Skip to main content

Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats 


-10 


HARVARD    UNIVERSITY 

Library  of  the 

Museum  of 

Comparative  Zoology 


aulletin  of  the 


f^x.  ■■''>  ■-- . 
•>■-'.  '■.■'■- 


seum   o 


omparative 


m^ 


oology 


Volume  135 
1966-1967 


'^ 


") 


HARVARD   UNIVERSITY 

CAMBRIDGE,   MASSACHUSETTS  02138   U.S.A. 


CONTENTS 

Page 

No.  1.  The  Postcranial  Skeleton  of  the  Giant  Permian  Pelycosaur  CotijJo- 
rhynchus  lomeii  B>  J.  Wilhs  Stovall,  Llewellyn  I.  Price,  and 
Alfred  Sherwood  Romer.    September,  1966  1 

No.  2.     The   Stromateoid   Fishes:     Systematics   and   a   Classification.    By 

Richard  L.  Haedrich.    January,  1967  31 

No.  3.  Morphology  and  Relationships  of  the  Holocephali  with  Special 
Reference  to  the  Venous  System.  By  Barbara  J.  Stahl.  January, 
1967  - ' 141 

No.  4.     A  Review  of  the  Mesochrysinae  and  Nothochrysinae  (Neuroptera: 

Chrysopidae).    By  Phillip  A.  Adams.    February,  1967  215 

No.  5.     Marine  Nematodes  of  the  East  Coast  of  North  America.  I.  Florida. 

By  Wolfgang  Wieser  and  Bruce  Hopper.  April,  1967    239 

No.  6.     The   Ameiva    (Lacertilia,   Teiidae)    of   Hispaniola.     III.   Ameiva 

taeniura  Cope.    By  Albert  Schwartz.   April,  1967  345 

No.  7.     New  Cyclopoid  Copepods  Associated  with  Polychaete  Annelids  in 

Madagascar.  By  Arthur  G.  Humes  and  Ju-Shey  Ho.  April,  1967  —  377 

No.  8.     Proterochompso    banionuevoi   and   the   Early   Evolution   of   the 

Crocodilia.  By  William  D.  Sill.  April,  1967 415 

No.  9.     Sihcified  Silurian  Trilobites  from  Maine.    By  H.  B.  Whittington 

and  K.  S.  W.  Campbell.   June,  1967  447 


^^^m^mmm-'m 


uUetin  OF  THE 


The  Postcranial  Skeleton  of  the  Giant  Permian 
Pelycosaur  Cotylorhynchus  romeri 


J.  WILLIS  STOVALL,  LLEWELLYN   I.  PRICE,  AND 
ALFRED  SHERWOOD  ROMER 


Museum  of  Comparafive  Zoology,   Harvard   University 


HARVARD   UNIVERSITY 

CAMBRIDGE,  MASSACHUSETTS,   U.S.A. 


VOLUME    135,   NO.    1 
SEPTEMBER  22,    1966 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Bremora  1952- 

MEMoms  1864-1938 

JoHNSONL\,  Department  of  Mollusks,  1941- 

OccASiONAL  Papers  on  Mollusks,  1945- 


Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Davve  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).   $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.  $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.   ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©  The  President  ond  Fellows  of  Harvard  College  1966. 


THE  POSTCRANIAL  SKELETON  OF  THE 

GIANT  PERMIAN  PELYCOSAUR  COTYLORHYNCHUS  ROMBRI 

J.   WILLIS  STOVALL,   LLEWELLYN   I.   PRICE,  AND  ALFRED  SHERWOOD   ROMER 


Introductory  note.  In  1937  fragmentary 
remains  of  a  gigantic  new  pelycosaur  from 
the  Oklahoma  Pennian  were  brought  to  the 
attention  of  Dr.  Stovall  of  the  University  of 
Oklahoma,  who  shortly  after  published  a 
preliminary  account  of  the  animal,  as  Cotij- 
Jorhijnchus  romeri,  aided  by  notes  and  draw- 
ings b\'  Llewellyn  I.  Price  and  myself  ( Sto- 
vall, 1937 ) .  Subsequently,  many  additional 
specimens  of  this  form  were  found;  these 
were  prepared  under  Dr.  Stovall's  direc- 
tion, and  studies  and  drawings  were  made 
of  much  of  the  material  by  Mr.  Price.  It 
was  planned  that  the  material  be  described 
in  a  joint  paper  by  Stovall  and  Price,  but 
owing  to  a  variety  of  circumstances,  this 
was  never  written.  Nearly  thirty  years  have 
passed,  and  except  for  a  brief  notice  and 
figures  of  the  skull  in  the  "Review  of  the 
Pelycosauria"  ( Romer  and  Price,  1940: 
419-421),  no  further  account  of  Coty- 
lorhynchus  has  appeared.  Meanwhile  Dr. 
Stoxall  has  died,  and  Mr.  Price  is  now 
resident  in  Brasil.  Adding  to  the  need  for 
description  is  the  fact  that  Dr.  E.  C.  Olson 
has  found  additional  remains  of  Coty- 
lorhynclms,  and  a  number  of  related  fomis 
have  been  discovered  in  a  variety  of  locali- 
ties and  horizons  in  Texas  and  even  in 
Russia  (Olson.  1962:  24-47).  In  this  situa- 
tion, it  has  been  agreed  that  I  should  pub- 
lish a  general  account  of  the  postcranial 
anatomy  of  Cotylorhynchus.  But  although 
I  alone  should  be  held  responsible  for  any 
inaccuracies  or  misinterpretations  that  this 


paper  may  contain,  I  have  felt  that  the 
paper  should  be  considered  as  a  joint  effort, 
to  give  credit  to  Dr.  Stovall  for  his  work 
in  the  collection  and  supervision  of  prep- 
aration of  the  material,  and  to  Mr.  Price  for 
his  invaluable  notes  and  drawings  of  the 
Oklahoma  materials. 


— Alfred  Sherwood  Romer 


MATERIALS 


Although  a  broad  band  of  continental 
Permian  beds  extends  across  Oklahoma 
from  the  Kansas  border  to  the  Red  River, 
relatively  few  vertebrates  have  been  dis- 
covered in  the  Lower  Permian  of  the  state 
— quite  in  contrast  to  the  situation  to  the 
south,  in  northern  Texas.  This  relative  pau- 
cit>'  of  finds  appears  to  be  due  to  the  gen- 
erally higher  rainfall  and  consequent  better 
vegetative  covering  in  Oklahoma,  and  to 
the  more  level  topography  of  Oklahoma,  in 
contrast  to  the  Texas  area  where  the  pres- 
ence of  a  number  of  limestones  and  heavy 
sandstones  makes  for  a  more  rugged  to- 
pography and  the  development  of  poten- 
tially fossiliferous  "breaks." 

The  discovery  of  abundant  remains  of  a 
large  new  type  of  caseid  pelycosaur,  Coty- 
lorhynchus romeri,  was  in  consequence  a 
major  event  in  the  history  of  paleontological 
work  in  Oklahoma.  As  noted  in  Stovall's 
paper  (1937:  308),  the  first  find  of  Coty- 
lorhynchus material  was  made  by  Dr.  W.  S. 
Strain    (then   a   graduate   assistant   at   the 


Bull.  Mus.  Comp.  Zool.,  135(1):  1-30,  September,  1966        1 


2        Bulletin  Miisriiw  of  Comparative  Zoology,  Vol.  135,  No.  1 


University  ot  Oklahoma).    The  site  lay  in 
the   Hennessey   shales,   4%    miles   west   of 
Navina,  Logan  County.   The  type  specimen 
consisted  of  an  incomplete  skull  and  jaw,  a 
front  foot,  and  intercla\icle.    Sliortly  after, 
a  nearly  complete  postcranial  skeleton  was 
found  at  about  the  same   horizon,   in   the 
Hennessey  Formation,  but  close  to  the  city 
of  Norman  and  hence  some  50  miles  or  so 
from  the  region  of  the  first  discovery,  and 
during   the    years    1937-1939    a    very    con- 
siderable number  of  further  Cotijlorhijnchus 
specimens  were  collected  in  the  Hennessey 
shales  of  the  Norman  district  and  prepared 
by  laborers  under  Works  Progress  Admini- 
stration funds.   Most  of  this  material  is  now 
in    the    Stovall    Museum    at    Norman,    luit 
specimens  were  also  obtained  for  the  mu- 
seums in  Chicago,  New  York,  Washington, 
and  Cambridge.    Preparation  of  the  mate- 
rials at  Norman  was  carried  out  under  the 
general  direction  of  Dr.  Stovall,  and  notes 
and  figures  on  much  of  the  material  were 
made  by  Mr.  Price.    In  the  preparation  of 
the   present   description,    Romer   has    seen 
all   tliis  material,   and   has   utilized  Price's 
notes  and  figures  in  addition  to  his  own  ob- 
servations.    A    word    of   caution    must   be 
given,  however.   Although  prc>paration  was 
carried   out  under  Dr.   Stovall's   direction, 
it  is  not  improbable  that,  with  a  consider- 
able number  of  workers  and  with  a  number 
of  specimens  undergoing  simultaneous  prep- 
aration,   a    certain    amount    of    inaccuracv 
may  have  occurred  in  the  restoration  and 
as.sociatioii    of   materials.     Further,    in    the 
time  that  has  since  elapsed,  some  of  the 
specimens  have  been  moved  several  times 
for  storage  and  re-cataloging   and   further 
chances  of  error  have  crept  in. 

The  reptilian  remains  from  the  Hennessey 
.shales  of  the  Norman  region  appear,  curi- 
ously, to  pertain  almost  entirely  to  Cotij- 
lorlnjncluis,  with  very  few  remains  of  other 
reptiles  and  amphibians — a  situation  in 
strong  contrast  to  the  usual  mixed  assem- 
blages generally  present  in  typical  Texas 
redbeds  deposits.  As  far  as  can  be  seen, 
the  Cotylorhynchus  materials  from  the  Nor- 


Table  1 


Specimen  No. 

Humerus 

Radius 

Femur 

Tibia 

4-0-6 

352 

166+ 

311 

4-0-2  (1249) 

312 

195 

4-0-4 

343 

188 

305 

197 

CiN'HM  272 

327+ 

138+ 

308 

127+ 

AMNH  7517 

190 

1250   (?) 

232+ 

298 

170 

4-0-16 

265 



285 

USNM 

263 

123+ 

251  + 

147+ 

MCZ  3416 

246 

239 

135 

4-0-13 

197 

115 

4-0-3 

191 

110 

man  area  can  be  assigned  to  a  single  species 
as  far  as  morphological  characters  are  con- 
cerned.^ There  is  however,  considerable 
variation  in  size.  In  reptiles  generally,  early 
growth  is  rapid,  followed  by  a  slower  if 
steady  growth;  the  natural  result  is  that  the 
greater  part  of  specimens  recovered  in  the 
case  of  fossil  forms  should  represent  rela- 
tively young  mature  animals,  with  a  small 
percentage  of  overly  large  specimens  repre- 
senting indi\'iduals  which  were  exception- 
ally long-lived  and  consequently  exception- 
ally large,  and  a  small  percentage  of  young 
and  immature  individuals.  The  present  ma- 
terials tend  to  show  a  rather  greater  size 
range  than  is  common.  As  a  rough  index  to 
size,  I  list  the  length  (in  mm)  of  major  limb 
elements  in  a  number  of  the  better  speci- 
mens (Table  1). 

These  specimens  are  deposited  in  the 
following  institutions:  Chicago  Natural 
History  Museum  ( CNHM ) ;  American  Mu- 
seum of  Natural  History  (AMNH);  United 
States  National  Museum  (USNM);  Museum 
of  Comparati\'e  Zoology,  Harvard  ( MCZ ) . 
Other  specimens  (numbered)  are  in  the 
University  of  Oklahoma  collection. 

As  can  be  seen  from  this  table,  the  first 
five  specimens  listcxl  appear  to  be  of  rela- 
tively uniform  large  size.  Below  this  there 
is  a  sharp  drop  to  three  specimens  which 
are  20  per  cent  or  so  smaller  than  the  first 

^  Dr.  Olson  informs  mo  that  there  is  a  possible 
specific  tUfferencc  Ix-lwecn  the  Norman  material 
and  the  type  from  Navina. 


COTYLORHYNCHUS  SKELETON  •  StovaU,  Price,  and  Rotner 


group  but  appear  to  be  essentially  mature; 
below,  with  the  MCZ  specimen  as  a  tran- 
sition, there  is  a  drop  to  small  and  seem- 
ingly immature  specimens  such  as  4-0-3. 
Possibly  there  is  a  specific  difference  be- 
tween the  first  and  second  groups.  Equally 
possible,  however,  the  contrast  is  a  sexual 
one;  in  Dimetrodon  limhatiis,  for  example, 
there  is  a  seemingly  clear-cut  size  difference 
of  10  per  cent  between  the  sexes  (Romer 
and  Price,  1940:  341-342). 

In  typical  Texas  redbeds  localities,  fossil 
reptile  remains  are  most  commonly  found 
in  disarticulated  condition,  and  when  found 
articulated  appear  to  have  undergone  de- 
position in  a  variety  of  poses.  One  gets  the 
impression  that  this  is  the  result  of  stream 
action,  and  that  many  of  the  "bone  pockets" 
represent  back  eddies  in  streams  in  which 
cadavers  brought  downstream  tended  to 
collect  (and  decompose).  But  in  the  Hen- 
nessey shales  of  the  Nomian  region,  many 
of  the  specimens  show  clearly  that  the 
CotylorhyncJnis  individuals  were  generally 
buried  in  articulated  and  undisturbed 
fashion,  right  side  up,  with  the  limbs  spread 
outward  at  the  sides.  This  strongly  suggests 
that  we  are  here  dealing  with  entombment 
of  quite  another  sort.  South  African  Per- 
mian pareiasaurs,  equally  large  and  clumsy 
herbivores,  are  typically  preserved  in  simi- 
lar fashion  (Watson,  1913).  There  are  two 
possible  interpretations.  Watson,  in  the 
case  of  the  pareiasaurs,  implies  death  from 
starvation,  and  suggests  a  covering  of  the 
cadavers  in  situ  by  loess-like  materials.  I 
would  favor  an  alternative  interpretation  in 
both  cases — that  the  animals  were  bogged 
down  and  entombed  in  swamps. 

Study  has  been  greatly  handicapped  by 
the  nature  of  the  specimens.  Due  in  part, 
perhaps,  to  a  rather  spongy  structure  of  the 
skeletal  elements,  and  to  the  nature  of  the 
sediments  in  which  they  were  embedded, 
much  of  the  material  has  been  subjected  to 
crushing  and  distortion,  with  a  consequent 
limitation  of  the  accuracy  with  which  res- 
toration and  illustration  can  be  made.  In 
great  measure  description  and  illustration 


of  structures  given  here  are  based  on  a 
synthesis  of  a  number  of  specimens.  Wher- 
ever possible  the  illustrations  are  based  on 
specific  examples,  although  frequently  with 
the  addition  from  other  individuals  of  de- 
tails missing  or  obscured  in  the  specimen 
primarily  utilized. 

It  became  apparent,  even  from  the  ma- 
terial described  in  Stovall's  preliminary 
paper,  that  CotijJorhijnchus  was  a  giant  rel- 
ative of  Casea,  a  small  Clear  Fork  Texas 
pelycosaur  described  by  Williston  (Willis- 
ton,  1910:  590-592;  1911:  111-131,  etc.; 
Romer  and  Price,  1940:  412-419).  Despite 
cranial  differences,  Casea  and  a  number  of 
other  Texas  pelycosaurs  show  such  a  large 
number  of  diagnostic  postcranial  similari- 
ties to  Edaphosaurus  that  Romer  (Romer 
and  Price,  1940:  21,  366-378)  felt  justified 
in  including  these  forms  with  Edaphosaurus 
in  a  common  suborder  Edaphosauria.  As  in 
Casea,  the  CotylorJiynchus  postcranial  skele- 
ton agrees  in  almost  every  particular  with 
the  characteristics  assigned  to  the  Edapho- 
sauria as  a  whole.  In  consequence,  in  the 
description  which  follows,  comparisons  are, 
in  general,  with  Casea  or  other  edapho- 
saurian  types. 

VERTEBRAL  COLUMN 

The  general  characteristics  of  the  Coty- 
lorhynchus  vertebral  column  are  almost  ex- 
actly those  cited  in  the  "Review  of  the 
Pelycosauria"  for  the  suborder  Edaphosauria 
as  a  whole: 

"The  number  of  presacral  vertebrae  is  sometimes 
reduced.  The  dorsal  vertebrae  are  moderately  to 
greatly  elongated;  the  cervicals  are  small  in  all 
dimensions.  The  dorsal  centra  are  spool-shaped, 
with  rounded  bottoms,  and  lack  any  trace  of  ven- 
tral keel  or  lip.  Intercentra  are  small.  The  dorsal 
transverse  processes  are  moderately  elongated  and 
rise  from  a  more  antero-dorsal  position  on  the  arch 
than  in  other  pelycosaurs.  The  zygapophysial 
surfaces  are  large,  moderately  tilted,  and  extend 
farther  laterally  than  in  sphenacodontids.  The 
neural  arches  are  not  excavated  above  the  dia- 
pophyses.  .  .  .  The  atlas  centrum  reaches  the  ven- 
tral surface  of  the  column." 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135.  No.  1 


Onl>-  a  very  few  qualifications  on  this  tenninal,  are  present  in  seemingly  articu- 
definition  need  be  made  to  fit  Cotylorlujn-  lated  fashion.  However,  there  is  a  suspi- 
chiis.  (1)  It  was  stated  that  the  presacral  cionsly  sharp  drop  in  the  size  of  centra 
count  is  "sometimes"  reduced.  At  the  time  following  element  27,  and  a  photograph  of 
this  was  written,  the  figure  was  known  the  specimen  made  during  preparation 
only  in  Casca  where  apparently  24  or  25  shows  that  the  last  16  vertebrae  were  not 
presacrals  were  present  ( Romer  and  Price,  part  of  the  block  containing  the  main  part 
1940:  417).  It  is  now  known  that  Edapho-  of  the  material,  although  they  may  well 
saiirus  boanerges  has  likewise  reduced  the  pertain  to  the  same  individual.  Measure- 
primitive  pelycosaur  number  from  27  to  23  ments  suggest  that  there  is  a  gap  here  of 
presacral  vertebrae  (Shuler  and  Witter,  about  10  vertebrae,  to  raise  the  probable 
1942).  It  is  thus  a  reasonable  inference  that  total  count  to  about  53.  In  the  CNHM 
presacral  reduction  was  general  in  the  sub-  specimen,  37  caudals  are  present  in  articula- 
order,  and  reduction  is  present  in  Coty-  tion  with  the  trunk;  following  a  gap  of 
lorhynchus.  (2)  "Intercentra  are  small."  about  23  cm  in  the  slab  there  are  eight 
Few  are  present  in  any  Edaphosaurus  ma-  further,  much  smaller,  vertebrae,  the  last 
terial;  only  a  single  intercentrum  was  found  apparently  terminal  or  sub-terminal.  The 
in  the  Casea  material  (in  the  sacral  region);  gap  is  of  such  length  as  to  suggest  that  10 
none  are  known  in  Cotylorhynchus  except  vertebrae  are  missing  here,  giving  a  prob- 
for  the  atlas-axis.  Small  gaps  between  the  able  total  length  of  55  vertebrae.  Tlris  is  a 
ventral  edges  of  the  centra  in  all  three  reasonable  pelycosaurian  number, 
genera  suggest  that  tiny  intercentra  were  Dorsal  veriehrae.  Such  a  dorsal  vertebra 
present  in  cartilaginous  fonn.  (3)  "The  as,  for  example,  vertebra  12  of  4-0-6  (Figs. 
atlas  centrum  reaches  the  ventral  surface  of  IC,  2C ) ,  exhibits  a  typical  pelycosaurian 
the  column."  This  appears  to  be  true  of  and,  further,  edaphosaurian  character.  The 
Edaphosaurus,  and  WilUston  (1911:  pi.  centra  are  edaphosauroid — spool-shaped 
XIV )  restores  this  centrum  as  reaching  the  structures  without  any  of  the  tendency  seen 
ventral  surface  in  Casea.  As  described  be-  in  ophiacodontoids  and  sphenacodontoids 
low,  however,  it  does  not  reach  this  surface  for  development  of  a  ventral  keel.  The 
in  Cotylorhynchus,  and  Williston's  material  centrum,  as  in  Casea,  is  relatively  short  and 
of  Casca  was  obviously  imperfect  and  open  stout,  its  length,  averaging  about  60  mm  in 
to  misinterpretation.  large  individuals,  being  but  about  IV2  times 

The   Cotylorhynchus  romcri  column  in-  the  end  height  of  38-40  mm.    The  ventral 

eludes  definitely  25  and  probably  26  pre-  surface  is  broad,  essentially  flattened,  but 

sacral  vertebrae,   3  sacrals,   and   about  55  slightly  convex;  at  the  margins  of  this  area 

caudals.    It  has  proved  difficult  to  deter-  the  sides  curve  upward  sharply  and,  above 

mine  the  presacral  count  in  Co^y/or/if/nc/H/.s.  this  point,  are  slightly  in-pinched.    There 

There  are  several  specimens  in  which  dor-  is  no  visible  suture  between  arch  and  cen- 

sals,   lumbars,   and   sacrals   are   present   in  trum.  On  the  sides  of  the  centrum,  at  about 

well-connected  series;   generally,  however,  the  level  where  the  base  of  the  neural  arch 

the  cervicals  are  poorly  preserved  or  absent,  pedicel  would  be  expected,  there  is  a  low 

In  4-0-6,  however,  the  cervicals  are  present,  longitudinal  ridge,  rounded  at  its  summit, 

There  are  definitely  25  well-preserved  pre-  running  lengthwise  from  a  point  well  below 

sacral  vertebrae,  and  apparently  one  further  the  upper  margin  of  the  centrum  anteriorly 

poorly   preserved   one.    Several  specimens  back  to  a  somewhat  higher  position  pos- 

show   a   sacrum   of   three   vertebrae.     The  teriorly.  Above  this  point  the  upper  part  of 

tail  is  preserved  in  but  few  cases.    In  the  the  surface  of  the  centrum  and  the  lower 

mounted  slab  of  4-0-2   (1249),  a  hind  leg  lateral  surface  of  the  neural  arch  form  a 

and   tail,   43   caudals,   the   last   apparently  longitudinal   depression   extending   length- 


CoTYLORHYNCHUS  SKELETON  •  StovdU,  Piicc,  and  Romcr        5 


Fig.    1.      Lateral   views  of   representative  vertebrae,    in   side   view.     A,  atlas-axis,  X  %;  8/   vertebra  6,  X   Vi;  C,  vertebra   12, 
X    Vv   D,   vertebra  21,  X    Vv   f'    proximal   caudal,  X   'A- 


wise  beneath  the  base  of  the  transverse 
process. 

The  ends  of  the  centrum  are  of  a  typical 
pelycosaurian  nature — essentially  circular  in 
outline,  with  thickened  edges.  At  either 
end  this  circle  contains  a  cone-shaped  de- 
pression for  the  notochord,  the  two  cones 
connected  by  a  small  foramen.  Both  an- 
teriorly and  posteriorly  the  rim  of  the  cen- 
trum is  somewhat  thickened  laterally  for 
the  articulation  of  the  capitulum,  which 
was  thus  inter-central  in  position. 

The  transverse  processes  are  highly  de- 
veloped in  this  region,  extending  outward 
on  either  side  to  a  distance  from  the  mid- 
line approximately  equal  to  the  vertebral 


length.  In  edaphosauroid  fashion  the  trans- 
verse processes  arise  from  a  far  anterior 
position  on  the  neural  arch,  and  their  front 
margins  extend  almost  directly  outward  from 
the  level  of  the  prezygapophyses.  There  is 
in  this  region  no  ventral  expansion  of  the 
process,  which  is  thin  dorsoventrally  but 
broad  anteroposteriorly;  there  is  thus  no 
continuum  in  articular  areas  between  ca- 
pitulum and  tuberculum.  The  transverse 
process  here  is  directed  almost  straight  lat- 
erally, with,  however,  a  slight  upward  and 
forward  tilt,  rather  than  the  downward  slant 
of  most  pelycosaurs  ( Cosea  excepted ) .  The 
base  of  the  process  is  very  broad,  extending 
nearly  the   whole   length  of  the   vertebra. 


6        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.    2.      Posterior    views    of    the    same    vertebrae    as    those    of   Figure  1. 


The  broad  articular  surface  for  the  tuber- 
cukxm  faces  ventrally  and  only  slightly  lat- 
erally on  the  under  surface  of  the  process. 
Since  the  tuberculum  is  little  raised,  the 
proximal  portion  of  the  rib  continues  out- 
ward in  the  line  of  the  transverse  process. 
The  zygapophyses  are  normally  constructed, 
with  a  very  slight  median  tilt  of  the  articular 
surfaces.  The  neural  arches,  as  in  edapho- 
saurs  generally,  lack  the  lateral  excavation 
seen  in  sphenacodontoids  above  the  trans- 
verse processes.  The  neural  spines,  as  in 
Casea,  are  slender  transversely  but  long 
anteroposteriorly,  and  are  low  (although 
not  as  low,  relatively,  as  in  the  last-named 
genus).    In  such  mid-dorsals  as  are  com- 


pletely preserved  the  rugose  end  of  the 
spine  is  expanded  laterally  on  either  side 
and  indented  medially,  to  give  a  somewhat 
bifurcate  appearance,  much  as  in,  for  ex- 
ample, some  parts  of  the  En/ops  column. 

CcrvicaJs.  Progressing  forward  through 
the  anterior  dorsals  into  the  cervical  series, 
the  vertebrae  are  increasingly  lightly  built, 
as  tends  to  be  true  of  pelycosaurs  generally, 
but  more  especially  of  edaphosauroids,  such 
as  CotylorJiyncJws;  in  which  the  head  is 
relatively  small.  The  centra  decrease  steadily 
in  diameter  and  length  anteriorly,  and  the 
width  tends  to  equal  the  length.  In  the 
6th  vertebra  of  4-0-4,  for  example  (Figs. 
IB,  2B),  the  length  is  about  50  mm,  the 


CoTYLORHYNCHUS  SKELETON  •  StovciU,  Price,  and  Romer        7 


width  about  45  mm.  The  flattened  ventral 
surface  of  the  centra  persists  into  the  cer- 
vical region.  In  the  cervicals  the  transverse 
processes  are  shorter  and  lightly  built,  in 
correlation  with  the  decreased  size  of  the 
ribs.  Anteriorly,  the  cleft  in  the  summit  of 
the  neural  spines  disappears,  although  some 
distal  expansion  may  persist;  the  spines  be- 
come shorter  and  somewhat  more  rounded 
— ovoid — in  section.  Proceeding  forward, 
the  direction  of  the  transverse  processes 
gradually  shifts.  In  a  typical  dorsal  it  points 
some\\'hat  forward  and  upward,  but  in  the 
cervicals  the  processes  become  somewhat 
ventrally  and  posteriorly  directed,  in  cor- 
relation with  the  direction  of  the  ribs  in  the 
relatively  slender  neck  region.  Even  as  far 
forward  as  vertebra  3,  the  capitular  facet  is 
high  up  on  the  rim  of  the  centrum,  showing 
little  of  the  tendency  to  descend  anteriorly, 
seen  in  most  pelycosaurs.  The  cervical 
zygapophyses  are,  like  the  dorsals,  widely 
separated  and  with  essentially  horizontal 
articular  surfaces. 

Atlas-axis.  The  atlas-axis  complex  is  pres- 
ent and  well  preserved  in  4-0-6,  and  is 
present  also  in  4-0-4  (Figs.  lA,  2A).  A 
facet  on  the  atlas  neural  arch  indicates  the 
presence  in  life  of  a  proatlas.  The  atlas 
arch  is  bipartite  and  typically  pelycosaurian, 
with  a  long  posterior  process  articulating 
with  the  prezygapophysis  of  the  axis  and 
with  a  spur  running  farther  back  along  the 
side  of  the  axis  neural  arch.  There  is  a 
short  but  distinct  transverse  process  for  rib 
attachment,  turned  strongly  downward  and 
backward.  The  atlantal  intercentrum  is 
highly  developed  for  cranial  articulation; 
laterally  there  is  a  distinct  facet  for  the 
rib  capitulum.  As  is  well  known,  the  atlas 
centrum  is  variable  in  development  in  pely- 
cosaurs, reaching  the  ventral  surface  in 
Dimetrodon,  for  example,  but  excluded  from 
this  surface  bv  the  second  intercentrum  in 
Ophiacodon.  The  situation  in  Cotyloihyn- 
chiis  is  not  too  clear.  In  both  specimens 
in  which  the  atlas-axis  is  preserved,  there 
is  a  distinct  ventral  gap  between  inter- 
centra  1  and  2,  but  the  atlas  centrum  does 


not  reach  the  surface  here;  possibly  this  gap 
may  have  been  filled  by  a  ventral  car- 
tilaginous continuation  of  the  bone. 

Intercentrum  2  is  very  well  developed 
and  bears  a  distinct  rib  facet.  The  axis  cen- 
trum and  arch  are  built  in  typically  pely- 
cosaurian— and  especially  edaphosaurian — 
fashion.  The  transverse  process  slants 
strongly  downward  and  to  a  slight  degree 
backward.  The  neural  spine  is,  for  an  axis, 
rather  short,  but  is  stout,  long  anteropos- 
teriorly,  and  expanded  in  a  3-pronged  ar- 
rangement posteriorly  toward  the  summit. 

Posterior  dorsals.  Passing  backward  along 
the  dorsal  series,  the  vertebral  centra  in- 
crease somewhat  in  length  to  about  verte- 
brae 17-19,  beyond  which  a  progressive 
reduction  takes  place  so  that  beyond  this 
point  (as  is  also  the  case  anteriorly)  width 
exceeds  length.  In  4-0-4,  for  example,  the 
length  of  vertebra  21  (Figs.  ID,  2D)  is  51 
mm,  the  breadth  about  55  mm.  An  increase 
in  central  width,  however,  persists,  width 
reaching  its  maximum  in  the  sacral  region, 
and  the  posterior  dorsals  are  the  most  mas- 
sive vertebrae  in  the  column.  In  the  more 
posterior  presacrals  the  neural  spines  are 
increasingly  long  fore-and-aft  and  increas- 
ingly narrow  in  transverse  diameter,  al- 
though the  arches  are  broadly  expanded  at 
the  spine  bases.  In  the  posterior  dorsal 
series  the  two  members  of  each  pair  of 
prezygapophyses  are  joined  by  a  horizontal 
ridge  of  bone  bridging  the  space  between 
them  and  forming  a  strengthening  trans- 
verse element  wliich  crosses  the  front  end 
of  the  vertebra  between  the  ribs  of  either 
side.  Progressing  backward,  the  transverse 
processes  are  increasingly  stout  but  decrease 
in  length.  In  the  last  members  of  the  series 
the  processes  are  fused  with  the  ribs.  In 
the  posterior  vertebrae  the  area  of  attach- 
ment of  the  rib  capitulum  moves  upward 
and  backward  to  an  oval  area  on  the  an- 
terodorsal  surface  of  the  side  of  the  cen- 
trum below  the  transverse  process.  Here, 
as  in  the  transverse  process,  fusion  with  the 
rib  takes  place  in  the  last  few  members  of 
the  dorsal  series. 


8        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.  3.  Left,  ventral  view  of  last  presacral,  the  three  sacrals,  and  first  two  caudal  vertebrae,  X  'A-  '?ighf,  obove,  lateral 
view  of  mid-caudal  vertebrae,  X  Vi.  Right,  below,  posterior  and  ventral  views  of  a  mid-caudal  vertebra  and  a  mid- 
caudal    chevron,   X   'A- 


Sacrals  (Fig.  3).  Three  sacral  vertebrae 
are  present,  as  noted  in  the  description  of 
the  ribs.  The  centra  are  broad  but  appear 
to  be  shallow  dorsoventrally  ( an  effect  per- 
haps due  to  crushing).  In  one  specimen,  at 
least,  the  centra  of  the  first  and  second 
sacrals  appear  to  be  fused;  whether  the  last 
two  were  fused  is  uncertain.  On  the  first 
sacral  the  zygapophyses  are  still  well  sep- 
arated, but  on  the  third  vertebra  the  two 
members  of  each  pair  are  much  closer  to- 
gether, a  situation  transitional  to  that  in 
the  tail.  There  appears  to  be  no  ossification 
of  apposed  zygapophyses.  The  transverse 
processes — firmly  fused  to  the  ribs — are 
short,  but  very  stout,  and  are  continuous 
with  a  large  area  on  the  centrum  represent- 
ing the  capitular  attachment.  A  depression 
which  represents  the  point  of  junction  of 
tubercular  and  capitular  areas  is  seen  in 
some  specimens;  it  cannot  be  detennined 
whether  an  arterial  foramen  perforates  the 


structure.  The  neural  spines  of  the  first  two 
sacrals  are  similar  to  those  of  the  "lumbars"; 
that  of  sacral  3  is  shorter  fore-and-aft  and 
less  compressed  from  side  to  side. 

Caudah  (Figs.  IE,  2E,  3).  The  most 
anterior  caudals  resemble  the  last  sacral 
closely.  The  breadth  of  the  centrum  is 
greater  than  the  length  or  height;  the  rib 
attachment  areas  are  short  but  greatly  ex- 
panded, with  a  groove,  presumably  for  an 
intersegmental  artery,  marking  the  line  of 
distinction  between  tubercular  and  capitu- 
lar components.  More  posteriorly,  the  trans- 
verse processes  become  greatly  reduced  in 
size  (with  a  concomitant  reduction  in  rib 
size),  and  process  and  rib  have  essentially 
disappeared  by  vertebra  12,  although  a 
slightly  projecting  ridge  is  present  as  far  as 
vertebra  20.  Meanwhile,  the  centra  have 
begun  a  change  in  their  proportions,  so  that 
by  mid-length  of  the  tail  the  length  is  con- 
siderably greater  than  the  width  and  the 


CoTYLORHYXCHUS  SKELETON  •  StovciU,  Piicc,  and  Roiuer        9 


8 


5  4        3        2 


U 


Fig.    4.      Above,    right    ribs    1-8,    seen    from    the    posterior    aspect,  X  Vi-     Below,    the    heads    of    right   and    left    ribs    from    the 
same   mid-dorsal   segment  of  a   single   individual,    to   show  contrasts  in   post-mortem  distortion,  X  'A- 


height  also  exceeds  the  \\'idth.  The  zygapo- 
ph)'sial  pairs  have  come  to  He  close  to  each 
other  near  the  mid-Hne,  and  their  articular 
surfaces  have  become  sharply  tilted  to  a 
plane  close  to  the  vertical.  There  is,  further, 
a  gradual  reduction  in  height  and  stoutness 
of  the  neural  spines,  so  that  by  the  time 
the  mid-caudals  are  reached,  the  spine  is 
a  low  nubbin  projecting  a  short  distance 
above  tlie  postzygapophyses.  The  neural 
arch  as  a  whole  is  much  reduced,  and  in  the 
most  posterior  part  of  the  tail  the  material 
shows  httle  evidence  of  any  structure  above 
the  centra. 

The  first  of  the  chevrons  appears,  in  the 
CNHM  specimen  (No.  272),  between  the 
3rd  and  4th  caudals;  they  continue  back 
as  far  as  vertebra  36,  at  least  in  one  speci- 
men. The  first  elements  are  stoutly  built, 
with  a  length  equal  to  two  centra;  as  is 


normal,  the  length  decreases  posteriorly, 
although  but  slowly.  In  one  specimen,  as  in 
pelycosaurs  generally  ( and  in  Casea ) ,  there 
is,  in  the  proximal  elements,  a  basal  inter- 
central  component  connecting  the  two  arms 
of  the  chevron;  but  in  the  CNHM  specimen 
not  even  the  most  proximal  chevrons  have 
a  proper  intercentrum,  there  being  merely 
an  enlargement  of  each  of  the  two  proximal 
ends  of  the  chevron  structure.  Although 
both  of  the  adjacent  centra  are  bevelled 
for  reception  of  a  chevron,  the  centrum 
anterior  to  the  chevron  has  the  most  ob- 
vious articulations;  it  bears  ventrally  a  pair 
of  well  developed  facets,  each  supported  by 
a  longitudinal  ridge  on  the  centrum.  In  the 
tail  the  bottom  of  the  centrum  shows  a 
longitudinal  depression,  bounded  on  either 
side  by  these  ridges. 


10        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.  5.      The  proximal   part  of  right  ribs  9-21,  seen   from  the  posterior   aspect,   X   'A- 


CoTYLORHYNCHUS  SKELETON  •  Stovdl,  Piicc,  and  Romer        11 


RIBS 

As  in  all  known  pelycosaurs,  ribs  were 
present  on  every  vertebra  from  the  atlas  to 
and  including  the  proximal  caudals.  For 
most  parts  of  the  series,  specimens  4-0-6 
and  CNHM  272  furnish  the  best  material. 

Dorsal  ribs  (Fig.  5).  In  a  typical  dorsal 
rib  the  shaft  extends  outward  (and  in  life 
somewhat  upward)  from  the  capitulum, 
the  articular  surface  of  which  is  an  oval, 
narrow  anteroposteriorly,  occupying  the  en- 
tire head  of  the  rib.  The  tuberculum  in  the 
mid-dorsals  does  not  project  to  any  marked 
degree  from  the  shaft;  it  apposes  to  the 
transverse  process  a  concave  articular  sur- 
face with  an  oval  shape,  broader  distally, 
the  length  about  twice  the  width. 

The  proximal  end  of  the  rib  appears  to 
run  distally  slightly  upward  to  and  past  the 
tubercular  region.  Shortly  beyond  this  point 
the  rib  curves  markedly  outward  and  down- 
ward to  encase  the  flank.  The  degree  of 
curvature  in  life  is  difficult  to  detennine, 
due  to  variable  post-mortem  crushing  and 
compression;  see,  for  example,  the  marked 
contrast  between  the  two  members  of  a  rib- 
pair  from  the  same  segment  of  a  single  in- 
dividual shown  in  Figure  4.  In  most  in- 
stances there  is  a  considerable  segment  of 
the  rib,  lying  beyond  the  point  of  proximal 
cunature,  which  is  nearly  straight;  this  in- 
cludes about  half  the  total  rib  length.  Tliis 
segment  appears  to  have  been  directed  di- 
agonally outsvard  and  downward.  Distally, 
beyond  this  straight  section,  the  rib  curves 
gradually  inward  toward  the  mid-line.  Al- 
though seldom  is  a  complete  rib  length 
preserved,  it  would  appear  that  the  greatest 
length  of  a  rib  (measured  from  the  tuber- 
culum) is  about  the  length  of  13  dorsal 
centra;  two  mid-dorsals  of  CNHM  272  mea- 
sure 721  and  718  mm.  In  such  dorsals  as  are 
completely  preserved  the  distal  end  of  the 
rib  exhibits  a  cupped  tenuination  with 
which  the  cartilaginous  section  of  the  rib 
presumably  articulated. 

Despite  the  handicap  noted  as  due  to 
post-mortem  distortion,  articulation  of  ribs 


and  \  ertebrae  gives,  on  the  average,  a  pic- 
ture closely  comparable  to  that  gi\en  by 
Williston  (1911:  pi.  XVII,  fig.  1)  for  a 
Casea  mid-dorsal,  and  shows  the  presence 
of  a  very  broad,  barrel-like  trunk,  with  a 
probable  transverse  diameter  of  about  60 
cm — i.e.,  about  two  feet. 

Except  for  the  rather  flattened  capitular 
region,  typical  dorsal  ribs  are  essentially 
oval  in  section,  and  so  tilted  in  life  that  the 
long  axis  of  the  oval  slants  downward  and 
forward  well  over  45°,  with  the  two  major 
surfaces  essentially  anterior  and  posterior 
in  position.  The  anterior  margin  is  in  all 
typical  dorsals  smoothly  rounded.  On  the 
posterior  surface  a  ridge  develops  at  a  point 
somewhat  distal  to  the  tuberculum  and 
rather  toward  the  ventral  margin.  This  ridge 
extends  outward  along  the  straight  lateral 
segment  of  the  rib,  gradually  approaching 
the  ventral  margin  of  the  rib  and,  decreas- 
ing in  prominence,  disappears  at  about  the 
point  where  the  median  curvature  begins. 

