﻿Aruba, 
  Curaqao, 
  and 
  Bonaire. 
  329 
  

  

  no 
  yellow 
  feathers 
  are 
  to 
  be 
  seen 
  on 
  the 
  throat, 
  and 
  only 
  

   a 
  few 
  scanty 
  feathers 
  on 
  the 
  upper 
  chin 
  are 
  of 
  a 
  pale 
  yellow. 
  

   In 
  nearly 
  all 
  my 
  specimens 
  these 
  feathers 
  are 
  of 
  a 
  more 
  or 
  less 
  

   reddish-brown 
  colour, 
  but 
  this, 
  I 
  believe, 
  is 
  due 
  only 
  to 
  the 
  

   juice 
  of 
  some 
  fruit, 
  as 
  in 
  one 
  rather 
  clean 
  specimen 
  they 
  are 
  

   pale 
  yellow. 
  

  

  (3) 
  While 
  the 
  whole 
  of 
  the 
  band 
  of 
  the 
  wing 
  is 
  yellow 
  in 
  

   C. 
  ochroptera, 
  and 
  only 
  a 
  few 
  scanty 
  red 
  feathers 
  are 
  some- 
  

   times 
  to 
  be 
  seen 
  next 
  the 
  body, 
  the 
  cubital 
  edge 
  in 
  C. 
  roth- 
  

   schildi 
  is 
  bright 
  scarlet^ 
  more 
  or 
  less 
  mixed 
  with 
  yellow 
  

   outwards, 
  but 
  not 
  to 
  a 
  great 
  extent. 
  

  

  (4) 
  The 
  3'ellow 
  shoulder-patch 
  is 
  very 
  much 
  smaller, 
  and 
  

   often 
  quite 
  restricted 
  and 
  mixed 
  with 
  red. 
  The 
  outer 
  bend 
  

   of 
  the 
  wing 
  is 
  not 
  pure 
  yellow, 
  but 
  yellowish 
  green. 
  

  

  (5) 
  The 
  rump 
  and 
  abdomen 
  show 
  less 
  or 
  no 
  blackish 
  edges 
  

   to 
  the 
  feathers, 
  and 
  the 
  abdomen 
  is 
  less 
  distinctly 
  tinged 
  

   with 
  blue. 
  

  

  I 
  may 
  add 
  that 
  the 
  skulls 
  of 
  C. 
  rothschihli 
  appear 
  to 
  be 
  

   decidedly 
  smaller, 
  and 
  that 
  the 
  bills 
  are 
  generally 
  thinner 
  

   and 
  the 
  wings 
  somewhat 
  shorter 
  ; 
  but 
  these 
  are 
  not 
  very 
  

   decisive 
  characters 
  for 
  distinguishing 
  this 
  species, 
  as 
  they 
  are 
  

   not 
  quite 
  constant. 
  In 
  other 
  respects 
  C. 
  rothscJtildi 
  re- 
  

   sembles 
  C. 
  ochroptera. 
  Were 
  it 
  not 
  for 
  the 
  red 
  cubital 
  edge 
  

   and 
  the 
  less 
  bluish 
  tinge 
  of 
  the 
  abdomen, 
  C. 
  rothschildi 
  m.\^h.t 
  

   be 
  said 
  to 
  resemble 
  a 
  young 
  stage 
  of 
  C. 
  ochroptera. 
  

  

  The 
  amount 
  of 
  scarlet 
  at 
  the 
  base 
  of 
  the 
  outer 
  rectrices 
  

   varies, 
  and 
  is 
  sometimes 
  spread 
  over 
  both 
  webs 
  of 
  the 
  second 
  

   and 
  third 
  pairs 
  of 
  the 
  outer 
  rectrices. 
  

  

  In 
  one 
  quite 
  adult 
  female 
  (No. 
  202 
  of 
  my 
  collection) 
  some 
  

   bluish 
  feathers 
  are 
  visible 
  on 
  the 
  forehead 
  ; 
  they 
  are 
  perhaps 
  

   the 
  remains 
  of 
  the 
  immature 
  plumage, 
  which 
  I 
  do 
  not 
  

   know. 
  

  

  Both 
  Aruba 
  and 
  Bonaire 
  must 
  have 
  received 
  their 
  Amazons 
  

   from 
  the 
  continent. 
  On 
  Aruba, 
  which 
  is 
  so 
  close 
  to 
  the 
  

   mainland, 
  they 
  have 
  not 
  become 
  specialized, 
  and 
  very 
  likely 
  

   fresh 
  immigrants 
  might 
  from 
  time 
  to 
  time 
  fly 
  over 
  to 
  that 
  

   island. 
  Bonaire, 
  however, 
  is 
  remote 
  enough 
  to 
  produce 
  a 
  

   new 
  insular 
  form. 
  

  

  