﻿Qli) 
  rnoCEEDINGS 
  OF 
  THE 
  XATIOXAL 
  MUSEUM. 
  

  

  located 
  are 
  joiued, 
  we 
  have 
  a 
  curve 
  whicli 
  brings 
  out 
  the 
  relationship 
  

   between 
  these 
  i)ercentages 
  and 
  areas. 
  

  

  Ul)oii 
  examining 
  the 
  curve 
  thus 
  constructed 
  it 
  becomes 
  imniediately 
  

   evident 
  that 
  there 
  is 
  a 
  general 
  relationship 
  between 
  the 
  size 
  of 
  the 
  

   islands, 
  and 
  the 
  percentage 
  of 
  genera 
  represented 
  by 
  two 
  or 
  more; 
  

   species, 
  for 
  the 
  curve 
  begins 
  at 
  zero 
  for 
  the 
  smallest 
  islands, 
  and 
  its 
  

   general 
  trend 
  is 
  upward 
  until 
  it 
  Anally 
  reaches 
  the 
  80 
  per 
  cent 
  mark 
  in 
  

   the 
  largest 
  island 
  — 
  Luzon. 
  Jli^^umerous 
  irregularities 
  are 
  noticeable, 
  

   however, 
  and 
  the 
  more 
  conspicuous 
  of 
  these 
  are 
  of 
  decided 
  interest. 
  

  

  It 
  is 
  self-evident 
  that 
  only 
  a 
  general 
  correspondence 
  between 
  area 
  

   and 
  amount 
  of 
  differentiation 
  could 
  be 
  expected. 
  Size 
  is 
  no 
  doul)t 
  

   directly 
  important, 
  since 
  room 
  is 
  afforded 
  for 
  numerous 
  individuals 
  of 
  

   the 
  species 
  represented, 
  and 
  the 
  probability 
  of 
  the 
  occurrence 
  of 
  o[)poi 
  

   tune 
  and 
  important 
  individual 
  variations 
  is 
  correspondingly 
  increased: 
  

   but 
  for 
  our 
  present 
  purpose 
  I 
  believe 
  that 
  the 
  size 
  of 
  islands 
  is 
  chietly 
  

   important 
  in 
  that 
  it 
  serves 
  as 
  a 
  rough 
  index 
  of 
  the 
  i)robable 
  diversity 
  

   of 
  conditions 
  existing 
  upon 
  them. 
  The 
  occurrence 
  of 
  extensive 
  high 
  

   lands, 
  of 
  undisturbed 
  forest 
  and 
  of 
  fresh-water 
  lakes 
  and 
  streams, 
  as 
  

   well 
  as 
  of 
  extensive 
  open 
  lowlands, 
  must 
  be 
  taken 
  into 
  consuleration 
  it 
  

   we 
  are 
  to 
  get 
  to 
  the 
  bottom 
  of 
  the 
  matter. 
  Were 
  it 
  possible 
  to 
  give 
  

   each 
  of 
  these 
  factors 
  its 
  due 
  value 
  in 
  constructing 
  our 
  curve, 
  and 
  t(» 
  

   introduce, 
  as 
  well, 
  another 
  important 
  factor, 
  namely, 
  the 
  completeness 
  

   and 
  length 
  of 
  duration 
  of 
  separation 
  from 
  neighboring 
  islands, 
  I 
  

   believe 
  that 
  the 
  irregularities 
  would 
  disappear. 
  

  

  For 
  instance, 
  Bohol, 
  though 
  an 
  island 
  of 
  850 
  square 
  kilometers, 
  has 
  

   110 
  highlands 
  and 
  its 
  forest 
  has 
  seemingly 
  been 
  wiped 
  out. 
  The 
  very 
  

   low 
  level 
  of 
  the 
  curve 
  at 
  this 
  point, 
  then, 
  finds 
  its 
  explanation 
  in 
  a 
  uni- 
  

   formity 
  of 
  conditions 
  unfavorable 
  to 
  the 
  differentiation 
  of 
  numerous 
  

   species, 
  or 
  to 
  their 
  continued 
  existence 
  after 
  they 
  have 
  become 
  differ 
  

   entiated. 
  

  

  It 
  will 
  be 
  noted 
  that 
  the 
  curve 
  is 
  much 
  broken 
  at 
  its 
  origin, 
  although 
  

   it 
  runs 
  low 
  on 
  the 
  whole. 
  This 
  irregularity 
  is 
  largely 
  due 
  to 
  our 
  scanty 
  

   knowledge 
  of 
  the 
  islands 
  in 
  question. 
  For 
  Lapac 
  our 
  conclusions 
  

   are 
  drawn 
  from 
  but 
  two 
  genera, 
  for 
  Fuga 
  from 
  four, 
  for 
  Cagayan 
  Sulu 
  

   from 
  five, 
  and 
  for 
  Camiguin 
  from 
  three. 
  Manifestly, 
  in 
  dealing 
  with 
  

   such 
  small 
  numbers 
  the 
  addition 
  or 
  subtraction 
  of 
  a 
  single 
  genus 
  even 
  

   makes 
  a 
  great 
  variation 
  in 
  the 
  percentage. 
  ]S"o 
  collections 
  approaching 
  

   completeness 
  have 
  ever 
  been 
  made 
  on 
  these 
  islands, 
  and 
  the 
  irregularity 
  

   of 
  the 
  curve 
  is 
  exactly 
  what 
  would 
  be 
  expected 
  from 
  the 
  scanty 
  hap- 
  

   hazard 
  collecting 
  on 
  whicli 
  it 
  is 
  based. 
  

  

  We 
  are 
  indebted 
  to 
  Mr. 
  Everett 
  for 
  nearly 
  all 
  that 
  we 
  know 
  cf 
  Sibutu, 
  

   and 
  he 
  tells 
  us 
  little 
  about 
  its 
  surface. 
  It 
  would 
  be 
  interesting 
  to 
  know 
  

   whether 
  the 
  conspicuous 
  rise 
  in 
  the 
  curve 
  for 
  this 
  island 
  is 
  correlated 
  

   with 
  a 
  comparatively 
  great 
  diversity 
  of 
  conditions. 
  Tawi 
  Tawi, 
  at 
  any 
  

   rate, 
  is 
  well 
  wooded 
  and 
  well 
  watered; 
  the 
  curve 
  rise^. 
  Siqnijor 
  is 
  not 
  

   well 
  wooded 
  nor 
  well 
  watered, 
  and 
  is 
  of 
  comparatively 
  recent 
  origin 
  ; 
  

   the 
  curve 
  falls. 
  The 
  surfaces 
  of 
  Guimaras 
  and 
  Sulu 
  are 
  diversified, 
  and 
  

  

  