There  is  little  expansion  of  typical  dorsal 
ribs  at  any  region.  The  diameter  of  the  rib 
is  greatest  a  short  distance  distal  to  the 
tuberculum;  beyond  this  region  there  is 
generally  a  very  gradual  decrease  to  a  mini- 
mum of  about  two-thirds  of  the  maximum. 

Progressing  anteriorly,  from  such  a  typi- 
cal dorsal  as  that  described,  to  rib  9,  a 
series  of  gradual  changes  may  be  observed. 
The  length  decreases,  rib  9  in  4-0-6,  for 
example,  being  but  430  mm  in  length,  i.e. 
about  60  per  cent  as  long  as  the  longest  rib, 
and  the  ribs  are  more  lightly  built.  In  an- 
terior dorsals,  in  contrast  with  mid-dorsals, 
the  tuberculum  projects  markedl\-  above 
the  general  outline  of  the  dorsal  rib  margin, 
and,  in  contrast,  the  capitulum  is  sharply 
turned  downward  medially  from  the  line  of 
the  shaft.  There  is  here  only  a  slight  curva- 
ture beyond  the  tubercular  region;  when 
the  rib  is  articulated  with  the  vertebra,  it  is 
obvious  that  its  shaft  runs  more  directly 
downward  than  in  the  typical  dorsals,  in- 
dicating a  narrowing  of  the  bod>'  in  the 
"chest"  region,  as  the  level  of  the  shoulder 
girdle  is  approached.   Distal  to  the  straight 


12 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  Xo.  1 


descending  portion  of  the  shaft,  there  is  a 
final  segment  curving  somewhat  inward. 
The  total  distance  from  the  tuberculum  to 
the  beginning  of  the  median  curvature  is 
little  less  than  in  rib  14,  indicating  that  the 
"chest"  is  nearly  as  deep  as  the  belly,  al- 
though narrower. 

The  longitudinal  ridge  found  on  the  pos- 
terior rib  surface  changes  in  position  as  one 
progresses  fonvard.  It  has  shifted  upward, 
so  that  its  proximal  portion  lies  along  the 
dorsal  margin  of  the  rib,  and  its  distal  end 
lies  at  about  the  middle  of  the  posterior 
surface,  rather  than  toward  its  ventral  mar- 
gin. It  has,  further,  moved  proximally,  so 
that  its  proximal  end  has  attained  the  distal 
margin  of  the  tubercular  projection,  and  its 
distal  end  extends  only  part-way  down  the 
vertical  segment  of  the  rib.  Distally  the 
anterior  dorsal  ribs  remain  sub-circular  in 
section.  Proximally,  however,  the  posterior 
surface  becomes  much  hollowed  out,  with 
the  proximal  part  of  the  longitudinal  ridge 
forming  a  very  prominent  projection  dor- 
sally.  As  far  as  can  be  determined  in  the 
usually  crushed  condition  of  the  specimens, 
the  most  anterior  dorsal  ribs  are  somewhat 
expanded  distally — notably  rib  9  in  4-0-6. 

Cervical  ribs  (Fig.  4).  In  the  few  pre- 
viously known  pelycosaurs  in  which  the 
distal  ends  of  the  ribs  have  been  well  pre- 
served, it  appeared  that  rib  8  was  stout 
distally  and  presumably  had  a  sternal  con- 
nection, establishing  7  as  the  probable  num- 
ber of  non-sternal,  i.e.,  cervical,  ribs.  In 
4-0-6,  however,  rib  8,  although  elongate 
(with  a  length  of  about  360  mm)  and 
resembling  the  dorsal  rib  following  it  in 
most  regards,  tapers  to  a  point  distally.  It 
thus  failed,  obviously,  to  reach  the  sternum 
and  must  be  considered  technically  to  be  a 
cervical. 

In  4-0-6,  remains  of  all  the  cervical  ribs 
are  present  (but  the  third  is  poorly  pre- 
served). Although  the  rib-tips  are  incom- 
plete, it  is  clear  that  all  were  slender  and 
pointed  distally.  From  rib  8  forward  to  rib 
4  there  is  a  sharp  and  steady  diminution  in 
length,  and  tlic  preserved  portions  of  the 


most  anterior  ribs  indicate  that  these  ribs 
were  shorter  still.  In  correlation  with  the 
wide  separation  of  the  two  points  of  attach- 
ment to  the  vertebrae,  tuberculum  and 
capitulum  are  strongly  divergent,  giving  a 
V-shape  to  the  proximal  part  of  the  rib. 
The  slender  distal  portions  of  the  ribs  ap- 
pear to  be  somewhat  compressed  antero- 
posteriorly. 

In  the  cervicals,  as  in  the  most  anterior 
dorsals,  the  rib  shaft  runs  straight  distally 
from  the  tuberculum,  indicating  a  narrow 
neck  region;  the  tuberculum,  as  far  as  pre- 
served in  these  ribs,  retains  the  somewhat 
distinct  character  seen  in  the  most  anterior 
dorsals.  On  rib  8  the  longitudinal  ridge 
retains  the  character  seen  in  the  rib  follow- 
ing for  much  of  the  proximal  half  of  the 
shaft,  fonning  a  sharp  dorsal  margin  of  the 
rather  thin  rib,  but  it  fails  to  reach  the 
tuberculum.  On  rib  7,  the  ridge  is  present 
on  the  middle  third  of  the  shaft;  on  the 
more  anterior  ribs  it  has  disappeared.  It 
is  in  a  sense  replaced  by  a  thin  dorsal  flange 
extending  distally  on  rib  7  a  short  distance 
outward  from  the  tuberculum,  but  not  con- 
tinuous distally  with  the  typical  dorsal  ridge. 
This  flange  is  present  in  reduced  form  on 
rib  6;  more  anteriorly,  as  far  as  can  be  seen 
from  the  material,  no  noticeable  structures 
are  present  on  the  rib  shafts.  Ribs  6-8 
appear  to  have  lain  beneath  the  scapula; 
these  reduced  flanges,  presumably  func- 
tioning for  the  origin  of  serratus  muscula- 
ture, contrast  with  the  much  greater  flange 
development  seen  in  many  early  tetrapods. 

Posterior  dorsal  ribs  (Fig.  6).  Proceed- 
ing backward  along  the  dorsal  series,  the 
posterior  dorsal  ribs  as  far  as  about  rib  20 
appear  to  be  essentially  similar  to  more 
anterior  dorsals  in  character,  and  show  little 
decrease  in  length;  however,  the  proximal 
end  of  the  longitudinal  ridge  described  for 
the  anterior  dorsals  retreats  distally  to  a 
small  extent  in  this  region.  Rib  20  becomes 
broader  proximally;  in  the  posterior  ribs 
there  is  a  gradual  approximation  of  tuber- 
cular and  capitular  heads  so  that,  from 
about  rib  22  back,  capitular  and  tubercular 


COTYLORHYNCHUS  SKELETON  •  StovaU,  Pitcc,  and  Roiyier        13 


26 


25 


24 


23 


22 


Fig.    6.      Left,    posterior   presacral    ribs   of   the    right   side,  X  'A-     Rig^^i,    left    clavicle,    external    and    internal    views,  X  Vi- 


areas  are  essentially  fused  into  a  single 
articular  surface.  A  decrease  in  length  of 
ribs  is  not  marked  until  approximately  rib 
23,  which  is  notably  shorter  than  that  pre- 
ceding it  (with  a  length  in  CNHM  272, 
as  preserved,  of  413  mm)  and  ribs  24-26 
are  increasingly  short,  the  last  being  about 
158  mm  long  in  CNHM  272.  As  in  pely- 
cosaurs  generally,  the  freedom  of  rib  artic- 
ulation with  the  vertebrae  decreases  pos- 
teriorly; the  last  3  short  ribs  appear  to  be 
well  fused  and  immovable  and  hence  are  to 
be  considered  as  lumbars,  and  the  next  pre- 
ceding may  also  have  been  immovable  in 
life,  although  in  4-0-4  a  suture  between  rib 
and  vertebra  was  seen  during  preparation 
for  the  fourth  presacral. 

All  the  more  posterior  ribs  tend  to  have 
shafts  which  are  relatively  broad  and  thin 
for  most  of  their  length;  the  last  5,  however, 
definitely  taper  to  a  point  distally,  and  thus 
lack  a  sternal  connection,  and  the  same 
may  be  true  of  rib  21.  As  far  as  segment 
23  the  ribs  continue  to  be  curved,  turning 
downward  and  backward  in  life.  Ribs  24- 
26,  however,  are  nearly  straight  and  di- 
rected laterally,  their  length  being  little 
more  than  that  of  the  proximal  nearly 
straight  segment  of  rib  23.  The  breadth  of 
the  proximal  part  of  the  ribs  increases  pos- 


teriorly to  a  maximum  in  rib  24,  which  is 
very  broad  in  proportion  to  its  length. 

Sacral  ribs  (Fig.  3).  In  correlation  with 
the  width  of  the  trunk  and  pelvic  region, 
the  sacral  ribs  are  longer  than  in  sphena- 
codonts  and  ophiacodonts.  Three  sacral 
ribs  are  present.  These  are  tightly  fused 
to  their  vertebrae  over  a  broad  area;  this 
includes  the  short  transverse  processes  which 
arise  from  much  of  the  lateral  surface  of  the 
centra.  No  clear  sutures  are  seen,  but  dor- 
sally  a  rugose  anteroposterior  ridge  marks 
the  line  of  fusion  of  rib  and  transverse  proc- 
ess. The  rib  head  is  deep  dorsoventrally, 
as  well  as  anteroposteriorly.  Distinction  be- 
tween capitular  and  tubercular  attachment 
is  indicated  by  depressions  on  both  anterior 
and  posterior  surfaces  about  half  way  down; 
these  depressions  have  not  been  excava- 
ted, but  may  have  been  connected  by  a 
canal  for  the  intervertebral  artery. 

The  first  sacral  rib  is  short  but  stout. 
Beyond  the  head  it  contracts  somewhat  in 
width  and  extends  outward  horizontally  and 
slightly  posteriorly  to  terminate  in  a  rela- 
tively thin  expanded  blade  extending  down- 
ward and  apposed  laterally  to  the  inner 
face  of  the  ilium.  The  second  rib  extends 
directly  laterally;  it  is  similar  to  the  first 
but  slightly  less  developed.    The  third  rib 


14 


Bulletin  Muscitm  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.    7.      Left,   external,    and   right,   internal    views   of   the    left  scapulocoracoid.     Broken   lines  on  the  internal   view  outline  the 
portion    preserved    in    No.    4-0-6.     Center,    dorsal,    right    lateral,   and   anterior  views  of  the   interclavicle.  X    \/b- 


is  essentially  accessory  in  nature;  it  is  more 
slenderly  built  and  curves  forward  to  but- 
tress the  second  rib  as  well  as  apposing  the 
ilium  with  an  only  slightly  expanded  tip. 

The  first  two  sacral  ribs  are  fairly  com- 
parable to  those  of  Casca.  In  that  genus, 
however,  sacral  rib  3  is  well  developed,  and 
Cotylorhyndnis  here  exhibits  a  condition 
which  is  less  advanced.  Edapliomiini.s 
shows  an  intermediate  condition  in  the  de- 
velopment of  sacral  rib  3. 

Caudal  ribs.  As  in  pelycosaurs  gener- 
ally, ribs  are  present  in  the  proximal  tail 
region.  Tlie  first  five  show  a  fused  attach- 
ment to  transverse  process  and  centrum  with 
a  pattern  similar  to  that  of  the  sacrals,  al- 
though \\'ith  a  steadily  diminishing  attach- 
ment area;  the  heads  are  pierced  antero- 
posteriorly  by  canals  for  the  intervertebral 
artery.  The  first  caudal  is  comparable  to 
the  third  sacral  in  most  regards,  but  it  ex- 
tends directly  laterally,  with  a  slight  distal 
posterior  curvature,  to  terminate  in  a  pointed 
tip.  Posterior  to  this,  all  the  caudal  ribs,  as 
in  pelycosaurs  generally,  continue  to  extend 
outward  in  a  horizontal  plane  and  to  taper 
to  distal  extremities.  In  caudals  2-5  a  back- 


ward curvature  of  the  distal  end  is  pro- 
nounced. These  ribs  show  a  steady  decrease 
in  length  and  stoubiess.  Rib  6  is  notably 
shorter,  with  little  development  of  a  distal 
curved  segment,  and  from  this  point  back 
the  caudal  ribs,  fused  to  the  transverse  proc- 
esses, are  laterally  projecting,  tapering, 
and  pointed  structures  which  decrease  to 
small  nubbins  and  disappear,  except  for 
low  rugosities,  beyond  vertebra  11.  In  gen- 
eral the  caudal  ribs  are  comparable  to  those 
of  Casea. 

GIRDLES 

Shoulder  girdle.  A  nearly  complete  scap- 
ulocoracoid, as  seen  from  the  inner  side, 
is  present  in  4-0-6,  and  the  figure  is  based 
primarily  on  this  specimen.  The  outer  sur- 
face is  not  available  in  this  specimen,  but 
several  others  show  this  aspect.  The  scapu- 
lar blade,  incomplete  in  4-0-6,  is  better 
preserved  in  other  specimens;  its  exact 
height,  however,  is  difficult  to  determine, 
due  to  the  lack  of  distinctive  features  in  this 
area.  The  blade  in  large  specimens  may 
have  been  somewhat  more  developed.    In 


COTYLORHYNCHUS    SKELETON    •    StOVall,    PlicC,    Olid   Roiiwr  15 


all  available  specimens  the  dorsal  margin  of 
the  blade  shows  an  unfinished  surface,  so 
that  there  may  have  been  a  considerable 
cartilaginous  suprascapula.  No  sutures  be- 
tween the  presumed  three  elements  could 
be  determined. 

The  scapulocoracoid  (Fig.  7)  is  edapho- 
saurian  in  general  character.  It  is  short  dor- 
soventrally,  broad  anteroposteriorly,  with 
much  the  proportions  of  Lupeosaurus 
( Ccisca,  presumably  in  relation  to  its  smaller 
size,  has  a  girdle  of  more  slender  build). 
The  scapular  blade  is  very  short  and  broad, 
its  breadth  distally  due  to  a  backward  flare 
of  the  posterior  margin  greater  than  is  seen 
in  any  other  pelycosaur.  Below  the  point 
of  cla\'icular  attachment  the  margin  of  the 
scapula  curves  out  widely  anteriorly,  to 
give  great  breadth  to  the  lower  part  of  the 
blade,  as  in  Edaphosaurus  and  Lupeosaurus. 
In  contrast  to  Edaphosaurus  and  Lupeosau- 
rus, but  in  agreement  with  Casea,  Nito- 
satirus,  and  Mycterosaurus,  there  is  no 
supraglenoid  foramen.  There  is,  as  in  all 
pelycosaurs,  a  screw-shaped  glenoid  cavity; 
it  is  here,  as  in  other  edaphosaurians,  re- 
markably deep  anteriorly.  Below  its  anterior 
end  is  the  external  opening  of  the  supra- 
coracoid  foramen.  Posteriorly,  the  coracoid 
region  shows  little  development  of  a  tu- 
bercle for  origin  of  the  coracoid  head  of  the 
triceps.  On  the  inner  surface,  the  upper 
portion  of  the  subcoracoscapular  fossa  is 
shallow,  in  correlation  with  the  absence  of 
a  supraglenoid  foramen. 

No  cleithrum  is  present  in  the  available 
material.  Much  of  the  clavicles  and  the 
interclavicle  are  present  in  4-0-6  and  in  the 
type  (Figs.  6,  7).  In  contrast  to  Edapho- 
saurus and  sphenacodonts,  there  is  little 
ventral  expansion  of  the  clavicle  (the  clavicle 
of  Casea  is  unknown ) .  Its  upper  portion  is 
well  grooved  posteriorly  to  clasp  the  anterior 
edge  of  the  scapula.  Tlie  blade  of  the  in- 
terclavicle is  unusuallv  broad  and  short, 
and  there  is  no  development  of  the  longi- 
tudinal ridge  seen  on  the  ventral  surface  of 
the  shaft  in  most  pelycosaurs.  As  preserved, 
the  head  turns  sharply  upward  on  the  shaft. 


in  contrast  to  the  gentler  curvature  found 
in  pelycosaurs  generally,  so  that  the  sur- 
faces apposed  to  the  clavicles  are  essen- 
tially in  a  vertical  plane.  The  head  of  the 
interclavicle  is  short  and  T-shaped,  in  con- 
trast to  the  diamond-shaped  head  of  many 
pelycosaurs,  and  is  nearly  completely  oc- 
cupied by  the  pair  of  crescent-shaped  sur- 
faces for  the  clavicles.  These  surfaces,  which 
extend  far  laterally,  are  covered  by  rugose 
striations;  the  areas  on  the  clavicles  which 
meet  them  are  similarly  rugose,  and  set 
in  distinctly  below  the  general  level  of  the 
"inner"  surfaces  of  that  bone.  Obviously 
the  contact  bet\veen  clavicles  and  inter- 
clavicle was  an  intimate  one. 

Pelvic  girdle.  Pelvic  girdle  material 
(Fig.  8)  is  relatively  poorly  represented. 
The  Cotylorhynchus  skeletons  were  nearly 
all  buried  with  the  dorsal  surface  upward; 
the  ilia  in  consequence  have  been  generally 
subject  either  to  damage  by  crushing  dur- 
ing entombment  or  to  erosion  prior  to  dis- 
covery, and  this  element  is  well  preserved 
in  only  a  few  cases.  The  iHac  blade  is 
moderately  high,  as  in  sphenacodonts  and 
other  edaphosaurians,  and  in  contrast  to 
ophiacodonts,  and  agrees  with  other  eda- 
phosaurians in  that  ( in  contrast  with  sphen- 
acodonts) there  is  little  posterior  elonga- 
tion. There  is  a  narrow  but  well-defined 
longitudinal  area  for  muscle  attachment 
at  the  top  of  the  inner  surface,  above  the 
areas  for  the  sacral  ribs.  The  first  sacral 
appears  to  have  been  in  contact  with  a  de- 
pressed area  at  the  anterior  margin  of  the 
inner  surface  of  the  blade,  the  second  pre- 
sumably apposed  to  a  flat  area  posterior  to 
this;  still  farther  posteriorly,  a  well-marked 
internal  depression  received  the  small  distal 
end  of  the  third  sacral. 

The  acetabular  region  is  of  a  typical  pely- 
cosaurian  nature,  with  the  usual  primitive 
dorsal  buttress.  In  most  specimens  sutures 
between  the  three  peKic  elements  are  not 
clearly  seen;  in  one  specimen  part  of  the 
sutures  could  be  made  out  on  the  external 
surface,  and  in  several  instances  lines  of 
striae   on   the  inner  surface  indicated  the 


16        BuUf'lin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.   8.      Right  pelvic  girdle  in  dorsal  and  lateral  views,  X    '/3- 


areas   of  fusion   of  ilium   with   pubis   and 
ischium. 

The  puboischiadic  plate  is  of  very  large 
size,  with  an  anteroposterior  length  in  4-0-4, 
for  example,  of  380  mm,  a  depth  below  and 
internal  to  the  base  of  the  ilium  of  112  mm 
and  a  breadth  of  the  pubis,  measured  at 


right  angles  from  the  front  end  of  the  pubic 
symphysis,  of  173  mm.  There  is  but  a  slight 
indication  in  the  material  of  the  develop- 
ment of  a  pubic  tubercle,  such  as  is  found 
in  Edaphosaurus,  Nitosatinis,  and  Casea. 
The  great  development  of  the  puboischiadic 
plate  is  comparable  to  the  type  of  structure 


CoTYLORHYNCHUS  SKELETON  •  StovciU,  Piicc,  and  Roiuer        17 


seen  in  Edaphosaiirus  and  Casea,  and  is  and  well  preserved  in  a  single  specimen, 
associated  with  bodily  breadth,  present  in  The  pose  was  obviously  that  common  to  all 
the  pelvic  region  as  well  as  farther  forward,  pelycosaurs,  with  humerus  and  femur  pro- 
The  plate  was  strongly  tilted  outward  at  jecting  nearly  straight  outward  horizontally, 
somewhat  more  than  a  45°  angle,  so  that  it  and  with  the  lower  limb  segment  essentially 
is  not  seen  to  any  great  degree  in  side  view,  vertical  in  position.  Front  and  hind  legs 
Internally,  the  puboischiadic  plate  ex-  appear  to  be  nearly  equal  in  length,  but 
hibits,  as  in  pelycosaurs  generally,  a  gently  with  the  humerus  slightly  shorter  in  over- 
hollowed  out  area,  in  which  the  bone  is  all  length  than  the  femur.  In  all  pely- 
relatively  thin,  along  the  middle  third  of  its  cosaurs  the  lower  limb  is  much  shorter  than 
length.  Anterior  and  posterior  to  this,  on  the  proximal  segment,  but  this  relative 
pubis  and  ischium  respectively,  thickened  shortness  is  very  marked  in  Cofylorhyn- 
bony  areas— essentially  supporting  struts—  chits.  Here  radius  and  tibia  are  only  ap- 
descend  from  the  iliac  region  to  the  sym-  proximately  60  per  cent  as  long  as  humerus 
physis.  The  presence  of  these  thickened  and  femur,  respectively.  In  Ophiacodon  the 
areas  is  reflected  in  the  symphysis,  which  radius  is  about  77  per  cent  of  the  humeral 
is  thickened  in  both  pubic  and  ischiadic  length,  the  tibia  about  83  per  cent  of  the 
regions,  with  a  relatively  thin  intermediate  length  of  the  femur;  in  Dimetrodon  the 
zone.  Anterior  to  the  ridge  descending  the  comparable  figures  are  82  and  83  per  cent, 
pubis  is  a  very  broad  area  of  origin  for  The  C otijlorhynchm  proportions,  however, 
puboischiofemoralis  internus,  which  is  but  are  comparable  to  those  in  other  edapho- 
slightly  turned  outward  from  the  general  sauroids;  in  Edaphosaiirus  hoanerges  we 
plane  of  the  internal  surface  of  the  plate,  find  figures  of  62  and  57  per  cent.  Like 
Within  this  area  is  the  usual  internal  open-  other  edaphosauroids,  Cotijlorhynchus  was 
ing  of  the  obturator  foramen  which  opens  "low-slung." 

externally  below  the  pubic  portion  of  the  Humerus  (Fig.  9).     The  humerus  is  of 

acetabulum.    There   is   a   slight   indication  the  primitive  tetrahedral  type,  although  the 

in  certain  specimens  of  a  small  area  along  shaft  is  stouter  than  in  typical  (and  smaller) 

the   dorsal  margin  of  the  ischium   for  an  pelycosaurs.    As  figured,  the  "twist"  of  the 

ischiotrochantericus  origin.  ends  on  one  another  is  extreme;  this,  how- 
ever, appears  to  be  due  to  crushing  of  the 

LIMB  ELEMENTS  (none  too  well  preserved)  specimens  upon 

In  almost  all  instances  the  limb  elements  ^T^^^^'^'  *!^^  t^^^'"?,"  ''  l^^'^^^;  specimen  4-0-3 

1         1                 1    J       J  ri  j.i.        11  snows  the    twist    to  nave  been  one  or  ap- 

nave  been  crushed  and  flattened  dorsoven-  .      ^  ,     ^.^o      *     •        i             i      .          i 

.11            i.1    ,.  ^1                          ..11              1.  proximatelv  90  .    As  is  sphenacodonts  and 

trally,  so  that  they  are  essentially  two-di-  i     i          'a                n      ^.u             •      i 

.       ,           .        .             ,            ^  edaphosauroids  generally,  the  proximal  ar- 

mensional-a  situation  making  correct  re-  ^^^^^i^^.     ^^^^.^^^^     ^^^^^^^     posteriorly     well 

construction  difficult.  ^^^^^^  ^^^^^  ^^^  ^^^^^^^j  ^^^^f^^^  ^f  ^j^^  ^^^^ 

The  major  elements  differ  markedly  from  j^    ^^^    ^erv    considerable    proximo-distal 

those  of  httle  Casea,  and  to  a  lesser  extent  breadth  the  entepicondyle  is  edaphosauroid. 

from  those  of  Edaphosaiirus  and  Lupeosau-  The    entepicondylar    foramen    appears    to 

rus,  in  the  fact  that  they  are  very  stoutly  have  been  unusually  large, 

built.  This  is,  of  course,  a  feature  associated  The  region  of  the  ectepicondyle  and  the 

with   the   large   size   and   great   weight   of  supinator  process  is  imperfect  in  most  speci- 

Cotylorhynchus.   In  most  regards,  however,  mens.  In  small  specimens,  such  as  University 

they  are  not  only  typically  pelycosaurian  but  of  Oklahoma  specimens  bearing  the  num- 

show  a  number  of  distinctive  edaphosauroid  bers  4-0-3  and  2.3-38,  in  which  the  distal  end 

features.  In  few  instances  do  we  find  all,  or  is  well  preserved,  the  supinator  process  is 

most,  of  the  major  Hmb  elements  present  broad  and  close  to  the  ectepicondyle  but 


18 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.    9.      Right   humerus,    viewed    in    dorsal    and    ventral    aspects  in  the  plane  of  the  distal   end,  X   'A- 


separated  from  it  by  the  noteh  typical  of 
pelycosaurs.  Tliis  separation,  however,  is 
obviously  a  growth  stage  only,  for  in  4-0-16 
and  a  further  University  of  Oklahoma  speci- 
men of  uncertain  number  the  notch  is  closed 
and  an  ectepicondylar  foramen  present.  Tlie 
foramen  is  otherwise  found  in  pelycosaurs 
only  in  Edaphosaurus,  and  it  is  stated  to  be 
absent  in  Casca.  In  Caseci,  however,  the 
gap  between  the  tip  of  the  supinator  proc- 
ess and  the  ectepicondyle  is  small,  and 
may  well  have  been  bridged  in  cartilage. 
It  is  obvious  that  the  ectepicondylar  fora- 
men has  developed  more  than  once,  in 
parallel  fashion,  in  early  reptiles;  its  pres- 
ence here  may  perhaps  be  correlated  with 
changes  in  limb  mechanics  and  muscula- 
ture, due  to  increased  size,  rather  than  attrib- 
uted to  inheritance  from  a  basal  edapho- 
sauroid  ancestor. 

Radiu.s  (Fig.  10).  The  radius  is  pre- 
served (although  not  too  well  preserved) 
in  several  instances.    As  noted  above,  it  is 


short,  with  a  length  rather  less  than  three- 
fifths  that  of  the  humerus.  Although  the 
effect  is  in  all  specimens  accentuated  by 
crushing,  the  bone  was  obviously  relatively 
thin  dorsoventrally,  as  in  pelycosaurs  gener- 
ally. As  in  the  case  of  other  limb  bones  of 
Cotylorliynchus,  the  radius  is  broad  in  pro- 
portion to  its  length. 

Tlie  proximal  articular  surface,  where 
preserved,  has  the  appearance  of  an  oval, 
thin  dorsoventrally;  presumably  it  was  sub- 
circular  in  life.  The  dorsal  (extensor)  sur- 
face of  the  shaft  is  convex  in  section;  the 
ventral  surface  apparently  was  flattened.  A 
rugose  area  for  ligament  or  muscle  attach- 
ment is  visible  on  the  lateral  edge  of  the 
dorsal  surface  just  below  the  head  of  the 
bone.  From  the  head  the  bone  (when  un- 
crushed)  constricts  to  a  somewhat  thinner 
shaft.  A  short  distance  below  the  head, 
however,  a  ridge  arises  on  the  medial  sur- 
face of  the  bone.  Proximally  it  begins  some- 
what  toward   the   ventral   surface;    it    ex- 


COTYLORHYNCHUS  SKELETON  •  Stovdll,  Price,  and  Romer        19 


Fig.    10.      Left,    left    ulna    in    extensor    and    flexor    aspects.      Right,  comparable  views  of  tfie   radius.     Right,    below,    proximal 
and   distal    surfaces   of    radius    (dorsal    aspect    above).  X    '/s- 


panels,  however,  to  attain  the  lateral  margin 
and  nuis  do\\ai\vard  mueh  of  the  length 
of  the  bone.  Distally,  the  ridge  is  absorbed 
in  a  distal  expansion  of  the  bone  leading 
toward  the  temiinal  articulation.  As  in 
pelycosaurs  generally,  the  distal  end  of  the 
bone  is  somewhat  curved  ventrally,  so  that 
the  oval  distal  articular  surface  (for  the 
radiale)  faces  somewhat  ventrally  as  well 
as  distally.  The  lateral  margin  of  the  shaft 
is  also  rather  thin,  but  there  is  no  develop- 
ment of  a  projecting  ridge.  Medioventrally 
there  is  an  abiaipt  out-turning  of  the  lateral 
margin  above  the  articular  surface. 

Ulna  (Fig.  10).  The  olecranon  appears 
to  have  become  well  developed  at  a  rela- 
tively early  stage  of  growth,  for  it  is  nearly 
complete  ( although  ^^'ith  a  small  unfinished 
terminal  surface)  in  University  of  Okla- 
homa specimen  N-7-37,  a  small  specimen. 
As  preserved,  the  head  of  the  ulna  is  thin 
where  seen,  but  this  is  presumably  an  effect 


due  to  crushing;  very  probably  the  head 
in  life  was  as  thick  as  in  Edaphosaurus.  As 
in  the  case  of  the  radius,  the  bone  is  very 
short  compared  with  the  humerus.  It  is, 
further,  exceedingly  broad;  the  breadth  of 
the  distal  end  in  one  complete  specimen 
measures  about  40  per  cent  of  the  length  of 
the  bone  as  measured  from  the  lower  mar- 
gin of  the  sigmoid  notch;  and  the  proximal 
width,  across  the  notch,  is  about  50  per  cent 
of  the  length.  These  figures  are  far  in  ex- 
cess of  those  of  other  groups,  in  which  the 
highest  figures  available  to  me  are  29  per 
cent  and  39  per  cent  for  a  specimen  of 
Ophiacodon. 

Femur  (Fig.  11).  The  femur  is  typically 
edaphosaurian  in  nature,  closely  compar- 
able in  every  major  way  to  a  well-preser\'ed 
Lupcosaurus  femur  in  the  Harvard  collec- 
tion and  likewise  comparable,  except  for  its 
stouter  build,  to  the  femora  of  Casea  and 
Edaphosaurus.    As   in   other   edaphosaurs. 


20 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  Xo.  1 


Fig. 


Right    femur,    In    ventral    and    dorsa!    views;    at    right,   proximal   and  distal   viev/s   {dorsal   aspect  above).  X    /3- 


the  curvature  of  the  shaft  characteristic  of 
sphenacodonts  is  absent.  Particularly  char- 
acteristic is  the  ventral  trochanteric  system. 
There  is  a  well-developed  internal  tro- 
chanter from  which,  in  contrast  to  sphena- 
codonts and  ophiacodonts,  a  ridge  descends 
the  under  side  of  the  shaft  diagonally  to- 
ward the  external  condyle.  There  is  little 
indication  of  a  distinct  fourth  trochanter 
along  this  ridge,  and  likewise  little  develop- 
ment of  the  posterior  proximal  branch  of 
the  Y-shaped  ridge  system,  the  intertro- 
chanteric fossa  thus  being  shallow  posteri- 
orly. In  sphenacodonts,  and  to  a  lesser 
degree  in  some  ophiacodonts,  the  proximal 
articular  surface  extends  along  the  proximal 
portion  of  the  posterior  margin  of  the  shaft; 
here,  as  in  other  edaphosaurs,  this  surface 
is  confined  to  the  proximal  end  of  the  bone. 
Proximally,   on   the   dorsal   surface   of   the 


shaft,  there  is  sometimes  seen,  toward  the 
posterior  margin,  a  rugose  area  for  muscle 
attachment.  Unique  is  the  presence  at  the 
anterior  margin  of  a  very  distinct  rugose 
ridge,  about  25  mm  long,  likewise  presum- 
ably for  muscle  attachment;  this  is  clearly 
seen  in  two  specimens.  The  external  condyle, 
as  in  other  edaphosauroids,  projects  very 
markedly  beyond  the  internal  (medial) 
one,  and  in  a  well  developed  specimen  the 
tip  of  this  condyle  markedly  overhangs  the 
articular  surface  for  the  tibia  below  it,  as 
it  does  in  other  edaphosaiuoid  femora  in 
which  ossification  is  well  advanced. 

Tibia  (Fig.  12).  The  tibia  is,  as  noted 
earlier,  relatively  short,  with  a  length  only 
three-fifths  or  less  that  of  the  femur.  As 
with  other  limb  bones,  the  tibia  is  very 
broad,  notably  its  head.  The  width  of  the 
head  in  one  specimen  is  about  57  per  cent 


COTYLORHYNCHUS    SKELETON    •    StOVClll,    PlicC,    Ciud   RoTlier  21 


Fig.    12.      Lefl,    extensor   aspect   of   right   tibio;    center,    extensor    and    flexor    aspects    of    right    fibula;    right,    above,    proximal 
and  distal   surfaces  of  tibia,   and  below,   distal    surface  of  fibula,    extensor  surface   above.  X    '/s- 


the  length  of  the  bone,  and  the  distal  width 
nearly  33  per  cent  of  the  length.  The  closest 
approach  to  these  proportions  is  in  Edapho- 
saunis,  in  which  these  two  figures  approxi- 
mate 50  per  cent  and  30  per  cent.  In  all 
other  kno\vn  pelycosaurs  the  figures  are 
much  lower — Dimetrodon  limbatiis,  for  ex- 
ample, gi\'ing  figures  of  38  per  cent  and  22 
per  cent,  Ophiocodon  37  per  cent  and  26 
per  cent.  The  two  articular  surfaces  of  the 
head  are  distinctly  separated  and  set  off 
from  one  another  at  a  considerable  angle. 
As  in  other  ophiacodonts,  the  lateral  femoral 
articular  area  is  relatively  narrow  dorso- 
ventrally.  The  cnemial  crest  is  little  de- 
veloped. As  in  Edaplwsaurus,  and  in  con- 
trast with  most  other  non-edaphosaurian 
pelycosaurs,  the  distal  articular  surface 
curves  strongly  toward  the  lateral  side  of 
the  bone. 

Fibula  (Fig.  12).  As  in  the  case  of  the 
tibia,  the  fibula  is  very  short  as  compared 
with  the  femur,  and  is  very  broad  distally. 
Tlie  mean  breadth  here  in  three  specimens 
is  38  per  cent  of  the  length.  This  figure  is 
comparable  in  Edaphosoiirus,  but  pely- 
cosaurs   generally    have    a    much    slimmer 


fibula,  with  distal  widths  in  Dimetrodon 
and  Ophiacodon,  for  example,  of  20  per 
cent  and  29  per  cent  of  the  length. 

FEET 

Mantis  (Figs.  13,  14).  As  noted  above, 
the  specimens  of  Cotylorhynchus  are  not 
infrequently  found  in  articulated  fashion, 
and  the  feet  are  sometimes  well  preserved. 
We  may  note,  for  example,  well  preserved 
front  feet  in  4-0-1,  4-0-6  and  4-1-S2,  hind 
feet  in  4-0-10,  4-0-2  (1249),  and  both  front 
and  hind  feet  in  the  Chicago  skeleton.  Be- 
tween the  various  specimens  nearly  all  fea- 
tures of  carpus,  tarsus  and  digits  are  seen. 
As  in  the  case  of  the  major  limb  bones,  the 
feet  are  broad  and  short,  and  thus  differ 
considerably  at  first  glance  from  those  of 
most  pelycosaurs.  Study,  however,  shows 
that,  apart  from  questions  of  proportions 
related  to  the  size  of  the  animals  the  feet 
are  typically  pelycosaurian  and,  despite  the 
contrast  in  shape,  resemble  closely  those 
of  Casea. 

The  manus  has  the  usual  pelycosaur  ele- 
ments, including  a  pisiforme,  two  centralia 
and   a   series   of   five   distal   carpals.    Tlie 


22 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


Fig.  13.  Left,  left  manus  of  No.  4-0-6;  right,  restored  left  menus.  X  'A-  Abbreviations  for  Figs.  13-16:  o,  astragalus;  c, 
centralia;  col,  calccneum;  F,  fibula;  i,  intermedium;  p,  pisiforme;  R,  radius;  r,  radiale;  T,  tibia;  U,  ulna;  u,  ulnare;  1-5, 
distal    carpals    or    tarsals;    /-V,    digits. 


radiale  has  a  very  deep,  essentially  square, 
proximal  articular  surface  for  the  foot  of 
the  radius.  The  intermedium  is  short  and 
broad,  with  well  developed  processes  on 
both  radial  and  ulnar  sides  at  mid-height, 
and  a  broad  proximal  articular  surface  for 
the  ulna.  The  ulnare  is  far  shorter  than  in 
most  pelycosaurs,  but  comparable  in  pro- 
portions to  that  of  Casea.  The  proximal  end 
is  much  less  convex  in  outline  than  in  most 
pelycosaurs;  it  forms  a  nearly  continuous 
articular  surface  for  the  very  broad  ulna; 
this  surface  extends  medially  from  a  con- 
tact with  the  intermedium  to  a  lateral  facet 
for  the  pisiforme.  Tliis  last  element,  as 
usual,  is  a  thin  plate.  In  4-0-6  the  bone  is 
curved  sharply  toward  the  ventral  surface 
at  its  outer  margin;  this  may,  however,  be 
an  effect  of  crushing.  The  medial  or  proxi- 
mal centrale  is  again  relatively  short.  The 
usual  arterial  gap  is  present  between  proxi- 
mal centrale,  intermedium  and  ulnare.  The 
lateral  centrale  is  not  too  well  preserved;  it 
is,  as  in  pelycosaurs  generally,  a  small  ele- 
ment when  viewed  from  the  exterior  sur- 


face, short  proximodistally,  broad  mediola- 
terally.  The  articulated  feet  suggest  that  it 
was  placed  well  in  toward  the  center  of  the 
manus,  with  an  unossified  gap  between 
radiale  and  distal  carpal  1. 

Distal  cai-pal  1  appears  to  be  essentially 
a  simple  rectangle  in  dorsal  outline,  short 
proximodistally  but  broad  mediolaterally, 
covering  the  entire  width  of  the  head  of 
metacarpal  I.  Element  2  is  longer  but  less 
broad,  its  width  less  than  the  overall  width 
of  its  metacarpal;  its  lateral  border  is 
straight,  its  proximal  and  medial  borders 
a  continuous  curve.  Element  3  is  about 
as  broad  as  2,  but  longer  proximodistally. 
As  in  pelycosaurs  generally,  4  is  by  far  the 
largest  of  the  distal  series,  with  a  width 
double  that  of  element  3  and  a  somewhat 
greater  length  proximodistally.  Its  proxi- 
mal end  has,  as  in  pelycosaurs  generally, 
two  articular  faces  at  somewhat  of  an  angle 
to  each  other,  a  laterally  tilted  surface  for 
apposition  to  the  ulnare,  a  shorter  medially 
tilted  face  for  the  proximal  centrale.  Distally 
the  bone  articulates  broadlv  with  the  ex- 


COTYLORHYNCHUS  SKELETON  •  StovdU,  Piicc,  and  Romer        23 


Fig.    14.      leh,   left  manus  of  No.  4-1-S2;   nghl,    left  pes  of  No.    4-0-10.  X    '/j.     Abbreviations    as    in    Fig.    13. 


panded  head  of  metacarpal  IV.  Element  5 
is  narrow  proximodistally  but  is  expanded 
mediolaterally  to  meet  the  entire  breadth  of 
metacarpal  V.  Proximomedial  and  proxi- 
molateral  surfaces  meet  element  4  and  the 
ulnare,  respectively. 

