﻿GOATFISH 
  GENUS 
  UPENEUS 
  — 
  LACHNER 
  499 
  

  

  The 
  number 
  of 
  spines 
  in 
  the 
  spinous 
  dorsal 
  fin 
  is 
  either 
  7 
  or 
  8 
  

   in 
  any 
  particular 
  species. 
  Only 
  one 
  abnormally 
  developed 
  specimen 
  

   of 
  a 
  species 
  normally 
  with 
  seven 
  had 
  6 
  spines. 
  The 
  difference 
  of 
  one 
  

   spine 
  among 
  the 
  species 
  is 
  associated 
  with 
  the 
  presence 
  or 
  absence 
  

   of 
  the 
  minute 
  first 
  spine. 
  No 
  variability 
  was 
  found 
  in 
  this 
  character 
  

   in 
  any 
  species 
  in 
  more 
  than 
  300 
  specimens 
  examined. 
  This 
  small 
  

   spine 
  is 
  located 
  near 
  the 
  first 
  enlarged 
  one, 
  often 
  partially 
  embedded, 
  

   and 
  may 
  be 
  entirely 
  overlooked 
  without 
  employing 
  some 
  probing 
  

   and 
  a 
  microscope. 
  The 
  variability 
  of 
  this 
  character 
  as 
  given 
  by 
  

   Fowler 
  (1933, 
  pp. 
  322, 
  341) 
  is 
  erroneous. 
  

  

  Counts 
  of 
  the 
  pectoral 
  fin 
  rays 
  for 
  10 
  species 
  from 
  various 
  locali- 
  

   ties 
  in 
  the 
  Indo-Pacific 
  are 
  given 
  in 
  table 
  1 
  . 
  The 
  range 
  of 
  this 
  charac- 
  

   ter 
  in 
  any 
  species 
  did 
  not 
  exceed 
  four 
  rays. 
  An 
  inspection 
  of 
  the 
  data 
  

   in 
  the 
  table 
  shows 
  that 
  the 
  species 
  are 
  divisible 
  into 
  two 
  groups 
  and 
  

   that 
  the 
  modes 
  of 
  the 
  frequency 
  distributions 
  fall 
  on 
  13 
  or 
  14 
  in 
  one 
  

   group 
  and 
  on 
  16 
  in 
  the 
  other, 
  except 
  for 
  U. 
  parvus, 
  where 
  only 
  6 
  speci- 
  

   mens 
  were 
  available 
  for 
  study. 
  It 
  is 
  unfortunate 
  that 
  descriptive 
  

   accoimts 
  of 
  certain 
  nominal 
  species 
  did 
  not 
  include 
  this 
  character; 
  

   with 
  it, 
  a 
  more 
  reliable 
  interpretation 
  would 
  have 
  been 
  possible. 
  

  

  The 
  TOTAL 
  number 
  of 
  gillrakers 
  of 
  260 
  specimens 
  are 
  arranged 
  

   from 
  lowest 
  to 
  highest, 
  respectively, 
  for 
  nine 
  species 
  from 
  the 
  Indo- 
  

   Paciiic 
  region 
  in 
  table 
  1. 
  The 
  tenth 
  species, 
  parvus, 
  represents 
  the 
  

   single 
  form 
  in 
  American 
  waters. 
  These 
  data 
  represent 
  specimens 
  from 
  

   various 
  localities. 
  The 
  range 
  of 
  this 
  character 
  for 
  each 
  species 
  is 
  low. 
  

  

  The 
  usual 
  count 
  of 
  the 
  gillrakers 
  reported 
  in 
  the 
  literature 
  is 
  that 
  

   of 
  the 
  lower 
  limb 
  only. 
  The 
  rudiments 
  were 
  seldom 
  included. 
  Some- 
  

   times 
  they 
  were 
  partly 
  included 
  but 
  without 
  explanation 
  as 
  to 
  how 
  

   they 
  were 
  distinguished 
  from 
  the 
  developed 
  rakers, 
  making 
  it 
  impos- 
  

   sible 
  to 
  interpret 
  the 
  count. 
  To 
  form 
  a 
  more 
  reliable 
  basis 
  for 
  the 
  

   interpretation 
  of 
  data 
  in 
  the 
  literature, 
  the 
  raker 
  count 
  of 
  both 
  limbs 
  

   and 
  the 
  raker-rudiment 
  relationship 
  are 
  recorded 
  separately 
  in 
  tables 
  

   2, 
  4, 
  and 
  5. 
  It 
  is 
  apparent 
  from 
  these 
  data 
  that 
  the 
  number 
  of 
  rudi- 
  

   ments 
  and 
  rakers 
  differs 
  for 
  different 
  species 
  and 
  that 
  the 
  addition 
  of 
  

   developed 
  rakers 
  and 
  rudiments 
  of 
  the 
  upper 
  and 
  lower 
  limbs 
  to 
  form 
  

   the 
  total 
  count 
  displays 
  the 
  greatest 
  differences 
  among 
  the 
  species. 
  

  

  Species 
  having 
  a 
  minute 
  first 
  dorsal 
  spine 
  can 
  be 
  divided 
  into 
  two 
  

   subgroups 
  on 
  the 
  basis 
  of 
  the 
  total 
  number 
  of 
  gillrakers 
  (see 
  key, 
  

   p. 
  508). 
  The 
  number 
  of 
  giErakers 
  in 
  these 
  two 
  divisions 
  also 
  has 
  

   a 
  positive 
  relationship 
  to 
  the 
  number 
  of 
  pectoral 
  fin 
  rays. 
  

  

  The 
  species 
  are 
  divisible 
  into 
  two 
  groups 
  on 
  the 
  basis 
  of 
  color 
  of 
  

   peritoneum: 
  Those 
  in 
  which 
  it 
  is 
  light 
  to 
  silvery 
  (first 
  four 
  species 
  

   and 
  the 
  last 
  species, 
  table 
  1), 
  and 
  those 
  in 
  which 
  it 
  is 
  light 
  brown 
  to 
  

   blackish 
  (all 
  other 
  species, 
  table 
  1). 
  A 
  positive 
  relationship 
  occurs 
  

  

  