34 
ORIGIN  OP  VARIETIES  IN  PLANTS. 
which  cultivation  has  caused  to  vary,  but  from  as  many  original 
types  as  there  are  perceptible  varieties." 
Although  the  absurdity  of  the  latter  view  seems  evident 
enough,  Decaisne  proceeds  to  refute  it,  by  the  record  of  his  ob- 
servations and  experiments.  He  has  raised  seedlings  from  four 
very  different  varieties  of  Pears,  and  each  of  the  four  gave  rise 
to  a  considerable  number  of  new  varieties,  as  different  from 
each  other  and  from  their  mother,  as  she  was  from  the  greater 
part  of  our  old  varieties.  "  It  is  not  only  in  the  fruit  that  trees 
raised  from  the  same  seed  differ  ;  but  also  in  the  time  of  their 
ripening,  general  appearance,  and  in  the  form  of  their  leaves,  . 
"  Some  have  spines,  some  have  none,  some  have  slender  wood,  in 
some  it  is  stout  and  coarse.  Upon  some  of  the  seedlings  of  the 
old  Poire  d'  Angleterre  the  variation  went  so  far  as  to  produce 
lobed  leaves,  like  those  of  the  Hawthorn  or  of  Pyrus  Japoniea. 
Everything  varies  in  the  Pear-tree,  even  to  its  sap.  As  proof 
of  this  observe  the  very  different  success  of  the  graft,  accord- 
ing to  the  stocks  employed.  All  the  varieties  and  races  of 
Pear-trees  bear  grafting  upon  a  Pear-tree,  that  is  to  say,  upon 
the  wild  Pear-tree  ;  but  all  will  not  take  upon  the  Quince,  as 
for  example  the  Ranee,  Qlairgeau,  Bosc,  JDuchesse  de  Mars, 
&c."  The  flowers  also  showed  very  striking  variations.  As  to 
size  of  fruit  the  transition  is  said  to  be  perfect  from  the  wild 
Pear  of  Blidah,  which  is  only  the  size  of  a  pea,  to  the  Belle 
d*  Anguine  and  dy  Amour,  which  vie  with  a  melon  of  middling 
size. 
That  these  extreme  forms  belong  to  different  species  which 
have  crossed  and  recrossed  thousands  of  times  with  each  other, 
producing  fertile  hybrids,  &c,  Decaisne  will  not  allow  to  be 
probable.  But  it  would  be  difficult  to  disprove  it.  He  admits 
the  crossing,  but  maintains  that  the  constant  fertility,  after 
every  conceivable  cross,  argues  identity  and  not  diversity  of 
species. 
«  Does  the  graft,  as  some  people  maintain,  alter  the  charac- 
ter of  the  variety  V  He  concludes  it  does  not.  "  The  notion 
that  fruit-trees  degenerate  because  they  are  propagated  by  graft- 
ing is  an  error  which  must  be  exposed.  There  is  no  single  fact 
to  prove  it.  Those  which  have  been  cited  depend  upon  totally 
different  causes,  first  and  foremost  among  which  are  climate,  un- 
usable soil,  and  very  often  bad  cultivation  or  a  neglect  of  pru- 