In  the  metapodials  (and  in  the  phalanges) 
the  shortness  and  breadth  of  the  elements 
and  the  almost  complete  absence  of  a  dis- 
tinct shaft  region  give  the  foot  a  clumsy  ap- 
pearance. There  is  a  steady  increase  in 
length  from  metacarpal  I  to  metacarpal  IV, 
and  metacarpal  V  is,  exceptionally,  some- 
what longer  still.  Metacarpal  I  appears  to 
have  a  broad,  flat  head;  in  metacarpals  II 
and  III  the  proximal  articular  surface  is  a 
concave  area  not  occupying  the  full  width 
of  the  bone.  In  metacarpal  IV  the  proximal 
articulation  is  a  concavity,  but  a  very  broad 
one.  In  metacarpals  II  and  III  there  is  a 
pronounced  expansion  of  the  head  toward 
the  lateral  side,  and  in  IV  this  lateral  ex- 
tension is  very  pronounced.  In  metapodial 
V  the  proximal  articulation  does  not  appear 
to  be  cupped;  it  is  tilted  so  that  the  outer 


margin   is  much  more  proximal   than   the 
medial. 

The  phalanges,  hke  the  metapodials,  are 
short  and  massive  in  appearance.  The  na- 
ture and  structure  of  the  articulations  of  the 
elements  is  in  general  of  a  typically  pely- 
cosaurian  type  (Romer  and  Price,  1940: 
167-169).  From  the  massive  build  of  the 
feet  and  the  presumed  herbivorous  habits 
of  the  animal,  one  would  expect  the  toes  to 
terminate,  like  those  of  diadectids  or  parei- 
asaurs,  in  hoof-like  structures.  Instead, 
however,  there  are  (as  in  Casea)  long  and 
powerful  bony  claw  supports,  suggesting 
that  in  life  the  animal  did  considerable  dig- 
ging for  its  food  supply. 

Pelycosaurs,  in  which  good  articulated 
feet  are  known,  typically  have  a  phalangeal 
fomiula  in  the  manus  of  2-3-4-5-3.^  Coty- 
lorhynchus,  as  is  definitely  proven  by  the 
material,  has  the  surprisingly  low  formula 
of  2-2-3-3-2 — even  lower  than  in  typical 
therapsids  and  rivalled  for  reduction  among 

^  Edaphosatinis,  formerly  in  doubt,  is  now  known 
to  have  this  formula  also. 


24        Bulletin  Museum  of  Comparative  Zoology,  Vol  135,  No.  1 


Fig.    15.      Dorsal    and   ventral   views   of   right    pes    of    CNHM    272;    tfie   clav/ed    toes    are    strongly    flexed    and    bent    under    the 
tarsus.  X  'A-     Abbreviations    as    in    Fig.    13. 


Paleozoic  reptiles  only  by  the  pareiasaurs. 
In  Williston's  material  of  Casca,  the  manus 
was  nearly  complete  bvit  for  the  most  part 
disarticulated.  Williston,  not  expecting  re- 
duction, utilized  the  material  available  to 
give  a  manus  with  the  typical  reptilian  for- 
mula (1911:  fig.  13).  To  do  this  he  was  forced 
to  assume  that  a  number  of  elements  were 
missing  from  the  toes.  However,  new  ma- 
terial described  by  Olson  (1954)  shows  that 
the  formula  of  the  manus  was  2-3-3-4-3 — a 
definite  reduction,  although  not  as  marked 
as  in  its  giant  relative.  The  series  of  ele- 
ments present  in  Williston's  specimen  was, 
thus,  actuallv  nearly  complete. 

Pes  (Figs.  14,  15,  16).  Although  the 
material  of  the  hind  foot  is  not  as  good  as 
that  of  the  manus,  nearly  all  the  structure 
can  be  clearly  made  out.  There  is  a  series 
of  typical  tarsal  elements — astragalus  and 
calcaneum  proximally,  and  five  distal  tar- 
sals; there  is  a  lateral  centrale,  but  whether 
a  small  medial  centrale  was  present  is  un- 
certain. The  proximal  tarsal  elements  are 
relatively  short,  as  compared  with  those 
of  most  other  pelycosaurs  except  for  the 
ophiacodonts;  this  presumably  in  relation  to 
ponderous  build.  The  astragalus  appears 
to  have  had  a  relatively  flat  facet  for  the 
tibia.  There  is  a  typical  arterial  notch  be- 
tween artragalus  and  calcaneum. 

As  in  the  manus,  the  reduced  phalangeal 
formula  of  2-2-3-3-2  was  present.   Williston 


attempted  to  restore  the  foot  of  Casea  with 
the  primitive  formula.  But,  as  his  descrip- 
tion suggests,  little  of  the  material  was  actu- 
ally articulated  and  it  seems  reasonable  to 
believe  that  Casea  had  a  reduced  pha- 
langeal foiTuula  in  the  pes  similar  to  that  in 
the  manus;  if  so,  Williston's  foot  material 
was  nearly  complete. 

ABDOMINAL  RIBS 

In  agreement,  it  would  seem,  with  the 
fact  that  in  the  Edaphosauria  generally  the 
gastralia  are  little  developed,  no  trace  of 
abdominal  ribs  was  discovered  during  prep- 
aration of  the  materials  except  in  one  in- 
stance. Here  there  were  found  numerous 
slender  elements  with  tapering  ends.  The 
maximum  length  as  preserved  was  65  mm; 
the  widths  3  to  5  mm.  Tlie  material  was 
not  sufficient  to  determine  their  arrange- 
ment, although  they  were  presumably  ar- 
rayed in  the  usual  series  of  V-shaped  seg- 
ments along  the  course  of  the  abdomen. 

RESTORATION 

A  lateral  view  of  a  restoration  is  shown 
in  Figure  17,  based  on  the  series  of  larger 
specimens.  The  general  appearance  is  com- 
parable to  that  sho\Mi  in  restorations  of  its 
smaller  relative,  Caseo  (Williston,  1911: 
frontispiece;  Romer  and  Price,  1940:  fig. 
71),  except  for  the  somewhat  more  massive 
build  of  Cotylorhynchus  associated  with  its 


CoTYLORHYNCHUS  SKELETON  •  Stovdll,  Price,  and  Ronier        25 


Fig.    16.      Restored    left   pes,  X   Vj-     Abbreviations   as   in    Fig. 
13. 


larger  size,  and  the  absurdly  small  size  of 
the  head.  The  lateral  view  does  not,  of 
course,  give  proper  emphasis  to  the  great 
breadth  of  the  barrel-like  trunk,  commented 
on  previously,  and  well  shown  in  Wil- 
liston's  photograph  of  the  mounted  Casea. 
The  build  of  CotyloihyncJuis  is  in  agree- 
ment with  the  portrait  of  a  generalized 
edaphosaurian  given  by  Romer  and  Price 
(1940:  377): 

"We  find  a  tiny  head  armed  \\ith  a  powerful 
battery  of  blunt  teeth,  the  trunk  a  large,  broadly 
rounded  barrel,  the  legs  spread  out  broadly,  but 
the  lower  segments  so  short  that  the  belly  cannot 
have  been  far  clear  of  the  ground.  Such  an  ani- 
mal was  obviously  not  a  carnivore,  and,  in  the 
discussion  of  habits,  we  have  already  cited  data 
supporting  Williston's  belief  that  these  reptiles 
were  herbivores.  The  enormous  storage  capacity  of 
the  abdomen  further  suggests  that  the  food  was 
probably  of  a  bulky,  watery  nature,  low  in  nutri- 
tive value,  so  that  it  was  necessary  for  the  animal  to 
ingest  large  quantities.  The  curiously  small  head  of 
the  advanced  edaphosaurs  is  matched  among  later 
plant-eating  reptiles  by  the  sauropods,  whose  food 
may  have  been   of  a  comparable   type." 

Cotijlorhyncluis  romeri  exceeds  in  bulk 
any  of  the  known  pelycosaurs  from  the 
typical  Texas  Wichita  and  Clear  Fork  red- 
beds  deposits,  and  is  exceeded  only  by  its 
presumed  descendant,  C.  hancocki  from  the 


t—C-^ 


TS 


r^^ 


c 
o 

E 


c 
-c 


o 
U 


o 
o 


26        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


San  Angelo  (Olson  and  Beerbower,  1953; 
Olson,  1962:  28-45).  In  the  stndy  of  pely- 
cosauis  generally,  Romer  ( Romer  and  Price, 
1940;  Romer,  1948)  used  as  an  index  to 
relative  size — and  \\eight — an  "orthometric 
linear  unit"  based  on  the  dimensions  of 
dorsal  vertebrae.  At  the  time  of  publication 
of  the  "Review  of  the  Pelycosauria"  such 
infonnation  as  was  then  available  concern- 
ing C.  romer i  suggested  that  this  unit 
was  approximately  8.32,  and  this  appears 
still  to  be  a  reasonable  figure.  Closest  to 
Cotylorliynchu.s  romeri  among  Wichita  and 
Clear  Fork  pelycosaurs  were  the  large  ter- 
minal member  of  the  Ophiacodon  phylum, 
O.  major.,  with  an  orthometric  linear  unit  of 
7.37,  and  tiie  large  tenninal  Clear  Fork 
Dimcirodon  grandis,  at  7.61.  Since  the 
Casea  unit  is  but  3.30,  the  average  linear 
measurements  of  elements  of  Cotylorhyn- 
chus  should  be  approximately  two  and  one- 
half  times  that  of  Casea,  and  the  weight 
more  than  15  times  as  great.  With  an  esti- 
mated weight  of  about  331  kg — roughly 
about  one-third  of  a  ton — Cotylorhynehu.s 
was  the  giant  of  its  times. 

RELATIONSHIPS 

When  Casea  was  first  described  it  oc- 
cupied an  isolated  position  among  pely- 
cosaurs— so  isolated,  indeed,  that  Watson 
(1917:  173)  suggested  that  it  was  not  a 
pelycosaur  at  all.  Possibly  Tricha.murus  (cf. 
Romer  and  Price,  1940:  422-423)  is  a  rela- 
tive, but  until  the  discovery  of  Cotylo- 
rhynchus  no  further  members  of  the  family 
Caseidae  were  recognized.  In  recent  years, 
however,  Olson  (1962:  24-47;  and  earHer 
papers )  has  added  a  number  of  new  fonns 
to  the  group  from  the  middle  and  upper 
Clear  Fork  fonnations  and  the  lower  part 
of  the  Pease  River  group,  including  further 
species  of  Casea  and  Cotylorhynchiis,  and 
the  new  genera  Caseoides,  Ca.^eo))sis,  and 
An<ie}osaurus;  further,  the  caseids  are  now 
known  to  have  ranged  widely,  for  Ennato- 
sauriis  of  the  Russian  Kazanian  is  clearly  a 
caseid  (as  is  possibly  the  poorly  known 
Phreatophasma) . 


The  caseids  are  the  last  major  group  of 
pelycosaurs    to    appear    in    the    geological 
record.    Casea  broilii,  the  earliest  acknowl- 
edged member  of  tlie  family,  only  appears 
at  about  the  Arroyo-Vale  boundary  in  the 
Clear  Fork  group;  other  forms  occur  in  the 
later  Clear  Fork  formations  and  the  roughly 
equivalent  Hennessey  of  Oklahoma,  or  the 
still  later  Pease  River  group  of  Texas  and 
the  Russian  Kazanian.   Although  some  mil- 
lions of  years  must  be  allowed  for  the  de- 
velopment  of   caseid   specializations,    it   is 
quite  possible  for  the  family  to  have  evolved 
from  some  more  generalized  group  during 
Wichita  and  early  Clear  Fork  times.    Until 
recently  clues  as  to  caseid  ancestry  were 
few.    Tricha.satirus  of  the  Arroyo  Forma- 
tion and  Glaucosaiirus  of  the  Clyde  Forma- 
tion of  Texas  were  suggested  by  Romer  and 
Price    (1940:    421-423)    as   presumed  eda- 
phosauroids  possibly  related  to  the  caseids. 
In  the  first-named  genus  the  skull  is  un- 
known.^   The   second  is   represented   only 
by  a  single  small  skull  which  is  extremely 
short-faced  and  with  an  isodont  dentition; 
this  suggests  possibilities  of  relationship  to 
the  caseid  pedigree.   As  to  a  more  remote 
ancestry,     it    was     suggested    by     Romer 
(Romer,  1937;  Romer  and  Price,  1940:  405- 
412)   that  Mycterosaurus  and   Nito.saiiriis, 
small    early    Permian    pelycosaurs    which 
seemed  to  be  primitive  in  many  ways  but 
showed  definite  edaphosaurian   characters 
in  the  postcranial  skeleton,  might  represent 
the  ancestral  stock  of  the  caseids  and  per- 
haps of  the  edaphosaurians  as  a  whole. 

In  recent  years  several  new  finds  have 
added  somewhat  to  the  picture.  Vaughn 
(1958)  has  described  as  Colobomycter  an 
imperfect  skull  from  the  Fort  Sill  quarry 
(an  Arroyo  equivalent  in  Oklahoma),  and 
considers,  reasonably,  that  its  characters  in- 
dicate that  "it  provides  a  good  structural  if 
not  an  actual  ancestor  for  the  family  Casei- 
dae."   Fox   (1962)   has  described  as  Delo- 


1  A  toothplate  provisionally  referred  to  this  genus 
(Romer  and  Price  1940:  423)  is  now  known  to  per- 
tain to  the  cotylosaur  Labidosaurikos. 


COTYLORHYNCHUS  SKELETON  •  Stovull,  Piicc,  and  RoTTier        27 


rhynchus  three  isolated  maxillae  from  this  saurus  had  more  teeth  than  Oedaleops.    In 

same  quarry;  the  bone  itself  is  caseid-like,  the  latter  genus  the  dentary  is  not  known 

but  the   dentition   is   primiti\e   and   hence  from  associated  material.    In  the  two  max- 

Fox  classes  it  amongst  the  presumably  an-  illae  of  Oedaleops  found  by  Langston,  the 

cestral  Nitosauridae  rather  than  placing  it  tooth  counts  appear  to  be  IS  andl6;  in  the 

in   the   Caseidae.    As   noted   by   Langston  incomplete  Nifosaunis  maxilla  15  teeth  and 

(1965)   there  is  little  to  distinguish  Delo-  alveoli  are  present,  and  the  total  count  was 

rhynchus  from  its  quarr>'-mate  Colobomyc-  probably  about  18.    (4)  There  appears  to 

ter.  Langston  (1965)  has  recently  described  be  no  significant  difference  in  the  height  of 

as  Oedaleops  a  small  pelycosaur  from  the  the  maxilla  bet^^'een  Oedaleops  and  Nito- 

New  Mexican  Permian  represented  by  two  .sa»;f/.s— particularly  if  the  ob\'ious  crushing 

skulls  and  other  fragmentary  material.    As  undergone  by  the  Oedaleops  skull  be  taken 

Langston  points  out,  the  Oedaleops  skull  into  account. 

is  of  a  type  quite  surely  expected  in  a  caseid  The  one  possibly  valid  generic  distinction 

ancestor,  but  differs  from  proper  members  lies  in  the  greater  isodontv  of  the  maxillary 

of  that  family  in  that,  for  example,  the  face  dentition  in  Oedaleops.  In  the  type  there  is 

is  not  as  abbreviated  and,  most  especially,  a  modest  development  of  a  "canine"  pair 

the  dentition  is  primitive;  as  proper  for  a  at  maxillary  positions  2  and  3;  in  a  second 

primitive  pelycosaur  of  any  sort,  the  teeth  specimen  the  third  tooth  is  large;   in  the 

are  sharp-pointed  and  somewhat  recurved  A' /fo.sfl«/f/5  maxilla  a  maximum  is  not  gained 

and  \\ith  a  modest  development  of  a  maxi-  until  we  reach  teeth  4  and  5.    Considering 

mum  tooth  size  in  the  canine  region,  in  one  the  constant  tooth  replacement  characteristic 

specimen,  at  least.   As  possibly  attributable  of  reptiles   and   the   consequent   continual 

to  Oedaleops,  Langston  describes  a  number  changes  in  the  aspect  of  a  dentition,  this  one 

of  small  postcranial  elements  found  in  the  feature  seems  hardly  safe  ground  for  generic 

same  quarry.   Of  these,  the  ilium,  as  Langs-  distinction. ^ 

ton  notes,  is  of  a  very  primitive  type,  cer-  Although  the  material  of  the  genera  dis- 
tainly  not  expected  in  a  pre-caseid.  Other  cussed  above  is  quite  incomplete,  in  most 
elements,  most  notably  the  scapulocoracoid  instances,  it  appears  that  in  this  we  have 
(lacking,  significantly,  the  supraglenoid  at  least  the  beginnings  of  a  phyletic  series 
foramen),  are  comparable  to  those  of  Nito-  leading  from  such  a  primitive  but  edapho- 
saiinis  and  caseids.  They  may  well  pertain  sauroid  pelycosaur  as  Myctewsauriis  up- 
to  A  itosaurus.  ward  toward  the  caseid  condition.  A  com- 
Moreover,  may  not  Oedaleops  and  Nito-  plicating  and  confusing  factor  in  the  sit- 
saitnis,  contemporaries  from  the  same  region  nation,  however,  was  introduced  by  Watson 
and  horizon,  be  identical?  Of  Oedaleops  (1954:  356)  with  his  suggestion  that  £of/?{/m 
we  ha\'e  no  certain  knowledge  of  postcranial  might  be  related  to  caseid  ancestry.  Both 
material;  of  N itosaurus  we  have  no  skull  Vaughn  and  Langston  have  adopted  this 
material  except  of  maxilla.  Langston  briefly  point  of  view,  placing  such  forms  as  Col- 
mentions  this  possibility  but  says  that  Nito-  obomycter  and  Oedaleops  in  the  Eothv- 
saurus  had  longer  jaws,  a  more  slender  j-ididae,  and  Langston  goes  to  the  extreme 
dentary,  considerably  more  teeth  (sub-  ^f  excluding  the  Nitosauridae  from  anv  re- 
isodont  m  fonii)  and  a  higher  maxillary  lationship  to  the  Caseidae,  despite  the  num- 
bone.  But  1 )  we  do  not  have  a  complete  ^  i  i  •  •£•  '  ^  .  , 
•  „.  ■  v/  /T\  .1  11  erous  and  surelv  significant  postcranial  re- 
aw  in  A /fo.rai/n/.s;  (2)  the  seeming  slender-            i,            u  ^  u      i_ 

r  .1      TVT-x  1  .  ,    1  ,  semblances  between  the  two. 

ness  of  the  N itosaurus  dentary  is  probablv  t^i     i      .  r  j     •       r-  ^7      • 

1^.1  r  i.u     ^i-  •     1  ■      /    r  ^  he  basic  reason  tor  considering  Eothiiris 

due  to  loss  of  the  thm  lower  margin   (cf.  

Romer  and  Price:    fig.   70,  and  Langston:  1  a      \.    a-    ^  a,.-    v  ..    n     •    t 

r.       ^.,  °  '  °  '■A.  subordinal  distinction,  actually,  in  Langston  s 

tig.  2a  ) ;  ( 3 )  there  is  no  evidence  that  Nito-  chart,    page  43. 


28        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


as  a  possible  relative  of  the  caseids  is,  of 
course,  the  fact  that  Eothyris,  like  the 
caseids,  is  remarkably  short-faced.  This  in 
itself  is  no  more  a  valid  reason  for  associat- 
ing them  than  would  be  the  association  of 
the  sphenacodont  Secodontosaiirns  with  the 
ophiacodont  Varanosaurus  because  they 
are  both  extremely  long-snouted.  If  the 
Eothyris-cAseid  relationship  is  to  be  sub- 
stantiated, more  positive  reasons  must  be 
developed. 

Langston  (1965:  21)  cites  fourteen  points 
in  which  Oedaleops  and  Eothyris  are  in 
partial  or  complete  agreement.  As  he  says, 
this  seems  to  be,  at  first  sight,  an  imposing 
list.  Included,  of  course,  is  the  fact  that  the 
face  is  short,  not  necessarily  meaningful, 
and  the  fact,  of  no  systematic  value,  that 
both  are  small.  Correlated  with  small  size, 
and  hence  likewise  without  other  necessary 
significance,  is  the  relatively  large  size  of 
the  orbits  and  of  the  pineal  foramen.  A 
number  of  other  common  features  are  such 
as  are  liable  to  be  present  in  any  relatively 
primitive  pelycosaur,  including:  (1)  rela- 
tively flat  skull;  (2)  jaw  articulation  on  a 
level  with  the  toothrow,  as  in  the  Ophia- 
codontia,  the  primitive  sphenacodont  Var- 
anops,  and  Mycterosoiirus;  (3)  outward 
slope  of  cheek  plates  (contrasting  with 
sphenacodonts ) ;  ( 4 )  a  primitive  long  lacri- 
mal; (5)  nonnal  relationship  of  roofing 
bones;  (6)  an  unusually  large  supratemporal; 
(7)  a  long,  tapering  postorbital  (as,  for  ex- 
ample, in  the  ophiacodont  Varanosaurus, 
and  the  sphenacodont  Varanops);  (8) 
some  indications  in  the  tabular-supratem- 
poral  region  of  the  otic  notch  of  ancestral 
types;  (9)  a  differentiated  dentition,  as  in 
most  pelycosaurs  except  Edaphosaurus  and 
caseids. 

The  two  genera,  thus,  are  short-faced, 
small  in  size,  and  have  both  retained  various 
primitive  characters.  Little  remains  of  the 
original  fourteen  points  which  can  be  con- 
strued as  positive  indication  of  relationship. 
Langston  cites  "relative  position  of  orbits 
and  pineal  opening,"  but  there  does  not 
appear  to  be  any  unusual  condition  fiere  in 


either  case.  With  regard  to  "enlarged  nares 
and  obtuse  rostrum,"  the  nares  in  Eothyris 
do  not  appear  to  be  any  larger,  propor- 
tionately, than  in  many  other  pelycosaurs, 
and  the  rostrum  does  not  appear  to  be  any 
more  obtuse  than  in  pelycosaurs  generally. 
The  Eothyris  prefrontals  are  said  to  be  "in- 
flated" (i.e.,  somewhat  expanded),  but  this 
does  not  seem  significant. 

There  is  thus  little  positive  reason  to 
associate  Eothyris  with  caseid  ancestry,  and 
one  very  strong  objection — the  dentition. 
In  all  early  pelycosaurs — indeed,  in  nearly 
all  primitive  tetrapods  generally — there  is  a 
trend  for  the  development  of  somewhat 
enlarged  teeth  near  the  front  of  the  maxilla 
as  an  incipient  "canine"  region.  In  sphena- 
codonts and  their  therapsid  descendants 
this  trend  is  accentuated;  in  edaphosaurians 
— both  Edaphosaurus  and  the  caseids — 
there  is  an  opposite  trend  toward  isodonty. 
In  such  forms  as  Oedaleops  there  is  but  a 
mild,  essentially  primitive,  development  of 
a  canine  "maximum"  such  as  might  be  ex- 
pected in  any  relatively  primitive  pely- 
cosaurs. In  Eothyris,  on  the  other  hand,  we 
find  the  greatest  exaggeration  of  canine  tusks 
to  be  found  in  any  pelycosaur.  In  their 
dentition,  Eothyris  and  the  caseids  have 
evolved  in  such  diametrically  opposite 
directions  that  it  is  difficult  to  believe  that 
they  are  at  all  closely  related. 

But  even  if  (as  is  not  too  probably  the 
case)  Eotliyris  should  prove  to  be  related  to 
the  caseids,  it  is  dangerous  to  base  hypoth- 
eses of  broader  relationships  on  the  "family 
Eothyrididae,"  as  has  been  done  by  Langs- 
ton and,  to  a  lesser  degree,  by  Vaughn. 
As  I  have  pointed  out  (Romer  and  Price, 
1940:  247;  Romer,  1956:  676),  this  family 
is  a  purely  provisional  one,  set  up  to  receive 
fonus,  presumably  highly  predaceous,  which 
have  marked  canine  development  but  are 
not  members  of  the  Sphenacodontia.  Any 
unity  the  group  might  have  is  based  on 
this  dental  feature — which  is,  of  course, 
the  one  point  in  which  all  of  them  notably 
differ  from  the  trend  toward  isodonty  ex- 
pected   in    caseid   ancestors.     There    is    no 


COTYLORHYNCHUS  SKELETON  •  Stovoll  Pricc,  and  Romer        29 


evidence  that  any  form  assigned  to  this 
family,  other  than  Eofhyiis,  was  short-faced. 
Such  postcranial  material  as  can  be  as- 
sociated with  any  of  the  genera  included 
in  this  provisional  family  lacks  any  features 
indicative  of  caseid  relationships,  and  such 
indications  of  systematic  position  as  are 
shown  by  Stcrcorhachis,  StcrcophaUodon, 
and  Bcddwinonus  suggest  the  Ophiacodon- 
tia.  With  regard  to  Eothyris,  the  lack  of 
postcranial  data  is  a  stumbling  block. 

On  the  assumption  that  the  Ophiacodontia 
represent  the  basal  stock  of  the  Pelyco- 
sauria,  it  is  a  reasonable  assumption  that 
the  caseids  are  of  ultimate  ophiacodontian 
derivation;  but  that  any  of  the  "eothyrids" 
are  connecting  links  is  very  dubious. 

Romer  and  Price  (1940:  366-376)  pointed 
out  numerous  skeletal  features  which  Eda- 
phosaiiriis  and  Casea,  as  two  extremes,  have 
in  common,  and  advocated  their  being 
placed  in  a  common  suborder.  There  is, 
however,  a  seeming  difficulty  in  assuming  a 
common  ancestry,  because  of  chronological 
factors.  The  caseids  seem  quite  surely  to 
have  specialized  from  primitive  ancestors 
at  a  late  date,  for  no  form  attributable  to 
this  stock  is  known  earlier  than  fairly  early 
Wichita  times;  on  the  other  hand,  Edapho- 
saunis  had  already  evolved  by  the  late 
Pennsylvanian.  If  both  Edaphosourm  and 
the  caseids  evolved  from  essentially  primi- 
tive edaphosaurians,  such  as  the  nitosaurs, 
this  must  have  been,  as  Langston  (1965:  58) 
notes,  a  very  bradytelic  group,  and  nitosaurs 
should  have  been  in  existence  in  the  Penn- 
sylvanian. 

This  appears  to  have  been  the  case.  It 
seems  highly  probable  that  Pctrohcosaums, 
from  the  Pennsylvanian  Garnett  shales  of 
Kansas  (Lane,  1945;  Peabody,  1949,  1952), 
is  a  primitive  edaphosauroid. 

Lane,  and  Peabody  at  first,  concluded 
that  this  small  reptile  was  a  pelycosaur.  The 
latter  author,  however,  impressed  by  cranial 
resemblances  to  Prolaceita  (with  which  he 
was  famihar),  argued  that  Petrolacosaunis 
was  an  eosuchian — a  primitive  diapsid. 
There  is  no  morphological  proof,  one  way 


or  another,  of  this  suggestion.  There  was  a 
lateral  temporal  opening,  but  the  cheek  is 
poorly  preserved,  and  whether  an  addi- 
tional upper  opening  was  present  cannot  be 
determined.  Apart  from  the  possible  l)ut 
unproven  diapsid  nature  of  the  temporal 
region,  there  is  no  reason  to  assign  Petroh- 
cosaurus  to  the  Eosuchia.  Peabody  points 
out  a  number  of  common  features  of  the 
skulls  in  Petrolacosaunis  and  Prohceria, 
but  these  are  essentially  primitive  characters 
which  could  have  been  inherited  by  both 
from  captorhinomorph  cotylosaurian  an- 
cestors. The  time  of  appearance  of  Petrola- 
cosmirus  is  one  at  which  an  eosuchian  is 
hardly  to  be  expected.  Tliere  are  no  traces 
of  any  diapsid  in  the  Lower  Permian,  and 
no  sure  evidence  in  the  Middle  Permian;  the 
first  certain  diapsid  is  Upper  Permian  in 
age — a  full  period  after  Petrolacosaunis — 
and  Watson  ( 1957 )  has  argued  that  diapsids 
were  only  then  evolving  from  millerettid 
cotylosaur  derivatives. 

If,   then,   we   abandon  the  possible   but 
improbable  suggestion  that  Petrolacosaurus 
was  a  precocious  diapsid,  all  the  features 
of  this  little  reptile  agree  with  the  assump- 
tion that  we  are  dealing  with  a  pelycosaur, 
and  several  characters  point  strongly  to  the 
suggestion  that  it  is   a   primitive   edapho- 
saurian.    Tlie  skull  is   primitive   and   gen- 
erahzed,  as  it  is  in  such  archaic  ophiacodonts 
as  Clepsijdrops  and  Varanosaurus,  such  a 
sphenacodontian  as  Varanops,  and  such  a 
nitosaurian   as   Mycterosaurus.    Diagnostic 
features,    however,    can   be   found   in   the 
postcranial  skeleton.    (1)  The  postcervical 
vertebral  centra  are  rounded  ventrally  as 
in  ophiacodonts  and  edaphosaurs,  in  con- 
trast to  the  keeled  sphenacodonts.    (2)  In 
the  carpus  the  ulnare  is  short,  in  contrast 
to  advanced  sphenacodonts  and  as  in  ophia- 
codonts  and   edaphosaurs    (however,   Var- 
anops among  the  sphenacodonts  also  has  a 
short  ulnare).    (3)  In  the  tarsus  there  is  a 
broad,   higlily   developed,   medial   centrale 
fonning  the  sole  connection  between  astrag- 
alus and  distal  tarsals  1  and  2;  edaphosaurs 
and  sphenacodonts  are  similar,  but  ophia- 


30        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  1 


codonts  have  two  small  centralia  of  sub- 
equal   size   in   this   position.     (4)    On   the 
femur,   the   ventral   system   of   trochanters 
and  ridges  is  one  found  among  pelycosaurs 
in  edaphosaurs  only  (cf.  Romer  and  Price, 
1940:   fig.  37).    Distal  to  the  internal  tro- 
chanter an  adductor  crest  slants  diagonally 
across  the  bone  towards  the  external  tro- 
chanter.   In  sphenacodonts  there  is  a  pro- 
nounced fourth  trochanter  but  no  adductor 
crest;  in  opliiacodonts  the  crest  descends 
the  external  margin  of  the  bone.    ( 5 )  Both 
Lane   and   Peabody   ascribe   to   Petrolaco- 
■sauni.s  a  pelvis   (University  of  Kansas  no. 
1425)   which  is  very  distinctively  edapho- 
saurian.   It  is  of  a  type  strongly  contrasting 
wdth  that  of  any   other  reptilian   group — 
most    notabh'    in    the    shovel-shaped    iliac 
blade,  tall  but  without  a  marked  posterior 
projection.    Were  this  pelvis  definitely  as- 
sociated, Petwhicosxninis-  could  be  assigned 
to  the  Edaphosauria  without  hesitation.    It 
was,  however,  found  isolated,  and  since  a 
specimen  of  Edaphosainus  has  been  found 
in  the  quarry,  this  pelvis  may  pertain  to 
that  genus.    Several  pelves  with  low,  long 
iha  of  ophiacodont  type  are  also  present  in 
the  material;  but  these,  too,  lack  association, 
and  may  belong  to   an   ophiacodont,   also 
present  in  the  Garnett  material. 

There  are,  thus,  in  the  postcranial  skele- 
ton of  Petwlacosaiinis  a  number  of  features 
which  strongly  indicate  that  this  genus  be- 
longs to  a  group  of  archaic  edaphosaurians 
from  which  both  Edaphosaurus  and,  at  a 
much  later  time,  the  caseids  may  have 
arisen. 

Preparation  of  material  for  this  paper 
was  aided  by  a  grant  (No.  GB  500)  from 
the  National  Science  Foundation. 

LITERATURE  CITED 

Fox,  R.  C.  1962.  Two  new  pelycosaurs  from  the 
Lower  Permian  of  Oklahoma.  Univ.  Kansas 
Publ.  Mus.  Nat.  Hist.,  12:  297-307. 

Lane,  H.  H.  1945.  New  mid-Pennsylvanian  rep- 
tiles from  Kansas.  Trans.  Kansas  Acad.  Sci., 
47:  381-396. 

Langston,  W.  1965.  Oeduleops  campi  (Rep- 
tilia:  Pelycosauria ) .    A  new  genus  and  species 


from  the  Lower  Permian  of  New  Mexico,  and 
the  family  Eothyrididae.  Bull.  Texas  Mem. 
Mus.,  9:  1-46. 

Olson,  E.  C.  1954.  Fauna  of  the  Vale  and 
Choza.  7.  Pelvcosauria:  family  Caseidae. 
Fieldiana,  Geology,   10:    193-204. 

.  1962.  Late  Permian  terrestrial  verte- 
brates, U.S.A.  and  U.S.S.R.  Trans.  Amer. 
Philos.  Soc,  (N.S.)  52  (part  2):  1-196. 

Olson,  E.  C.  and  J.  R.  Beerbower,  Jr.  1953. 
The  San  Angelo  Formation,  Permian  of  Texas, 
and  its  vertebrates.    Jour.  Geol.,  61 :  381-423. 

Peabody,  F.  E.  1949.  Mid-Pennsylvanian  pely- 
cosaurs from  Kansas.  Bull.  Geol.  Soc.  Amer., 
60:  1913. 

.      1952.     Pctrolaco.miints  kansensis  Lane,  a 

Pennsylvanian     reptile     from     Kansas.     Univ. 
Kansas,  Palcont.  Contrib.  Vertebrata,  1:  1-41. 

Romer,  A.  S.  19.37.  New  genera  and  species  of 
pelvcosaurian  reptiles.  Proc.  New  England 
Zool.  Club,  16:  89-96. 

.     1948.     Relative  growth  in  pelvcosaurian 

reptiles.    Robert  Broom  Commemorative  Vol- 
ume.  Pp.  45-55. 

1956.     Osteology   of   the   reptiles.     Chi- 


cago, Univ.  Chicago  Press,  xxi  -f-    772  pp. 
Romer,  A.  S.  and  L.  I.  Price.     1940.     Review  of 

the  Pelycosauria.  Geol.  Soc.  Amer.,  Spec.  Pap. 

No.  28:  1-538. 
Shuler,  E.  W.  and  R.  V.  Witter.     1942.     The 

mounted  skeleton  of  Edaphosaurus  boanerges 

Romer    at    Southern     Methodist     University. 

Field  and  Laboratory,   10   (2):    140-144. 
Stovall,    J.    W.     1937.     Cotijlorhynchus    romeri, 

a   new    genus    and    species    of   pelvcosaurian 

reptile  from  Oklahoma.    Amer.  Jour.  Sci.,  (5) 

34:  308-313. 
Vaughn,  P.  P.      1958.     On  a  new  pelycosaur  from 

the  Lower  Permian  of  Oklahoma,  and  on  the 

origin  of  tlie  family  Caseidae.    Jour.  Paleont., 

32:  981-991. 
Watson,  D.  M.  S.     1913.     The  Beaufort  Beds  of 

the    Karroo    System    of    South    Africa.    Geol. 

Mag.,  (5)   10:  388-393. 
.      1917.     A    sketch    classification    of    the 

pre-Jurassic  tetrapod  \'ertebrates.    Proc.  Zool. 

Soc.  London,  1917:   167-186. 
.      1954.     On    Bolosaurus    and    the    origin 

and     classification     of    reptiles.      Bull.     Mus. 

Comp.  Zool,  111:  295-449. 

1957.     On   Millerosaunis  and   the   early 


historv  of  the  sauropsid  reptiles.   Philos.  Trans. 

Roy.  Soc.  London,  (B)  240:  325-400. 
WiLLisTON,  S.  W.     1910.     New  Permian  reptiles: 

rhachitomous     vertebrae.      Jour.     Geol.,     18: 

585-600. 
.      1911.     American     Permian    vertebrates. 

Chicago,  Univ.  Chicago  Press,  145  pp. 
{Received  27  September  1965.) 


SulLetln  OF  THE 

Museum   of 

ComparaHve 

Zoology 


The    Stromateoid    Fishes:    Systematics   and 

a    Classification 


RICHARD  L.  HAEDRICH 


M««exmi  'Of"  ConipernBtlve  Zook>gy,  Cambridge,  Massachusetts 
-  l5titutoi^j_JipiP9io  Generole,  Universita  di  Pisa,  Italy  -  . 


HARVARD  UNIVERSITY 

CAMBRIDGE,  MASSACHUSETTS,  U.S.A. 


VOLUME  135,  NUMBER  2 
JANUARY  27,  1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

MEMoms  1864-1938 

JoHNSONiA,  Department  of  MoUusks,  1941- 

OcxziAsiONAL  Papers  on  Mollusks,  1945- 


Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).   $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.  $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.   ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


@  The  President  and  Fellows  of  Harvard  College  1967. 


THE  STROMATEOID  FISHES:   SYSTEMATICS  AND  A  CLASSIFICATION' 


RICHARD    L.    HAEDRICH- 


TABLE  OF  CONTENTS 

Historical  introduction  31 

Methods  35 

Material   38 

Systematic  section  43 

Suborder  Stromateoidei   44 

Key  to  stroniateoid  families  52 

Family  Centrolophidae  53 

Key  to  centrolophid  genera  54 

Genus  Hypcro>iIijphe  54 

Genus  Schedoplulus  58 

Genus  Centrolopfnis  62 

Genus  IcichtJiys  65 

Genus   Seriolella   69 

Genus  Psenopsis  72 

Family  Nomeidae  76 

Key  to  nomeid  genera 77 

Genus   Cttbiceps  78 


Genus  Nomeus 81 

Genus  Psenes 84 

Family  Ariommidae  88 

Genus   Ariomma  90 

Family  Tetragonuridae  94 

Genus  Tetragonurus  96 

Family   Stromateidae   98 

Key  to  stromateid  genera  99 

Genus  Stronuiteus  99 

Genus  Peprilus  103 

Genus   Pampus   108 

Evolutionary  trends  in  the  Stromateoidei  113 

Distribution  of  the  Stromateoidei  122 

Acknowledgments  127 

Summary   128 

Literature  cited  129 


HISTORICAL  INTRODUCTION 

The  Stromateoidei  are  a  small  suborder 
of  the  perciform  fishes,  characterized  pri- 
marily by  toothed  saccular  outgrowths  in 
the  gullet  immediately  behind  the  last  gill 
arch.  The  stromateoids  are  all  marine,  pe- 
lagic, and  widely  distributed  in  the  tem- 
perate and  tropical  oceans  of  the  world. 
Most  species  are  rare  and  infrequently  seen, 
but  a  few  form  the  basis  of  fisheries.  Adult 
stromateoids  range  from  less  than  a  foot  to 
over  four  feet  in  length. 


^  This  paper  is  based  on  a  thesis  presented  to 
Harxard  University  in  partial  fulfillment  of  the  re- 
quirements for  the  Ph.D.  in  Biology.  Contribution 
No.  1685  from  the  Woods  Hole  Oceanographic 
Institution. 

-  Woods  Hole  Oceanographic  Institution,  and 
Museum  of  Comparative  Zoology,  Harvard  Univer- 
sity. 


Certain  stromateoids  were  recognized  in 
classical  times.  Stromateiis  was  the  name 
applied  by  the  Greeks  of  Egypt  to  a  fish 
probably  from  the  Red  Sea.  The  name,  de- 
rived from  the  word  for  a  brightly  colored 
rug,  may  have  referred  to  the  fish's  shape 
and  coloration.  Later,  however,  Rondelet 
( 1554 )  used  the  name  for  a  similar  Medi- 
terranean fish  known  in  the  contemporary 
Roman  vernacular  as  fiatola.  Linnaeus 
(1758)  described  the  same  fish  as  Stro- 
mateus  fiatola. 

