.  Am.  Jour.  Pharrn.  \ 
February,  1898.  J 
Testing  of  Formaldehyde, 
9* 
Per  Cent. 
Without  methyl  alcohol  .... 
With  pure     ;<        "  .... 
"     commercial  methyl  alcohol 
37'4>  37  "4 
37  "5.  37*6 
36'8,  36-9 
According  to  these  figures,  commercial  methyl  alcohol  may  lower 
the  result  slightly,  while  pure  methyl  alcohol  and  acetone  exert, 
practically,  no  influence.  This  method,  then,  is  the  only  thoroughly 
reliable  one  of  all  those  tried,  the  only  drawbacks  being  a  slight 
inaccuracy  introduced  through  loss  of  ammonia  by  volatilization, 
and  the  necessity  of  frequently  restandardizing  the  ammonia  solu- 
tion, which  loses  strength  rapidly.  These  considerations  suggested 
to  the  writer  a  modification  of  this  method,  which  obviates  the  need 
of  keeping  on  hand  a  standard  solution  of  ammonia,  and  reduces  the 
volatilization  of  ammonia,  during  the  manipulation,  to  a  minimum. 
Ammonia  Method  Modified. — Dissolve  2  grammes  of  pure,  neutral 
ammonium  chloride  in  25  c.c.  of  water  and  introduce  it  into  a  flask 
provided  with  a  well-fitting  stopper.    Add  2-25  grammes  of  the 
n 
sample,  and  then  run  in  from  a  burette  25  c.c.  of  _  potassium  (or 
sodium)  hydrate.  Stopper  the  flask  at  once  and  put  it  aside  for 
one-half  hour.    Then  add  a  few  drops  of  rosolic  acid  solution  and 
n 
determine  the  excess  of  ammonia  with  -  sulphuric  acid,  each  c.c.  of 
1 
n 
-  potassium  hydrate  consumed  indicating  0-5  per  cent  of  formalde- 
hyde. 
The  results  obtained  by  this  modification  agree  closely  with  those 
obtained  by  the  hydroxylamine,  fixed-alkali,  and  ammonia  methods. 
The  reactions  involved  are  as  follows  : 
NH4C1  -  KOH  =  NH4OH  +  KC1 ;  4NH4OH  -  6CH20  = 
4NH4C1  +  4KOH  +  6CH20  —  N4(CH2)6  +  4KCI  -  10H.O. 
The  ammonia  combines  with  the  formaldehyde  nearly  as  fast  as  it 
is  liberated,  and,  consequently,  has  no  chance  to  volatilize,  and  the 
final  excess  is  so  small  that  the  odor  is  barely  perceptible. 
Sample  I. — Colorless,  contains  a  white,  flocculent  precipitate 
probably  para-formaldehyde,  which  increases  on   keeping,  more 
N4(CH2)6  +  ioH20, 
or. 
EXAMINATION  OF  COMMERCIAL  SAMPLES. 
