdic  acid  dissolved  in  sulphuric  acid  is 
added  to  the  fluid  containing  alcohol,  but  Hager  (abs.  Y.  B.  P.,  1877, 
285)  failed  entirely  to  obtain  the  reaction.  Besides  these  general  tests 
a  great  many  others  have  been  proposed  for  use  in  special  cases  ;  thus 
Boettger  (abs.  Y.  B.  P.,  1873,  J^4)  recommends  solid  caustic  potash 
for  detecting  alcohol  in  ether,  and  anhydrous  glycerin  for  its  detection 
in  essential  oils.  Fuchsin,  tannin,  anilin  red,  jalap  resin  and  many 
other  substances  have  also  been  similarly  employed,  but  scarcely  one 
of  the  tests  enumerated,  it  is  worthy  of  mark,  is  capable  of  being 
employed  to  make  even  an  approximate  quantitative  determination  of 
the  ethyl  hydrate. 
Some  time  ago,  when  examining  a  solution  containing  aldehyd,  I 
was  struck  by  the  remarkable  delicacy  of  the  caustic  potash  reaction  ; 
I  found  that  an  aqueous  solution  containing  one  one-thousandth  part  of 
pure  aldehyd,  when  boiled  with  a  fragment  of  solid  caustic  soda  or 
potash  and  allowed  to  stand  for  a  little  while,  exhibited  a  distinct  yellow 
color.  If  a  solution  containing  *5  per  cent,  of  aldehyd  be  thus  treated, 
the  liquid  becomes  deep  yellow,  and  a  flocculent  yellow  precipitate 
gradually  subsides,  and  this  solution  when  diluted  first  with  a  little 
alcohol  and  then  with  50  volumes  of  water  still  exhibits  a  yellow  tint 