The  oceanic  fish  pompilus  was  sacred  to 
the  Greeks.  As  pompilm  accompanied  ships, 
it  brought  a  calm  sea  (Gesner,  1560).  Pom- 
pilus has  been  equated  with  Centrolophiis 
(Gunther,  1860),  and  Gesner's  figure 
(1560:113)  certainly  is  of  this  fish.  Thomp- 
son (1947),  however,  presents  evidence  that 
pompilus  is  the  pilot  fish  Naucrates.  Cuvier 

Bull.  Mus.  Comp.  Zool.,  135(2) :  31-139,  January,  1967        31 


32        Bulletin  Museum  of  Cotnpamtive  Zoology,  Vol.  135,  No.  2 


and  V^ilenciennes  (1833),  citing  Cetti's 
"Historia  Naturale  di  Sardagni"  ( 1777 ) , 
suggested  that  }wmpilus  might  be  a  tuna. 
And  while  the  subject  of  Ovid's  heroic  Hnes 

"Tuque  comes  ratiuin,   tractique  per  aequora 

sulci 

Qui  semper  spumas  sequeris,  pompile,  nitentes" 

could  well  ha\'e  been  the  centrolophid 
Schedophihis  ovalis,  it  seems  more  likely 
that  the  poet  is  referring  to  Coryphaena, 
the  dolphin.  The  classical  name  has  been 
used  in  Coryphaena  pompilus  Linnaeus, 
1758,  an  unrecognizable  fish;  in  Pompiltis 
Lowe,  1839,  a  synonym  of  Ceiitrolophus 
Lacepede,  1803;  and  in  Pompilus  Minding, 
1832,  a  synonym  of  Naucmtes  Rafinesque, 
1810. 

Identical  figures  of  stromateoids  were 
published  by  the  Renaissance  compilers 
Belon  (1553),  Rondelet  (1554),  Gesner 
(1560),  and  Aldrovandi  (1613).  Illustra- 
tions of  clearly  recognizable  species  appear 
in  these  works  on  the  facing  page  with  im- 
possible monsters.  Nonetheless,  the  infor- 
mation gathered  by  these  men  was  to  prove 
very  useful  to  later  authors,  and  was  con- 
sidered authoritative  by  many.  Some  of 
their  work,  for  example,  can  be  found  al- 
most word  for  word  in  Cuvier  and  Valen- 
ciennes ( 1833 ) . 

John  Ray's  studies  (Willughby,  1686) 
are  marked  by  their  care  and  attention  to 
detail.  His  anatomical  work  revealed  for 
the  first  time  the  structure  most  character- 
istic of  the  stromateoid  fishes,  the  peculiar 
pharyngeal  sacs.  Ray  mistakenly  believed 
that  the  sacs  constituted  a  second  stomach. 
Nonetheless,  care  that  was  not  to  reappear 
for  several  centuries  is  apparent  in  his  de- 
scription of  the  sacs  of  Stromateiis  (p.  156) : 

"In  palato  duo  oblonga  ossicula  aspera  .  .  . 
Nam  prime  duos  habet  ventriculos;  primum 
retro  cor,  prope  ipsiun  os  situm,  quem  echinum 
nun  ab  re  dixeris:  carnosus  enim  est,  &  apopli- 
ysibus  longis,  asperis,  crebris,  pellis  erinacei 
fere  in  modum  intus  consitur.  ( Appendices 
hae  in  sex  radios  divaricantur  cylindriae  supra 
centrum  stellae  erectae.    D.  Will.)" 


During  the  first  half  of  the  nineteenth 
century  a  majority  of  the  stromateoids  were 
described  and  characterized  reasonably 
well.  The  "Regne  Animal"  (Cuvier,  1817) 
and  "Histoire  Naturelle  des  Poissons"  (Cu- 
vier and  Valenciennes,  1833)  were  especially 
valuable.  Cuvier  (1817)  provided  the  basic 
arrangement  which  was  expanded  upon  in 
the  later  "Histoire  Naturelle." 

Cuvier  and  Valenciennes  (1833:381) 
added  to  Ray's  description  of  the  pharyn- 
geal sacs  of  Stromateus  fiatola: 

"A  I'exterieur,  cette  partie  presente  la  forme 
d'une  bourse;  les  epines  dont  elle  est  armee 
sont  de  different  grosseur;  les  plus  grandes 
sont  un  pen  en  forme  fuseau;  les  petites  garnis- 
sent  les  intervalles  des  grandes.  Chacune  de 
ces  epines  s'attache  a  la  veloutee  par  sept  ou 
huit  ravines  ou  fibres  disposees  en  etoile." 

They  continue  later  with  their  own  obser- 
vations on  the  sacs  of  Stromateus  condidu.s 
{— Pampus  argenteu.s)  (p.  392): 

"Immediatement  apres  les  os  pharyngiens 
vient  un  oesophage  en  forme  de  sac  ranfle  et 
chamu,  rond,  un  peu  bilobe  .  .  .  garnies  .  .  . 
de  grosses  epines  osseuses  .  .  ." 

and  of  Rhombus  xanthurus  ( =  Peprilus 
pom)  (p.  406): 

".  .  .  un  oesophage  charnu,  arme  interieure- 
ment  de  dents  osseuses  coniques,  les  unes  plus 
grandes,  les  autres  plus  petites  .  .  ." 

Cuvier  and  Valenciennes  not  only  recog- 
nized this  unique  structure  in  "les  Stro- 
matees"  but  also  described  a  similar  struc- 
ture in  "les  Centrolophes."  Concerning 
Ccntrolophus  pom})ilus  (=C.  niger)  they 
wrote  ( p.  339 ) : 

"Le  pharynx  du  centrolophe  presente  une 
particularite  remarquable,  qui  donne  au  com- 
mencement de  leur  oesophage  un  armure 
puissante.  Entre  les  os  pharyngiens  .  .  .  I'os 
superieur  du  quatrieme  arceau  porte  plusiers 
appendices  alonges  et  garnis  de  dents  semb- 
lables  ...  la  partie  laterale  du  pharynx  a  de 
profondes  cannelures  osseuses  et  dentees  .  .  ." 

They  continue,  observing  that  this  seems  to 
be: 

".   .   .  quelque  analogic  avec  les  epines  dont 
le  meme  cavite  est  armee  dans  les  stromatees." 


Stromateoid  Fishes  •  Haedrich        33 


Here,  for  the  first  time,  a  relationship  be- 
tween the  centrolophids  and  the  stromateids 
was    indicated. 

Gi.inther  recognized  a  unifying  character 
here  and,  in  his  Catalogue  (1(S60:355), 
noted  that  in  the  Stromateina  "tooth-hke 
processes  extend  into  the  oesophagus."  The 
"Catalogue"  provided  keys  to  the  scombrid 
group  Stromateina  and  to  the  two  genera 
Giinther  included  in  it,  Stiomatcus  ("ven- 
trals  none  in  an  adult  state")  and  Ccntio- 
Jophiis  ("ventrals  well  developed").  The 
other  groups  in  his  family  Scombridae  were 
the  Scombrina,  Cyttina,  Coryphaenina,  and 
Nomeina,  the  last  composed  largely  of 
stromateoids.  The  diagnostic  pharyngeal 
sacs  of  the  Nomeina  remained  to  be  dis- 
covered, for  they  \\ere  not  mentioned  in 
Giinther's  account. 

Discussing  the  limits  and  arrangement  of 
the  scombroids.  Gill  ( 1S62 )  corrected  some 
of  Giinther's  omissions.  In  doing  so,  how- 
ever, he  l)roke  up  the  convenient  group 
Nomeina,  and  added  little  to  the  classifica- 
tion. The  Stromateina,  though  mentioned, 
were  not  defined. 

"An  Introduction  to  the  Study  of  Fishes" 
(Giinther,  1880)  was  essentially  the  same 
as  the  earlier  "Catalogue"  in  its  treatment 
of  the  stromateoids.  Here,  however,  each 
group  was  given  full  family  status  as  Stro- 
mateidae  and  Nomeidae.  The  close  relation- 
ship of  the  two  was  still  not  indicated,  and 
Giinther  continued  to  allocate  a  number  of 
stromateoid  genera  to  other  families. 

Although  Giinther's  (1880)  publication 
added  almost  nothing  to  the  classification, 
it  had  an  important  incidental  effect. 
Whether  the  cause  was  Giinther's  failure 
to  have  noticed  Gill's  earlier  ( 1862 )  paper 
or  whether  it  was  his  casual  accounting 
cannot  be  said.  At  any  rate,  Giinther's  treat- 
ment of  the  stromateoids  soon  occasioned  a 
vitriolic  blast  from  Gill.  In  his  "Notes  on 
the  Stromateidae,"  Gill  (1884)  united  the 
fonns  scattered  by  Giinther  under  the  sin- 
gle family  Stromateidae,  still  omitting 
Nomeiis  but  including,  albeit  reservedly, 
Psenes  and  Cuhiceps.  The  definition  of  the 


family  noted  "the  gill-rakers  of  the  upper 
segment  of  the  last  branchial  arch  enlarged 
and  dentigerous  or  sacciform,  and  project- 
ing back\\'ards  into  the  oesophagus"  (p. 
665).  Gill  furthermore  recognized  a  basic 
dichotomy  in  the  group  by  dividing  the 
family  into  two  subfamilies,  the  Stro- 
mateinae  and  the  Centrolophinae: 

"These  are  distinguished  by  differences  in 
the  development  of  the  vertebrae,  the  former 
[Stromateinae]  having  14-15  abdominal  and 
17-21  caudal  vertebrae,  and  the  latter  [Centro- 
lophinae] 11  alxloniinal  and  14  caudal  verte- 
lirae;  these  differences  are  supplemented  by 
variations  in  the  degree  of  complexity  of  the 
peculiar  appendages  representing  and  homol- 
ogous with  the  gill-rakers  of  ordinary  fishes, 
developed  from  the  last  branchial  arch,  and 
extending  into  the  oesophagus   (p.  654)." 

He  also  observed  that  the  Centrolophinae 
have  normally  persistent  pelvic  fins,  while 
those  of  the  Stromateinae  are  lost  with 
growth.  Gill  considered  the  Centrolophinae 
to  be  the  most  generalized  type;  the  Stro- 
mateinae he  thought  more  specialized. 

"Spolia  Atlantica"  of  Liitken  (1880)  con- 
tained accounts  of  the  genera  Psenes.  Cuhi- 
ceps, Stwmoteus,  and  Schedophihis.  The 
discussion  of  relationships  was  carefully 
done,  and  the  listing  of  included  species 
was  especially  good.  Unfortunately,  the 
work  was  in  Danish,  and  has  apparently 
been  little  used  by  subsequent  investiga- 
tors. 

Fordice  ( 1884 )  reviewed  the  American 
species  of  the  Stromateidae.  No  mention 
was  made  of  the  pharyngeal  sacs.  Only  two 
genera,  Strotiiateus  and  Leirus  ( =  Sclwdo- 
pJiilus)  were  mentioned,  and,  again,  the 
division  was  based  on  the  presence  or  ab- 
sence of  pelvic  fins.  Fordice  provided  keys 
and  neat  synonymies  of  most  American 
stromateids.  His  jiaper  was  essentially  an 
extension  of  the  foundation  laid  do\\'n  by 
Jordan  and  Gilbert's  ( 1882 )  "Synopsis  of 
the  Fishes  of  North  America,"  a  work  which 
erroneously  reported  (p.  448)  for  the 
Nomeidae,  "No  tooth-like  processes  in  the 
oesophagus." 

Relying  heavily  on  the  work  of  Gill,  Jor- 


34        Bulletin  Museum  of  Comparative  Zoologij,  Vol.  135,  No.  2 


clan  and  (Tilbcrt,  and  Fordice,  Jordan  and 
EveiTnann's  ( 1896 )  "Fishes  of  North  and 
Middle  America"  provided  a  syndiesis  of 
current  thoughts  on  stromateoid  classifica- 
tion. The  Centrolophidae  were  considered 
a  family  apart  from  the  Stromateidae,  "dif- 
fering in  appearance  and  in  the  smaller 
number  of  \'erte]:)rae,  although  agreeing  in 
the  possession  of  teeth  in  the  oesophagus" 
(p.  964).  Nonwu.s  and  Pscnc.s,  in  the  family 
Nomeidae,  remained  distinct,  and  no  men- 
tion of  a  relationship  \\  ith  the  stromateids 
was  made. 

"Oceanic  Ichthyology"  ( Goode  and  Bean, 
1896)  drew  on  Jordan  and  Gilbert,  Gill, 
and  Giinther.  for  much  of  its  information. 
An  unexplained  but  correct  innovation  was 
the  inclusion  of  Icichthijs  among  the  stro- 
mateoids.  The  treatment  of  the  group  was 
extremely  casual;  genera  were  shuffled  into 
families  more  or  less  randomly  without 
checking  familial  characters.  The  Nome- 
idae constituted  almost  the  same  unnatural 
group  as  set  up  by  Giinther  (1860),  with 
still  no  realization  of  its  relationships. 
Goode  and  Bean's  account  confused,  rather 
than  improved,  the  stromateoid  classifica- 
tion. Fortunately,  it  has  been  disregarded 
by  most  subsequent  workers. 

The  first,  and  the  only,  world-wide  re- 
vision of  the  stromateoids  was  that  of 
Regan  (1902).  Regan  gave  the  group  its 
modern  dimensions  by  adding  the  genera 
^'Nomcus,  Cuhiceps,  Pscncs,  Bathy.scriola, 
and  SeriolcUo,  all  of  which  have  a  toothed 
oesophagus  exactly  similar  to  that  of  a 
Centrolophus"  (p.  117).  His  definition  of 
the  family  was  based  largely  on  osteology, 
and  made  important  contributions.  His 
warnings  of  the  pitfalls  of  allometry  and 
of  the  unusual  ubiquity  of  certain  char- 
acters recognized  a  recurrent  problem.  Re- 
gan treated  the  group  as  one  family,  the 
Stromateidae,  but  disregarded  the  conve- 
nient subfamilial  distinction  made  earlier  by 
(;ill  (1884).  Norman's  much  later  "Draft 
Synopsis"  ( 1957 )  differs  from  Regan  only 
in  this  one  respect,  for  Norman  recognized 
two  families  based  on  the  first  couplet  of 


Regan's  key  to  genera,  "ventral  fins  present" 
[Centrolophidae],  or  "ventral  fins  absent" 
[Stromateidae].  Citing  correspondence  with 
Boulenger,  Regan  suggested,  for  the  first 
time,  the  affiliation  of  Tetrcifionunis  to  the 
stromateoids. 

Boulenger  was  much  impressed  by  one 
of  Regan's  diagnostic  characters,  the  loose 
attachment  of  the  pelvic  bones  to  the  pec- 
toral arch  in  all  stromateids.  In  his  system- 
atic account  of  the  fishes  for  the  "Cam- 
bridge Natural  History"  (1904),  he  re- 
moved the  stromateids  from  the  scom- 
broids, where  all  previous  workers  had 
placed  them,  and  ranked  them  among  the 
Percesoces.  Holt  and  Byrne  (1903),  using 
the  same  argument,  also  considered  the 
stromateids  to  be  allied  with  the  Percesoces. 
Although  in  error  with  respect  to  the  rela- 
tionships of  the  group,  their  account  of  local 
British  and  Irish  species  was  otherwise 
carefully  done,  and  showed  broad  under- 
standing. 

Although  the  presence  of  teeth  in  sac- 
cular outgrowths  in  the  gullet  had  long 
been  used  as  a  diagnostic  character  for  the 
stromateoids,  no  one  since  John  Ray  had 
investigated  the  structure  of  this  peculiar 
feature.  Gilchrist  (1922)  examined  teeth 
from  the  sacs  of  several  South  African 
stromateoids.  He  noted  differences  be- 
tween species,  but,  although  he  recognized 
the  value  of  the  teeth  in  taxonomy,  he  did 
not  indulge  in  systematic  speculations.  The 
sacs  had  previously  been  referred  to  as 
"oesophageal";  Gilchrist  pointed  out  that 
they  were  "not  strictly  oesophageal,  but 
.  .  .  derived  from  .  .  .  pharyngeal  epithe- 
lium .  .  ."  (p.  254).  Later,  in  an  incisive 
review,  Barnard  (1948)  corrected  some  of 
Gilchrist's  errors,  and  extended  his  work 
by  examining  more  species  and  publishing 
more  illustrations. 

Biihler's  (1930)  monograph  on  the  diges- 
tive system  of  the  stromateoids  pointed  out, 
independently  from  Ciilchrist,  the  pharyn- 
geal origin  of  the  toothed  sacs.  Biihler 
proposed  the  terai  "Rachensiiche"  [=  pha- 
ryngeal   sacs]    to    replace    the    misleading 


Stromateoid  Fishes  •  Haedrich        35 


on   any 

ryngeal 

detailed 

(1965), 

of    the 

change 

changes 

Regan's 


"oesophageal  sacs"  commonly  in  use.  His 
work  was  done  primarily  with  serial  micro- 
scopic sections,  allowing  examination  and 
description  of  great  detail.  It  was  a  sub- 
stantial contribution  to  understanding  the 
origin,  nature,  and  probable  function  of  the 
teeth  in  the  pharyngeal  sacs.  For  details 
morjohological  aspect  of  the  pha- 
sacs,  Biihler's  work,  or  the  recent 
extension  of  this  by  Isokawa  et  al. 
should  be  consulted.  Other  parts 
digestive  system  were  noted  to 
in  rough  correspondence  with 
in  the  Rachensiiche,  and  within 
( 1902 )  framework  Biihler  pro- 
posed two  subfamilies,  the  Lirinae,  corre- 
sponding to  Norman's  ( vide  supra )  Cen- 
trolophidae,  and  the  Stromateinae,  corre- 
sponding to  NoiTnan's   Stromateidae. 

The  work  of  Gilchrist,  Barnard,  and 
Biihler  offered  sound  characters  for  the 
stromateoid  classification.  But,  because 
each  study  dealt  with  only  a  limited  array 
of  characters,  the  observations  could  not 
be  properly  or  safely  interpreted. 

My  work  has  dealt  primarily  with  skeletal 
characters.  I  have  looked  at  the  soft  anat- 
omy only  cursorily,  and  have  found  little 
of  use  except  in  a  most  general  way.  The 
study  has  involved  only  Recent  fishes. 

My  conclusions  are  largely  based  on  the 
presence  or  absence  of  pelvic  fins,  whether 
the  dorsal  fin  is  separated  or  continuous, 
the  presence  or  absence  of  certain  teeth, 
the  number  of  vertebrae,  the  number  of 
branchiostegal  rays,  and,  in  particular,  the 
structure  of  the  caudal  region  and  the  de- 
velopment of  the  papillae^  in  the  pha- 
ryngeal sacs.  The  comparative  morphology 
of  these  characters  not  only  provides  a 
reasonable  separation  of  the  suborder  into 
five  families,  but  also,  because  the  char- 
acters change  in  a  correlated  fashion,  it 
suggests  the  course  of  evolution  in  the 
stromateoids.     In    the    trunk    and    caudal 


^  This  temi  implies  the  unit  composed  of  a  bony 
base  with  teeth  seated  upon  it.  It  is  adopted  here 
in  conformance  with  past  usage  (Biihler,  1930; 
Barnard,  1948). 


region  the  number  of  vertebrae  increases, 
while  the  elements  in  the  tail  become  fused 
and  reduced,  and  the  pelvic  fins  are  lost. 
In  the  branchial  region,  the  number  of 
branchiostegals  decreases,  while  the  papil- 
lae of  the  pharyngeal  sacs  become  in- 
creasingly more  complex.  The  present 
geographical  distributions  of  the  different 
taxa  support  the  conclusions  based  on 
anatomical  evidence. 

I  propose  for  the  stromateoids  a  hierarchy 
of  five  families  and  fourteen  genera,  as 
follows : 

Order  Percifomies 
Suborder  Stromateoidei 
Family  Centrolophidae 

Hijperoglijphe 

Schcdophihts 

Cenirolophus 

Icichthtjs 

Seriolella 

Psenopsis 
Family  Nomeidae 

Cubiccps 

Nomeiis 

Psenes 
Family  Ariommidae 

Ariomma 
Family  Tetragonuridae 

Tetmgonurus 
Family  Stromateidae 

Strotnateus 

Peprihis 

Pampus 

METHODS 

Measurements  were  made  point-to-point 
with  a  pair  of  fine-point  dial  calipers.  A 
dissecting  microscope  with  an  eyepiece 
dial  micrometer  was  used  for  a  few  very 
small  specimens.  Measurements  routinely 
made  were: 

Total  length  (TL),  from  the  tip  of  the 
snout  to  the  farthest  tip  of  the  caudal  fin. 

Standard  length  (SL),  from  the  tip  of 
the  snout  to  the  caudal  fin  base. 

Length  of  head,  from  the  tip  of  the 
snout  to  the  hindmost  point  on  the  oper- 
cular membrane,  usually  immediately  above 
the  pectoral  fin. 

Length  of  pectoral  fin,  from  the  base 
of  the  uppermost  ray  to  the  farthest  tip 
of  the  fin. 


36 


Bulletin  Museum  of  Comparative  Zoologij.  Vol.  135,  No.  2 


Length  of  pel\'ic  fin,  from  the  base  of 
the  most  anterior  ray  to  the  farthest  tip 
of  the  fin. 

Longest  Di  spine,  from  the  base  of  the 
spine  to  its  extremity. 

Predorsal  distance,  from  the  tip  of  the 
snout  to  the  base  of  the  first  element  of 
the  dorsal  fin. 

Preanal  distance,  from  the  tip  of  the 
snout  to  the  base  of  the  first  element  of 
the  anal  fin. 

Maximum  depth,  the  greatest  depth  of 
the  body,  exclusive  of  fleshy  or  scaly  fin 
bases. 

Depth  of  peduncle,  the  least  depth  of 
the  caudal  peduncle. 

Snout,  from  the  tip  of  the  snout  to  the 
anterior  margin  of  the  orbit. 

Eye  diameter,  the  greatest  distance  be- 
tween the  fleshy  margins  of  the  eye. 

Length  upper  jaw,  from  the  symphysis 
of  the  premaxillaries  to  the  hindmost  point, 
often  covered  by  the  lacrimal  bone,  on  the 
maxillary. 

Interorbital  width,  the  least  distance 
between  the  bony  rims  over  the  eyes. 

The  measurements  used  for  showing  allo- 
metry  were  length  of  head,  length  of  pec- 
toral fin,  length  of  pelvic  fin,  predorsal 
distance,  preanal  distance,  and  maximum 
depth.  These  were  expressed  as  a  percent- 
age of  standard  length.  This  percentage 
was  plotted  against  standard  length  follow- 
ing the  method  advocated  by  Parr  ( 1956 ) . 
The  measurements  of  snout,  eye  diameter, 
length  of  upper  jaw,  and  interorbital  width 
were  expressed  as  a  percentage  of  length 
of  head. 

Counts  were  made  \\'ith  a  fine  needle, 
usually  under  low  magnification  on  a  dis- 
secting microscope.  For  extremely  small 
specimens,  median  finray  counts  were 
made  more  easily  using  transmitted,  polar- 
ized light.    Counts  routinely  made  were: 

D,  total  dorsal  fin  elements,  spines  indi- 
cated by  Roman  numerals,  rays  by  Arabic. 
In  some  cases  it  was  impossible  to  distin- 
guish between  spines  and  rays;  these  counts 


are  followed  by  the  expression  "total  ele- 
ments."  The  last,  double  ray  of  both  dorsal 
and  anal  fins  was  counted  as  one  element. 

A,  total  anal  fin  elements. 

P,  total  pectoral  fin  elements,  one  side, 
spine  not  distinguished  from  rays. 

Gill  rakers,  the  total  number  of  rakers 
on  the  first  arch,  one  side.  Expressed  as 
number  on  upper  limb  plus  one,  if  at  angle, 
plus  number  on  lower  limb  (e.g.  8  +  1-1-  17). 

Lateral  line  scales,  the  number  of 
scales  along  the  position  normally  occupied 
by  the  lateral  line,  one  side,  terminating  at 
the  caudal  fin  base.  In  many  stromateoids 
the  tubed  scales  end  on  the  peduncle,  but 
the  count  was  nonetheless  continued  to  the 
caudal  base.  Often  the  deciduous  scales  are 
lost  and  scale  pockets  must  be  counted. 
Lateral  line  scales  is  a  difficult  count  and 
cannot  be  made  on  most  specimens. 

Counts  less  routinely  made  were: 

Branchiostegal  rays  (BR),  left  side. 

Vertebrae,  number  of  precaudal  verte- 
brae plus  the  number  of  caudal  vertebrae, 
including  the  hypural  plate  (  =  1 ) .  Almost 
all  vertebral  counts  were  made  from  radio- 
graphs. Determination  of  the  first  caudal 
vertebra  is  hence  somewhat  subjective. 
Where  skeletal  material  has  been  used,  the 
count  is  followed  by  the  expression  "skel." 

All  measurements  and  counts  confomi 
with  the  standards  of  Hubbs  and  Lagler 
(1958). 

The  osteology  was  studied  primarily  with 
radiographs  and  cleared-and-stained  prep- 
arations. With  the  use  of  soft  X-rays,  speci- 
mens as  small  as  20  mm  SL  could  be  suc- 
cessfully radiographed. 

Small  fishes,  usually  no  longer  than  65 
mm  TL,  were  cleared  and  stained.  Gill 
arches  and  the  associated  pharyngeal  sacs 
were  dissected  from  larger  fish,  usually 
around  190  mm  TL,  and  were  also  cleared 
and  stained.  To  remove  the  arches  and 
sacs,  cuts  were  made  between  the  hyal  and 
opercular  series,  the  tongue  and  dentaries, 
and  the  last  gill  arch  and  pectoral  girdle. 
The  unit  so  freed  was  carefully  discon- 
nected from  the  base  of  the  neurocranium. 


Stromateoid  Fishes  •  llacdrich        37 


URONEURAL 
URONEURAL  2 

HYPURAL   6 
HYPURAL    5 

HYPURAL  4 


EPURAL  3 
EPURAL  2 
EPURAL 


UROSTYLAR  VERTEBRA 


PRETERMINAL  VERTEBRA 


PRETERMINAL  VERTEBRA  2 


PRETERMINAL  VERTEBRA  3 


HYPURAL  3 
HYPURAL  2 

HYPURAL 


AUTOGENOUS 
HAEMAL  SPINES 

Figure   1.     The    bones   of   the   caudal    skeleton,    schematic   drawing  of  the  basic  centrolophid  type. 


and  the  pharyngeal  sacs  pulled  forward 
from  between  the  cleithra.  The  esophagus 
was  cut  behind  the  sacs,  and  the  complete 
branchial  apparatus  removed.  This  was 
divided  sagitally,  and  one  half,  usually  the 
right,  was  cleared  and  stained.  A  more 
legible  preparation  \\  as  obtained  if  the  gill 
filaments  were  stripped  off  prior  to  treat- 
ment. Teeth  from  the  pharyngeal  sacs  were 
examined  by  macerating  the  sac  in  potas- 
sium hydroxide  until  the  muscle  was  trans- 
lucent, staining,  and  allowing  the  sac  to 
disintegrate.  Individual  teeth  could  then 
be  picked  out  and  examined. 

Excellent  results  were  obtained  follow- 
ing the  clearing-and-staining  method  of 
Clothier  (1950)  modified  from  Hollister 
( 1934 ) .  The  entire  clearing-and-staining 
procedure  takes  about  two  months. 

A  few  complete  skeletons  were  prepared, 
mostly  from  fresh  specimens,  by  picking  the 
meat  carefully  from  the  bones.   Skulls  were 


prepared  by  boiling  the  specimen  until  it 
fell  apart.  Only  partial  dissections  were 
made  in  some  cases.  Whenever  possible, 
skeletons  in  museum  collections  were  ex- 
amined. Scales  were  stained  in  alizarin, 
blotted  dry,  and  mounted  in  balsam  on 
slides. 

Drawings  of  cleared-and-stained  prepara- 
tions were  made  through  a  Wild  Dissecting 
Microscope  with  a  camera  lucida  attach- 
ment. All  caudal  skeletons  were  drawn  at 
25  power,  using  transmitted  light.  Draw- 
ings of  the  branchial  arches  were  made  at 
6  power,  using  transmitted  light  to  draw 
the  orientation  of  the  bones,  and  reflected 
light  to  draw  the  arrangement  and  structure 
of  the  bases  of  the  papillae.  These  draw- 
ings were  redrafted  on  tracing  paper  and 
the  final  drawing  was  made  with  reference 
to  the  specimen  through  the  scope  without 
camera    lucida.    All   anatomical   drawings. 


38        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


EPIBRANCHiALS 


CERATOBRANCHIALS 


HYPOBRANCHIAL  ,.<^S 


INFERIOR 
HYPOHYAL 


PHARYNGEAL  SAC 


ESOPHAGUS 


CERATOHYAL 


BRANCHIOSTEGAL  RAYS 


EPIHYAL 


Figure    2.      Elements    of    the    branchial    region,    schematic    drawing    of    the    basic    stromateoid    type. 


though    made    from    specific   preparations, 
are  semi-diagrammatic. 

The  various  elements  in  the  drawings  of 
the  caudal  skeleton  are  identified  in  Figure 
1.  Those  of  the  branchial  region  are  shown 
in  Figure  2.  Nomenclature  of  the  caudal 
skeleton  follows  Gosline  (1960,  1961);  that 
of  the  head  and  branchial  region  follows 
Mead  and  Bradbury  (1963). 

MATERIAL 

The  specimens  examined  are  arranged 
below  under  the  classificatory  scheme  here- 
in proposed.  Museum  and  institutional 
names,  in  alphabetical  order  under  each 
species,  are  abbreviated  as  follows: 

ABE      —Collection  of  Dr.  T.  Abe,  I'okyo 
AMS      — Australian  Museum,  Sydney 


ANSP  — Academy  of  Natural  Sciences  of  Phila- 
delphia 

BC  — University  of  British  Columbia,  Van- 
couver 

BCF  — Bureau  of  Commercial  Fisheries,  Bio- 
logical Laboratory,  Washington,  D.  C. 

BMNH — British  Museum  (Natural  History), 
London 

BOC  — Bingham  Oceanographic  Collection, 
Yale   University 

CF  — Danish  Carlsberg  Foundation,  Char- 
lottenlund  Slot 

CNHM— Chicago  Natural  History  Museum 

CTS       — Guinean  Trawling  Survey 

nOE'    — International  Indian  Ocean  Expedition 

MCZ  — xMuseum  of  Comparative  Zoology,  Har- 
vard University 

MNHN — Museum  National  d'Histoire  Naturclle, 
Paris 

^  These  specimens  will  ultimately  be  catalogued 
in  the  MCZ. 


Stromateoid  Fishes  •   Uacdrich 


39 


NRF      — Nankai    Regional    Fisheries    Research 

Laboratory,  Kochi 
NTU      — Department    of    Zoology,    Faculty    of 

Science,  Tokyo  University 
SAM      — South  African  Museum 
SIO        — Scripps   Institution  of  Oceanography 
SU         — Natural     History     Museum,     Stanford 

Uni\ersity 
USNM  — United      States      National      Museum, 

Washington,  D.  C. 
WHOF — Woods  Hole  Oceanographic  Institution 
ZMC      — Zoological  Museum,   Copenhagen 

The  number  of  specimens,  the  range  of 
standard  lengths,  and  the  locaHty  appear 
in  that  order  within  the  parentheses.  Types 
are  marked  with  an  asterisk  (*),  the  spe- 
cies name  follo\\'ing  within  the  parentheses. 
Specimens  radiographed  are  marked  with 
a  dagger  ( t ) ;  specimens  that  have  been 
cleared  and  stained  are  marked  CS. 

Hyperoghjphe 

H.  antarctica  ( Carmichael ) :  tAMS  IB.  3825 
(I,  170  mm,  near  Sydney,  N.S.W.).  *t  BMNH 
1855.  9.  19.  2  ( 1,  108  mm,  coasts  of  Australia, 
Diagravima  porosa  Richardson).  tSAM  23592 
(1,  105  mm.  Cape  Point,  South  Africa). 

H.  hythitcs  (Ginsburg):  CNHM  46408  (1,  12 
mm.  Ocean  Spring,  Miss.).  *tUSNM  157776 
(  1,  197  mm,  off  Pensacola,  Fla.,  Palimiricli- 
thijs  hijthitcs  Gimhing) ;  *tUSNM  157778  ( 1, 
187  mm,  Mississippi  Delta,  paratype  Pali- 
nurichthiis  hifthites  Ginsburg).  WHOI  (1,  204 
umi,  OREGON  sta.  3762);  WHOI  (1,  203 
mm,  OREGON  sta.  4011);  WHOI  (7,  188- 
213  mm,  OREGON  sta.  4030). 

H.  iaponica  (Doderlein):  ABE  58-258,  60-107 
to  111,  60-113  to  116.  60-139,  60-225,  60-370, 
60-709,  60-744,  60-775,  61-452,  61-458,  61- 
461  to  464,  61-500  to  511,  62-302,  62-303, 
62-336  to  .368,  63-435,  63-476,  63-447,  63- 
480,  63-541  to  544,  63-547,  63-668,  63-708, 
6.3-728,  63-872  (35,  35-233  mm,  Manazuru, 
Japan);  ABE  64-2201  and  2202  (2,  400,  415 
mm,  Tokyo  market);  ABE  2236  (1,  695  mm, 
Tokyo  market);  ABE  plankton  collection  (1,  20 
mm,  East  China  Sea).  CNHM  59428  (1,  420 
mm,  Tokyo). 

H.  percifonna  (Mitchill):  tMCZ  36624  (2,  159. 
209  mm.  Woods  Hole,  Mass.).  WHOI  (47, 
101-198  mm,  40°10'N  69°30'W,  gill  arch  CS); 
WHOI  (1,  50  mm,  38°37'N  69°24'W,  CS). 

Schedophilus 

S.  griscolineaius  (Norman):  tBMNH  1936.8. 
26.  1068-9  (2,  258,  258  mm,  South  Atlantic); 


*tBMNH  1936.8.26.1070-1  (2,  193,  196 
mm,  South  Atlantic,  Palinurichthys  grheo- 
Uneatus  Norman ) . 

S.  macukitus  Giinther:  *tBMNH  48.  3.  16.  150 
(1,  37  mm,  China  Seas,  S.  Duiculatus  Gimther). 

S.  huttoni  (Waite):  tZMC  (1,  ca.  40  mm, 
34°24'S  94M5'W). 

S.  maniioratus  Kner  and  Steindachner:  *tHam- 
burg  Museum  H464  (1,  37  mm,  Siidsee,  S. 
luaiDiorafus  Kner  and  Steindachner).  tZMC 
(2,  28  and   46   mm,   GALATHEA  sta.   176). 

S.  medusophagus  Cocco:  *BMNH  60.3.18.3 
(I,  ca.  405  mm,  stuffed,  Cornwall,  Centro- 
lopJius  britannicits  Giinther).  tCF  (7,  59- 
188  mm,  North  Atlantic).  tUSNM  163880 
(1,  333  mm,  Kitty  Hawk).  WHOI  (19,  5.5- 
435  mm.  North  Atlantic,  one  CS). 

S.  ovalis  (Cuvier  and  Valenciennes):  *tBMNH 
1860.7.17.2-3  (2,  171,  186  mm,  Madeira, 
Leiws  bennettii  Lowe).  *MNHN  264.4.  1.2 
( 1,  299  mm,  Nice,  Centrolophus  avails  Cuvier 
and  Valenciennes);  *MNHN  264.5.1.2  (1, 
144  mm,  Canary  Islands,  Crius  hcrthclotti 
Valenciennes ) . 

S.  pcmarco  (Poll):  BCF  928  (1,  95  mm,  4°31'S 
10°53'E).  tGTS  (11,  103-217  mm,  Gulf  of 
Guinea,  gill  arch  CS ) . 

Centrolophus 

C.  niger  Lacepede:  *tBMNH  1862.6.14.16 
( 1,  290  mm,  Madeira,  Schedophilus  elongatus 
Johnson);  BMNH  50.9.7.3  (1,  .385  mm,  no 
locality);  BMNH  55.9.  19.  1461  (1,  124  mm, 
Hasler  Coll.);  BMNH  80.12.11.1  (1,  183 
mm,  Falmouth);  BMNH  1934.8.8.67  (1, 
362  mm,  SW  Ireland);  BMNH  1934.8.8. 
68-9  (2,  498,  560  mm,  SW  Ireland).  tCF 
(2,  147,  204  mm,  DANA  sta.  4205,  gill  arch 
CS).  MCZ  .34246  (1,  443  mm,  42°10'N 
66°45'W);  MCZ  37983  (1,  284  mm,  Province- 
town);  MCZ  37984  (1,  490  mm.  Sable  Island 
Bank).  *tMNHN  264.2.2.1  (1,  287  mm, 
Fecamp,  Centrolophus  niger  Lacepede); 
*MNHN  264.2.2.2,  264.2.2.3  (3,  169-207 
mm,  Nice,  Centrolophus  morio  Cu\'ier  and 
Valenciennes);  MNHN  264.2.2.9  (1,  180 
mm,  Naples);  *MNHN  264.  4.  2.  1  (1,  121  mm, 
Marseille,  Centrolophus  valenciennesi  Moreau). 
tUSNM  44440  (1,  189  mm,  Dennis,  Mass.); 
USNM  48367  (1,  39  mm,  Naples);  tUSNM 
48906  (1,  265  mm.  North  Truro,  Mass.); 
USNM  49335  (1,  220  mm,  Genoa).  WHOI 
(1,  1065  mm,  40°13'N  65°45'W). 

IciclitJujs 

I.  lockingtoni  lordan  and  Gilliert:  ABE  63-526, 
63-527,  63-529,  63-530,  63-548,  63-549,  63- 
555  ( 12,  72-173  mm,  Manazuru,  gill  arch 
CS).     BC    53-99A    (1,    164   mm,    Vancouver 


40        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135.  No.  2 


Island);  BC  59-652  (1,  132  mm,  Vancouver 
Island).  *SU  7442  (1,  65  mm,  Monterey  Bav, 
Schedophilu.s  hcathi  Gilbert);  SU  17346  (1, 
42  mm,  Seal  Beach);  SU  22955  (1,  41  mm, 
Catalina  Island);  SU  22971  (1,  26  mm, 
Monterey  Bay);  SU  41028  (8,  15-78  mm, 
Monterey  Bay,  CS).  *tUSNM  27397  (1,  159 
mm,  Point  Reyes,  Icichthij.s  lockingtoni); 
*tUSNM  89398  (3.  28-65  mm,  Monterey 
Bay,  Centrolophu.s  califomicus  Hobbs).  WHOI 
(2,  17,  83  mm,  Monterey  Bay,  received 
through  Giles  Mead). 

SerioIc'Ua 

S.  }>r(nii(i  Giinther:  tAMS  lA.  10170  (1,  233 
mm,  Sydney  Harbour,  New  South  Wales ) ; 
tAMS  1.10333  (1,  152  mm,  40  miles  W 
Kingston,  South  Australia).  *BMNH  (1,  298 
mm.  New  Zealand,  stuffed,  Neptomenu.s 
hrama  Giinther). 

S.  punctata  (Bloch  and  Schneider):  tAMS  I. 
10840  (1,  192  mm.  Oyster  Bay,  Tasmania); 
tAMS  1.14747  (1.  220  mm,  Portobello). 
*tBMNH  1869.  2.  24.  42^44  (3,  234-243  mm, 
Tasmania,  Neptoiuenus  doJntIa  Giinther). 
USNM  176915  (2,  190,  195  mm,  Queensland, 
gill  arch  CS);  tUSNM  176968  (1,  149  mm. 
New  South  Wales);  tUSNM  177109  (2,  197, 
203  mm.  New  South  Wales). 

S.  pnrosa  Guichenot:  tUSNM  176478  (1,  218 
mm,  Tictoc  Bay,  Chile);  tUSNM  176535  (1, 
198  mm,  Puerto  Auchemo,  Chile);  tUSNM 
176593   (3,   197-203  mm,  Auellon,  Chile). 

S.  viohcea  Guichenot:  tMCZ  17239  (2,  430, 
445  mm,  Callao,  Peru).  *SU  9590  (1,  262 
mm,  Callao,  Peru,  paratype  Ncptomenus  cras- 


sti-s   Starks ' 


rUSNM    53465    (1,   265   mm. 


