396 
Formaldehyde  and  Cane  Sugar, 
j  Am.  Jour,  that-til. 
\      August,  1909. 
dchyde  had  been  added,  and  that  the  aniline  acetate  test  gives  posi- 
tive results  for  furfuraldehyde  in  every  solution  except  those  pre- 
pared with  formaldehyde  alone,  in  which  it  could  not  possibly  have 
been  present  except  as  an  impurity. 
From  the  nature  of  the  reaction  it  would  be  fair  to  conclude  that 
if  formaldehyde  were  produced  under  such  conditions  as  have  been 
described  it  would  be  found  in  a  concentrated  form  in  the  residue. 
The  residue  in  the  flasks,  therefore,  after  the  distillation  of  the  plain 
cane  sugar,  the  levulose,  and  the  two  lots  of  acidulated  cane  sugar 
solutions,  were  all  tested  for  formaldehyde  by  the  Rimini  test,  carry- 
ing alongside  blanks  prepared  from  part  of  the  same  residue,  to 
which  formaldehyde  in  the  proportion  of  1-100,000  had  been  added 
to  ascertain  if  there  was  any  substance  present  which  would  inter- 
fere with  the  detection  of  formaldehyde  under  these  conditions. 
These  tests  resulted  negatively  in  each  case  where  the  plain  residue 
was  tested  and  positively  in  each  case  where  formaldehyde  was 
known  to  be  present. 
Check  experiments  were  then  made  with  the  distillate  from  bran 
containing  furfuraldehyde  freshly  prepared  and  it  was  found  that 
when  this  distillate  was  very  much  concentrated  there  was  a  slight 
observable  difference  in  the  color  of  the  Hehner  test  as  compared 
with  that  produced  by  formaldehyde  alone,  but  that  diluted  solutions 
of  furfuraldehyde  could  not  be  distinguished  from  formaldehyde, 
either  in  the  intensity  or  the  color  of  the  reaction. 
A  number  of  samples  of  commercial  jellies  and  several  of  home 
origin  were  then  distilled  and  the  distillate  tested  by  the  foregoing 
methods.  In  every  case  the  Hehner  test  gave  positive  results,  as 
did  also  the  aniline  acetate  test  for  furfuraldehyde,  while  in  not  a 
single  instance  could  even  a  faint  reaction  be  obtained  by  the  Rimini 
test,  either  in  the  distillate  or  in  the  concentrated  residue  left  after 
the  distillation. 
The  conclusion  is  obvious,  therefore,  that  cane-sugar  solutions  do 
not  develop  formaldehyde  when  boiled  under  ordinary  conditions, 
but  that  furfuraldehyde  is  produced,  which  reacts  in  such  a  manner 
with  the  Hehner  test  as  to  deceive  the  analyst  who  relies  upon  it 
alone,  without  confirmation  by  the  Rimini  test. 
