Pedigree  Seed  Corn. 
129 
young  plant  daring  the  earlier  period  when  it  depends  for 
nutrition  on  its  own  endosperm  content.  In  the  following 
year,  when  the  progeny  of  both  parcels  of  seed  were  resown 
I found  no  significant  difference  of  yield,  and  there  was  of  course 
no  evidence  that  the  racial  character  had  been  changed  by  the 
effects  of  environment.  This  instance  is  only  of  interest  in 
this  particular  connection  as  illustrating  one  of  the  precautions 
necessary  with  reference  to  changes  of  locality  when  making 
“ variety  tests  ” of  yield.  It  may  be  noted  also,  however,  in 
passing, that  effects  of  “change  of  seed”  (well  recognised  in  all 
ages  in  a vague  and  general  way  to  be  of  great  importance),  is 
evidently  a matter  calling  for  more  systematic  study.'  It  is 
certain  that  change  of  locality  may  lead  with  mixed  races  to  a 
complete  alteration  of  the  racial  character  of  the  aggregate.  On 
the  other  hand  it  may  possibly  give  a result  the  first  year  after 
transferrence,  due  merely  to  physical  differences  in  the  seed 
not  affecting  racial  character,  this  latter  being  probably  the 
only  result  to  be  expected  in  the  case  of  a pure  race. 
Lastly,  What  effects  of  natural  selection  may  be  expected 
with  a hybrid  cereal  race  ? 
It  is  doubtful  at  present  whether  we  can  be  quite  certain 
that  any  single  plant  selected  from  the  progeny  resulting  from 
cross-fertilisation,  as  the  starting  point  for  a new  race,  will  give 
us  a pure  race  which  is  constant  in  the  same  sense  as  when 
there  has  been  no  crossing.  Even  if  the  Mendelian  hypothesis 
of  unit  characters  is  fully  accepted,  there  are  evidently  so  large 
a number  of  “ characters  ” which  may  be  genetically  different 
to  a greater  or  less  extent  in  the  original  pair  of  parent  plants, 
that  some  amount  of  “ splitting,”  especially  in  the  characters 
which  are  not  obvious  (or  which  normally  fluctuate  widely 
with  environment),  may  go  on  but  pass  unnoticed.  Further 
experience  will  no  doubt  lead  to  greater  certainty  in  extracting 
completely  fixed  individuals  from  amongst  hybrids,  and  no  one 
can  see  fields  of  some  cereals  now  in  general  cultivation,  which 
originally  started  from  hybrids,  and  especially  some  of  Professor 
Biffen’s  hybrid  wheats  without  feeling  sure  that  they  are — at 
any  rate  for  all  practical  purposes — uniform.  If,  however, 
genetic  (as  distinct  from  fluctuating)  variations  do  take  place, 
these  will,  as  the  result  of  crowding  out,  be  either  established 
or  eliminated  in  the  aggregate  in  the  course  of  time,  according 
as  the  conditions  are  favourable  or  not  to  the  survival  of  the 
individual  plants  in  which  they  occur. 
On  the  whole  it  seems  probable  that  with  aggregates 
resulting  from  the  progeny  of  hybrids,  there  will  be  more  risk 
' Lisle  considered  that  “ the  changing  of  all  seed  whatsoever  is  of  as 
much  use  and  service  as  half  the  dung  suflScient  for  a crop.”  Observations 
in  Agriculture,  1767,  page  83. 
VOL.  70. 
K 