Callao,  Peru,  Neptoiuenus  cra.s-s-us  Starks); 
tUSNM  77513  (1,  130  mm,  Mellendo,  Peru); 
tUSNM  77593  (1,  150  mm,  Mellendo,  Peru, 
gill  arch  CS);  tUSNM  77611  (1,  173  mm, 
Callao,  Peru);  tUSNM  77625  (1,  213  mm, 
Callao,  Peru).   tZMC  (1,  87  mm,  14°S  77°W). 

Pse7io])sis 

P.  anoniala  ( Temminck  and  Schlegel);  ABE  60- 
1232,  63-752,  63-1141  (3,  93-119  mm,  Mana- 
zuru  market);  ABE  62-656  (1,  40  mm,  Japan, 
CS);  ABE  61-590  (1,  160  mm,  Tsubaki); 
ABE  64-1223  to  1225  (25,  43-84  mm,  32^09' 
N  123°15'E);  ABE  64-1972  to  1959  (8,  17- 
52  mm,  Kozu);  ABE  64-2014  to  2017,  64- 
2142  to  2144,  64-2148  to  2150  (14,  7.5-93 
mm,  off  Misaki);  ABE  64-2037  to  2039  (3, 
30-42  mm,  Amakusa  Island);  ABE,  plankton 
collection  (9,  7-28  mm.  East  China  Sea); 
ABE  (1,  150  mm,  Tokyo  market,  gill  arch 
CS).    BC  56-29  (1,  136  mm,  Tokyo  market); 


BC  59-555  (1,  138  mm,  Aberdeen  market). 
CNHM  57288  (3,  91-124  mm,  Kobe).  tMCZ 
1186  (2,  122,  143  mm,  Kanagawa);  MCZ 
31150  (1,  125  mm,  Yenosima).  tUSNM 
6424  (1,  149  mm.  Hong  Kong);  tUSNM 
49465,  71131,  151829  (3,  146-156  mm, 
Tokyo  market);  tUSNM  59618  (1,  141  mm, 
Matsushima  Bay);  tUSNM  177426  (2,  132, 
142  mm,  Taipei  market). 

P.  ctjanea  Alcock:  *tBMNH  1890.11.28.9  ( 1, 
120  mm,  Ganjam  Coast,  India,  Bathuseriola 
cijanea  Alcock);  tBMNH  1937.6.28.1-8  (9, 
103-110  mm,  off  Cananore). 

Psenopsis  sp.:  tUSNM  98818  (1,  132  mm. 
Mare  Island,  Dutch  East  Indies). 

Cubiceps 

C.  athenae  Hacdrich:  *tMCZ  42974  ( 1,  68  mm, 
38°36'N  71°24'W,  Cubiceps  athenae  Hae- 
drich).  *tUSNM  198058  (1,  81  mm,  28°54'N 
88°18'W,  paratype  Cubiceps  athenae  Hac- 
drich). WHOI  (1,  33  mm,  24"2rN  8I°15'W, 
CS). 

C.  caeruleus  Regan:  *tBMNH  1913.12.4.28- 
29  (2,  86,  95  mm.  Three  King's  Isl.,  New 
Zealand,  Cubiceps  caeruleus  Regan);  BMNH 
1926.6.30.50  (1,  282  mm.  Lord  Howe  Isl., 
Tasman  Sea ) . 

C.  capensis  (Smith):  *BMNH  (1,  905  mm. 
South  Africa,  Atimnstoma  capensi.^  Smith, 
stuffed);  tBMNH  1925.10.14.1-4  (3,  all 
ca.  160  mm,  70  mi.  WNW  Saldanha  Bay, 
South  Africa). 

C.  carinatus  Nichols  and  Muri^hy:  CNHM 
61958  (5,  93-99  mm,  83°15'W  5n5'N); 
tCNHM  61939  (2,  both  110  mm,  125  mi.  SW 
Cape  Mala,  Panama).  SIO  63-538  (1,  101 
mm,  2I°35'N  107°00'W);  SIO  63-882,  63- 
888,  63-892  (4,  65-101  mm,  Golfo  de  Te- 
huantepec);  SIO  63-1027  (1,  91  mm,  13°33' 
N  95°59'W);  SIO  H  49-77  (1,  87  mm,  off 
El  Salvador);  SIO  H  52-351  (2,  90,  94  mm, 
13°45'N  99°22'W).  ZMC  (1,  57  mm,  11°52' 
N  97°19'W);  ZMC  (2,  64,  71  mm,  12n4'N 
97°46'W);  ZMC  (1,  70  mm,  12°38'N  98°14' 
W);  ZMC  (1,  72  mm,  13°00'N  98MrW); 
ZMC  (1,  71  mm,  13°4rN  97°34'W). 

C.  gracilis  Lowe:  tBMNH  63.12.12.7-8  (2, 
143,  150  mm,  Madeira);  BMNH  1960.12. 
19.8  (1,  165  mm,  5r51'N  13°43'W).  CF 
(5,  40-57  mm,  DANA  sta.  855);  CF  (2,  50, 
59  mm,  DANA  sta.  856);  CF  (26,  5-35  mm, 
DANA  sta.  939,  one  CS);  CF  (1,  69  mm, 
DANA  sta.  1372);  CF  (1,  ca.  70  mm, 
DANA  sta.  1378);  CF  (1,  77  mm,  DANA 
sta.  1380);  CF  (1,  43  mm,  DANA  .sta. 
4017,  CS);  CF  (4,  43-76  mm,  DANA  sta. 
4185);  CF  (18,  16-80  mm,  DANA  sta. 
4192);  CF  (37,  11-61  mm,  DANA  sta.  4195); 


Strom ATEOiD  Fishes  •  Hacdrich        41 


CF  (50,  10-58  mm,  DANA  sta.  4197). 
*MNHN  42-29  (1,  195  mm,  Sete,  Trachelo- 
cirrhus  meditcrraneiis  Doumet).  WHOI  (8, 
20-51  mm,  DELAWARE  63-4  sta.  3);  WHOI 
12,  2.5-51  mm,  39°27'N  27°35'W);  tWHOI 
(1,  92  mm,  DELAWARE  63-4  sta.  13); 
tWHOI  (2,  75,  81  mm,  DELAWARE  63-4 
sta.   16). 

C.  longimanus  Fowler:  *ANSP  55058  (1, 
42  mm,  Durban,  Natal,  Culnceps  lonfiitnamis 
Fowler).  IIOE  (1,  29  mm,  ANTON  BRUUN 
6  sta.  338). 

C.  pauchmliatus  Gimther:  ABE  57-347  (1,  120 
mm,  Manazuru,  Japan);  ABE  10832  (1,  80 
mm,  Kochi,  Japan).  *ANSP  68380  (1,  124 
mm,  40  mi.  S  Christmas  Isl.,  Line  Islands, 
Cuhiceps  nesiotes  Fowler).  *tBMNH  1870. 
8.31.124  (1,  113  mm,  Misol,  Molucca  Isl., 
Cuhiceps  pauciradiatus  Giinther).  SIO  60- 
216  (1,  93  mm,  10°26'N  128°22'W). 

C.  squamiceps  (Lloyd):  ABE  57-348,  59-37, 
61-843,  62-13,  62-106  (5,  84-154  mm,  Mana- 
zuru, Japan);  ABE  64-1348  (1,  264  mm, 
6°15'N  164°10'W);  ABE  10833  (1,  159  mm, 
Kochi,  Japan). 

ISlomeus 

N.  gronovii  (Gmelin):  ABE  64-1280  (1,  68 
mm,  Kushimoto,  Japan);  ABE  plankton  col- 
lections (5,  8-38  mm,  Japan).  BCF  1002 
( 1,  61  mm,  3°41'N  0"05'E  ) ;  BCF  1003  ( 1,  40 
mm,  GERONIMO  4-155).  BOG  602  (6,  17- 
101  mm.  Key  West  Harbor);  BOG  3361  (21, 
10-118  mm,  Atlantic  Ocean);  BOG  3515  (22, 
13-149  mm,  Gulf  of  Mexico);  tBOG  3516 
(8,  31-153  mm,  Gulf  of  Mexico);  BOC  3517 
(14,  16-133  mm,  28"07'N  89°53'W);  BOC 
3518  (11,  13-114  mm,  Atlantic  Ocean). 
tCNHM  (1,  225  mm,  OREGON  sta.  1178). 
IIOE  (4,  20-45  mm,  2^20'N  65°54'E);  IIOE 
(9,  12-41  mm,  8°00'S  65"00'E).  tMCZ  35327 
(2,  144,  154  mm,  39°27'N  70°38'W  gill  arch 
GS).  *MNHN  264.6.2.3  (3,  largest  ca. 
42  mm,  seas  of  Java,  Nomeus  peronii  Guvier 
and  Valenciennes).  SIO  60-263  (4,  22-40 
mm,  5°18'N  160°05'W);  SIO  61-84  (2,  20, 
25  mm,  5°58'S  149°31'W);  SIO  61-87  (1, 
33  mm,  1°32'S  148°39'W);  SIO  61-89  (3, 
32-41  mm,  5°32'N  146°09'W).  WHOI  (2, 
45,  66  mm,  17°00'N  65  05'W,  CS). 

P  series 

P.  arafurensis  Giinther:  ABE  60-101  (1,  72 
mm,  Manazuru,  Japan);  ABE  62-651  (1,  35 
mm,  Japan,  CS)";  ABE  64-212  (1,  150  mm, 
Nagasaki);  ABE  64-1767  (1,  92  mm,  Komat- 
subara,  Japan);  ABE,  plankton  collection  (2, 
19,    37    mm.    East    China    Seas).      *tBMNH 


1889.  7.  20.  55  ( 1,  30  mm,  China  Seas,  Fsenes 
arafurcmus  Giinther).  IIOE  (  1,  25  mm,  2°20' 
S  64°54'E);  IIOE  (1,  20  mm,  4°0rS  65°02' 
E).  MCZ  41550  (5,  14-18  mm,  10°52'N 
29°26'W).  WHOI  (1,  20  mm,  41°33'N  54° 
55'W). 
P.  hemmli  Rossignol  and  Blache:    *MNHN  264. 

9.1.1  (1,    63    mm,    3°38'S    9°22'E,    Psenes 
hcmirdi  Rossignol  and  Blache);  *MNHN  264. 

9.1.2  (2,  63,  65  mm,   1°55'S   8°30'E,   para- 
types,  Psenes  henardi  Rossignol  and  Blache ) . 

P.  cijanophrifs  Guvier  and  Valenciennes:  *t 
BMNH  1871.  7.  20.  156  ( 1,  111  mm,  Manado, 
Philippines,  Cuhiceps  mtdtiradiatus  Ciiinther). 
CF  ( 1,  28  mm,  Dansk  Vestindien  sta. 
132,  CS).  *tMNHN  264.9.2.4  (1,  117 
mm.  New  Ireland,  Bismarck  Archipelago, 
Psenes  cijanophnjs  Guvier  and  Valenciennes). 
CNHM  "46409  (1,  80  mm,  25"N  89°W). 
WHOI  (1,  38  mm,  CRAWFORD  62,  CS). 
tWHOI  (2,  119,  128  mm,  OREGON  sta. 
3715,  gill  arch  CS). 

P.  maculatus  Liitken:  ABE  64-1226  (1,  20  mm, 
Japan):  ABE  plankton  collection  (1,  20  mm, 
"East  China  Sea).  CF  (1,  48  mm,  15°31'N 
18°05'W).  tMCZ  41122  (1,  59  mm,  40M9'N 
64°57'W).  WHOI  (1,  27  mm,  38°26'N  68° 
15'W);  WHOI  (1,  35  mm,  41°36'N  60^30' 
W).  *tZMC  (2,  57  and  72  mm,  39^00'N 
34°10'W,  P.  maculatus  Liitken). 

P.  pcllucidus  Liitken:  ABE  59-172,  60-102,  60- 
106,  gill  arch  CS.  60-140  and  141,  61-21, 
61-82  and  83,  61-459  (13,  72-147  mm,  Mana- 
zuru, Japan);  ABE  63-1064  (1,  89  mm, 
Tokyo  market).  tBGF  957  (1,  130  imn, 
4°07'S  10°23'E).  CF  (1,  60  mm,  15°31'N 
18°05'W);  CF  (1,  39  mm,  AGENT  PETER- 
SEN sta.  769,  CS).  CNHM  5285  (1,  54  mm. 
Bermuda);  tCNHM  49189  (1,  131  mm,  Ber- 
muda); nCNHM  57097  (1,  193  mm,  Oki- 
nawa, Icticus  ischanus  Jordan  and  Thomp- 
son). *tUSNM  49745  (1,  67  mm,  Newport, 
Rhode  Island,  Psenes  eduardsii  Eigenmann). 
SU  43310  (1,  93  mm,  Bermuda).  WHOI 
(1,  25  mm,  38°38'N  68°50'W).  *tZMC  (1, 
38  mm,  Surabaya,  P.  peUucidus  Liitken). 

Ariomma 

A.  africana  (Gilchrist  and  von  Bonde):  *t 
BMNH  1927.  12.  6.  45  (1,  166  mm,  Agulhas 
Bank,  South  Africa,  co-t\pe,  Psenes  africanus 
Gilchrist   and   von    Bonde ) . 

A.  hondi  Fowler:  *ANSP  52528  (1,  79  mm, 
Grenada,  British  West  Indies,  Arioninia  hondi 
Fowler ) . 

A.  doUfusi  (Ghabanaud):  *tBMNH  1931.  4.  16. 
1  (1,  112  mm.  Gulf  of  Suez,  co-type,  Cuhi- 
ceps doUfusi  Ghabanaud ) . 

A.    evermanni    Jordan    and    Snyder:       *tUSNM 


42        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


57783  ( 1,  156  mm,  Honolulu,  Hawaii,  Arioiii- 
ma  evcrmanni  Jordan  and  Snyder). 

A.  irulica  (Day)":  BC  59-555  (1,  151  mm, 
Aberdeen  market,  Hong  Kong).  *tBMNH 
1889.2.1.3255-6  (2,  74,  90  mm,  Madras. 
India,  Psene.s-  imlicm  Day).  NTU  51941  to 
51946  (6,  123-181  mm,  near  Hainan  Island, 
sill  arch  CS). 

A.  Ittrida  Jordan  and  Snyder:  *SU  8441  (1,  ca. 
190  mm,  Honolulu,  Hawaii,  paratype  Ariomma 
Itirida  Jordan  and  Snyder).  *tUSNM  51400 
(1,  166  mm,  Honolulu,  Hawaii,  Ariomma  hirida 
Jordan  and  Snyder);  tUSNM  109418  (1,  193 
mm,  Honolulu). 

A.  mdimu  (Ginsburg):  *tUSNM  157779  ( 1, 
154  mm,  Mississippi  Delta,  Cubiceps  melanus 
Ginsburg). 

A.  multisquumis  (Marchal):  *MNHN  264.7.2. 
1  (type),  264.7.2.2  (paratype)  (2,  159, 
147  mm,  Ivory  Coast,  Parcictihiccps-  miilti- 
sqiiamis  Marchal). 

A.  nigriargenteus  (Ginsburg):  *MCZ  .37183  (1, 
113  mm.  Sandwich,  Massachusetts,  paratype 
Cuhiceps  nigriurgenieus  Ginsburg).  *tUSNM 
151954  (1,  190  mm,  off  Cape  Romain,  South 
Carolina,  Cuhiceps  nigriargcnteus  Ginsburg). 

A.  regidm  (Poey):  tUSNM  197110  (2,  142, 
150  mm,  British  Guiana).  tWHOI  (1,  136 
mm,  29°59'N  87°06'W). 

Ariomma  sp.  Western  North  Atlantic:  MCZ 
40259  (1,  116  mm,  Provincetown,  Massachu- 
.setts);  MCZ  40498  (1,  115  mm.  Province- 
town,  Massachusetts).  SU  57297  (1,  86  mm, 
Bermuda).  WHO!  (3,  87-121  mm,  OREGON 
sta.  3725);  WHOI  (2,  103,  109  mm,  ORE- 
GON sta.  3733);  WHOI  (3,  122-134  mm, 
OREGON  sta.  4014);  WHOI  (1,  143  mm, 
9°03'N  81^22^);  WHOI  (2,  141,  149  mm, 
9°13'N  80°44'W);  WHOI  (3,  139-140  mm, 
16°45'N  8r27'W);  WHOI  (3,  131-140  mm, 
28°57'N  88°4rW);  WHOI  (2,  125,  133  mm, 
29°07'N  88°34'W);  WHOI  (3,  1,30-140  mm, 
28°54'N  88°51'W,  gill  arch  CS);  WHOI  (1, 
23  mm,  CRAWFORD  62  sta.  29,  CS);  WHOI 
(1,  28  mm,  24°N  81  °W,  CS). 

Ariomma  sp.  Japan:  ABE  59-404,  59-408,  60- 
124,  60-144,  60-224,  60-478  and  479,  60-1611, 
61-1188,  62-738,  62-955,  62-1387,  62-1628 
(13,  99-226  mm,  Manazuru,  Japan).  NRF 
1441  (1,  ca.  800  mm,  Bonin  Islands). 

Tctra^oniirus 

T.  atlanticm  Lowe:  *tBMNH  ( 1,  197  mm, 
Madeira,  Tctragonurus  ailanticas  Lowe). 
MCZ  41726  (1,  21  nun,  39°47'N,  70'32'W, 
CS);  MCZ  41791  (1,  66  mm,  39°4rN  69° 
54'W,  CS).  WHOI  (2,  15,  20  mm,  tropical 
Atlantic). 


T.  cuvicri  Risso:  CNHM  64218  (1,  340  mm, 
46°5rN,    I55°00'W,    gill    arch    CS). 

Stromotcus 

S.  Inasilicnsis  Fowler:  *ANSP  11354  (1,  276 
mm,  Rio  Grande  do  Sul,  Brazil,  Stromatcus 
hra.silicnsi.s  Fowler).  tMCZ  4599  (3,  254- 
285  mm,  Rio  Grande  do  Sul,  Brazil). 

S.  ficitola  Linnaeus:  tBMNH  87.  3.  2.  30  ( 1,  76 
mm.  Lower  Congo).  tGTS  (6,  200-240  mm, 
7°20'N  I2°40'W).  tMCZ  16729  (1,  257  mm, 
Mediterranean?).  tSU  1537  (1,  44  mm, 
Palermo,  Italy).  tUSNM  (I,  177  mm.  Fed. 
Fish.  Serv.  Nigeria,  No.  4046). 

S.  steUattis  Cuvier:  tUSNM  Ace.  No.  167496  (2, 
230,  237  mm,  42°29'S  72°46'W).  *MNHN 
264.11.3.1-2  (2,  144,  164  mm,  Valparaiso, 
Chile,  Stromatetis  maculatus  Cuvier  and  Va- 
lenciennes). tUSNM  176474  (3,  178-250  mm, 
Calbucco,  Chile);  tUSNM  176494  (1,  193 
mm,  41°52'S  73°53'W). 

Feprilus 

P.  alcpidotus  (Linnaeus):  tUSNM  127352, 
127353  (2,  122,  141  mm,  Grand  Terre,  La.); 
tUSNM  23215  (1,  130  mm.  Bay  Chaland, 
La.).  WHOI  (1,  39  mm,  SILVER  BAY  sta. 
4331,  CS). 

P.  palometa  ( lordan  and  Bollman):  *tUSNM 
41136  (5,  53-61  mm,  Perlas  Isl;  Panama, 
Stromatcus  palometa  Jordan  and  Bollman; 
tUSNM   .50.337    (2,   136,   169  mm,   Panama). 

P.  paru  (Linnaeus):  tMCZ  4600  (1,  55  mm, 
Brazil);  tMCZ  41064  (2,  48,  57  mm,  Port-au- 
Prince,  Haiti).  *MNHN  (1,  93  mm,  Rio  de 
Janeiro,  Sescrinus  xautJitirus  Quoy  and  Gai- 
mard ) . 

P.  simillimus  (Ayres):  tMCZ  26875  (1,  110 
mm,  San  Diego,  Calif.).  tSU  48000  (1,  82 
mm,  Oceanside,  California). 

P.  smjdcri  Gilbert  and  Starks:  *tBMNH  1903. 
5.  15.  190  (1,  217  mm,  Panama,  co-type 
Peprilus  sntjderi  Gilbert  and  Starks).  *tUSNM 
50448  (  1,  189  mm,  Panama,  PcprUus  smjdcri 
Gilbert  and  Starks).  tUSNM  76796  (1,"  178 
mm,  Panama  City). 

P.  triacanthus  (Peck):  tABE  64-1920,  1924, 
1930  (3,  80-110  mm,  Florida).  tWIIOI  (7, 
157-195  mm,  SILVER  BAY  sta.  4104); 
tWHOI  (4,  127-155  mm,  40°0rN  7r23'W); 
tWHOI  (1,  120  mm.  Woods  Hole,  gill  arch 
CS);  WHOI  (I,  36  mm,  BEAR  188  sta.  371, 
CS). 

Pampus 

P.  argenteus  (Euphrascn):  tABE  64-1231,  12.33 
(2,  10,5,  147  mm,  Bangkok);  tABE  64-1929 
(I,  1,33  mm.  Hong  Kong);  tABE  19,36,  1937 


Stromateoid  Fishes  •  Haedrich        43 


(2,  48,  72  mm,  off  mouth  of  Sliiota  River, 
Ariake  Sound,  Kyushu,  CS);  tABE  64-1964 
(1,  156  mm,  East  China  Sea);  tABE  P  4347 
(1,  92  mm,  Sarawak,  North  Borneo).  tllOE 
(6,  65-161  mm,  Bombay).  tUSNM  44900 
(1,  299  mm,  Japan). 

P.  chinemis  (  Euphrasen  ) :  tABE  64-1229  (1, 
89  mm,  Bangkok);  tABE  P  2211,  P  4319 
(2,  77,  112  mm,  Sarawak,  North  Borneo). 
tCNHM  15917  (1,  74  mm,  Batavia,  Java). 
tMCZ   16772   (1,  84  mm,  Singapore). 

P.  echinogaster  ( Basilewsky ) :  tABE  64-902, 
64-906,  64-909,  64-911,  64-926,  64-1192  (5, 
165-207  mm.  East  China  Sea);  tABE  64-1112 
(1,  221  mm,  Tokyo  market);  tABE  1743  ( 1,  ca. 
180  mm,  Tokyo  market,  gill  arch  CS);  tABE 
64-1934  (1,  158  mm,  south  of  Poi  Toi  Isl., 
Hong  Kong).  tCNHM  55810  (3,  81-106  mm, 
Chinnampo,  Korea).  tSU  22292  (3,  77-87, 
Chinnampo,  Korea).  tUSNM  75941  (1,  236 
mm,  Japan?). 

In    addition    to    stromateoid    fishes,    the 
following  non-stromateoids  were  examined: 

Apolectidae 

Apolectus     [=  Parastromateus]     niger     MCZ 

15912  (3,  Singapore). 
Arripididae 

Arripis  trutia  ZMC  lA.VAl^  (3,  New  Zealand). 
Atherinidae 

Menidia  sp.  WHOI   (2,  Morehead  City). 
Carangidae 

Curanx  hippos  WHOI  ( 1,  Woods  Hole),   hkw- 

cratcs  ductor  WHOI  (1,  ATLANTIS  sta.  219). 

Selar    crutucnoptliahuus    WHOI     ( 1,    Woods 

Hole).    Trachinotus  glaucus  WHOI    (1,  Tru- 

jillo,  Honduras ) . 
Coryphaenidae 

Conjphaena   equiselis   WHOI    (1,   OREGON 

sta.  1297). 
Ephippidae 

Pkitax  occUaim  MCZ  2748   (1,  Manila). 
Icosteidae 

Icosfetis  aenigmaticus  BC  63-98   (1,  Alaska); 

BC  64-12   (1,  Spiller  Channel).    MCZ  34915 

(1,  California).    SU  1171   (1,  Pacific  Grove); 

SU  25640  ( 1,  Monterey  Bay).    *USNM  27398 

(1,  Point  Reyes);   USNM  75159    (1,   Pacific 

Grove ) . 
Girellidae 

Boops    vulgaris    MCZ    21706     (1.     France). 

Girella  nigricans  MCZ  10775   (1,  Cahfornia). 
Kuhliidae 

KuJdia  main  MCZ  29367   (1,  Papeete). 
K\'phosidae 

Kyphosus  sectatrix  WHOI    (2,   32°08'N   67° 

lO'W).     Pimelepterm   hoscii   MCZ   2610    ( 1, 

Florida ) . 


Monodactylidae 

MonodacUjlus  argcntcus  MCZ  34101  (1,  Dar 
es  Salaam). 

Nematistiidae 

Nematistius  pcctoralis  BC  60-15  (1,  Acapulco 
market ) . 

Pomatomidae 

Pomatomus  saltatrix  MCZ  16941  (5,  New  Jer- 
sey).  WHOI    (5,   Woods    Hole). 

Scomberesocidae 

Scomberesox  saurus  WHOI  (1,  40°12'N  62° 
54'W). 

Scombridae 

Auxis  thazard  WHOI  (1,  41°16'N  .57°37' 
W).  Scomber  scomber  WHOI  (1,  Quisset, 
Mass.). 

Scorpididae 

Scorpls  calif orniensis  MCZ  4896  (2,  Cali- 
fornia ) . 

Tlieraponidae 

Autlsthes  puta  WHOI  (2,  Australia,  received 
through  James  Moulton).  Pelates  sp.  WHOI 
(2,  Australia,  received  through  James  Moul- 
ton).   Therapon  jarbita  MCZ  24823  (2,  Java). 

Zaproridae 

Zaprora  silcnus  BC  61-573  ( 1,  Cross  Sound, 
Alaska). 

SYSTEMATIC  SECTION 

In  the  classification  which  follows,  the 
suborder,  the  families,  and  the  genera  are 
characterized.  Categories  below  the  genus 
are  not.  In  a  number  of  cases,  subgeneric 
division  is  certainly  called  for.  This  action 
is  not  taken  here,  but  is  reserved  for  critical 
treatment  in  a  planned  series  of  mono- 
graphs. Likewise,  the  proper  sorting  of  spe- 
cies is  left  for  the  future.  As  a  preliminary 
step  in  clarifying  the  confusion  which  sur- 
rounds stromateoid  classification,  however, 
lists  of  nominal  species  are  included  under 
each  genus.  Species  synonymies,  whenever 
given,  are  to  be  considered  tentative. 

The  synonymy  of  the  suborder  is  fairly 
complete  and  is  intended  as  a  guide  to  most 
works,  especially  those  of  a  faunistic  nature, 
that  include  references  to  stromateoids. 
Distributional  notes  concerning  single  or 
very  few  species,  however,  have  not  been 
included.  The  synonymies  of  each  family 
contain  only  major  references.  Family 
names  have  been  used  with  such  confusion 
in  the  past  that  complete  synonymies  would 
be  essentially  meaningless. 


44        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Included  under  each  taxon  are:  synon- 
ymy, diagnosis  or  brief  characterization, 
description,  distribution  (suborder,  fami- 
hes),  natural  history  (suborder,  genera), 
relationships,  and  key  to  included  taxa  or 
list  of  nominal  species.  An  asterisk  (*) 
precedes  the  names  of  species  of  which  I 
have  seen  the  types. 

Suborder  STROMATEOIDEI 

Stromatini.  Rafinesque,  1810:39  {Stronwtcus, 
Luvarus). 

Stromateini.  Bonaparte,  1846:76  {Siwmatcm, 
Peprilus,   Luvarus,   Kurtus). 

Stromatiniae.  Swainson,  1839:177  (Seserinus, 
Stromateus,    Peprilus,    Kurtus,    Kcris). 

Stromateina.  Giinther,  1860:397  (+  Nomeina  p. 
387,  clef.,  Stromateus,  Ccntrolophu.s,  Gastero- 
chisma,  Nomcus,  Ctibiceps,  Neptomenus, 
Phtijsieth us,   Ditrema ) . 

Stromateidae.    Gill,  1884:66.5   (def.,  gen.  synopsis, 
Centrolophus,  Schcdophilus,  Lirus,  Stromateus, 
Stromateoides,  Psenopsis).   Day,  1875:246  (  + 
Nomeidae,  p.  237.  descr.;  India).    Jordan  and 
Gilbert,   1882:449   (key,  descr.;   North  Amer- 
ica).   Fordicc,  1884:  311  (key,  synon.;  Amer- 
ica).   Collett,   1896:26   ( -f   Nomeidae,  p.  31. 
descr.;    eastern    Atlantic).    Goode    and    Bean, 
1896:213  (descr.;  oceanic  spp.).    Regan,  1902: 
117      (major     revi.sion,     Nomeus,     Cuhiceps, 
Psenes,     SerioleUa,     Psenopsis,     Centrolophus, 
Lirus,    Stromateus,    Peprilus,    Stromateoides). 
Holt  and  Byrne,  1903:71   (key,  descr.;  British 
Isles).      Boulenger,     1904:64.3     (popular    ac- 
count).   Smith,  1907:221    (key,  descr.;  North 
Carolina).  Miranda-Ribeiro,  1915  (key,  descr.; 
Brazil).    Gilchrist,   1922:249   (papillae).    Gil- 
christ and  von  Bonde,   1923:1    (descr.;  South 
Africa).      Meek    and     Hildebrand,     1925:407 
(key,  descr.;  Panama).    Biihler,  19.30:62   (di- 
gestive    system).      Fowler,     1936:658     (key, 
descr.;  West  Africa).    Fowler,   1941:152   (  + 
Nomeidae,  list;  Brazil).    Fowler,  1944:78   (  + 
Centrolophidae,  p.  79;  Nomeidae,  p.  80.  descr.; 
Chile).    Hildebrand,  1946:416  (descr.;  Peru). 
Barnard,  1948:394  (descr.,  sacs  and  papillae; 
South  Africa).    Smith,  1949:302  (key,  descr.; 
South    Africa);    1949a: 8.39    (revision;    South 
Africa).     Lozano    y    Rey,    1952:648    (descr.; 
Iberia).    Mori,   1952:138   (+   Nomeidae,  hst; 
Korea).   Bigelow  and  Schroeder,  1953:363  {  + 
Centrolophidae,  p.  369,  descr.;  Gulf  of  Maine). 
Herre,   1953:258    (+   Nomeidae.   list,   synon.; 
Philippines).  Tchang  et  al,  19.55:195  (descr.; 
Gulf  of  Pechili,  Yellow  Sea).    Poll,   19.59:125 
(descr.;  West  Africa).    Blache,  1962:70  (hst; 
West  Africa).    Lowe,   1962:694   (list;   British 


Guiana).    Chu  et  al,  1963:407   (descr.;  East 
China  Sea). 

[Stromateidae.]  Liitken,  1880:513,  521  (disc,  gen., 
oceanic  spp.).  Giinther  1889:10  (disc,  gen., 
descr.;  CHALLENGER).  Nobre,  1935:332 
(descr.;  Portugal).  Okada,  Uchida,  and 
Matsubara,  1935:123  (descr.,  ill;  Japan). 
Kamohara,    1940:173    (descr.;    Japan). 

Nomeifonnes.    Gregory,   1907:. 502    (relationships). 

Stromateiformes.  Jordan,  1923:182  (list,  fam., 
gen.).  Jordan,  Evermann,  and  Clark,  1930: 
226   (list,  synon.;   North  America). 

Centrolophidae.'  Fowler,  1928:138  (descr.;  Pa- 
cific); 1931:325  (descr.;  add.  Pacific);  1934: 
403  (descr.;  add.  Pacific);  1949:75  (descr.; 
add.  Pacific).  Norman,  1937:115  (+  Stro- 
mateidae, p.  118.  descr.;  Patagonia).  Sanz- 
Echeverria,  1949:151  (otoliths).  Tortonese, 
19,59:.57   (revision;  Gulf  of  Genoa). 

Stromateoids.  Gregory,  1933:306  (skull  of  Rhom- 
l)us  [=  Peprilus] ). 

Stromateoidei.  Berg,  1940:.324  (def.,  fam.  synop- 
sis); 19.55:247  (def.,  fam.  synopsis).  Bertin 
and  Arambourg,  1958:2441  (fam.,  gen.  synop- 
sis). Munro,  1958:  117  (descr.;  Australia). 
Duarte-Bello,  1959:119  (list;  Cuba).  Gosline 
and  Brock,  1960:281  (descr.;  Hawaii).  Chu 
et  al,  1962:759  (descr.;  South  China  Sea). 
Scott,  1962:142  (key,  descr.;  South  Australia). 

Stromateoidea.  Blegvad  and  L0ppenthin,  1944:178 
(descr.;  Iranian  Gulf).  Beaufort  and  Chap- 
man, 1951:85  (descr.;  Indo-Australian  Archi- 
pelago). Norman,  1957:216  (fam.,  gen. 
syTiopsis,  after  Regan).  Herald,  1961:243 
(popular  account).  Marshall,  1964:398  (key, 
descr.;  Eastern  Austraha). 

Stromateoidae.  Clemens  and  Wilby,  1961:230 
(de.scr.;   British   Columbia). 

Distinctive  characters.  There  is  no  mis- 
taking the  "stromateoid  look."  Though  the 
characters  given  in  the  diagnosis  are  the 
only  ones  that  absolutely  identify  a  member 
of  the  suborder,  these  fishes  nonetheless 
have  a  physiognomy  that  nine  times  out  of 
ten  says  "Stromateoid!"  to  an  experienced 
ichthyologist.  Once  recognized,  the  stro- 
mateoid expression  is  not  likely  to  be  for- 
gotten. It  is  a  fat-nosed,  wide-eyed,  stuf fed- 
up  look,  smug  and  at  the  same  time  appre- 
hensive. Some  stromateoids  might  even  be 
accused  of  a  certain  prissiness. 

The  stromateoid  look  results  from  the  ex- 
panded lacrimal  bone  all  but  covering  the 
maxillary,  the  slightly  underslung  lower  jaw 
shutting  within  the  upper,  the  large  cen- 


Stromateoid  Fishes  •  Haedrich        45 


trally  located  eye  rimmed  with  adipose  tis- 
sue, and  the  protruding,  inflated,  naked, 
and  pored  snout  and  top  of  the  head. 

Diagnosh.  Perciform  fishes  with  toothed 
saccular  outgrowths  in  the  gullet  immedi- 
ately behind  the  last  gill  arch.  Small  teeth 
approximately  uniserial  in  the  jaws. 

Description.  Body  slender  to  deep,  com- 
pressed or  rounded.  Dorsal  fin  single  or 
double;  dorsal  spines  present,  very  weak  in 
some  species.  One  to  three  anal  spines, 
never  separated  from  the  rays.  Dorsal  and 
anal  fins  coterminal.  Pelvic  fins  present  or 
absent.  Rays  in  pectoral  fin  16  to  25.  Body 
scaled,  snout  and  top  of  head  naked.  Scales 
usually  thin,  cycloid,  deciduous,  but  very 
weakly  ctenoid  in  some  nomeids  and 
Schcdophilus  medusophagtis,  and  heavy, 
keeled,  and  adherent  in  Tefragonurus. 
Scales  usually  covering  bases  of  median 
fins.  Lateral  line  present,  the  scales  with 
simple  tubes,  except  no  tubed  scales  in 
Tetragonurus.  No  bony  scutes  or  keels  as- 
sociated with  lateral  line.  Usually  a  well 
developed  subdennal  mucous  canal  system 
communicating  to  the  surface  through  small 
pores  liberally  scattered  over  head  and 
body.  Eyes  small  to  large,  lateral,  not  enter- 
ing into  profile  of  head.  Nostrils  double, 
the  anterior  round,  the  posterior  usually  a 
vertical  slit.  Jaw  teeth  small,  simple  or 
minutely  cusped,  arranged  more  or  less  in 
a  single  series,  close-set  or  spaced.  Teeth 
on  vomer,  palatines,  and  basibranchials 
present  or  absent.  No  teeth  on  entoptery- 
goid  or  metapterygoid.  Small  teeth  usually 
present  on  inner  edge  of  gill-rakers.  Gills 
4,  a  slit  behind  the  fourth.  Gill-rakers  10 
to  20  on  lower  limb  of  first  arch.  Well  de- 
veloped pseudobranch  usually  present,  but 
absent  in  Pampus;  rudimentary  gill-rakers 
under  pseudobranch  commonly  present. 
Gill-membranes  usually  free  from  isthmus, 
but  united  in  Pampus. 

Skeleton  never  strongly  ossified.  Epiotic 
forked,  orbitosphenoid  absent,  15  principal 
branched  rays  in  caudal  fin.  Pelvic  fin, 
when  present,  with  one  spine,  five  rays.  A 
bony  bridge  partially  covering  the  anterior 


vertical  canal  of  the  ear.  Opercular  bones 
thin,  denticulate  or  entire,  never  with  strong 
spines,  except  moderate  preopercular  spines 
in  Schedophilus.  No  bony  stay  for  the  pre- 
opercle.  Five  to  seven  branchiostegal  rays. 
Lacrimal  bone  usually  enlarged,  often  cov- 
ering maxillary  almost  completely.  Lacri- 
mal absent  in  Pampus.  Premaxillary  scarcely 
if  at  all  protractile.  A  slender  supramaxil- 
lary  present  or  absent.  Pelvic  bones  not 
finnly  attached  to  coracoid.  Vertebrae  25, 
26,  or  29  to  60,  including  hypural.  Caudal 
skeleton  with  two  to  six  hypural  elements, 
two  or  three  epural  elements,  two  auto- 
genous haemal  spines,  except  three  in 
Icichthys,  hypuropophysis  present  on  first 
hypural.  Sometimes  two  but  usually  three 
free  interneurals  ahead  of  dorsal  fin,  but 
six  or  more  in  Icichthys.  Lower  pharyngeal 
bones  not  united,  partially  supporting  pha- 
ryngeal sacs.  In  the  sacs,  numerous  simple 
teeth  on  irregularly-shaped  or  stellate  bony 
bases  [=  papillae]  seated  in  the  muscular 
walls,  arranged  in  longitudinal  bands  or 
not. 

Distribution.  All  stromateoid  fishes  are 
marine.  They  are  found  in  the  three  major 
oceans  of  the  world  on  the  high  seas,  over 
the  continental  shelves,  and  in  large  bays. 
Most  species  live  in  tropical  and  temperate 
waters,  but  a  few  occur  in  colder  areas.  No 
stromateoids  have  been  reported  from  the 
Arctic  Ocean,  the  Bering  Sea,  the  Okhotsk 
Sea,  the  Baltic  Sea,  the  Black  Sea,  and  the 
Antarctic  Ocean. 

Natural  history.  The  association  between 
stromateoid  fishes  and  medusae  or  siphono- 
phores  is  widely  known  and  well  docu- 
mented. Mansueti  (1963)  has  reviewed  the 
literature  concerning  this  unusual  associa- 
tion. His  lists  of  all  fish  species  involved 
is  made  up  largely  of  stromateoids,  and  in- 
cludes centrolophids,  nomeids,  tetragonu- 
rids,  and  stromateids.  Only  one  ariommid 
has  been  reported  from  under  a  jellyfish,  a 
.36-mm  Ariomma  indica  taken  in  a  305-mm 
diameter  ctenophore  in  Durban  Bay,  Natal 
(Fowler,   1934a). 


46        BiiUctin  Museum  of  Couipaidtive  Zoology,  Vol.  135,  No.  2 


Plate   1.      A  young  stromateoid,  Psenops/s  onomo/o,  under  a  medusa.     Iwago  photo. 


