﻿PROCEEDINGS 
  OF 
  THE 
  UNITED 
  STATE5 
  NATIONAL 
  MUSEUM 
  

  

  i33ueJ 
  ff^jKyL 
  Q*Ml 
  hy 
  the 
  

  

  SMITHSONIAN 
  INSTITUTION 
  

   U. 
  S. 
  NATIONAL 
  MUSEUM 
  

  

  Vol. 
  % 
  Wathington: 
  1944 
  No. 
  3187 
  

  

  A 
  EEVISIOX 
  OF 
  THE 
  AMERICAN 
  CLIXGFISHES, 
  FAMILY 
  

   GOBIESOCIDAE, 
  WITH 
  DESCRIPTIONS 
  OF 
  NEW 
  GEN- 
  

   ERA 
  AND 
  FORMS 
  

  

  By 
  Leonard 
  P. 
  Schultz 
  

  

  Recently, 
  while 
  attempting 
  to 
  identify 
  some 
  specimens 
  of 
  cling- 
  

   fishes 
  from 
  the 
  fresh 
  waters 
  of 
  Venezuela 
  and 
  Colombia, 
  South 
  

   America, 
  it 
  became 
  clear 
  to 
  me 
  that 
  the 
  American 
  Gobiesocidae 
  were 
  

   in 
  a 
  state 
  of 
  confusion 
  greater 
  than 
  I 
  had 
  suspected 
  from 
  previous 
  

   work 
  on 
  the 
  group. 
  A 
  search 
  of 
  the 
  literature 
  did 
  not 
  reveal 
  any 
  

   attempt 
  to 
  strai<i:hten 
  out 
  the 
  classification 
  or 
  nomenclature 
  of 
  this 
  

   family. 
  Mew 
  species 
  have 
  been 
  described 
  with 
  regularity, 
  but 
  in 
  

   many 
  cases 
  the 
  doscribers 
  have 
  ignored 
  the 
  same 
  species 
  long 
  ago 
  

   named 
  from 
  nearly 
  the 
  same 
  localities. 
  I 
  have 
  made 
  no 
  attempt 
  in 
  

   this 
  revision 
  to 
  record 
  all 
  the 
  miscellaneous 
  references 
  to 
  American 
  

   Gobiesocidae, 
  but 
  I 
  have 
  included 
  the 
  most 
  important 
  contributions. 
  

   To 
  record 
  every 
  reference 
  in 
  the 
  literature 
  would 
  require 
  one 
  to 
  

   examine 
  the 
  specimens 
  in 
  most 
  of 
  the 
  important 
  museums 
  of 
  the 
  

   worM, 
  which 
  is 
  not 
  possible 
  at 
  the 
  present 
  time. 
  However, 
  that 
  will 
  

   eventually 
  iiave 
  to 
  be 
  done 
  if 
  the 
  identifi«ations 
  recorded 
  in 
  numerous 
  

   instances 
  are 
  to 
  be 
  corrected. 
  

  

  It 
  was 
  found 
  necessary 
  to 
  dissect 
  tlic 
  skin 
  away 
  from 
  the 
  front 
  of 
  

   all 
  median 
  fins 
  in 
  order 
  to 
  count 
  all 
  the 
  fin 
  rays. 
  Vei*y 
  few 
  authors 
  

   lia\t' 
  counted 
  tlie 
  first 
  one 
  or 
  two 
  rays 
  at 
  the 
  beginning 
  of 
  both 
  anal 
  

   and 
  doisal 
  fins. 
  The 
  shoi-t 
  stubby 
  ray 
  on 
  the 
  dorsal 
  edge 
  of 
  I 
  he 
  

   pectoral 
  fin 
  also 
  is 
  included 
  in 
  my 
  counts. 
  (See 
  table 
  L) 
  

  

  The 
  coloration 
  is 
  variable 
  and 
  cannot 
  be 
  used 
  to 
  separate 
  species 
  

   except 
  to 
  a 
  limited 
  extent. 
  

  

  47 
  

   G04148 
  — 
  «4 
  1 
  

  

  