548  History  of  Argon  and  Helium.       { A™oveZe™m™' 
found  that  nitrogen  extracted  from  chemical  compounds  is  about  0  5 
per  cent,  lighter  than  atmospheric  nitrogen.  With  Professor  Ram- 
say he  then  made  nitrogen  in  six  different  ways,  and  determined  the 
density  of  each  kind  of  nitrogen.  Let  us,  for  convenience,  call  this 
nitrogen  obtained  from  chemical  substances,  "  chemical  nitrogen  " 
to  distinguish  it  from  "  atmospheric  nitrogen,"  as  this  term  had  been 
understood  prior  to  the  autumn  of  1894. 
The  chemical  nitrogen  was  obtained  : 
(1)  From  nitrous  oxide,  and  gave  a  density  of  2*2990 
(2)  From  nitric  oxide,  and  gave  a  density  of  2*3001 
(3)  From  ammonium  nitrite  purified  at  a  red  heat,  and  gave  a 
density  of    .  ,  2-2987 
(4)  From  ammonium  nitrite  purified  in  the  cold,  and  gave  a 
density  of  2*2987 
(5)  From  urea  by  means  of  sodium  hypobromite,  and  gave  a 
density  of  2*2985 
(6)  From  air  by  means  of  hot  magnesium  and  subsequent  liber- 
ation as  ammonia  and  conversion  into  nitrogen  by  means 
of  calcium  hypochlorite,  and  gave  a  density  of  2*29918 
The  mean  of  all  these  densities  is  2*2990,  which  is  hence  the  density 
of  chemical  nitrogen,  while,  as  a  mean  of  three  experiments,  the 
density  of  atmospheric  nitrogen  was  found  to  be  2-3102. 
The  difference  was  naturally  at  first  attributed  to  impurities,  but 
none  such  in  form  of  a  lighter  gas  could  be  found  in  the  chemical 
nitrogen.  The  next  explanation  was  that  some  of  the  nitrogen 
molecules  of  the  chemical  nitrogen  had  been  split  up  into  atoms. 
Both  kinds  of  nitrogen  were  then  tested  in  this  particular  by 
subjecting  them  to  the  silent  electrical  discharge,  and  both  were 
found  to  retain  their  volume  unaltered,  thus  dispelling  the  possibility 
of  this  explanation.  There  remained,  then,  only  as  the  explanation 
of  the  discrepancy  in  weight,  the  fact  that  one  of  the  gases  was  a 
mixture  of  several  gases.  To  assume  that  chemical  nitrogen  was  a 
mixture  was  contrary  to  experience  and  to  Stas'  work.  Hence  the 
hypothesis  was  made  that  atmospheric  nitrogen  contains  some- 
thing else  than  nitrogen.  The  identification  of  "phlogisticated  air" 
with  the  ^constituent  of  nitric  acid  is  due  to  Cavendish,  whose 
method  consisted  in  operating  with  electric  sparks  upon  a  short  col- 
umn of  gas  confined  with  potash  over  mercury  at  the  upper  end  of 
an  inverted  U  tube1.    The  repetition  of  Cavendish's  experiments 
1  Cavendish — Experiments  on  Air;  Phil.  Trans.,  75,  372  (1785). 
