j£ypenmental  Xegtslatjon  23  \
jf*  _  ,  ,  %  \
other.  We  need,  in  fact,  what  the  cherriists  call
a  “blind  experiment.”  Suppose,  for  instance,
that  an  agricultural  chemist  or  a  scientific  farmer*  './
wished  to  ascertain  the  effect  of  a  new  kind  of
manure  ;  would  it  be  rational  for  him  to  spread
the  manure  over  all  his  available  land  ?  Would
it  not,  then,  be  doubtful  whether  the  increase  or
decrease  of  yield  were  due  to  the  manure  or  the
character  of  the  seasons  ?  In  this  case  his  neighbour’s ­
  crops  might,  to  some  extent,  furnish  the
“  blind  experiment,”  showing  what  had  been  the
ordinary  yield.  But,  of  course,  the  obvious  mode
of  procedure  is  to  spread  the  new  manure  over
a  part  only  of  each  experimental  field,  so  that
the  difference  of  the  crops  on  the  different  patches
brings  out,  in  a  most  unquestionable  way,  the
effect  of  the  manure.  Not  only  is  the  smaller
experiment,  in  a  logical  point  of  view,  far  better
than  the  larger  one,  but  it  is  possible  to  try  many
concurrent  small  experiments  upon  a  farm  of
moderate  extent.
I  maintain  that,  if  our  legislators  are  to  act
rationally,  they  will,  as  far  as  possible,  imitate  the
agricultural  chemist.  The  idea,  for  instance,  of
obliging,  or  even  allowing,  all  the  boroughs  in
the  kingdom  simultaneously  to  adopt  the  Gothenburg ­
  plan,  would  be  ridiculous  and  irrational.
The  cost  and  confusion  which  would  arise  from
a  sudden  general  trial  must  be  very  great  ;  many
years  would  elapse  before  the  result  was  apparent.