Young  stromateoids  are  pelagic,  and  it  is  living  host,  and  descend  to  deeper  layers, 

not  surprising  that  it  is  the  young  stages  the  adult  habitat.    Stromateoids  also  hover 

that   are   found   with   jellyfishes.     As   they  beneath     flotsam     and     Sarf],os.s-iim     weed 

grow  older.  th(>  fishes  desert  their  surface-  ( Besednov,   1960).    It  is  this  charactf^istic 


Stromateoid  Fishes  •  Hacdrich 


NOUEUS     BRONOVII 

J  100  120  140 


200  220 


PREANAL      DISTANCE 


PREDORSAL     DISTANCE 


MAXIMUM       DEPTH 


_l_ 


_J_ 


_L_ 


J_ 


"T  I  1 1 1 

LENGTH     OF     HEAD 


_i- 


-J_ 


1          I 

1      1      1 

I     I      i      1      1     1 

■ 

,  * 

- 

•    .     '  •  ••*, 

'    ' 

1      ' 

1      1      1 

LENGTH    OF     PECTORAL     FIN 

1 1 1                    1                    !                    1 

1                    I                -    , 

1        I     ■-] 

1                 1 

1 

- 

••  • 

LENGTH 

OF    PELVIC 

FIN 

- 

— 1 1 1 1 

1_  .1                1                 1 

1                1 

1    • 

STANDARD   LENGTH  IN  MILLIMETERS 


Figure  3.      Scatter    diagram    of    the    allometric    growth    in    the 
nomeid  Nomeus  gronovn. 


habit  which  gives  the  barrelfish,  Hypcro- 
gjyphe,  its  common  name.  Young  stromat- 
eoids  typically  have  a  banded  pattern, 
whereas  adults  tend  to  be  plain.  Undoubt- 
edly, the  banding  is  protective  coloration 
for  the  fishes  during  that  period  of  their 
lives  when  they  live  in  the  shifting  shadows 
under  a  jellyfish. 

Shelter  is  not  all  the  jellyfish  provide. 
Many  stromateoids  have  been  observed 
actively  feeding  on  their  hosts  {Schedo- 
philus.  Maul  1964;  Nomem,  Kato,  1933; 
Peprihis,  Mansueti,  1963,  photo  p.  63).  The 
diagnostic  pharyngeal  sacs  of  stromateoids 
may  have  been  perfected  partially  in  re- 
sponse to  this  sort  of  diet.  Tetragonurus, 
one  of  the  most  highly  specialized  stromat- 
eoids with  great  slicing  teeth  and  a  long 
pharyngeal  sac,  may  feed  almost  entirely 
on  soft-bodied  medusae,  ctenophores,  and 
salps  (Grey,  1955). 


SCHEDOPHILUS    MEDUSOPHAGUS 

60 80 100  120  140 


PREANAL    DISTANCE 


PREDORSAL    DISTANCE 


S    35 


MAXIMUM    DEPTH 


to 

5 

S    551-    •'■ 


"1  I  I 

LENGTH   OF  HEAD 


LENGTH  OF   PECTORAL  FIN 


STANUARD  LENGTH  in  MILLIMETERS 


Figure  4.  Scatter  diagram  of  the  early  growth  in  the  soft- 
spined  centrolophid  Schedoph/lus  medusophogus,  a  fish 
which    grows    to   500    mm    SL. 


Stromateoids  do  show  high  resistance  to 
the  toxins  of  jellyfish  (Lane,  1960;  Maul, 
1964 )  but  they  are  by  no  means  immune  to 
it  (Garman,  1896;  Totton,  1960).  Besides 
the  relatively  high  resistance  to  the  toxins, 
simple  avoidance  of  the  tentacles  and  the 
characteristic  hea\'y  coating  of  slime  prob- 
ably are  important  in  allowing  the  fishes 
to  swim  with  impunity  under  their  hosts. 

Because  of  the  efficient  shredding  of  food 
by  the  papillae,  stomach  contents  of  stro- 
mateoids are  largely  unidentifiable.  Shred- 
ded transparent  tissues,  probably  from  jelly- 
fish, ctenophores,  and  salps,  predominated 
in    stomachs    examined.    But    I    have    also 


48        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


PSENOPSIS   ANOMALA 

20 

40 

60            80 

100             120             140             160 

leo 

1 

1 

1                1 

1                  1                   1                   1 

70 

- 

# 

• 

PREftNAL     DISTANCE 

- 

60 

>■ 

•• 

.*  '•'*•'•• 

■  ': 

- 

50 

- 

1 

1 

1                1 

1            1            1            1 

1 

- 

-.4 


1 r 

•  •  •  ** 


"T 1 1 r 

PREDORSAL    DISTANCE 


J I I L- 


>u  55 


1 

. , 

1 
\ 

1          1 

1 

1                 1                 1                 1 

• 

•'    1 

1 

1                 1 

1 

MAXIMUM    DEPTH 

1                1                1               1 

T 1 1 r 

LENGTH  OF  HEAD 


■••  <: . 


1 

1 

1          1 

1            1            1 

1 

1 

1 

1            1 

LENGTH  OF    PECTORAL 
1                 1                 1 

FIN 
1 

.J 

1 

1      1 

1      1      1      1      1 

i: 

•• 

1 

■>. 

1 

1      1 

LENGTH  OF  PELVIC    FIN 

'■'■•.".* 

1                1                1                1                1 

- 

40  60  80  100  120  140  160  180 

STANDARD  LENGTH  in  MILLIMETERS 

Figure  5.      Scatter  diagram  of  the  growth   in   the  hord-spined 
centrolophid   Psenops/s  onomala. 


found  remains  of  fishes,  large  crustaceans, 
and,  rarely,  squids. 

With  growth,  marked  changes  occur.  The 
fish  are  no  longer  in  the  immediate  upper 
layers,  but  tend  to  live  deeper  in  the  water. 
The  typically  banded  pattern  of  the  young 
fish  gives  way  to  the  plainer  colors  of  the 
adult.  The  most  pronounced  changes  are 
in  relative  proportions.  Allometry  is  the 
rule  among  stromateoids.  Generally,  the 
relative  length  of  the  head,  length  of  the 
pelvic  fins,  predorsal  distance,  and  preanal 
distance  increase  rapidly,  then  decrease 
with  growth.  The  relative  length  of  the 
pectoral  fin  and  the  maximum  depth  may 
increase  steadily,  or  increase  and  then  de- 
crease. Allometry  is  especially  marked  in 
nomeids   (Fig.  3),  most  of  which  have  a 


similar  pattern  of  growth,  and  centrolophids 
(Figs.  4,  5).  The  growth  curves  given  for 
Tetraiionurm  (Grey,  1955)  are  fairly  regu- 
lar, and  rarely  show  a  range  of  variation  ex- 
ceeding ten  per  cent.  McKenney's  ( 1961 ) 
curves  for  Ariomma,  perhaps  based  on  two 
species,  likewise  show  regular  and  little 
variation.  The  pectoral  fin  of  the  stroma- 
teids  may  relatively  lengthen  or  shorten 
with  age  and  the  fin  lobes  may  be  very 
long  in  the  young,  but  the  allometry  is  in 
general  not  remarkable  in  this  family. 

The  largest  stromateoids  are  the  centro- 
lophids, many  of  which  reach  three  feet  in 
length.  A  large  CentroJophus  or  Hijpero- 
glyphc  will  exceed  four  feet.  A  few  no- 
meids attain  two  feet,  but  most  are  smaller. 
There  is  a  diminutive  species  group  in  the 
family;  its  members  rarely  exceed  six  inches. 
Tetragonurids  are  said  to  reach  two  feet  in 
length  (Fitch,  1951).  The  largest  reported 
ariommid  was  almost  three  feet  long 
(Klunzinger,  1884),  but  most  seem  to  be 
much  smaller.  A  large  stromateid  barely 
exceeds  18  inches. 

Stromateids,  the  subjects  of  substantial 
fisheries  in  the  eastern  United  States,  India, 
China,  and  Japan,  and  ariommids,  with 
commercial  potential  but  as  yet  unfished, 
are  schooling  fishes.  The  centrolophids 
Psenopsis  and  Hijperoiijijphe,  commercial 
fishes  in  Japan,  and  Seriolella,  fished  to  a 
lesser  extent  in  Peru,  Ghile,  and  Australia, 
are  also  found  in  shoals.  Numerous  young 
specimens  of  the  more  oceanic  stromateoids 
are  often  taken  by  dipnet  collecting,  but 
data  is  too  sparse,  and  large  specimens  too 
few,  to  know  whether  these  fishes  occur  in 
schools  or  not. 

Little  is  known  of  the  breeding  of  stro- 
mateoids. The  eggs  are  pelagic.  Those  of 
Vcprihis  triacanthiis  are  described  as  buoy- 
ant, transparent,  and  containing  a  large  oil 
globule  (Bigelow  and  Schroeder,  1953). 
All  species  seem  dioecious.  There  is  some 
evidence  that  the  sexes  may  be  slightly  di- 
morphic, with  respect  to  color  and  or  rela- 
tive proportions. 

Rclatiunships.  The  stromateoid  fishes  are 


Stromateoid  Fishes  •  Haedrich        49 


a  well-defined  unit.  Pharyngeal  sacs  are 
the  diagnostic  character.  These  are  present 
in  all  species,  and  are  readily  apparent  on 
dissection.  A  somewhat  similar  organ  is 
found  in  Dorosoma,  a  clupeoid  (Miller, 
1964),  and  in  Trisotrophis,  a  serranid  (Kata- 
yama,  1959). 

Within  the  group,  there  is  a  broad  spec- 
trum from  primitive  to  highly  advanced 
forms.  Between  existing  families,  there  is 
evidence  of  phyletic  relationships.  The 
natural  coherence  of  the  group  and  the  ten- 
dency toward  direct  internal  lines  from  the 
generalized  to  the  specialized  condition, 
makes  it  unlikely  that  the  stromateoids  have 
given  rise  to  other  groups.  Tetragominis,  a 
highly  specialized  form,  is  sometimes  placed 
in  a  suborder  of  its  own,  in  which  case  it 
would  be  considered  a  derived  group. 
Tetragomirus  is  certainly  far  removed  from 
the  mainstream  of  stromateoid  evolution, 
but  its  degree  of  divergence  is  probably  not 
great  enough  to  warrant  subordinal  recog- 
nition. It  retains  the  characteristic  sacs,  and 
is  here  considered  the  sole  genus  of  the 
stromateoid  family  Tetragonuridae. 

Giinther  (1880)  considered  the  stromat- 
eoids a  subdivision  of  the  scombroids. 
Little  subsequent  attention  has  been  paid 
to  the  relationships  of  the  suborder,  with 
the  exception  of  Boulenger  (1904)  and  Holt 
and  Byrne  ( 1903 )  who  found  cause  for  in- 
cluding them  in  the  Percesoces.  General 
classifications  have  continued  to  place  them 
near  the  Scombroidei.  There  is  a  close  re- 
semblance between  some  stromateoids  and 
the  carangids,  a  group  standing  near  the 
base  of  the  scombroid  stem  ( Suzuki,  1962 ) . 

Freihofer's  ( 1963 )  survey  of  patterns  of 
the  ramus  lateralis  accessorhis  (a  facial 
nerve  complex)  in  fishes  indicated  that  the 
stromateoids  might  be  related  to  entirely 
different  groups.  His  pattern-10  teleosts 
are  a  novel  assemblage,  composed  of  stro- 
mateoids ( centrolophids,  nomeids,  stroma- 
teids),  pomatomids,  kyphosids,  scorpidids 
(excluding  Scorpis),  arripidids,  girellids, 
theraponids,  and  kuhliids.  The  nematistiids 
have  a  reduced  pattern  10. 


The  pattern  of  a  nerve  complex,  because 
of  its  basically  conservative  nature,  should 
be  a  strong  taxonomic  character  in  dealing 
with  higher  categories.  The  common  pat- 
tern of  the  ramus  lateralis  aecessoriiis  in  the 
above  families  is  probably  good  cause  for 
considering  them  a  phylogenetic  unit.  But 
is  there  additional  evidence  for  lumping 
these  families  together?  The  stromateoids, 
almost  without  exception,  have  a  bony 
bridge  across  the  inside  of  the  anterior  ver- 
tical canal  of  the  ear.  Because  of  its  wide- 
spread occurrence  in  the  diverse  forms  of 
the  suborder,  I  consider  this  a  conservative 
character,  useful  at  higher  categories.  This 
bridge  is  also  present,  at  least  in  some  spe- 
cies and  at  some  stage  of  growth,  in  pat- 
tern-10 families.  The  bridge  is  absent,  how- 
ever, in  other  perciform  families  examined 
(see  Table  1  and  Material  examined,  p.  43). 

Many  pattern-10  families  have  character- 
istics common  to  basal  perciforms.  Many 
have  25  vertebrae,  seven  branchiostegal 
rays,  a  suborbital  shelf,  and  a  caudal  skele- 
ton near  to  the  basal  perciform  type  with 
six  hypurals  and  three  epurals.  In  most, 
the  shape  of  the  body  is  of  the  most  gen- 
eralized type.  The  general  impression  is  of 
an  older  group  of  fishes  which  have  man- 
aged to  remain  successful  without  di- 
verging too  far  from  the  basal  stock. 

The  common  ramus  lateralis  accessorius 
pattern  and  the  bridge  over  the  anterior 
vertical  canal  are  strong  evidence  for 
considering  that  the  relatively  specialized 
stromateoids  arose  from  somewhere  in 
this  relatively  undifferentiated  assemblage. 
Nonetheless,  it  is  unlikely  that  the  stromat- 
eoids are  the  descendents  of  any  living  pat- 
tern-10  family,  all  of  which  are  specialized 
in  some  respect.  With  the  present  imper- 
fect knowledge  of  the  comparative  osteology 
of  these  groups,  the  best  that  can  be  said 
is  that  all  share  a  common  heritage. 

The  fin  spines  of  stromateoids  are  not 
remarkably  developed,  and  the  teeth  are 
uniserial  in  the  jaws.  Pattern-10  families 
with  moderate-to-weak  fin  spines  and  a 
major  row  of  uniserial  teeth    ( some   have 


50        Bulleiin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Table  1.  Selected  characters  of  some  teleostean  families.    +  denotes  presence;  —  absence. 


RLA  PatternI 

AVC   Bridge 

Vertebrae- 

Sub- 
orbital 

Shelf" 

Branchi- 

ostegal 

Rays< 

Centrolophidae 

10 

+ 

25-60 

+ 

7 

Stroniateidae 

Nomeidae 

Ariommidae 

10 

10 

* 

+ 
+ 
+ 

30-48 
30-42 
30-33 

* 

* 

5-6 
6 
6 

Tetragonuridae 

* 

+ 

43-58 

* 

5-6 

Poniatoniidae 

10 

+ 

26 

+ 

7 

Kyphosidae 
Arripididae 
Scorpididae 
Girellidae 

10 

10 
10 
10 

+ 
+ 
+ 

25 
25 
25 

27 

+ 
+ 
+ 

7 
7 
7 
7 

Theraponidae 

Kuhliidae 

Nematistiidae 

10 

10 

reduced    10 

+ 

25 

25 

* 

+ 
* 

6 

6 

* 

Apolectidae 
Carangidae 
Scombridae 

9 

9 

none 

+ 

24 

24-27 
31-66 

* 

7 
7 
7 

Coryphaenidae 

* 

— 

30-33 

— 

7 

Atherinidae 

6 

— 

31-60 

± 

5-6 

Monodactylidae 
Ephippidae 

reduced  9 
9 

'~~ 

24 
24 

+ 
+ 

6 
6 

*  not  examined 

1  From   Freihofer   1963 

-  Largely  from  Regan   1913,  and  Berlin   and   Arambourg   1958 

"  From   Smith   and   Bailey    1962 

^  Largely  from  Giinther   1859,    1860,    1861 


very  reduced  rows  behind  the  major  row) 
are  the  pomatomids,  kyphosids,  and  scorpi- 
dids.  These  could  be  the  groups  most 
closely  associated  with  the  line  leading  to 
the  stromateoids.  The  arripidids,  girellids, 
theraponids,  and  kuhliids  have  teeth  in 
bands  and  heavy  fin  spines. 

There  is  a  strong  resemblance  between 
the  kyphosids  and  the  centrolophids,  the 
primitive  stromateoids.  In  both,  the  general 
rule  is  10  +  15  vertebrae,  a  caudal  skeleton 
with  six  hypural  and  three  epural  elements, 
a  perforate  ceratohyal,  an  expanded  lacri- 
mal, and  scaled  fin  bases.  Behind  the  major 
row  of  uniserial  teeth  in  KypJio.'iUs,  there 
are  rows  of  small  villiform  teeth  ( Moore, 
1962);  there  is  only  one  row  of  teeth  in 
centrolophids.      The     kyphosids,    however. 


lack  the  subocular  shelf  found  in  some 
centrolophids. 

The  kyphosids  are  today  one  of  the  most 
primitive  perciform  families.  They  retain 
the  teeth  on  the  ectopterygoid  and  endo- 
pterygoid  and  the  two  foramina  in  the  fa- 
cial wall  of  the  ])ars  jufi,ularis,  conditions 
associated  with  the  bcrvciform  level.  Pat- 
terson  (1964)  offers  evidence  for  the  deriva- 
tion of  the  kyphosids,  scoq^idids,  and  mono- 
dactylids  from  the  Cretaceous  polymixioid 
Benjcopsis.  If  this  is  the  case,  and  if  the 
stromateoids  arose  from  near  the  kyphosid 
stem,  a  direct  line  can  be  drawn  from  the 
berycoid  level  to  Pampiis,  one  of  the  most 
advanced  perciforms. 

The  carangids  were  probably  derived 
from  the  Cretaceous  dinopterygoid  Aipich- 
thys  (Patterson,  1964).  Their  development, 


Stromateoid  Fishes  •  Haedrich 


51 


Figure  6.      Parastromateus   niger,   drawing   of  a   15-inch   specimen,  from   Day,   1875. 


from  ber\'ciform  to  perciform  level,  has 
been  independent  of  the  line  which  pro- 
duced the  stromateoids.  The  resemblances 
between  certain  members  of  these  two 
groups  must  therefore  be  considered  an 
evolutionary  convergence. 

A  problem  remains  in  the  genus  Para- 
stromateus  {=  Apolectus)  (Fig.  6).  Bloch 
( 1795 )  described  the  sole  representative  of 
the  genus  as  a  species  of  Stromateus.  The 
shape  of  the  body  and  the  small  pelvics 
which  are  lost  with  age  were  good  cause 
for  this  action.  But  Parasfromatcus  lacks 
pharyngeal  sacs  and,  because  of  this,  Regan 
( 1902 )  removed  the  genus  from  the  stro- 
mateoids and  placed  it  in  the  carangids. 
More  extensive  comparisons  by  Apsangikar 
( 1953 )  supported  the  separation,  but  diver- 
gence from  the  carangids  was  noted  and 
the  new  subfamily  Parastromateinae  of  the 
Carangidae  proposed.  Suzuki  ( 1962 ) ,  in 
his  great  review  of  Japanese  carangids,  fol- 
lowed Jordan  (1923)  in  considering  Paro- 
stromateus  the  sole  representative  of  a 
monotypic  family. 


Parastromateus  is  a  pattem-9  teleost,  as 
are  the  carangids  (Freihofer,  1963),  but 
has  a  bony  bridge  over  the  anterior  vertical 
canal  of  the  ear,  as  do  the  stromateoids. 
The  only  reason,  other  than  the  bony 
bridge,  for  relating  Parastromateus  to  the 
stromateids  is  the  similarity  in  body  form. 
Parastromateus,  however,  has  10  +  14  ver- 
tebrae, while  all  stromateids  have  at  least 
13  +  17.  Even  if  Parastromateus  is  related 
to  the  stromateids,  the  relationship  is  at 
most  a  very  distant  one. 

Within  the  suborder  Stromateoidei,  there 
are  three  distinct  groups,  the  primitive 
centrolophids,  the  intermediate  nomeids 
with  their  specialized  off-shoots  the  tetra- 
gonurids  and  the  ariommids,  and  the  ad- 
vanced stromateids.  The  stromateids  are  an 
obvious  derivative  of  the  centrolophid  line. 
The  nomeids,  an  evolutionary  grade  above 
the  centrolophids,  have  evolved  parallel  to 
the  centrolophids,  but  from  an  earlier  com- 
mon ancestor.  The  probable  relationships 
between  the  stromateoid  families,  discussed 


52        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


TETRAGONURIDAE 


STROMATEIDAE 


Figure  7.      Dendrogram     showing     probable     relationships    of 
the   five    stromateoid   families. 


more  fully  in  the  family  accounts,  are  ex- 
pressed diagrammatically  in  Figure  7. 

Fossils.  The  Cretaceous  Omo.soma,  usu- 
ally considered  a  stromateid  (Arambourg, 
1954 )  has  been  shown  by  Patterson  ( 1964 ) 
to  be  a  polymixioid,  standing,  interestingly 
enough,  near  Bcnjcopsis.  Camnp,odcs  ceph- 
alus,  from  the  Monte  Bolca  beds  of  Austria, 
is  well  described  and  figured  by  Heckel 
( 1856 ) .  Though  it  looks  somewhat  like 
some  stromateoids  and  was  considered  a 
nomeid  by  Jordan  (1923),  the  diagnostic 
characters  are  missing  and  it  cannot  be  af- 
filiated with  this  group  with  any  certainty. 
AspidoJepis  Geinitz  1868,  based  on  a  scale, 
was  considered  a  stromateid  by  Jordan 
(1923).  But  the  scales  of  the  majority  of 
stromateoids  are  in  no  way  distinctive,  and 
thus  the  possible  relationships  of  this  fossil 
genus  cannot  be  determined.  Two  new 
fossil  genera  have  been  found  by  Bonde 
(1966)  in  the  lower  Eocene  Mo-clay  of 
Denmark. 

Key  tu   Stronuiteoid  Families 

1  (6j.  Two  dorsal  fins,  distinctly,  though 
scarcely,  separated,  the  first  usually 
with  ten  to  twenty  spines;  if  there  are 
fewer  than  ten  spines,  the  longest 
spine  is  about  the  same  length  as  the 
longest  dorsal  finray.  Pelvic  fins  al- 
ways present.  Vomer,  palatines,  and 
basibranchials  toothed  or  not.  2 


2  (5).    The    first    dorsal    fin    with    about    ten 

long,  slender  spines,  often  folded  into 
a  groove,  the  longest  spine  nearly  as 
long  as,  or  longer  than,  the  longest 
finray  in  the  second  dorsal.  Anal  fin- 
rays  14  to  30.  Scales  cycloid,  thin, 
deciduous.  Fleshy  lateral  keels  on 
peduncle  near  caudal  fin  base  absent 
or  only  slightly  de\eloped.  Vertebrae 
29  to  42 --     3 

3  (4).    Vomer,    palatines,    and    usually    basi- 

branchials with  small,  often  almost 
indistinguishable,  teeth.  Caudal  pe- 
duncle compressed,  its  least  depth 
greater  than  5  per  cent  of  the  stan- 
dard length,  without  lateral  keels. 
Usually  more  than  fifteen  rays  in  both 

the  dorsal  and  anal  fins.  

NOMEIDAE,  p.  76 

4  (3).    Vomer,   palatines,   and   basibranchials 

toothless.  Caudal  peduncle  square  in 
cross-section,  its  least  depth  less  than 
5  per  cent  of  the  standard  length, 
with  two  low  lateral  keels  on  each 
side  near  caudal  fin  base.  Fourteen  or 
fifteen  rays  in  both  the  dorsal  and 
anal  fins.  ARIOMMIDAE,  p.  88 

.5  (  2  ) .  The  first  dorsal  fin  with  ten  to  twenty 
short  spines,  the  longest  only  half  the 
length  of  the  longest  finray  in  the 
second  dorsal.  Anal  finrays  10  to  16. 
Scales  keeled,  heavy,  very  adherent. 
Modified  scales  form  two  well  de- 
veloped lateral  keels  on  each  side  of 
peduncle  near  caudal  fin  base.  Ver- 
tebrae 43  to  58.  

TETRAGONURIDAE,  p.  94 

6(1).  A  continuous  dorsal  fin,  or  two  dor- 
sal fins  scarcely  separated,  the  first 
with  less  than  ten  spines;  if  spines  are 
present,  the  longest  spine  is  less  than 
half  the  length  of  the  longest  dorsal 
finray.  Pelvic  fins  present  or  absent. 
Vomer,  palatines,  and  basibranchials 
toothless.    7 

7  (8).    Pelvic   fins   always  present.     None  or 

one  to  five  weak  spines,  or  five  to 
nine  stout  spines  precede  dorsal  fin- 
rays.  Anal  finrays  15  to  30.  Median 
fins  never  falcate;  their  bases  rarely 
the  same  length.  Jaw  teeth  all  conical, 
simple.  Supramaxillary  bone  usually 
present,  but  hard  to  find  in  some. 
Seven  branchiostegal  rays.    Vertebrae 

25  to  30  or  50  to  60.  

CENTROLOPHIDAE,  p.  53 

8  (7).    Pelvic    fins    never    present    in    adults, 

rarely  present  in  the  young.  No  stout 
spines  precede  dorsal  finrays,  but,  in 
.some  species,  five  to  ten  small  blade- 


Stromateoid  Fishes  •   Haedrich 


53 


like  spines  resembling  the  ends  of  free 
interneurals  protrude  ahead  of  the  fin. 
Anal  finrays  30  to  50.  Median  fins 
often  falcate;  their  bases  about  equal 
in  length.  Jaw  teeth  laterally  com- 
pressed, with  three  cusps.  No  supra- 
maxillary  bone.  Five  or  si.x  branchi- 
ostegal  rays.  Vertebrae  30  to  48.  __.. 
STROMATEIDAE,  p. 


98 


Family  CENTROLOPHIDAE 

Type  genus:    Cenfrolophus  Lacepede  1803 

Centrolophes.    Cuvier  and  Valenciennes,  1833:330 

( descr. ) . 
Centrolophinae.    Gill,    1861:34     (list);     1862:127 
(genera   listed);    1884:666-667    (def.,    gen.). 
Jordan  and  Gilbert,  1882:450  (name,  descr.). 
Centrolophidae.     Jordan     and     Evermann,     1896: 
962   (descr.;   North  America).    Jordan,   1923: 
182  (in  part,  list).   Nomian,  1937:115  (descr., 
relationships;  Chile).    Tortonese,  1959:57  (in 
part,  revision;  Gulf  of  Genoa). 
Lirinae.    Biihler,   1930:62   (in  part,  morph.,  diges- 
tive system ) . 
Nomeidae.    Berg,  1940:323  (in  part,  dist.);  1955: 
248  (part,  dist.).    Norman,  1957:503  (in  part, 
def.,  genera  listed). 
Diagnosis.  Stromateoid  fishes  with  pelvic 
fins  present  in  adults,  continuous  dorsal  fin, 
toothless  palate,  seven  branchiostegal  rays, 
and  six  hypural  bones  in  the  tail.  The  papil- 
lae in  the  pharyngeal  sacs  with  irregularly 
shaped  bases,   arranged   in   ten   to   twentv' 
longitudinal  bands. 

Description.  Body  slender  to  deep,  usu- 
ally somewhat  compressed.  The  rayed  por- 
tion of  the  continuous  dorsal  fin  preceded 
by  six  to  eight  short  stout  spines  in  Hijpcro- 
ghjphe,  Seriolella,  and  Psenopsis;  none  or 
one  to  five  thin  weak  spines  in  Cenfro- 
lophus, Schedophihis,  and  Icichtht/s.  In  the 
latter  group  and  in  Psenopsis  the  spines 
graduating  to  the  rayed  portion  of  the  fin; 
in  the  others  not.  Three  anal  fin  spines, 
not  separated  from  the  rays.  Pelvic  fins 
usually  attached  to  the  abdomen  by  a  thin 
membrane  and  folding  into  a  broad  shallow 
groove.  Head  conspicuously  naked,  usually 
covered  with  small  pores.  Scales  cycloid, 
but  with  minute  cteni  in  SchedopJiiJiis 
medusophagus,  and  usually  deciduous. 
Tubed  scales  of  lateral  line  extending  onto 
peduncle.     Margin    of    preopercle    usually 


moderately  denticulate,  but  spinulose  in 
most  young  stages  and  in  SchedophiJus. 
Opercle  thin,  with  two  flat,  weak  spines; 
the  margin  denticulate.  Seven  branchios- 
tegal rays.  Mouth  large,  maxillary  extend- 
ing at  least  to  below  eye.  A  nearly  uniserial 
row  of  small  conical  teeth  in  the  jaws; 
vomer,  palatines,  and  basibranchials  tooth- 
less. Supramaxillary  bone  present  in  most 
but  absent  in  Psenopsis.  Adipose  tissue 
around  eye  usually  not  conspicuously  de- 
veloped. Vertebrae  25  or  26  in  most  spe- 
cies, except  50  to  60  in  Icichthys.  Caudal 
skeleton  with  six  hypurals  and  usually  three 
epurals,  except  two  in  Icichthys.  Pharyn- 
geal sacs  with  irregularly  shaped  papillae 
in  ten  to  twenty  longitudinal  bands.  Teeth 
seated  directly  on  top  of  the  bony  base. 
Adults  one  to  four  feet  in  length,  usu- 
ally dark-colored  and  without  conspicuous 
pattern. 

Distribution.  Centrolophids  are  pelagic, 
usually  on  the  high  seas  and  over  the  edge 
of  the  continental  shelves,  although  Psenop- 
sis and  SerioJeUa  occur  in  shallow  water 
near  the  coast.  Some  are  found  in  tropical 
waters,  but  the  majority  are  fishes  of  tem- 
perate seas.  The  soft-spined  centrolophids 
(Cenfrolophus,  Icichthys,  and  SchedopJjihis) 
tend  to  be  more  oceanic  than  the  hard- 
spined  centrolophids  (Hyperogh/phc,  Seri- 
olella, and  Psenopsis) .  To  some  extent,  the 
distributions  of  these  two  groups  comple- 
ment each  other  (Figs.  52,  53). 

The  distribution  of  the  centrolophids  is 
in  part  a  relict  distribution.  Cenfrolophus 
is  bipolar,  found  in  the  North  Atlantic, 
South  Africa,  and  Southern  Australia  and 
New  Zealand.  Icichthys,  very  similar  in 
appearance  to  Cenfrolophus,  is  bipolar  in 
the  Pacific.  In  the  waters  from  Australia  to 
the  coasts  of  Chile,  the  endemic  genus 
Seriolella  has  evolved. 

No  centrolophids  occur  across  the  broad 
tropical  Pacific  or  Indian  Oceans. 

Relationships.  The  Centrolophidae  stand 
at  the  base  of  the  line  leading  to  the  Stro- 
mateidae.  Of  all  stromateoids,  they  are  the 
least  differentiated  from  the  percifomi  an- 


54        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


cestor.  Most  have  25  vertebrae,  the  well- 
known  basic  perciform  number.  Most  have 
a  snpramaxillary  bone.  There  are  seven 
branehiostegal  rays,  and  the  caudal  skeleton 
(Figs.  10,  12,  23)  is  of  the  basic  perciform 
type  (Gosline,  1961a).  In  this  sense,  the 
centrolophids  can  be  considered  the  most 
primitive  stromateoids. 

The  caudal  skeleton  of  the  nomeids,  with 
hypurals  2  +  3  and  4  +  5  of  the  basic  six 
fused,  could  easily  have  been  derived  from 
the  centrolophids.  The  stellate  papillae, 
also,  and  loss  of  one  branehiostegal  repre- 
sent a  grade  above  the  centrolophid  condi- 
tion. But  teeth  are  present  on  the  vomer, 
palatines,  and  basibranchials  of  the  no- 
meids, in  general  a  primitive  condition 
(Liem,  1963),  and  are  absent  in  the  cen- 
trolophids. This  makes  it  unlikely  that  the 
former  group  has  been  derived  from  the 
latter.  Rather,  the  two  must  represent  ap- 
proximately parallel  lines,  derived  from  an 
earlier  form  which  had  palatal  dentition. 

There  is  a  fairly  close  and  probably  direct 
relationship  between  the  advanced  centro- 
lophid genera  SeholcUa  and  P.scnopsis  and 
the  stromateids  Stromoteus  and  Pcprilus. 
Fishes  in  both  these  groups  have  well- 
ossified  sclerotic  bones,  minute  body  pores, 
slender  tapering  branchiostegals,  and  ex- 
tremely deciduous  scales. 

Key  to  Centrolo))hid  Genera 

1  (6).    Spines   of   the   dorsal   fin   weakly   de- 

veloped and  all  graduating  to  the 
dorsal  rays.   2 

2  (5).    Weak   dentieulations   on   preopercular 

margin.  Origin  of  dorsal  fin  usually 
well  behind  insertion  of  pectoral  fins, 
])ut  over  pectoral  insertion  in  very 
small  specimens.  Body  elongate,  maxi- 
mum depth  usually  less  than  30  per 
cent  of  the  standard  length.  3 

3  (4).    Total  elements  in  anal   fin   23  to  27. 

Scales  small,  very  deciduous,  pre- 
opercle  and  cheek  naked.  Scales  in 
lateral  line  160  to  230.  Vertebrae 
25 Centrolophu.s,  p.  62.  Fig.  13 

4  (3).    Total  elements  in  anal  fin  27  to  31. 

Scales  moderate  in  size,  not  especially 
deciduous,  present  on  preopercle  and 
cheek.     Scales   in   lateral   line    100   to 


130.    Vertebrae  50  to  60.  

Icichfhtjs,  p.  65.  Fig.   15 

5  (2).    Nine  to  fifteen  small   spines  on   pre- 

opercular margin.  Origin  of  dorsal 
fin  usually  before  insertion  of  pec- 
toral fins,  but  over  pectoral  insertion 
in  very  large  specimens.  Body  deep, 
maximum  depth  usually  greater  than 
.35  per  cent  of  the  standard  length.  __._ 
_.  Schedophihi.s,  p.  58.  Fig.  11 

6  ( 1 ) .    Five    to     nine     stout     dorsal     spines, 

shorter  than  and  not  graduating 
(graduating  slightly  in  Psenopsis)  to 
the  dorsal  rays.  7 

7  (8).    Dorsal  finrays   19  to  25;  anal  finrays 

14  to  21.  Preopercular  margin  spi- 
nulose.  Scales  not  especially  decid- 
uous. Lateral  line  arched  anteriorly, 
straightening  out  over  the  anal  fin. 
Adipose  tissue  around  eye  not  well  de- 
veloped. Sclerotic  bones  not  well 
ossified;  golden  iris  appears  as  a  com- 
plete ring.  -    Hyperoghjphe,  p.  54.  Fig.  8 

8  (7).    Dorsal  finrays  25  to  40;  anal  finrays 

18  to  30.  Preopercular  margin  entire 
or  finely  denticulate.  Scales  very  de- 
ciduous. Lateral  line  follows  dorsal 
profile.  Adipose  tissue  around  eye 
well  developed.  Sclerotic  bones  usu- 
ally well  ossified;  golden  iris  appears 

divided  by  a  vertical  bar.  9 

9(10).  Insertion  of  pelvic  fins  behind  inser- 
tion of  pectorals.  Supramaxillary  bone 
present.  At  least  seven  more  dorsal 
finrays  than  anal  finrays.  Usually 
eight  dorsal  spines,  the  third,  fourth, 

and  fifth  the  longest.  

Seriolella,  p.  69.  Figs.  18,  19 

10  (9).  Insertion  of  pelvic  fins  before  or  just 
under  insertion  of  pectorals.  Supra- 
maxillary bone  absent.  Number  of 
dorsal  finrays  never  exceeds  number 
of  anal  finrays  by  more  than  five. 
Five  to  seven  dorsal  spines,  increasing 

in  length  posteriorly. 

Psenopsis,  p.  72.  Fig.  21 

Genus  HYPEROGLYPHE  Gunther,  1859 
Figure  8 

Palinurus  DeKay,  1842:118.  (Type  species:  Cory- 
phcunia  pcrciformis  Mitchill,  1818:244,  by 
monotypy.  New  York  Harlior.  Preoccupied 
by   Pulinurns   Fabricius,    1798,    Crustacea.) 

Hypero^hjphc  Ciinther,  1859  (June):337.  (Type 
species:  *Diagramnia  porosa  Richardson,  1845: 
26,  l)y  monotypy.  Coasts  of  Australia.  A  syn- 
onym of  Perca  antarctica  Carmichael,  1818: 
501.) 


Stromateoid  Fishes  •  Hacdrich        55 


Figure  8.      Hyperoglyphe  perciiorma,   drawing   of  an  approximately  200-mm  specimen,  courtesy  of  the  Smitfisonian   Institution. 


Palimirichthys  Bleeker,  1859  (November): 22.  (Sub- 
stitute name  for  Palinurus  DeKay,  and  there- 
fore taking  the  same  type  species,  Conjphacna 
perciformis  Mitchill,  1818:244.) 

Palimirichthys  Gill,  1860:20.  (Substitute  name, 
proposed  independently  from  Bleeker,  for 
Palinums  DeKay,  and  therefore  taking  the 
same  type  species,  Coryphacna  perciformis 
Mitchill,  1818:244.) 

PammcJas  (Uinther,  1860:485.  (Substitute  name 
for  Palinurus  DeKay,  and  therefore  taking  the 
same  type  species,  Coryphacna  perciformis 
Mitchill,  1818:244.) 

Eurumetopos  Morton,  1888:77.  (Type  species: 
Eurumetopos  johnstonii  Morton,  1888:77,  by 
monotypy.  Tasmania.  A  synonym  of  Perca 
antarctica  Camiichael,   1818:501.) 

Tolcdiu  Miranda-Ribeiro,  1915:5.  (Type  species: 
Toledia  macrophihalma  Miranda-Ribeiro, 
1915:5,  by  monotypy.     Macahe,   Brazil.) 

Ocycrius  Jordan  and  Hubbs,  1925:226.  (Type  spe- 
cies: Centrolophus  joponicus  Doderlein  in 
Steindachner  and  Doderlein,  1885:183,  by 
original  designation.    Tokyo,   Japan. ) 

The  combination  of  less  than  25  dorsal 
finrays,  about  eight  short  spines  not  in- 
creasing in  length  to  the  rays  in  the  dor- 
sal fin,  toothless  palate,  pelvic  insertion 
under  pectoral  fin  base,  supramaxillary 
bone  present,  and  lateral  line  arched  an- 
teriorly straightening  out  over  the  anal  fin, 
distinguishes  Hyperoglyphe  from  all  other 


stromateoid  genera.  The  name,  a  feminine 
noun,  is  from  the  Greek  vvkfj,  above,  + 
y\v<f>yi,  groove,  in  reference  to  the  deep 
longitudinal  groove  in  the  roof  of  the 
mouth. 

Description.  Body  moderately  deep, 
maximum  depth  around  30  to  35  per  cent 
of  the  standard  length;  musculature  firm. 
Caudal  peduncle  broad,  of  moderate  length. 
Dorsal  fin  originating  over  or  a  little  behind 
insertion  of  pectoral  fins,  continuous,  six  to 
eight  short  stout  spines  not  graduating  to 
the  longer  rays.  The  longest  spine  half  the 
length  of  the  longest  ray.  Anterionnost  fin- 
rays  the  longest,  those  that  follow  shorter, 
19  to  25  finrays  in  all.  Anus  at  mid-body, 
in  a  slit.  Anal  fin  originating  a  little  behind 
middle  of  body,  three  spines  precede  the 
15  to  20  rays.  Pectoral  fin  rounded  in  the 
young,  pointed  in  adult.  Pelvic  fins  insert- 
ing under  end  of  pectoral  fin  base,  attached 
to  abdomen  by  a  small  membrane  and  fold- 
ing into  a  shallow  groove.  Caudal  fin  broad, 
emarginate  to  moderately  forked  in  adult. 
Scales  cycloid,  moderate  in  size,  somewhat 
deciduous,  covering  bases  of  median  fins. 
Lateral  line  arched  anteriorly,  straightening 
out  over  middle  of  anal  fin  and  extending 


56 


Bulletin  Miisciini  of  Comparative  Zoology,  Vol.  135.  No.  2 


onto  peduncle.  Skin  moderately  thick;  ex- 
tensive subdermal  canal  system  communi- 
cating to  the  surface  through  small  pores. 
Head  around  33  per  cent  of  the  standard 
length,  broad.  Top  of  head  not  scaled, 
pores  prominent,  naked  skin  projecting 
backward  over  nape.  Eye  moderate  to 
large,  no  adipose  tissue.  Nostrils  located 
near  tip  of  obtuse  snout,  large,  the  anterior 
round,  the  posterior  a  slit.  Angle  of  gape 
extending  below  eye.  Premaxillary  not  pro- 
tractile. Lacrimal  bone  partially  covering 
anterior  portion  of  upper  jaw  when  mouth 
is  closed,  end  of  maxillary  remaining  ex- 
posed. Supramaxillary  present.  Jaw  teeth 
very  small,  pointed,  uniserial,  close-set; 
vomer,  palatines,  and  basibranchials  tooth- 
less. Opercle  and  preopercle  thin;  opercle 
with  two  weak  flat  spines,  scaled,  margin 
very  finely  denticulate  or  entire;  preopercle 
not  scaled,  striated,  margin  \\'ith  numerous 
very  small  spinules.  Angle  of  preopercle 
rounded,  bulging  backward  slightly.  Gill- 
rakers  heavy,  slightly  longer  than  the  fila- 
ments, finely  toothed  on  inner  edge,  spaced, 
about  16  on  lower  limb  of  first  arch.  Seven 
branchiostegal  rays,  five  on  the  ceratohyal, 
two  on  the  epihyal.  Scapula  visible.  Verte- 
brae 10  +  15  =  25.  Stomach  a  simple  sac; 
intestine  long.  Pyloric  caeca  numerous,  in 
a  mass  resembling  a  raspberry. 

