oSobeTi9^2a"°' }        Estimation  of  Methyl  Alcohol.  721 
and  Reports  of  Inspector  of  Foods,  Local  Government  Board,  1909, 
N.  S.  No.  12;  also  Analyst,  34:  470,  1909).  The  method  in  question 
consisted  in  treating  the  solution  containing  the  formaldehyde  with 
I  per  cent,  solution  of  phenylhydrazine  hydrochloride,  then  adding 
potassium  ferricyanide,  and  afterwards  concentrated  hydrochloric 
acid,  when,  in  the  presence  of  the  aldehyde,  a  brilliant  fuchsine- 
like  color  was  produced,  which  could  be  detected  even  when  the 
concentration  of  formaldehyde  did  not  exceed  i  in  2,000,000.  This 
reaction  forms  the  basis  of  the  method  for  estimating  methyl  alcohol 
described  in  the  present  communication. 
Several  reagents  are  capable  of  oxidizing  methyl  alcohol  to 
formaldehyde,  but  most  of  these  interfere  with  the  application  of  the 
test  described  above.  This  was  not  found,  however,  to  be  the  case 
when  ammonium  persulphate  was  used.  If  an  excess  of  the  per- 
sulphate is  added,  the  formaldehyde  is  oxidized  further  to  carbon 
dioxide  and  water.  By  ascertaining  the  amount  of  persulphate 
that  must  be  added  to  a  given  solution  of  methyl  alcohol  in  water, 
so  that  after  completion  of  the  reactions  under  certain  standard 
conditions  of  experiment  formaldehyde  just  fails  to  be  detectable 
by  the  above-mentioned  reaction,  the  percentage  of  the  alcohol 
present  in  the  water  can  be  estimated  with  a  considerable  degree  of 
accuracy,  even  when  the  amounts  present  are  as  low  as  0.0005 
per  cent. 
When  ammonium  persulphate  is  heated  with  a  solution  of  methyl 
alcohol  three  reactions  can  take  place  simultaneously,  viz.,  the  de- 
composition of  the  persulphate  itself,  the  oxidation  of  methyl  alcohol 
to  formaldehyde,  and  the  oxidation  of  the  aldehyde.  It  is,  therefore, 
necessary  to  specify  accurately  the  conditions  under  which  the  es- 
timation should  be  carried  out,  as  is  the  case  in  other  oxidation 
methods,  such  as  that  of  Fehling's,  where  several  oxidation  processes 
take  place  at  the  same  time. 
The  method  of  carrying  out  the  estimation  is  as  follows :  Samples, 
each  of  5  Cc.  of  the  methyl  alcohol  solution  in  water  are  introduced 
into  a  series  of  test-tubes  (6  inches  by  ^/ 4  inch) ,  and  mixed  with  5  Cc. 
of  ammonium  persulphate  solution  of  varying  concentrations.  The 
test-tubes  containing  these  mixtures  are  then  inserted  into  a  water 
bath  containing  boiling  water,  in  which  they  are  kept  immersed  for 
10  minutes.  At  the  end  of  this  period  i  Cc.  of  the  mixture  is  pipetted 
out  from  each  tube  and  introduced  into  a  series  of  smaller  test-tubes, 
each  of  which  contains  i  Cc.  of  i  per  cent,  solution  of  phenyl-hy- 
