Am.^Jour.^pharm.|  Detevmmation  of  Ethyl  Nitrite.  185 
closer  results  might  be  obtaiued  by  saturating  the  soda  solution  with 
common  salt. 
Although,  up  to  the  present  time,  Eykman's  process  has  been  rightly 
regarded  as  that  furnishing  the  nearest  approach  to  accuracy  in  the 
assay  of  spirit  of  nitrous  ether  and  kindred  preparations  for  ethyl 
nitrite,  I  believe  the  following  method  will  practically  replace  it  in  the 
future.  It  suggested  itself  to  me  when  attempting  to  estimate  ethyl 
nitrite  by  the  process  of  Mr.  D.  B.  Dott  ("  Pharmaceutical  Journal,'^ 
[3],  XV,  200,  492,  592),  which  consists  in  treating  the  sample  with  an 
acidulated  solution  of  potassium  iodide,  and  estimating  the  nitrous 
compounds  from  the  amount  of  iodine  liberated,  as  ascertained  by 
titration  with  decinormal  solution  of  sodium  thiosulphate.  It  is  evi- 
dent that  the  ethyl  nitrite  may  be  estimated  either  from  the  amount  of 
iodine  set  free  or  from  the  volume  of  nitric  oxide  evolved  in  the 
reaction,  the  former  being  Dott's  method,  and  the  latter  the  one  to 
which  I  now  desire  to  call  attention. 
The  assay  might  no  doubt  be  made  in  an  apparatus  similar  to  that 
employed  by  Eykman,  but  I  have  found  the  following  method  of 
manipulation  far  more  simple  and  satisfactory  in  every  sense.  It  is 
the  outcome  of  a  large  number  of  experiments  made  to  test  the  con- 
stancy of  the  results  under  different  conditions  of  operating.  A  nitro- 
meter should  be  filled  with  strong  brine  and  5  cc.  of  the  samj^le  to  be 
tested  should  then  be  placed  in  the  cup  of  the  nitrometer,  and  allowed 
to  enter  through  the  tap,  taking  care  that  no  air  gets  in  at  the  same 
time.  Five  cc.  of  a  strong  solution  of  potassium  iodide  is  next 
allowed  to  enter,  and  this  is  followed  by  about  5  cc.  of  dilute  sulphuric 
acid.  Effervescence  immediately  ensues,  and  if  the  tube  be  vigorously 
agitated  at  intervals,^  the  reaction  is  complete  in  .five  minutes,  when 
the  level  of  the  liquid  in  the  two  limbs  of  the  nitrometer  is  adjusted, 
and  the  volume  of  nitric  oxide  gas  read  off.  If  the  volume  of  gas 
evolved  be  small,  another  5  cc.  of  the  sample  should  be  let  into  the 
nitrometer,  and  the  agitation  repeated.  The  calculation  is  the  same  as 
in  Eykman's  process,  the  reaction  being : 
(C2H,)N02  +  KI  -f-  H^SO,  =  (C2H,)0H  -f  KHSO,  +  I  +  NO. 
Thus,  75  parts  by  weight  of  ethyl  nitrite  evolve  30  of  nitric  oxide 
^  This  should  be  efltected  by  holding  the  tapped  tube  vertically  and 
agitating  the  contents  by  a  transverse  motion.  In  this  way  the  reaction  is 
readily  confined  to  the  upper  part  of  the  liquid  iu  the  closed  side  of  the 
nitrometer. 