Base  color  green-grey  or  blue-grey  to 
reddish  brown.  Back  dark,  sides  and  be- 
low lighter,  sometimes  silvery.  Head  dark, 
iris  a  golden  ring,  opercle  often  silvery. 
Median  fins  usually  darker  than  the  body. 
Color  pattern  irregularly  striped,  mottled, 
or  clear,  changeable  in  life.  Inside  of  mouth 
and  gill  cavity  light.  Peritoneum  light  with 
minute  dark  speckles. 

Natural  history.  Though  Ilypcro^hjphc 
occurs  throughout  the  world  and  is  fished 
commercially  in  Japan,  very  little  is  known 
of  its  habits.  The  young  commonly  occur 
under  flotsam,  but  usually  not  under  jelly- 
fish, in  surface  waters  near  the  edge  of 
the  continental  shelf.  The  larger  adults 
form  shoals  in  deep  water,  perhaps  fairly 
near  the  bottom. 


Figure  9.  Branchial  region  of  Hyperoglyphe  percilorma, 
drawing  of  a  cleared-and-stained  preparation  from  a  173-mm 
SL  specimen.     Elements  identified   in   Figure  2. 


Small  H.  ))crciforma  two  to  four  inches 
long  occur  off  the  New  England  coasts 
under  floating  objects  in  great  numbers 
during  the  summer.  By  fall,  these  fish  have 
doubled  or  even  tripled  their  size.  With 
the  approach  of  cold  weather  they  dis- 
appear. Only  recently  has  it  been  found 
that  adult  H.  pcrciforma  attain  three  feet 
in  length,  and  live  in  deep  water  off  the 
coast  of  west  Florida  (Schwartz,  1963).  This 
discovery  bears  out  an  earlier  suggestion 
by  Merriman  ( 1945 )  that  the  fish  observed 
off  New  England  were  the  young  of  a  much 
larger  bathypelagic  species.  In  Japan,  the 
large  adults  had  been  marketed  long  before 
the  young  were  first  discovered  (Abe,  1955). 

Bigelow  and  Schroeder  ( 1953 )  reported 
small  fishes  and  crustacean  remains  from 
stomachs  of  Hypcroiijiiplic  pcrcifornia.  The 
fish  may  also  feed  occasionally  on  barnacles 
(Cornish,  1874;  Holt  and  Byrne,  1903).  At 
times  //.  porosa  feeds  heavily  on  the  tuni- 
cate Pyrosoma  atlanticum  (Cowper,  1960). 

Rclation.sJiips.  Hyperoglyphe  is  the  cen- 
tral genus  of  the  Centrolophidae.  The  mem- 


Stromateoid  Fishes  •  Hacdrich        57 


EPURALS 


HYPURALS 


Figure   10.      Caudal   skeleton   of   Hyperoglyphe   perciforma,   drawing   of   a   cleared-and-stained    preparation    from    a    50-mm    SL 
specimen.     All   elements   identified  in   Figure  1. 


bers  of  this  genus  are  the  most  generalized 
fishes  in  the  entire  suborder,  and  are  prob- 
ably not  unlike  the  ancestral  form.  The 
relatively  low  number  of  median  finrays, 
the  stout  spines  in  the  median  fins,  the 
seven  blunt  branchiostegal  rays  (Fig.  9), 
the  25  vertebrae,  the  spiny  preopercle,  and 
the  large  size  attained,  are  all  basal  char- 
acters. The  caudal  skeleton  (Fig.  10)  is  of 
the  generalized  perciform  type. 

Hyperog,Iy])he  has  given  rise,  on  the  one 
hand,  to  the  more  oceanic  soft-spined  cen- 
trolophids,  through  Schedophilus  to  Cen- 
trolophus  and  Icichthys.   The  major  change 


has  been  the  softening  of  the  fin  spines  and 
of  the  tissues  in  general.  On  the  other  hand, 
Hyperoglyphe  has  given  rise  to  the  more 
coastal,  advanced,  hard-spined  genera  Scri- 
olella  and  Psenopsis.  The  change  in  this 
direction  has  been  one  of  slight  refinement 
in  the  branchial  region  and  a  tendency 
toward  fusion  of  elements  in  the  caudal 
skeleton. 

Species.  Hyperoglyphe  is  a  wide-ranging 
genus.  The  species  are  found  in  the  slope 
water  off  the  east  coasts  of  the  New  World, 
in  the  Gulf  of  Mexico,  near  St.  Helena  and 
Tristan  da  Cunha,  along  the  west  coast  of 


58        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Africa,  in  Australia-New  Zealand,  and  in 
Japan. 

There  is  little  problem  of  synonymy,  since 
the  species  are  fairh'  distinct  from  one  an- 
other. The  changes  that  occur  with  growth, 
however,  remain  a  stumbling  block.  The 
species  in  Hypew^h/phe  are: 

Hyperoglyphc  antarctica  (Carmichael, 
1818)  =  Pcrco  antarctica  Carmichael.  Tris- 
tan da  Cunha,  South  Africa,  southern  Aus- 
tralia, and  New  Zealand,  type  locality 
Tristan  da  Cunha.  D  VIII,  19-21.  A  III 
15-16.  P  18-20.  Gill-rakers  5+1  +  14. 
Vertebrae  10  +  15.  This  is  the  most  prim- 
itive species  in  Hypcro<jJyphc,  and  stands 
nearer  to  the  base  of  the  stromateoid  stem 
than  any  other  fish.  It  attains  a  very  large 
size.  McCulloch  (1914)  reports  a  specimen 
1072  mm  long.  The  low  median  finray 
counts,  large  mouth,  and  a  characteristic 
patch  of  scales  on  the  otherwise  naked 
occiput  distinguish  this  species  from  all 
others.  Synonyms  are:  '''Diaiiramma  porosa 
Richardson,  1845,  from  Australia;  Eiirume- 
topos  johmtonii  Morton,  1888,  from  Tas- 
mania; and  Scriolella  ampins  Griffin,  1928, 
from  Bay  of  Plenty,   New  Zealand. 

Uypcroiijyphc  hythitcs  (Ginsburg,  1954) 
=  *PaJinurichfhys  hythitcs  Ginsburg.  Gulf 
of  Mexico,  type  locality  off  Pensacola,  Flor- 
ida. D  VII-VIII,  22-25.  A  III  16-17.  P 
20-21.  Gill-rakers  6-7  +  1  +  15-16.  Verte- 
brae 10  +15.  Possibly  a  synonym  of  //. 
macrophthalma  (Miranda-Ribeiro,  1915). 
This  species  has  more  dorsal  finrays  and 
a  larger  eye  than  H.  pcrciforma  (Mitchill, 
1818). 

Hypcro'j^lyphc  pcrciforma  (Mitchill,  1818) 
=  Curyphaena  pcrciformis  Mitchill.  East 
coast  of  North  America,  Florida  to  Nova 
Scotia,  type  locality  New  York  Harbor.  D 
VII-VIII,  19-21.  A  III  15-17.  P  20-22. 
Gill-rakers  5-7  +  1  +  15-17.  Vertebrae  10  + 
15  (skel.).  This  fish  is  the  common  "barrel- 
fish"  of  the  offings  of  New  England.  Young 
specimens  have  followed  floating  logs 
across  the  Atlantic  to  the  British  Isles  ( Holt 
and  Byrne,  1903).  Pimeleptenis  cornu- 
hiensis  Cornish,  1874,  is  a  synonym  based 


on  a  specimen  which  floated  to  Cornwall 
in  a  box. 

Hypcro'^lyphc  japonica  (Doderlein  in 
Steindachner  and  Doderlein,  1885)  =  Ccn- 
troloplius  japonicus  Doderlein.  Seas  of  Ja- 
pan, type  locality  Tokyo.  D  VIII,  22-24. 
A  III  17-19.  P  21-23.  Gill-rakers  6-7  +  1 
+  15-16.  Vertebrae  10  +  15.  Tliis  fish  is 
the  "medai"  of  Japanese  fisheries  literature. 
It  is  the  subject  of  a  small,  deep,  hand-line 
fishery.  A  prol)able  synonym  is  *Linis 
paiicidens  Giinther,  1889,  based  on  three 
small  specimens  captured  by  the  CHAL- 
LENGER somewhere  between  New  Guinea 
and  Japan. 

Hyperoglyphc  mosclii  (Cunningham, 
1910)  =  Lcirus  mosclii  Cunningham.  St. 
Helena,  and  coasts  of  Angola  and  South 
Africa,  type  locality  St.  Helena.  D  VI,  23- 
25.  A  II i  19-20.  P  20-22.  Gill-rakers  about 
7  +  1  +  15.  The  type  is  described  as  having 
but  one  spine  and  25  rays  in  the  dorsal  fin. 
The  specimen  is  so  large  and  heavy  that  I 
was  unable  to  lift  it  and  its  container  off 
the  shelf  in  the  British  Museum,  and  hence 
did  not  get  to  examine  it  closely  during 
my  brief  visit  there.  At  such  a  large  size, 
the  first  five  spines  in  the  dorsal  may  be 
buried  in  the  skin,  where  Cunningham 
could  have  overlooked  them.  Probable 
synonyms  are  Palinurichthys  pringlci  Smith, 
1949,  and  PalinuriclitJu/s  matthewsi  Smith, 
1960,  both  from  South  Africa. 

Hyperoglyphc  macrophthalma  ( Miranda- 
Ribeiro,  1915)  =  Tolcdia  macrophthalma 
Miranda-Ribeiro.  Brazil,  type  locality  Ma- 
cahe.  D  VII,  26.  A  20  (from  Miranda- 
Ribeiro,  1915).  Known  from  a  single  speci- 
men 68  cm  long.  Possibly  a  synonym  of 
//.  mosclii  (Cunningham,  1910). 

Genus  SCHEDOPHILUS  Cocco,   1839 

Fi.uure  11 

Lcirus  Lowe,  1833:143.  (Type  species:  *  Lcirus 
hcnnettii  Lowe,  1833:143,  by  monotypy. 
Madeira,  Atlantic  Ocean.  Preoccupied  by 
Leirus  Dahl,  1823,  Coleoptera.  A  junior  syn- 
onym of  *Ccnirolo])}ius  ovalis  Cuvier  and 
Valenciennes,    1 833 :  346. ) 


Stromateoid  Fishes  •  Haedrich 


59 


■^-W^5?ir??:^, 


_ii_  ^^'•. 


*V.  ,.-••.-. ^V^Sig^  :  l?:';?  C-  -4; ;v^  .-i^iv  ■■;^-  i'-iv^-C^S^-.SySfeB- 


Figure    11.     Scfiedoph/lus  pemarco,  drawing  of  a  245-mm  specimen,  from  Poll,   1959. 


Schedophihts  Cocco,  1839^:57.  (Type  species: 
Schi'dophihis  medusophagus  Cocco,  1839:57, 
by  monotypy.    Messina. ) 

Mupiis  Cocco,  1840': 237.  (Type  species:  Mitpus 
imperialis  Cocco,  1840:237,  by  monotypy. 
Messina.  A  synonym  of  *Centrolophus  ovalis 
Cuvier  and  Valenciennes,   1833:346.) 

Lints  Agassiz,  1846:213.  (Emendation  of  Leirus 
Lowe,  1833:143,  and  therefore  taking  the 
same  type  species,  *Leinis  hennettii  Lowe, 
1833:143,  a  junior  synonym  of  *Ceniwlophus 
ovalis  Cuvier  and  Valenciennes,  1833:  346.) 

Crhis  Valenciennes,  1848:43.  (Type  species:  *Crius 
bertheloti  Valenciennes,  1848:45,  l>y  original 
designation.  Canary  Islands,  Atlantic  Ocean. 
A  synonym  of  *Centrolophus  ovalis  Cuvier 
and  Valenciennes,  1833:346.) 

Hoplocorijphis  Gill,  1862:127.  (Type  species: 
*SchedophiIus  maculatus  Giinther,  1860:412, 
by  original  designation.    Seas  of  China.) 

Eucrotiis  T.  H.  Bean,  1912:123.  (Type  species: 
Eucrotus  ventralis  T.  H.  Bean,  1912:123,  by 
monotypy.    Bemiuda,  Atlantic  Ocean. ) 

Ti/ii^j/a  Whitley,  1943:178.  (Type  species:  Tuhbia 
tasmanica  Whitley,  1943:179,  by  original 
designation.    Eastern  Tasmania.) 

The  combination  of  deep  body,  broad 
deep  head,  large  eye,  continuous  dorsal  fin 
with  weak   spines   graduating   to   the   rays 

^  Tortonese  (1959)  has  clarified  the  confusion 
surrounding  the  publication  dates  of  Cocco's  names. 


and  originating  before  the  pectoral  inser- 
tion, toothless  palate,  and  prominent  spines 
on  the  preopercular  margin  distinguishes 
Schedophihis  from  all  other  stromateoid 
genera.  The  name,  a  masculine  noun,  is 
from  the  Greek  axeSta,  raft,  +  c^iAos,  friend, 
in  reference  to  the  fish's  common  associa- 
tion with  floating  objects. 

Description.  Body  deep,  maximum  depth 
generally  greater  than  35  per  cent  of  the 
standard  length;  musculature  soft.  Pedun- 
cle fairly  broad,  short.  Dorsal  fin  originat- 
ing before  (or  over  in  very  large  specimens) 
insertion  of  pectoral  fins,  continuous,  three 
to  seven  weak  spines  graduating  to  the  23 
to  50  rays.  Anus  and  genital  pore  at  mid- 
body,  in  a  slit.  Anal  fin  originating  behind 
middle  of  body,  three  long  weak  spines  pre- 
ceding the  16  to  30  rays.  Median  fins  with 
compressed  fleshy  bases.  Pectoral  fin 
rounded  in  the  young,  pointed  in  adult, 
relative  length  decreasing  with  growth. 
Pelvic  fins  inserting  under  end  of  pectoral 
fin  base,  reaching  to  anus  in  young  and 
juveniles,  attached  to  abdomen  by  a  mem- 
brane and  folding  into  a  shallow  groove; 
relative  length  of  fin  decreasing  markedly 


60        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


with    growth.     Caudal    fin    broad,    forked,  take    small    erustaceans.     At    a    length    of 

Scales  small  to  moderate,  cycloid,  or  with  about  200  mm,   S.    meduso])Jia^us  deserts 

one  or  two  minute  cteni  in  young  Schedo-  its   coelenterate   companion,   and   descends 

/)/n7n.9  mc'dusophuiius,  deciduous,  covering  to  deeper  water. 

fleshy  bases  of  the  median  fins.  Lateral  line  Adult  Scliedophilus  appear  very  different 
arched  anteriorly,  straightening  out  about  from  the  younger  stages.  The  relative 
mid-body  and  extending  onto  peduncle,  length  of  the  paired  fins  is  greatly  de- 
Skin  thin;  extensive  subdermal  canal  system  creased,  the  body  is  much  more  elongate, 
communicating  to  the  surface  through  small  and  the  mottled  or  barred  pattern,  typical 
pores.    Head  soft,  broad  and  deep,  usually  of  juveniles,  is  gone. 

greater  than  25  per  cent  of  the  standard  Relatiun.ships.   ScJicdophiJus  provides  the 

length,   not   scaled,   naked   skin   projecting  link  between  the  soft-spined  and  the  hard- 

slightly  backward  over  the  nape.  Eye  large,  spined  centrolophids.    The  range  of  varia- 

no  adipose  tissue.    Nostrils  located  near  tip  tion  in  the  genus  is  great,  and  the  species 

of  obtuse  snout,  anterior  nostril  round,  the  grade  from  the  one  condition  to  the  other, 

posterior  a  slit.    Angle  of  gape  extending  The  caudal  skeleton  ( Fig.  12 )  is  most  like 

below    eye.    Premaxillary    not    protractile,  that  of  Centrolophus  and  Icichthijs.    The 

Lacrimal  bone  covering  anterior  portion  of  pharyngeal  sacs  and  teeth  are  intermediate 

upper  jaw  when  mouth  is  closed,  end  of  between   those   of   Centrolophus   and   Hy- 

maxillary  remaining  exposed.  Slender  supra-  peroii.lyphe.    Schedophilus  ovalis  has  fairly 

maxillary   present.    Jaw    teeth   very   small,  stout  spines  ahead  of  the  median  fins;  in 

pointed,    uniserial,    close-set;    vomer,   pala-  S.   mcdusophagus  the  spines  are  soft  and 

tines,  and  basibranchials  toothless.   Opercle  flexible. 

and  preopercle  thin;  opercle  with  two  weak  Schedopluhi.'i  is  derived  from  the  central 

flat  spines,  scaled,  margin  denticulate;  pre-  Jhjpero'^Jyphc  stock.   As  it  has  moved  into 

opercle  not  scaled,  margin  set  with  nine  to  a  more  oceanic  environment,  the  spines  on 

eighteen   prominent   spines,   angle   of   pre-  the    preopercle    have    become    more    pro- 

opercle  rounded,  bulging  back  slightly.  Gill-  nounced,  while  the  fin  spines  and  the  body 

rakers  heavy,  about  half  the  length  of  the  in  general  have  become  softer, 

filaments,  toothed  on  inner  edge,  spaced;  Species.   The  species  in  Schedophilus  are 

10  to  16  on  lower  limb  of  first  arch;  a  few  in  general  well  differentiated.    Almost  all 

rudimentary    rakers    present    under    large  descriptions  are  based  on  young  specimens, 

pseudobranch.    Seven  branchiostegal  rays.  The  large  adults  differ  greatly  in  appear- 

five  on  the  ceratohyal,  two  on  the  epihyal.  ance  from  the  young.    Adults  are  so  very 

Scapula  visible.    Vertebrae   10  +  15,   16  or  rarely  seen   that  only  three  have   entered 

20  =  25,  26  or  30,  or  12  +  17  =  29.   Stomach  the  literature,  two  of  them  assigned  to  other 

a  simple  sac;  intestine  long.    Pyloric  caeca  genera.  Unfortunately,  the  species  from  the 

numerous,  dendritic.  Australian  region  are  very  poorly  known. 

Base  color  brown,  bluish,  or  silvery.   Me-  Because  of  their  isolated   geographic  dis- 

dian    fins,    pectorals,    and    pelvics    usually  tribution,  critical  examination  of  these  spe- 

darker  than  the  body.    Color  pattern  irreg-  cies   will    doubtless   provide   much    insight 

ularly  striped,  mottled,  or  clear.    Young  of  into  the  evolution  of  the  soft-spined  centro- 

some  have  dark  vertical  stripes.  lophids. 

Natural  history.    Most  species  in  ScJwdo-  The  species  in  the  genus  are: 

/j/nVn.s' are  oceanic,  rare,  and,  consequently,  Schedophilus  ovalis  (Cuvier  and  Valen- 

little    is    known    concerning    their   biology,  ciennes,  1833)  =  ^Centrolophus  ovalis  Cu- 

The    young    of    S.    medusophw^us    occurs  vier    and    Valenciennes.     Eastern    Atlantic 

commonly    with    jellyfish.     The    fish    may  Ocean    from    Spain    to    South    Africa    and 

feed  very  largely  on  medusae,  but  will  also  Mediterranean  Sea,  type  locality  Nice.    D 


Strom ATEOiD  Fishes  •  HaedricJi        61 


EPURALS 


HYPURALS 


Figure   12.      Cauda!  skeleton  of  Schedophilus  medusophagus,  drawing  of  a  cleared-and-stained   preparation  from  a  39-mm  SL 
specimen.     All    elements   identified   in    Figure    1. 


VI-VIII,  30-32.  A  III  20-24.  P  21-22.  Gill- 
rakers  around  6+1  +  16.  Vertebrae  10  + 
15.  Silvery  to  greenish.  Synonyms  are: 
CentwJopluis  crassus  Cuvier  and  Valen- 
ciennes, 1833,  from  west  of  the  Azores; 
*Leirus  bennettii  Lowe,  1833,  from  Ma- 
deira; Mupiis  impcrkiUs  Cocco,  1840,  from 
the  Mediterranean;  *Crins-  bertheloti  Valen- 
ciennes, 1848,  from  the  Canary  Islands; 
Centrolophiis  rotund  icauda  Costa,  1866, 
from  Naples;  Centrolophus  porosissimus 
Canestrini,  1865,  and  Schedophilus  bottcri 
Steindachner,    1868,   from    Barcelona. 

Schedophihis  medusophagus  Cocco,  1839. 
Atlantic  Ocean  and  western  Mediterranean 


Sea,  type  locality  Messina.  D  44-50  (total 
elements).  A  28-31  (total  elements).  P 
18-21.  Gill-rakers  around  5  +  1  +  11.  Ver- 
tebrae 10  +  15.  Major  preopercular  spines 
usually  about  12.  Lateral  line  scales  160- 
230,  increasing  in  number  with  growth. 
Brown,  often  mottled.  The  report  of  this 
species  from  the  South  Pacific  (Giinther, 
1876),  is  undoubtedly  that  of  a  closely 
related  form,  Schedophilus  huttoni  (Waite, 
1910).  The  adult  of  S.  medusophagus  has 
long  been  known  under  the  name  ''Cen- 
trolophus britannicus  Giinther,  1860a. 

"^Schedophilus  maculatus  Giinther,  1860. 
China  Seas.    D  36  (total  elements).    A  27 


62        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


(total  elements).  P  19.  Gill-rakers  5  +  1 
+  13.  Vertebrae  10  +  15.  Major  preopercu- 
lar  spines  13.  This  speeies  is  known  only 
from  the  t\pe,  a  37-mm  SL  specimen  that  is 
soft  and  in  poor  condition. 

*SchedophiIus  marmomtiis  Kner  and 
Steindachner,  1866.  "Siidsee,"  presumably 
near  Australia.  D  38  (total  elements).  A 
27  (total  elements).  Vertebrae  12+17. 
This  species  is  usually  treated  as  a  synonym 
of  S.  maculatus  Giinther,  1860.  A  probable 
synonym  is  HopJocoryphis  pJii/.sdUanini 
Whitley,   1933,  from  New  South  Wales. 

Schcdophilii.s-  huttoni  (Waite,  1910)  = 
Centrolopliu.s  huttoni  Waite.  Seas  of  New 
Zealand,  eastern  Australia,  and  Tasmania, 
type  locality  Sumner,  New  Zealand.  D  57 
(total  elements).  A  38  (total  elements). 
Gill-rakers  5  +  12.  Vertebrae  10  +  20.  Lat- 
eral line  scales  in  the  776-mm  holotype  near 
240.  Brownish.  As  in  S.  mcdu.sophuiius, 
the  number  of  lateral  line  scales  probably 
increases  with  age.  A  probable  synonym 
is  Tuhhia  tosmanica  Whitley,  1943,  from 
Tasmania,  known  only  from  a  10-cm  speci- 
men reported  as  having  144  scales  in  the 
lateral  line. 

Schedophilu.s  ventmlis  (Bean,  1912)  = 
Eucrotus  vcntmlis  Bean.  Bermuda.  D  IV- 
VII,  31-34.  A  III  20-23.  P  22.  Gill-rakers 
around  5  +  1  +  16.  Vertebrae  10  +  15.  Ma- 
jor preopercular  spines  about  9.  The  type 
is  apparently  lost.  This  nominal  species  has 
been  synonymized  with  S.  ovalis  (Cuvier 
and  Valenciennes,  18.33)  by  Fowler  (1936). 

Schedo))hiJn.s  iiriseoUneatus  ( Norman, 
1937)  =  ^'PalinuriclitJuj.s  griseolineatus  Nor- 
man. Southern  Atlantic  Ocean,  type  local- 
ity 49'^00'S  6r58'W.  D  VII-VIII,  31-33.  A 
III  20-21.  P  19-21.  Gill-rakers  around  6  +  1 
+  14.  Vertebra(>  10  +  16.  Lateral  line 
scales  about  120.  Major  preopercular 
spines  around  14.  Blue-brown,  horizontally 
striped.  This  species  can  be  distinguished 
at  once  by  the  increased  number  of  caudal 
vertebrae.  The  large  specimens  which  Nor- 
man ( 1937)  doubtfully  referred  to  "Palintir- 
icJithys  caendeiis"  belong  to  this  species. 


Schcdopliilu.s  pcmarco  (Poll,  1959)  = 
Palinurichthijs  pemarco  Poll.  Gulf  of 
Guinea,  tropical  Atlantic  Ocean.  D  V-VII, 
23-26.  A  III  16-18.  P  19-22.  Gill-rakers 
around  5  +  1  +  16.  Vertebrae  10  +  15.  Lat- 
eral line  scales  about  95.  Major  preopercu- 
lar spines  15-19.  Blue-brown,  horizontally 
striped.  The  median  finray  counts  in  this 
species  are  lower  than  in  any  other. 

Genus  CENTROLOPHUS  Lacepede,  1803 
Figure  13 

Ccntrolophii.^  Lacepede,  1803:441.  (Type  species: 
Perca  ni^ro  Cinelin,  1788:132,  by  monotypy. 
"Rivers  of  Cornwall.') 

Acentrolophus  Nardo,  1827:28.  (Substitute  name 
for  CentroJophiis  Lacepede,  1803,  and  there- 
fore taking  the  same  type  species,  Perca  nigra 
Gmelin,  1788:132.  Centrolophiis  deemed  in- 
apphcable. ) 

Gijtnnucephahi.s  (non  Bloch,  1793:24)  Cocco, 
1838:26.  (Type  species:  Gymnocephalus 
mcs.sinen.sis  Cocco,  1838:26,  by  monotypy. 
Messina.  A  synonym  of  Perca  nigra  Cmelin, 
1788:132.)     (Vi^e  Jordan,  1923.) 

Pompihis  Lowe,  1839:81.  (Type  species:  *Ccntro- 
lophu.'i  morio  Cuvier  and  Valenciennes,  1833: 
342,  by  absolute  tautonymy,  C.  pompilus  [  = 
P.  pompihi.s]  Cuvier  and  Valenciennes,  1833: 
334,  considered  a  synonym.  Madeira.  A 
synonym  of  Perca  nigra  Cmelin,  1788:132. 
Preoccupied  in  Pompilus  Schneider,  1784, 
Cephalopoda. ) 

Centrolophodes  Gilchrist  and  von  Bonde,  1923:2. 
(Type  species:  Centrolojyhode.s  irwini  Gil- 
christ and  von  Bonde,  1923:3,  by  monotypy. 
South  Africa.  A  synonym  of  Perca  nigra 
Gmelin,    1788:132.) 

The  combination  of  elongate  body,  small 
head  with  prominent  pores,  continuous  dor- 
sal fin  with  very  weak  spines  graduating  to 
the  rays,  toothless  palate,  very  small  scales, 
and  160  to  230  scales  in  the  lateral  line, 
distinguishes  Centrolojdni.s  from  all  other 
stromateoid  genera.  The  name,  a  masculine 
noun,  is  from  the  Greek  Kei'Tfjov,  spine, 
+  Aoc/)os%  crest  of  a  helmet,  probably  in 
reference  to  the  manner  in  which  the  dor- 
sal fin  rises  from  the  back. 

De.^cription.  Body  elongate,  maximum 
depth  rarely  exceeding  30  per  cent  of  the 
standard  length  except  in  very  small  sp(X'i- 
m(>ns;   musculature  firm.    Peduncle   broad, 


Stromateoid  Fishes  •  Hacdrich 


63 


y^    " 


'"•••I'  ^z,-^?,,.  .^«'55r^ 


--%*** 


,fKluu£a> 


"^ 


#^>^!ai!ii,a6toaBa«.ss».ia«»-"*' 


Figure    13.      Cenfro/ophus   n/ger,    drawing   of   a   223-mm    specimen,   USNM  44440,  courtesy  of  tfie  Smithsonian   Institution. 


thick,  long.  Dorsal  fin  originating  a  little 
behind  insertion  of  pectoral  fins,  contin- 
uons,  about  fi\'e  very  weak  spines  graduat- 
ing to  the  32  to  37  rays.  Anus  and  genital 
pore  at  mid-body,  in  a  slit.  Anal  fin  orig- 
inating a  little  behind  middle  of  body, 
three  weak  spines  precede  the  20  to  23  rays. 
Pectoral  fin  rounded  in  the  young,  pointed 
in  adult,  relative  length  decreasing  slightly 
with  growth.  Pelvic  fins  inserting  under 
posterior  portion  of  pectoral  fin  base,  at- 
tached to  the  abdomen  by  a  small  mem- 
brane and  folding  into  a  shallow  groove. 
Caudal  fin  broad,  moderately  forked.  Very 
small  cycloid  scales,  deciduous,  covering 
fleshy  bases  of  the  median  fins.  Lateral 
line  slightly  arched  anteriorly,  straightening 
out  about  mid-body  and  extending  onto  pe- 
duncle; lateral  line  scales  around  190.  Skin 
fairly  thick;  extensive  subdermal  canal  sys- 
tem communicating  to  the  surface  through 
small  pores.  Head  usually  less  than  25 
per  cent  of  the  standard  length,  not  scaled, 
pores  very  prominent,  naked  skin  not  pro- 
jecting backward  over  the  nape.  Eye  of 
moderate  size,  no  adipose  tissue.  Nostrils 
near  tip  of  rounded  snout,  the  anterior 
round,  the  posterior  a  slit.  Angle  of  gape 
extending  below  eye.  Premaxillary  not  pro- 
tractile. Upper  jaw  covered  completely  by 
lacrimal  bone  when  mouth  is  closed.  Slen- 
der supramaxillary  present.  Jaw  teeth  small, 
pointed,    uniserial,    spaced,    increasing    in 


number  with  growth;  vomer,  palatines,  and 
basibranchials  toothless.  Opercle  and  pre- 
opercle  thin,  margins  finely  denticulate; 
opercle  with  two  weak  flat  spines,  scaled; 
angle  of  preopercle  rounded,  bulging  back 
slightly;  preopercle  and  cheek  not  scaled. 
Gill-rakers  heavy,  about  half  the  length  of 
the  filaments,  toothed  on  inner  edge, 
spaced,  about  13  on  lower  limb  of  first 
arch;  rudimentary  rakers  present  under 
large  pseudobranch.  Seven  branchiostegal 
rays,  five  on  the  ceratohyal,  two  on  the 
epihyal.  Scapula  prominent.  Vertebrae  10 
+  15  =  25.  Caudal  skeleton  with  six  hy- 
purals  and  three  epurals.  Stomach  a  simple 
sac;  intestine  long.  Pyloric  caeca  about  10, 
digitiform. 

Base  color  brown.  Range  is  from  russet 
through  chocolate  to  dark  bluish.  Median 
fins  and  peKics  darker  than  the  body.  No 
pattern,  hardK*  any  countershading  in 
adults;  young  have  three  or  four  dark  ver- 
tical stripes. 

Natural  history.  Young  Centrolophus 
have  been  taken  under  jellyfish  (Collett, 
1896)  and  swimming  with  Mola  (Munro, 
1958).  Some  described  as  "small"  were 
found  in  the  stomachs  of  bottom-living  hake 
trawled  west  of  the  British  Isles  (Blacker, 
1962 ) .  Presumably  these  Centrolophiis  had 
not  been  in  association  with  pelagic  medu- 
sae. While  young  fish  are  found  near  the 
surface,  the  large  fish  are  taken  at  depth. 


64 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure   14.      Branchial    region   of   Cenfro/ophus   n/ger,    drawing   o'  a  cleared-and-stained  preparation  from  a  190-mm  specimen. 
Elements  identified  in  Figure  2, 


In  the  North  Atlantic,  the  adults  seem 
widespread,  but  the  young  have  been 
found  only  in  the  eastern  Atlantic  and 
Mediterranean  areas.  A  spawned-out  fe- 
male, however,  has  been  caught  south  of 
New  England  (Templeman  and  Haedrich, 
1966). 

The  young  are  at  first  vertically  banded, 
but  by  the  time  they  are  about  100  milli- 
meters long  they  have  become  a  uniform 
brown.  Growth  is  very  rapid;  from  De- 
cember to  May,  five  months,  a  Mediterra- 
nean specimen  grew  from  20  to  170  milli- 
meters (Padoa,  1956).  Growth  is  regular 
and  the  allometry  is  not  marked.  The  num- 
ber of  jaw  teeth  does  increase,  however, 
from  about  17  in  a  150-millimeter  specimen 
to  near  100  in  one  of  1,200  millimeters. 

Centrolophus  is  one  of  the  largest  stro- 
mateoids    known.    Specimens    a    meter    or 


more  in  length  have  been  taken  in  Australia 
(Mees,  1962),  South  Africa  (Barnard,  1948), 
and  the  western  North  Atlantic. 

Autumn  spawning,  from  October  into 
winter,  is  indicated  by  the  occurrence  of 
eggs  and  very  small  fish  at  this  time  in  the 
Mediterranean  (Padoa,  1956)  and  by  the 
capture  of  a  large,  recently  spawned-out 
female  in  December  1963  in  the  western 
North  Atlantic.  Fraser-Brunner  (1935)  noted 
dimorphism  in  the  coloring  of  the  sexes,  the 
females  said  to  be  lighter  than  the  males. 
This  difference,  however,  is  not  always  ob- 
served. 

Lo  Bianco  (1909)  observed  young  Centro- 
lophus feeding  on  medusae,  but  Chabanaud 
and  Tregouboff  (1930)  found  that  their 
aquarium  specimen  preferred  small  fish  and 
plankton.  It  never  attempted  to  eat  the 
medusae  which  were  j^resent  in  the  tank. 


Stromateoid  Fishes  •  Hacdrich        65 


The  large  specimen  from  south  of  New 
England  was  taken  on  a  long-line  baited 
with  squid.  Fish  and  large  crustacean  re- 
mains occurred  most  often  in  stomachs  ex- 
amined, and,  on  one  occasion,  bits  of  po- 
tato and  an  onion  were  found. 

As  Nielsen  ( 1963 )  has  suggested,  in  re- 
porting the  seining  of  five  near  Skagen, 
Centrolophus  may  school.  Blacker  (1962) 
reports  several  hundred\\'eight  trawled  off 
Ireland.  Potentially  a  good  fish  with  fine 
white  meat,  those  offered  experimentally 
in  Milford  Market  found  no  sale  (Blacker, 
1962). 

Relationships.  Centrolophus  is  one  of  the 
most  primitive  stromateoids.  The  small 
pharyngeal  sac  with  few  rows  of  large  pa- 
pillae (Fig.  14),  the  heavy  blunt-ended 
branchiostegal  rays,  and  the  large  size  at- 
tained, are  all  primitive  characters.  Centro- 
troIopJius  shows  much  affinity  of  fonn 
towards  Icichthijs,  from  which  it  differs 
mainly  in  having  far  fewer  vertebrae. 

Centrolophus,  Icichthijs,  and  Schedopliilus 
are  the  soft-spined  centrolophids.  This 
group  is  in  general  a  little  more  primitive 
than  the  hard-spined  centrolophids,  Hijpero- 
ghjphe,  Seriolella,  and  Psenopsis.  The  soft- 
spined  centrolophids  usually  have  smaller 
sacs  with  fewer  papillae,  coarser  jaw  teeth, 
and  attain  a  larger  size  than  the  hard-spined 
centrolophids. 

Species.  Centrolophus  is  known  from  the 
Australian  region,  from  South  Africa,  and 
from  the  North  Atlantic,  where  numerous 
species  have  been  described.  The  counts  of 
the  Southern  Hemisphere  specimens,  of 
which  only  a  handful  are  known,  o\erlap  the 
range  of  those  for  the  North  Atlantic  spe- 
cies. Some  differences  may  exist  in  rela- 
tive proportions,  but  these  are  only  at  cer- 
tain stages  of  growth.  Lacking  comparative 
material,  the  safest  course  is  to  follow  Waite 
(1910)  and  Mees  (1962)  in  recognizing  but 
one  bipolar  species: 

^Centrolophus  niger  (Gmelin,  1788)  = 
Perca  nigra  Gmelin,  1788.  North  Atlantic, 
western  Mediterranean  Sea,  Adriatic  Sea, 
South  Africa,  southern  Australia,  and  New 


Zealand,  type  localitv  "Rivers  of  Cornwall." 
D  37-41  (total  elements).  A  III  20-23.  P 
19-22.  Gill-rakers  5-6  +  1  +  12-15,  usually 
19  total.  Vertebrae  10  +  15.  The  name 
Centrolophus  pompilus  (Linnaeus,  1758) 
is  often  used  for  this  fish.  Linnaeus's 
Conjphaena  pompilus,  however,  is  too 
poorly  characterized,  and  differs  too  much 
in  certain  respects  to  be  considered  the 
same  species.  Gmelin's  (1788)  Perca  nigra 
is  the  first  available  name.  Synonyms  from 
the  North  Atlantic  are:  Centrolophus  liparis 
Risso,  1826,  from  Nice;  Acentrolophus 
maculosus  Nardo,  1827,  from  the  Adriatic 
Sea;  ^Centrolophus  pompilus  Cuvier  and 
Valenciennes,  1833,  from  Marseille;  "^Cen- 
trolophus morio  Cuvier  and  Valenciennes, 
1833  (ascribed  to  Lacepede),  from  the 
Mediterranean  Sea;  "^Schedopliilus  elon- 
gatus  Johnson,  1862,  from  Madeira;  and 
"^Centrolophus  calenciennesi  Moreau,  1881, 
from  Marseille. 

The  two  species  described  from  the  South- 
ern Hemisphere,  here  considered  synonyms 
of  niger,  are:  Centrolophus  maoricus 
Ogilby,  1893,  Australia  and  New  Zealand. 
Counts  made  on  two  specimens  of  this 
nominal  species  fell  at  the  high  end  of  the 
range  for  C.  niger,  as  do  the  counts  for  one 
small  specimen  reported  by  Regan  ( 1914 ) . 

Centrolophus  incini  (Gilchrist  and  von 
Bonde,  1923)  =  Centrolophodes  incini  Gil- 
christ and  von  Bonde.  South  Africa.  The 
counts  reported  for  the  holotype  likewise 
fall  at  the  high  end  of  the  range  for  C. 
niger.  Mupus  bifasciatus  Smith,  1961,  based 
on  two  small  specimens,  is  almost  certainly 
the  same  fish.  There  is  little  cause  to  doubt 
that  the  South  African  and  Australian  forms 
belong  to  the  same  population. 

Genus  ICICHTHYS  Jordan  and  Gilbert,  1 880 
Figure   15 

Icichthijs  Jordan  and  Gilbert,  1880:305.  (Type 
species:  *Icichthys  lockingtoni  Jordan  and 
Gilbert,  1880:305,  by  original  designation. 
Point   Reyes,   California."! 

The  combination  of  elongate  soft  body, 
continuous  dorsal  fin  originating  well  be- 


66 


BiiUctin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure   15.      Icichthys  lockingtoni,  drawing  of  a  390-mm  specimen,  from    Parin,   1958. 


hind  pectoral  insertion,  toothless  palate, 
moderate  scales  covering  opercles  and 
cheek,  100  to  130  scales  in  the  lateral  line, 
and  50  to  60  vertebrae,  distinguishes 
Icichthys  from  all  other  stromateoid  genera. 
The  name,  a  masculine  noun,  is  from  the 
Greek  Hkoj,  to  yield,  +  Ixdv^i,  fish,  in  refer- 
ence to  the  fish's  flexible  soft  body. 

