AmAug.ri?92aruJ' }  American  Pharmaceutical  Association.  449 
three  over  70  per  cent.,  three  over  80  per  cent.,  and  one  9371  per  cent,  of 
amyl  nitrite.  The  preparation  must  be  preserved  with  the  utmost  care,  and 
particularly  protected  from  light.  A  specimen  assaying  65*35  per  cent.,  after 
having  been  kept  on  the  prescription  desk  for  three  months  contained  only 
39/60  per  cent,  of  amyl  nitrite  ;  and  another  sample  assaying  84*235  per  cent., 
after  one  week's  standing  in  a  white  glass  vial  was  reduced  to  71*32  per  cent, 
amyl  nitrite.  The  nature  of  all  the  impurities  present  was  not  determined  ; 
one  sample  contained  much  amyl  valerianate  and  some  butyl  nitrite.  Atten- 
tion is  directed  by  the  author  to  the  fact  that  the  addition  of  methyl  alcohol 
to  amyl  nitrite  converts  the  mixture  into  amyl  alcohol  and  gaseous  methyl 
nitrite  ;  and  that  the  addition  of  ethyl  alcohol  gradually  results  in  the  produc- 
tion of  ethyl  nitrite  and  amyl  alcohol. 
Economic  Percolation  was  the  title  of  a  paper  presented  by  Harry  Vin 
Arny,  in  which  the  results  of  numerous  observations  were  given,  showing  the 
loss  of  alcohol  by  evaporation  in  conducting  the  process  of  percolation.  In 
estimating  this  loss  the  amount  of  percolate  was  taken  into  consideration,  and 
the  marc  remaining  in  the  percolator  was  subjected  to  distillation  in  order  to 
recover  the  alcohol  as  far  as  practicable.  To  guard  against  this  loss  or  reduce 
it  to  a  minimum,  an  air-tight  percolator  was  constructed,  resting  upon  a 
receiver,  terminating  below  with  a  tapering  bottom,  provided  with  a  straight 
faucet  ;  the  two  parts  are  fastened  together  by  means  of  a  screw  joint,  and 
further  connected  by  means  of  a  tube  extending  from  the  top  of  the  percolator 
to  the  top  of  the  receiver.  In  order  to  facilitate  the  withdrawal  of  the  perco- 
late from  the  faucet,  the  top  of  the  receiver  is  provided  with  an  additional 
Orifice  through  which  air  is  admitted,  but  which  is  closed  with  a  screw  cap 
while  percolation  is  going  on. 
The  separation  of  strychnine  and  brucine  was  discussed  in  a  paper  by 
H.  W.  Snow.  Various  processes  for  estimating  the  strychnine  in  mixtures  of 
the  two  alkaloids  were  tried  without  satisfactory  results.  On  treatment  with 
nitric  acid,  sp.  grav.  1*056,  it  was  found  that  a  portion  of  the  strychnine  was 
destroyed.  The  author  regards  the  subject  as  a  most  discouraging  one,  and 
though  he  has  performed  many  experiments,  has  as  yet  nothing  but  negative 
results  to  offer.  It  should,  however,  be  stated  that  the  process  suggested  by 
Gerock  (Amer.  Jour.  Phar.,  1889,  p.  180)  has  not  been  tried  by  Mr.  Snow. 
Aromatic  spirit  of  ammonia. — A.  Conrath  observed  that  using  ammonium 
carbonate  picked  out  of  a  cask  opened  for  the  purpose,  a  preparation  could  be 
made  by  the  pharmacopceial  process,  without  containing  a  permanent  precipi- 
tate. Using  the  carbonate  as  ordinarily  met  with,  a  precipitate  usually  occurred, 
even  after  increasing  the  quantity  of  ammonia  water,  if  the  alcohol  had  been 
added  at  once,  but  on  leaving  the  alkalies  together  for  several  hours  before 
adding  the  alcohol,  no  precipitate  was  formed.  The  author  suggests  the  use 
of  bicarbonate  of  ammonium  as  leading  to  greater  uniformity  in  the  strength 
of  the  preparation. 
American  Potash. — Prof.  Iyloydhas  found  commercial  potash  of  very  uneven 
quality,  and  the  commercial  terms  of  first,  second  and  third  sorts  to  have  no 
definite  significance,  since  the  poorest  potash  is  sometimes  sold  as  first  sorts  ; 
the  adulteration  consists  chiefly  of  table  salt  and  lime.  In  the  casks  examined 
the  carbonate  and  hydrate  combined  assayed  sometimes  as  low  as  16-5  per  cent. 
KOH,  while  casks  were  met  with  assaying  sometimes  as  high  as  91  per  cent. 
