﻿G, 
  M, 
  Thomson, 
  — 
  On 
  the 
  Fertilization 
  of 
  some 
  New 
  Zealand 
  Orchids. 
  419 
  

  

  lie 
  very 
  closely 
  ensconced 
  in 
  the 
  anther 
  case. 
  For 
  fertilization 
  by 
  insects, 
  

   however, 
  the 
  parts 
  are 
  very 
  simply 
  fitted. 
  The 
  rostelliun 
  with 
  the 
  attached 
  

   caudicle 
  projects 
  to 
  a 
  small 
  point, 
  and 
  is 
  viscid 
  on 
  the 
  edge 
  and 
  mider-surface. 
  

   An 
  insect 
  visiting 
  the 
  flower 
  would 
  insert 
  its 
  head 
  or 
  proboscis 
  into 
  the 
  

   email 
  square 
  aperture 
  between 
  the 
  labellum 
  and 
  the 
  column, 
  and 
  in 
  with- 
  

   drawing 
  would 
  inevitably 
  touch 
  the 
  viscid 
  surface. 
  The 
  slightest 
  touch 
  

   brings 
  away 
  the 
  pollinia, 
  usually 
  all 
  four, 
  but' 
  sometimes 
  only 
  two. 
  Were 
  

   they 
  to 
  be 
  withdrawn 
  just 
  as 
  they 
  lie 
  on 
  the 
  summit 
  of 
  the 
  column, 
  they 
  

   would 
  hardly 
  be 
  in 
  the 
  position 
  to 
  strike 
  the 
  stigmatic 
  surface 
  of 
  another 
  

   flower 
  ; 
  but 
  in 
  being 
  withdrawn, 
  the 
  cap 
  of 
  the 
  anther 
  pulls 
  them 
  slightly 
  

   downwards 
  and 
  depresses 
  the 
  caudicle 
  considerably. 
  I 
  repeatedly 
  imitated 
  

   the 
  action 
  with 
  the 
  point 
  of 
  a 
  pencil 
  or 
  needle, 
  and 
  found 
  in 
  every 
  case 
  that 
  

   the 
  pollinia 
  came 
  away 
  readily, 
  and 
  were 
  depressed 
  considerably 
  below 
  a 
  

   right-angle 
  to 
  the 
  surface 
  to 
  which 
  they 
  adhered. 
  In 
  this 
  position 
  they 
  

   were 
  easily 
  placed 
  on 
  the 
  stigma 
  of 
  a 
  second 
  flower. 
  I 
  carefully 
  examined 
  

   91 
  flowers, 
  and 
  found 
  that 
  the 
  pollinia 
  had 
  been 
  removed 
  from 
  the 
  anthers 
  

   in 
  41 
  cases, 
  and 
  remained 
  intact 
  in 
  50 
  ; 
  this 
  too 
  in 
  bright, 
  warm 
  weather. 
  

   Of 
  course 
  the 
  plants 
  were 
  not 
  in 
  their 
  native 
  habitat, 
  which 
  might 
  account 
  

   partly 
  for 
  lack 
  of 
  the 
  usual 
  bush-frequenting 
  insects. 
  Those 
  spikes 
  pro= 
  

   minently 
  placed 
  on 
  the 
  plant 
  usually 
  had 
  the 
  pollinia 
  of 
  their 
  flowers 
  more 
  

   or 
  less 
  removed, 
  while 
  those 
  which 
  were 
  buried 
  among 
  the 
  leaves 
  had 
  not 
  

   as 
  a 
  rule 
  been 
  visited. 
  

  

  Tribe 
  Malaxe.e. 
  

   (2.) 
  DendroUnm 
  cunninghamii. 
  

  

  This 
  beautiful 
  orchid 
  has 
  its 
  flowers 
  evidently 
  fitted 
  for 
  cross-fertiliza- 
  

   tion. 
  The 
  upper 
  sepal 
  is 
  lanceolate 
  in 
  form, 
  and 
  is 
  the 
  smallest 
  in 
  the 
  

   whorl. 
  The 
  lateral 
  sepals 
  are 
  broad 
  at 
  the 
  base, 
  and 
  adnate 
  to 
  the 
  

   produced 
  base 
  of 
  the 
  column. 
  The 
  lateral 
  petals 
  are 
  linear 
  oblong. 
  

   The 
  labellum 
  is 
  widely 
  expanded 
  above 
  the 
  middle, 
  with 
  two 
  small 
  lateral 
  

   lobes, 
  and 
  bears 
  on 
  its 
  face 
  five 
  elevated 
  ridges 
  or 
  plates. 
  It 
  is 
  attached 
  

   to 
  the 
  base 
  of 
  the 
  column 
  by 
  a 
  short 
  and 
  very 
  elastic 
  claw. 
  The 
  column 
  

   stands 
  in 
  the 
  flower 
  exactly 
  like 
  the 
  letter 
  J, 
  being 
  produced 
  forward 
  at 
  the 
  

   base, 
  and 
  terminated 
  by 
  a 
  large 
  green 
  glandular 
  swelling. 
  "When 
  the 
  

   flower 
  is 
  open, 
  a 
  small 
  drop 
  of 
  nectar 
  is 
  always 
  found 
  at 
  the 
  base 
  of 
  

   this 
  swelling. 
  The 
  erect 
  portion 
  or 
  limb 
  of 
  the 
  column 
  is 
  rather 
  long. 
  

   The 
  anther 
  is 
  terminal, 
  and 
  encloses 
  four 
  narrow 
  and 
  flatfish 
  pollen 
  

   masses, 
  attached 
  in 
  pairs 
  to 
  a 
  strap-shaped 
  caudicle. 
  Tlie 
  stigmatic 
  

   surface 
  is 
  placed 
  slightly 
  below 
  it, 
  and 
  is 
  nearly 
  square. 
  "When 
  in 
  the 
  

   bud, 
  the 
  lower 
  surface 
  is 
  hollow^ed 
  into 
  a 
  deep 
  pit, 
  and 
  on 
  its 
  summit, 
  

   standing 
  directly 
  in 
  front 
  of 
  the 
  bases 
  of 
  the 
  pollinia, 
  is 
  the 
  rostellum, 
  

   "which 
  at 
  this 
  early 
  stage 
  is 
  membranous. 
  As 
  the 
  flower 
  opens, 
  the 
  cells 
  

  

  