122 
The  Use  of  Methyl  Alcohol. 
(  Am.  Jour.  Pharm. 
\      March,  1905. 
from  the  street-lamps  and  suck  cologne  from  the  penny-in-the-slot 
machine — but  I  cannot  substantiate  the  statement  recently  made 
by  the  Dairy  and  Food  Commissioner  of  Pennsylvania  that  methyl 
alcohol  is  a  common  adulterant  of  the  cheaper  grades  of  whiskies. 
I  learn  with  regret  that  the  test  for  methyl  alcohol  to  be  em- 
ployed in  the  new  pharmacopoeia  will  distinguish  but  2  per  cent, 
of  methyl  alcohol  in  ethyl  alcohol,  resorcinol  being  more  delicate 
for  the  final  test  than  phloroglucinol.  If  the  quantity  of  sub- 
stance at  hand  is  large,  fractional  distillation  will  reduce  the  deli- 
cacy of  this  test  many  fold.  There  is  one  peculiarity  in  testing 
formaldehyde  that  it  may  be  well  to  call  attention  to,  namely,  that 
the  iron  test  seems  to  be  much  more  delicate  when  performed  in 
milk  than  under  other  conditions.  In  a  U.  S.  Bulletin,  just  issued, 
the  Association  of  Agricultural  Chemists  recommends  that  the  fol- 
lowing methods  for  discovering  the  presence  of  formaldehyde  be 
given  trial.  These  tests  are  easy  of  application,  and  are  specially 
suited  to  the  detection  of  this  widely-used  preservative  in  milk. 
To  familiarize  oneself  with  Rimini's  method  take  15  c.c.  of  a  very 
dilute  formaldehyde  solution,  and  treat  with  1  c.c.  of  a  dilute  solution 
of  phenylhydrazin  hydrochlorid,  then  with  a  few  drops  of  freshly- 
prepared  sodium  nitroprussid  solution,  and  finally  with  concentrated 
caustic  soda  solution.  A  blue  color  is  formed,  which,  after  a  long 
time,  changes  to  red.  This  reaction  is  capable  of  indicating  formal- 
dehyde in  milk,  even  in  a  dilution  of  1  in  30,000.  Ferric  chlorid 
may  be  used  instead  of  the  sodium  nitroprussid,  to  be  followed  by 
a  concentrated  hydrochloric  acid,  in  place  of  the  caustic  soda.  By 
so  doing  a  red  color,  which  changes  after  some  time  to  orange 
yellow,  will  show  itself  in  the  presence  of  formaldehyde.  With 
meats  and  fats  the  formaldehyde  should  first  be  extracted  with 
alcohol  and  the  filtrate  tested.  Milk  may  also  be  shaken  with 
an  equal  volume  of  absolute  alcohol,  and  the  filtrate  tested.  When 
the  reagents  are  applied  to  dark  beers  the  coloration  may  be 
determined  by  the  color  of  the  froth,  just  as  one  can  do  in  the 
case  of  the  diazo  reaction  in  urine.  In  Rideal's  test  (Analyst,  1895), 
IOO  c.c.  of  milk,  suspected  to  contain  formaldehyde,  is  distilled,  and 
Schiffs  reagent  (a  colorless  solution  of  fuchsin  and  sodium  sulphite) 
is  added  thereto,  a  violet  red  color  denotes  formaldehyde.  Selig- 
mann  (  Zeitschr.  f.  Hyg.  u.  Infectionskrankh.,  Vol.  XIL,  No.  2,  p. 
325,  1905)  detects  1-40,000  parts  of  formaldehyde  in  milk,  as  fol- 
