recent  attack  upon  the  Atomic  Theory.  429 
Dissected  (Structural)  Formulae,"  in  which  he  attempts  to  prove 
that  "the  conceptions  involved  in  the  atomic  hypothesis  are 
both  unnecessary  and  insufficient."  The  following  remarks  are 
intended  to  show  that  whenever  the  author  has  the  choice  of 
two  ideas,  he  invariably  selects  the  one  in  accordance  with  the 
atomic  theory,  thus  himself  refuting  the  assertion  that  it  is  un- 
necessary. The  answer  to  the  argument  of  insufficiency  is  that 
the  theory  is  a  growing  one,  not  one  which  has  reached  its  limit ; 
and  the  fact  that  nothing  has  been  discovered  which  is  inconsis- 
tent with  it  is  strong  evidence  in  favour  of  its  truth. 
After  defining  the  terms  "  volume  "  and  "  vapour-density," 
the  author  deduces  the  volumetric  ratio  of  the  gaseous  consti- 
tuents of  a  compound  from  its  composition,  its  vapour-density, 
and  the  densities  of  its  constituents.  He  then  asserts  that  "  ex- 
periment shows  that  a  volume  of  the  vapour  of  almost  all  gase- 
nable  compounds  contains  the  gasefiable  elements  in  the  propor- 
tion of  1  I-,  § ,  4,  4,  .  .  .  .  volumes."  Taking  at  random  one  of 
the  most  recent  determinations,  including  all  the  data  required, 
let  us  see  whether  the  volumetric  ratio  of  the  constituents  of  the 
oxychloride  of  tungsten  conforms  to  this  so-called  experimental 
law.     Analysis  gave  the  percentage  numbers  : 
Tungsten  .  .  =  53*89 
Chlorine  .  .  .  =  41-11 
Oxvffen  =    5*00 
J  o  ... 
100*00 
Vapour- density  from' four  experiments  =172*4.  These  num- 
bers show  that  one  volume  of  the  oxychloride  contains  0*539 
volume  of  oxygen  and  1*996  volume  of  chlorine;  that  is,  the 
oxygen  occurs  in  the  ratio  ^  or  f_,  and  the  chlorine  in  the 
ratio  — j— .  These  numbers  approach  the  ratios  respectively 
J  and  4 ;  but  they  differ  sufficiently  from  them  to  compel  us  to 
adopt  one  of  the  alternatives,  either  they  are  the  correct  num- 
bers or  they  are  not.  If  they  are  the  correct  numbers,  then  the 
elements  do  not  combine  in  definite  volumetric  ratios ;  if  they 
are  not,  then  in  correcting  them  we  assume  the  atomic  theory. 
The  theory  of  the  atomic  constitution  of  matter  leads  us  to  reject 
the  experimental  numbers  and  to  adopt  the  simple  ratios;  but 
if  we  reject  that  theory,  we  are  bound  to  retain  the  numbers  ob- 
tained by  experiment.  The  assertion  of  Dr.  Wright,  that 
vapours  contain  the  elements  in  simple  ratios,  shows  that  he 
does  not  accept  the  first  alternative,  and  consequently  that  he 
adopts  (unconsciously  no  doubt)  the  atomic  theory  to  account 
for  the  discrepancy  'between  the  theoretical  and  experimental 