Description.  Body  elongate,  maximum 
depth  less  than  25  per  cent  of  the  standard 
length  except  in  small  specimens;  muscula- 
ture soft.  Peduncle  broad,  compressed,  of 
moderate  length.  Dorsal  fin  originating 
well  behind  insertion  of  pectoral  fins,  con- 
tinuous, a  few  very  weak  spines  graduating 
to  the  rays,  39-43  elements  in  all.  A  mid- 
dorsal  ridge  preceding  the  fin.  Anus  at 
about  mid-body.  Anal  fin  originating  slightly 
behind  middle  of  body,  three  weak  spines 
precede  the  rays,  27  to  32  elements  in  all. 
Median  fins  with  compressed  fleshy  bases. 
Pectoral  fin  rounded,  base  fleshy.  Pelvic 
fins  small,  inserting  directly  under  insertion 
of  pectoral  fins,  not  attached  to  abdomen 
with  a  membrane,  folding  into  an  insignifi- 
cant groove.  Caudal  fin  broad,  slightly 
rounded  or  emarginate.  Moderate  cycloid 
scales  with  prominent  circuli,  not  especially 
deciduous,  covering  bases  of  median  fins. 
Lateral  line  slightly  arched  anteriorly, 
straightening  out  over  anterior  part  of  anal 
fin  and  extending  onto  peduncle;  lateral 
line  scales  around  120.  Skin  fairly  thick; 
subdermal  canal  system  not  well  developed, 
pores  very  small.  Head  around  25  per  cent 
of  the  standard  length,   its  profile  sloping 


and  the  pores  not  prominent.  Top  of  head 
not  scaled,  naked  skin  not  projecting  back- 
ward over  the  nape.  Eye  of  moderate  size, 
no  adipose  tissue.  Nostrils  near  tip  of  trun- 
cate snout,  both  round.  Angle  of  gape  ex- 
tending below  eye.  Premaxillary  not  pro- 
tractile. Only  upper  margin  of  upper  jaw 
covered  by  lacrimal  bone  when  mouth  is 
closed.  Very  slender  supramaxillary  pres- 
ent. Jaw  teeth  minute,  pointed,  uniserial, 
close-set;  vomer,  palatines,  and  basibran- 
chials  toothless.  Opercle  and  preopercle 
thin,  both  well  scaled,  margins  with  very 
fine  spinules;  opercle  with  two  weak  flat 
spines;  angle  of  preopercle  rounded,  bulg- 
ing backward.  Cheek  scaled.  Gill-rakers 
heavy,  a  little  shorter  than  the  filaments, 
toothed  on  inner  edge,  spaced,  about  10  on 
lower  limb  of  first  arch.  Pseudobranch 
small.  Seven  branchiostegal  rays,  five  on 
the  ceratohyal,  two  on  the  epihyal.  Scapula 
not  prominent.  Vertebrae  50  to  60.  Caudal 
skeleton  with  three  autogenous  haemal 
spines,  six  hypurals,  and  two  or  three 
epurals.  Stomach  a  simple  sac;  intestine 
long.  Pyloric  caeca  about  10,  digitiform, 
slender. 

Color  in  preservative  tan  to  dark  brown, 
the  median  fins  and  pelvics  darker  than  the 
body.    No  pattern,  slight  countershading. 

Natural  Jiistory.  Young  Icichthys  are 
commonly  found  swimming  under  or  within 
medusae  (Jordan,  1923a;  Ilobbs,  1929; 
Fitch,  1949),  and  sometimes  appear  in  fair 
number  off  the  California  coast.  Large 
adults  have  been  taken  by  drift-nets  (Parin, 


Stromateoid  Fishes  •  Hacdrich 


67 


Figure  16.      Branchial    region  of  Icichthys  lockingtoni,  drawing  of  a  cleared-ond-stained  preparation  from  a  173-mm  specimen. 
Elements  identified  in  Figure  2. 


1958)  and  by  deep  trawl  ( Ueno,  1954),  but 
are  very  rare.  All  recorded  captures  are 
from  deep  water.  Icichthys  is  certainly 
oceanic,  and,  judging  from  its  soft  tissues, 
somber  color,  and  rare  occurrence,  it  may 
well  live  as  an  adult  in  the  bathypelagic 
realms.  Ueno's  (1954)  362-mm  SL  speci- 
men is  the  largest  known. 

Icichthys  is  found  in  cool  waters.  The 
appearance  of  twelve  small  specimens  off 
the  Cape  of  Manazuru,  Japan,  in  the  spring 
of  1963  corresponded  with  an  unusual  in- 
flux of  ca.  15°C  water  in  this  normally 
warmer  area  (Abe,  1963). 

Relationships.  Externally,  Icichthys  very 
closely  resembles  Centwlophus,  with  which 
it  has  been  synonymized  by  Parin  ( 1958 ) . 
However,  in  several  respects — the  scalation 
on  the  cheeks,  the  caudal  skeleton,  and  the 
greatly    increased    number   of   vertebrae — 


Icichthys  differs  from  Ccnirolophus  enough 
to  warrant  generic  recognition. 

The  structure  of  the  pharyngeal  sacs  (Fig. 
16,  cf.  Fig.  14)  and  the  general  appearance 
of  IcichtJiys  suggest  a  very  close  relation- 
ship with  Centrolophus.  Icichthys  has  lost 
an  epural  in  the  caudal  skeleton  (Fig.  17), 
and  is  almost  certainly  the  derived  form. 
But  Centrolophus,  having  lost  the  cheek 
scales  retained  in  Icichthys,  cannot  be  the 
direct  ancestor.  Both  must  have  branched 
from  a  common  stem.  It  is  perhaps  signifi- 
cant that  the  ranges  of  the  two  genera  com- 
plement each  other  nicely  (Fig.  52). 

Icichthys,  a  member  of  the  most  primi- 
tive group  of  stromateoids,  has  a  very  high 
number  of  vertebrae,  an  advanced  condi- 
tion. The  number,  between  50  and  60,  is 
slightly  more  than  twice  the  basic  perciform 
number,  25,  found  in  other  centrolophids. 


68 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


EPURALS 


HYPURALS 


AUTOGENOUS 

HAEMAL 

SPINES 

Figure   17.      Caudal    skeleton   of    Icichfbys    lockingtoni,    drawing  of  a  cleared-and-stained  preparation  from  a  43-mm  specimen, 
SU  41028.     All  elements  identified  in   Figure  1. 


The  number  of  elements  in  the  median  fins 
is  about  the  same  as  in  Centrolophus,  but 
there  are  more  than  twice  the  number  of 
free  intemeurals  ahead  of  the  dorsal  fin. 
The  evidence  is  at  least  suggestive  that 
Icichthys  may  have  arisen  by  polyploidy; 
chromosome  counts  would  be  most  instruc- 
tive. The  three  autogenous  haemal  spines 
in  the  tail  (Fig.  17),  in  contrast  to  the  two 
of  all  other  perciforms  (Gosline,  1961a), 
are  undoubtedly  a  by-product  of  the  in- 
creased number  of  vertebrae. 


S})ccics.  The  genus  is  restricted  to  the 
cooler  waters  of  the  North  Pacific  and  of 
New  Zealand,  from  whence  a  new  species  is 
being  described  (  Haedrich,  in  press ) .  Abe 
( 1963 )  reports  more  pyloric  caeca  and 
slightly  fewer  vertebrae  for  his  Japanese 
specimens  than  are  found  in  specimens 
from  off  California.  Many  more  specimens 
will  be  needed  to  see  whether  these  differ- 
ences are  significant.  From  knowledge  of 
Jcichthijs  apparent  bathypelagic  habitat,  it 


Stromateoid  Fishes  •  Haedrich 


69 


Figure   18.     Senolella  punctata,   an    elongate   species,   drawing  of  on  approximately  250-mm  specimen,  from  McCulloch,  191' 


seems  best  for  the  time  being  to  recognize 
but  one  North  Pacific  species: 

*IcicJit]iys  Jocking,toni  Jordan  and  Gilbert, 
1880.  Cahfornia  to  Japan,  type  locaHty 
Point  Reyes,  Cahfornia.  D  39-43  ( total  ele- 
ments). A  27-32  (total  elements).  P  18- 
21.  Gill-rakers  4-6  +  1  +  11-13,  usually  18 
total.  Vertebrae  56-60.  Synonyms,  both 
based  on  small  specimens  from  the  coast 
of  California,  are  *Schedo))JiiIus  hcatJii  Gil- 
bert, 1904,  and  "^Centrolophus  californiciis 
Hobbs,  1929. 

Genus  SERIOLELLA  Guichenot,  1848 

Figures   18,   19 

Seriolella  Guichenot,  1848:238.  (Type  species: 
Seriolella  porosa  Guichenot,  1848:239,  by  sub- 
sequent designation  of  Jordan,  1923:238. 
Chile. ) 

Neptomenus  Giinther,  1860:389.  (Type  species: 
Neptomemis  brama  Giinther,  1860:340,  by 
original  designation.    New  Zealand. ) 

The  combination  of  at  least  seven  more 
dorsal  than  anal  finrays,  short  stout  spines 
not  increasing  in  length  to  the  rays  in  the 
dorsal  fin,  toothless  palate,  pelvic  insertion 
behind  the  pectoral  insertion,  supramaxil- 
lary  bone  present,  and  lateral  line  following 
the  dorsal  profile,  distinguishes  Seriolella 
from  all  other  stromateoid  genera.  The 
name,  a  feminine  noun,  is  the  diminutive 
of   Seriola,    a    carangid    genus.    Ultimately 


from  the  Latin  seria,  an  oblong  earthen  ves- 
sel, it  doubtless  refers  to  the  shape  of  the 
fish. 

Description.  Body  moderately  deep  to 
elongate,  maximum  depth  25  to  40  per  cent 
of  the  standard  length,  compressed  but 
fairly  thick;  musculature  firm.  Peduncle 
stout.  Two  dorsal  fins,  the  first  originating 
over  or  slightly  behind  insertion  of  pectoral 
fins,  with  seven  to  nine  short  spines.  Usu- 
ally the  third,  fourth,  and  fifth  spines  are 
the  longest,  the  longest  spine  less  than  half 
the  length  of  the  longest  dorsal  finray. 
Second  dorsal  with  25  to  40  finrays,  the  an- 
teriormost  the  longest.  Anal  and  genital 
pore  slightly  before  or  behind  mid-body,  in 
a  slit.  Anal  fin  originating  at  or  behind 
mid-body,  three  spines  increase  in  length  to 
the  18  to  25  ravs,  the  anteriormost  finravs 
the  longest.  Number  of  dorsal  finrays  ex- 
ceeds number  of  anal  finrays  by  more  than 
seven.  Pectoral  fins  rounded  in  the  young, 
long  and  falcate  in  the  adult.  Pelvic  fins 
inserting  just  under  end  of  or  behind  pec- 
toral fin  base,  attached  to  the  abdomen  by 
a  small  membrane  and  folding  into  a  shal- 
low groove.  Caudal  fin  broad  and  forked. 
Large  cycloid  scales,  very  deciduous,  cov- 
ering fleshy  bases  of  the  median  fins.  Lat- 
eral line  moderately  high,  following  dorsal 
profile  and  extending  onto  peduncle.  Skin 
thin;    main    subdermal    canal   along    inter- 


"()        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure   19.     Ser/o/e//a  brama,   a  deep-bodied   species,  drawing  of  an  approximately  250-mm  specimen,  from  McCulloch,  191' 


muscular  septum  and  side  branches  usually 
visible,  pores  small.  Head  about  30  to  35 
per  cent  of  the  standard  length.  Top  of 
head  naked,  fine  canal  network  and  small 
pores  usually  visible,  naked  skin  projecting 
backwards  over  the  nape.  Eye  moderate 
to  large.  Adipose  tissue  around  eye  well 
developed  and  extending  forward  around 
the  nostrils.  Nostrils  near  tip  of  pointed  or 
truncate  snout,  small,  the  anterior  round, 
the  posterior  a  vertical  slit.  Maxillary  ex- 
tending below  eye  but  angle  of  gape  be- 
fore eye.  Premaxillary  not  protractile.  Lac- 
rimal bone  partially  covering  upper  jaw 
when  mouth  is  closed,  ventral  border  of 
premaxillary  and  end  of  maxillary  remain- 
ing exposed.  Supramaxillary  present.  Jaw 
teeth  small  to  minute,  pointed,  uniserial, 
close-set  or  slightly  spaced,  covered  later- 
ally by  a  membrane;  vomer,  palatines,  and 
basibranchials  toothless.  Opercle  and  pre- 
opcrcle  thin,  margins  entire  or  finely  den- 
ticulate; opercle  with  two  weak  flat  spines, 
scaled,  the  scales  covered  by  skin;  pre- 
opercle  not  scaled,  angle  rounded,  bulging 
backward.  Cheek  scaled,  the  scales  cov- 
ered by  thick  skin  and  not  visible  without 
dissection.  Gill-rakers  one-half  to  one-third 
the  length  of  the  filaments,  toothed  on  inner 


edge,  slightly  spaced,  14  to  18  on  lower 
limb  of  first  arch;  no  rudimentary  rakers 
under  the  small  pseudobranch.  Seven  bran- 
chiostegal  rays,  five  on  the  ceratohyal,  two 
on  the  epihyal,  the  tips  of  the  branchioste- 
gals  pointed.  Posterior  border  of  scapula 
free  from  the  body.  Vertebrae  10  +  15,  or 
11  +  14  =  25.  In  the  adult,  hypurals  2  -f  3 
and  4-1-5  closely  conjoined  or  even  par- 
tially fused,  three  epurals.  Sclerotic  bones 
well  ossified,  subocular  shelf  present  on 
second  suborbital.  Stomach  a  simple  sac; 
intestine  long.  Pyloric  caeca  numerous  and 
fonning  a  dendritic  mass. 

Color  in  preservative  brown  or  bluish, 
darker  above  than  below,  the  sides  some- 
times with  a  silvery  overlay.  Usually  a 
prominent  dark  blotch  on  the  shoulder  at 
the  beginning  of  the  lateral  line;  smaller 
spots  often  present  on  sides.  Fins  usually 
a  little  lighter  than  the  body,  but  black- 
edged.  Inside  of  mouth  and  gill  cavity  light. 

Natural  history.  In  contrast  to  most  other 
centrolophids,  the  species  of  ScrioIcUa  are 
coastal  fishes.  Schools  of  them  occur  from 
150  fathoms  in  towards  the  coasts,  and 
some  species  even  enter  estuaries  ( Munro, 
1958).  Others  live  in  kelp  beds,  apparently 
not  deeper  than  40  fathoms  (Scott,  1962). 


Strom ATEOiD  Fishes  •  Hacdricli        71 


\  ■»^"^<..£^*-   =^"  -     1     err     y^-    ■  -  s 


Figure  20.      Branchial   region  of  Seriolella  vio/aceo,  drawing  of  a   cleared-and-stained    preparation   from   a    188-mm    specimen, 
USNM  77593.     Elements   identified   in   Figure  2. 


Nichols  and  Murphy  ( 1922 )  report  a 
young  Peruvian  specimen  from  under  a 
jellyfish. 

Seriolella  is  the  subject  of  a  modest  fish- 
ery in  Chile  (Mann,  1953).  In  Peru,  nine- 
or  ten-inch  specimens  are  at  times  so  com- 
mon that  they  are  caught  by  jigging  ( Nich- 
ols and  Murphy,  1922).  These  fish  are  oc- 
casionally taken  by  fishennen  in  Australia 
and  New  Zealand,  but  apparently  are  not 
sought-after  commercial  species  there. 

Relationships.  Seriolella,  with  its  prob- 
able off-shoot  Psenopsis,  represents  the  ad- 


vanced condition  among  centrolophids. 
Seriolella  is  derived  from  a  Hypcroghjplie- 
like  stock,  with  which  it  shares  the  short 
stout  spines  in  the  dorsal  fin  and  the  fluted 
first  haemal  spine  curving  backward  to 
meet  the  first  interhaemal.  The  slender 
pointed  branchiostegal  rays  (Fig.  20),  the 
numerous  bands  of  small  papillae  in  the 
pharyngeal  sacs,  the  well  ossified  sclerotic 
bones,  and  the  partial  fusion  of  hypurals 
2  +  3  and  4  +  5  with  growth  are  all  ad- 
vanced characters,  and  approach  the  no- 
meid  grade.    S.  violacea,  from  Peru,  comes 


72 


Bulletin  Musctini  of  Comparative  Zoologtj,  Vol.  135,  No.  2 


near  to  bridging  the  gap  between  Ilijpcro- 
glyphe  and  Seriolella. 

ScriolcIIa  has  given  rise  to  Psenopsis.  The 
pharyngeal  sacs  and  caudal  skeleton  of  both 
are  very  similar.  Both  genera  have,  in  most 
species,  well  ossified  sclerotic  bones  and  a 
dark  blotch  on  the  shoulder.  Seriolella,  how- 
ever, is  closer  to  Hypcroiijiiphe  in  the  pos- 
session of  a  supramaxillary,  which  has  been 
lost  in  Psenopsis. 

Species.  Seriolella  is  restricted  to  the 
cool  temperate  waters  of  the  Southern 
Hemisphere.  About  a  dozen  species  have 
been  described;  the  majority  are  known  to 
me  only  from  published  descriptions.  I 
have  been  able  to  examine  only  a  few 
Seriolella,  most  of  them  from  South  Amer- 
ica. The  nominal  species  in  the  genus  are: 

Seriolella  punctata  (Bloch  and  Schneider, 
1801)  =  Scomber  puncfafus  Bloch  and 
Schneider.  Southern  Australia,  Tasmania, 
and  New  Zealand.  Elongate.  D  VI-VII, 
34-39.  A  III  21-24.  P  19-22.  Gill-rakers 
usually  6  +  1  +  14-15.  Vertebrae  10  +  15. 
Synonyms  are  "^'Neptomenus  dohula  Giin- 
ther,  1869,  from  Tasmania,  and  Neptotiienus 
hilincatus  Hutton,  1872,  from  Wellington 
Harbor,  New  Zealand. 

Seriolella  violacea  Guichenot,  1848.  Chile 
and  Peru,  type  locality  \^alparaiso.  Mod- 
erately deep.  D  VII-VIII,  25-28.  A  III 
18-20.  P  21-22.  Gill-rakers  5-7  +  1  +  16- 
18.  Vertebrae  11  +  14.  This  fish  is  the 
"cojinoba"  of  Chilean  fisheries  literature 
(Mann,  1953).  Synonyms  are  Centrolophus 
peruanus  Steindaehner,  1874,  from  Callao, 
Peru,  and  '"Neptomenus  crassus  Starks, 
1906,  also  from  Callao. 

Seriolella  porosa  Guichenot,  1848.  Chile 
and  Peru,  type  locality  Valparaiso.  Elon- 
gate. D  \T-VIII,  34-38.  A  III  22-23.  P 
19-21.  Gill-rakers  usually  6  +  1  +  14-15. 
Vertebrae  10  +  15.  This  species  has  the 
same  counts  as  S.  punctata  (Bloch  and 
Schneider,  1801),  and  was  synonymized 
with  S.  dohula  (Giinther,  1869)  [here  con- 
sidered =  S.  punctata]  by  Regan  (1902). 
It  is  unlikely  that  an  essentially  coastal  fish 
such  as  Seriolella  would  regularly  cross  the 


broad  expanse  of  ocean  between  South 
America  and  Australia.  With  closer  study 
S,  porosa,  S.  punctata,  and  possibly  S. 
dohula  will  probably  prove  distinct. 

Seriolella  hrama  (Giinther,  1860)  = 
Neptomenus  hrama  Giinther.  Southern 
Australia  and  New  Zealand,  type  locality 
New  Zealand.  Deep-bodied.  D  VI-VIII, 
26-33.  A  III  21-23.  P  20-21.  Gill-rakers 
7  +  1  +  16.  Vertebrae  10  +  15.  Neptomenus 
travale  Castelnau,  1872,  from  New  Zealand 
is  a  synonym. 

Seriolella  velaini  Sauvage,  1879.  Island 
of  St.  Paul,  Indian  Ocean.  Moderately  deep. 
D  VIII,  27.  A  III  20.    ( From  Regan,  1902. ) 

Seriolella  christopherseni  Sivertsen,  1945. 
Tristan  da  Cunha.  Atlantic  Ocean.  D  VI, 
28.  A  III  20.  Moderately  deep.  (From 
Sivertsen,  1945. ) 

Seriolella  noel  Whitley,  1958,  is  based  on 
one  battered  specimen  from  Sydney,  Aus- 
tralia, standard  length  331  mm.  The  counts 
given  are  D  X  +  31?;  A  2-30;  P  14;  gill- 
rakers  8  +  16;  lateral  line  scales  95  +  8. 
The  description  is  inadequate  to  tell  even 
to  what  genus  this  fish  belongs,  but  it  is 
decidedly  not  a  Seriolella.  The  ten  dorsal 
spines  indicate  it  may  belong  in  the  family 
Nomeidae. 

Genus  PSENOPSIS  Gill,  1862 

Figure  21 

Ps-cnopsis  Gill,  1862:127.  (Type  species:  Trachi- 
iwtus  anoinalus  Temminck  and  Schlegel,  1850: 
107,  by  monotypy.    Japan. ) 

Bathyscriola  Alcock,  1890:202.  (Type  species: 
*Bathi/seriola  ajanca  Alcock,  1890:202,  by 
monotypy.    Ganjam  Coast,  India. ) 

The  combination  of  dorsal  and  anal  fin- 
rays  in  almost  equal  numbers,  spines  grad- 
uating to  the  rays,  toothless  palate,  pelvic 
insertion  directly  under  the  pectoral  inser- 
tion, broad  forward  scoop  in  the  opercle 
below  the  second  opercular  spine,  and  no 
supramaxillary  distinguishes  Psenopsis  from 
all  other  stromateoid  genera.  The  name,  a 
feminine  noun,  is  from  the  Greek  i/zTJir/, 
Psenes  +  oxpis,  appearance,  drawing  atten- 


Stromateoid  Fishes  •  Hacdrich 


^^- 


Figure    21.     Psenops/s    cyoneo,    an    elongate    species,    drawing    of  a   139-mm   specimen,    BMNH    1890.    11.    28.   9,    from   Alcock, 
1892. 


tion  to   the   superficial   similarity   between 
these  two  genera. 

Description.  Body  moderately  deep  to 
deep,  maximum  depth  30  to  45  j^er  cent  of 
the  standard  length,  compressed  but  fairly 
thick;  musculature  soft.  Peduncle  short, 
deep,  and  compressed.  Dorsal  fin  originat- 
ing over  or  slightly  behind  insertion  of  pec- 
toral fins,  continuous,  with  five  to  seven 
short  spines  increasing  in  length  to  the  27 
to  32  rays.  The  last  spine  the  longest,  but 
less  than  half  the  length  of  the  longest  dor- 
sal finray.  Anal  and  genital  pore  well 
before  or  at  mid-body,  in  a  slit.  Anal  fin 
originating  well  before  or  slightly  behind 
mid-body,  three  spines  increase  in  length  to 
the  22  to  29  rays.  Number  of  dorsal 
finrays  never  exceeds  number  of  anal  fin- 
rays  by  more  than  five.  Pectoral  fins 
rounded  in  the  young,  usually  produced 
in  the  adult.  Pelvic  fins  inserting  directly 
under  origin  of  the  pectoral  fin,  attached 
to  the  abdomen  by  a  small  membrane  and 
folding  into  a  grooxe  which  reaches  to  the 
anus.  Caudal  fin  broad,  slightly  forked. 
Small  cycloid  scales,  very  deciduous,  cov- 
ering fleshy  bases  of  the  median  fins.  Lat- 
eral line  moderately  high,  following  dorsal 
profile  and  extending  onto  peduncle.  Skin 
very  thin;  main  subdennal  canal  along  inter- 
muscular septum  and  side  branches  clearh' 
visible,  canals  particularly  dense  on  back. 


pores  very  small.  Head  around  30  per  cent 
of  the  standard  length.  Top  of  head  naked, 
minute  pores  faintly  visible,  naked  skin  not 
projecting  or  projecting  only  slightly  back- 
wards over  the  nape.  Eye  moderate  to 
large.  Adipose  tissue  around  eye  developed 
and  extending  forward  around  the  nostrils. 
Nostrils  near  tip  of  truncate  snout,  mod- 
erate in  size,  the  anterior  round,  the  pos- 
terior a  slit.  Maxillary  extending  below  eye, 
angle  of  gape  at  anterior  border  of  eye. 
Premaxillary  not  protractile.  Upper  jaw 
covered  completely  by  lacrimal  bone  when 
mouth  is  closed.  Supramaxillary  absent. 
Jaw  teeth  minute,  pointed,  uniserial,  close- 
set,  covered  laterally  by  a  membrane;  vo- 
mer, palatines,  and  basibranchials  toothless. 
Opercle  and  preopercle  thin,  not  scaled, 
margins  entire  or  finely  denticulate;  opercle 
with  two  weak  flat  spines;  under  the  second 
spine  the  bone  is  3-shaped,  the  upper  in- 
dentation reaching  almost  to  the  preopercle 
and  covered  with  uncalcified  membrane; 
angle  of  preopercle  rounded,  bulging  back- 
wards significantly,  the  margin  scalloped 
in  very  small  specimens.  Gill-rakers  about 
half  the  length  of  the  filaments,  toothed 
on  inner  edge,  spaced,  about  13  on  lower 
limb  of  first  arch;  no  rudimentary  rakers 
under  small  pseudobranch.  Seven  bran- 
chiostegal  rays,  fi\e  on  the  ceratohyal,  t^^'o 
on   the    epihyal,   the   tips   of   the   branchi- 


74        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


EPURALS 


HYPURALS 


Figure    22.      Caudal    skeleton    of    Psenops/s    anomala,    drawing  of  a  cleared-and-stained  preparation  from  a  40-mm  specimen, 
ABE  62-656.    All  elements  identified  in   Figure  1. 


ostegals  pointed.  Scapula  visible.  Verte- 
brae 10  +  15  =  25.  In  the  adult,  hypurals 
2  +  3  and  4  +  5  closely  conjoined,  three 
epurals.  Sclerotic  bones  well  ossified,  sub- 
ocular  shelf  present  on  second  suborl^ital. 
Stomach  a  simple  sac;  intestine  long.  Py- 
loric caeca  very  numerous,  in  a  mass 
resem])ling    a    raspberry. 

Color  in  preservative  brown  or  bluish, 
deep-bodied  form  often  with  a  silvery  or 
whitish  overlay.  Deep-bodied  form  coun- 
tershaded,  others  uniform.  Usually  a  prom- 
inent ]>lack  spot  on  shoulder  at  beginning 
of  lateral  line.  Fins  a  little  lighter  than  the 
body.  Opercles  and  peritoneum  silvery  or 
blackish.  Inside  of  mouth  light,  gill  cavity 
dark. 

Natural  histonj.   Though  fished  commer- 


cially in  Japan,  very  little  is  known  of  the 
habits  of  these  fishes.  Young  Psenopsis 
have  been  reported  in  association  with 
medusae  (Shojima,  1961).  The  adults  of 
P.  anomala,  at  least,  live  nearer  the  coasts 
and  in  shallower  water  than  most  centro- 
lophids.  Large  schools  are  taken  by  near- 
shore  trap  nets  in  Japan.  Adult  specimens 
of  P.  cijanca  were  taken  off  Cananore  in 
Wj.  fathoms. 

Psenopsis  is  one  of  the  smaller  ccntro- 
lophids.  Specimens  of  180  mm  SL  are  fully 
mature.    Few  exceed  200  mm. 

Relationships.  Psenopsis,  with  Seriolella, 
is  the  most  evolutionarily  advanced  cen- 
trolophid.  The  slender  tapering  branchio- 
stegal  rays  and  the  conjunction  of  hypurals 
2  +  4    and   4  +  5    ( Fig.    22 )    \\'ith    growth 


Stromateoid  Fishes  •  Hacdrich 


75 


Figure    23.      Branchial    region    of    Pier\op%\%    anomala,    drawing    of   a    cleared-and-stained    preparation    from    a    150-mm    speci- 
men.    Elements  identified    in    Figure   2. 


approach  the  nomeid  grade.  The  pharyn- 
geal sacs  (Fig.  23)  are  larger,  and  there 
are  more  numerous  bands  of  small  papillae 
than  are  found  in  Hijperoglijphe  or  Ccntro- 
lophits.  The  well  ossified  sclerotic  bones, 
the  absence  of  a  supramaxillary  bone,  the 
smallish  mouth,  the  deciduous  scales,  and 
the  dorsal  fin  with  only  slightly  more  fin- 
rays  than  the  anal  suggest  that  Fscnopsis 
may  be  near  the  base  of  the  line  leading  to 
the  Stromateidae.  ScrioJella,  which  retains 
the  supramaxillary  lost  in  Psenopsis,  is  its 
closest  relative  within  the  centrolophids. 

Species.  Psenopsis  is  an  Indo-Pacific 
genus,  found  in  India,  Japan,  northwest 
Australia,  and  the  East  Indies.  There  are 
four   allopatric   species,    one    of   them    un- 


described.  Little  confusion  has  arisen  re- 
garding the  identification  of  these  fishes, 
and  there  are  no  problems  of  synonymy. 
The  species  are: 

Psenopsis  anomala  (Temminck  and  Schle- 
gel,  1850)  =  TracJunotus  anomahis  Tem- 
minck and  Schlegel.  China  and  southern 
Japan,  type  locality  Tokyo.  Deep-bodied. 
D  V-\TI,  27-32.  A  III  25-29.  P  20-23. 
Gill-rakers  usually  6  +  1  +  13,  12-15  on 
lower  limb  of  first  arch,  18-21  total.  Verte- 
brae 10  +  15  ( skel. ) .  This  species  is  the 
"ibodai"  of  Japanese  fisheries  literature, 
and  is  common  from  Hong  Kong  to  Tokyo 
and  into  the  Sea  of  Japan  as  far  north  as 
Hokkaido.  It  fomis  the  basis  of  an  impor- 
tant fishery.    P.  sliojimai  Ochiai  and  Mori, 


76        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


EPURALS 


HYPURAL  6 


HYPURAL  4+5  — 


HYPURAL   2  +  3  — 


Figure   24.     Caudal    skeleton    of   Nomeus    gronovii,    drawing    of  a   cieored-and-stained    preparation    from    an    87-mm   specimen. 
All  elements  identified  in   Figure  1. 


1965,  from  the  Sea  of  Japan  is  a  probable 
.synonym. 

Psenopsis  humero.su  Munro,  1958.  Dam- 
pier  Archipelago,  N.  W.  Australia.  Deep- 
bodied.  D  VII,  28.  A  III  25.  P  22.  Gill- 
rakers  12  on  lower  limb  of  first  arch  ( from 
Munro,  1958).  Probably  a  good  species, 
little  differentiated  from  P.  anumala. 

Fsenopslsciianea  (Alcock,  1890)  =  *B(ithy- 
seriola  cijanea  Alcock,  type  locality,  Ganjam 
Goast,  India.  Elongate.'  D  VI,  25-26.  A  III 
22-23.  P  20.  Gill-rakers  5+1  +  14.  Verte- 
brae 10  +  15. 

Family  NOMEIDAE 

Type  genus:    Nomeus  Cuvier,   1817 

Fasteur.s.      Cuvier     and     \'aleiicieiines,     1833:242 
( descT. ) . 


Nomeina.    Giinther,  1860:387   (in  part,  def.). 

Nonieidae.  Giinther,  1880:455  (in  part,  def.). 
Jordan  and  Gilbert,  1882:448  (descr.).  Jor- 
dan and  Evermann,  1896:948  (descr..  North 
America).  Jordan,  1923:183  (in  part,  hst). 
BerK,  1940:323  (in  part,  dist.);  1955:249  (in 
part,  dist.).  Norman,  1957:503  (in  part,  def., 
genera  hsted ) . 

Psenidae.    Auctorinn. 

Diagnosis.  Stromateoid  fishes  with  pelvic 
fins  present  in  adults,  two  dorsal  fins,  teeth 
on  vomer  and  palatines,  six  branchiostegal 
rays,  and  four  hypural  and  three  epural 
bones  in  the  tail.  The  papillae  in  the  pharyn- 
geal sacs  with  stellate  bases,  arranged  in 
about  five  broad  longitudinal  bands. 

Description.  Body  slender  to  deep,  com- 
pressed. Two  dorsal  fins,  the  first  with 
about    ten    slender    spines    folding    into    a 


Stromateoid  Fishes  •  HaedricJi 


I  I 


Figure   25.      Branchial    region   of  Nomeus   gronovii,    drawing   of   a   cleared-and-stained   preparation    from   a    187-mm   specimen, 
MCZ  35327.     Elements  identified  in  Figure  2. 


groove,  the  longest  spine  at  least  as  long 
as  the  longest  ray  of  the  second  dorsal  fin. 
One  to  three  anal  spines,  not  separated 
from  the  rays.  Soft  dorsal  and  anal  fins 
approximately  the  same  length.  Bases  of 
median  fins  sheathed  by  scales.  Pelvic  fins 
attached  to  the  abdomen  by  a  thin  mem- 
brane, folding  into  a  narrow  groove,  the 
fins  greatly  produced  and  expanded  in 
young  Nomeus  and  some  Psenes.  Scales 
small  to  very  large,  cycloid  or  with  ver\' 
small  weak  cteni,  thin,  extremely  deciduous. 
Lateral  line  high,  following  dorsal  profile 
and  often  not  extending  onto  peduncle. 
Skin  thin;  subdermal  mucous  canal  system 
well  developed  and  visible  in  most;  the 
main  canal  down  the  side  of  the  body  may 
be  mistaken  for  a  lateral  line.  Opercular 
and  preopercular  margins  entire  or  finely 
denticulate.  Opercle  very  thin,  with  two 
flat,  weak  spines.    Six  branchiostegal  rays. 


Mouth  small,  maxillary  rarely  extending  to 
below  eye.  Teeth  small,  conical,  or  cusped 
in  some  Psenes,  approximately  uniserial  in 
the  jaws,  present  on  vomer,  palatines,  and 
basibranchials.  Supramaxillary  absent.  Adi- 
pose tissue  around  eye  only  moderately 
developed  in  most.  Vertebrae  30  to  38,  41,  or 
42.  Caudal  skeleton  with  four  hypurals  and 
three  epurals.  Pharyngeal  sacs  with  papil- 
lae in  upper  and  lower  sections,  papillae  in 
five  to  seven  broad  longitudinal  bands. 
Bases  of  the  papillae  stellate,  teeth  seated 
on  top  of  a  central  stalk.  Adults  usually 
about  a  foot  long,  although  a  giant 
Cubiceps  may  exceed  three  feet.  Silver\' 
to  bluish-brown,  some  with  conspicuous 
striped  or  blotched  pattern. 

Distribution.  Nomeids  are  oceanic  fishes 
of  tropical  and  subtropical  waters.  They 
occur  in  the  Gulf  of  Mexico,  the  Caribbean 
Sea,  the  Atlantic  Ocean,  the  western  Medi- 


78 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


terranean  Sea,  the  Indian  Ocean,  and  across 
the  Pacific.  Numerous  in  the  waters  of  the 
Philippines  and  soutliern  Japan,  they  do  not 
seem  to  enter  the  shallow  South  China  Sea 
(Fig.  54). 

Relationships.  From  an  evolutionary 
standpoint,  the  nomeids  are  a  grade  above 
the  centrolophids.  There  are  more  verte- 
brae, fusions  have  occurred  in  the  hypural 
fan  ( Fig.  24 ) ,  a  branchiostegal  ray  has  been 
lost,  and  the  papillae  in  the  pharyngeal  sacs 
have  stellate  bases  (Fig.  25).  Nomeids 
have  teeth  on  the  palate  and  basibran- 
chials,  however,  which  precludes  their  deri- 
vation from  a  centrolophid.  Probably  both 
families  have  a  common  ancestor,  and  de- 
velopment has  been  somewhat  parallel. 
The  palatal  dentition,  lost  in  the  Centro- 
lophidae,  remains  in  nomeids.  The  Nome- 
idae  have  passed  through  the  centrolophid 
stage  without  leaving  living  representatives 
at  that  level. 

The  Nomeidae  have  given  rise  to  two 
other  families,  each  with  a  single  genus. 
The  tetragonurids,  a  very  highly  specialized 
group,  arose  early,  perhaps  from  the  same 
line  which  produced  Psencs.  The  similarity 
between  the  teeth  of  Tetragonurus  and 
Psenes  pcUucidus  is  striking,  but  need  not 
imply  too  close  a  relationship.  The  re- 
appearance of  characters  in  divergent  lines 
of  common  ancestry  is  not  an  unusual  phe- 
nomenon (Simpson,  1953),  and  seems  wide- 
spread in  stromateoids. 

The  ariommids  may  have  been  derived 
more  recently.  Superficially,  they  resemble 
nomeids  very  much,  but  the  teeth  on  the 
palate  have  been  lost,  further  fusions  have 
taken  place  in  the  caudal  skeleton,  and  the 
pharyngeal  sacs  are  strikingly  divergent. 
The  species  of  the  Cuhiceps  pauciradiatus 
group  may  share  a  common  ancestor  with 
the  Ariommidae.  These  show  a  tendency 
toward  the  ariommid  condition  in  the  re- 
duced palatal  dentition,  and  share  with 
them  the  very  large,  deciduous  scales  and 
the  extremely  slender  (sometimes  even  ab- 
sent )  bridge  over  the  anterior  vertical  canal 
in  the  ear. 


Key  to  Nonwid  Genera 

1  (4).    Body   elongate,    maximum   depth   us- 

ually less  than  .35  per  cent  of  the  stan- 
dard length,  greatest  in  small  speci- 
mens. Origin  of  dorsal  fin  behind,  or 
directly  over  in  very  small  specimens, 
insertion  of  pectoral  fins.  2 

2  (3).    Anal  count  I-III   14-25.    Insertion  of 

pelvic  fins  under  end  or  behind  base 
of  pectoral  fin.  An  oval  patch  of 
knoblike  teeth  on  the  tongue.  Ver- 
tebrae 30  to  33.  Cuhiceps,  p.  78.  Fig.  26 

3  (2).    Anal  coimt  I-II   24-29.    Insertion   of 

pelvic  fins  before  or  under  insertion 
of  pectoral  fin,  possibly  behind  in  very 
large   specimens.     No   patch   of   teeth 

on  the  tongue.    Vertebrae  41.  

Nomeits,  p.  81.  Fig.  27 

4  ( 1 ) .    Body  deep,   maximum   depth   usually 

greater  than  40  per  cent  of  the  stan- 
dard length,  but  possibly  less  in  very 
large  specimens.  Origin  of  dorsal  fin 
before,  or  directly  over  in  large  speci- 
mens, insertion  of  pectoral  fins.  

Psenes,  p.  84.  Fig.  28 

Genus  CUBICEPS  Lowe,  1843 

Figure  26 

Cuhieeps  Lowe,  1843:82.  (Type  species:  Seriola^ 
gracilis  Lowe,  1843:82,  by  subsequent  desig- 
nation of  Jordan  and  Evermann,  1896:950. 
Madeira. ) 

Atiniostoma  A.  Smith,  1849,  plate  XXIV.  (Type 
species:  AtiniDstoina  capensis  Smith,  1849, 
plate  XXIV,  by  monotypy.    South  Africa. ) 

Navarchiis  Filippi  and  Verany,  1859:187.  (Type 
species:  Navarchtis  suJcatus  Filippi  and  Ver- 
any, 1859:187,  by  monotypy.  Mediterranean. 
A  synonym  of  Cuhiceps  gracilis  Lowe,  1843: 
82. ) 

TraclielocirrJnis  Doumet,  1863:220.  (Type  species: 
Trachelocirrhus  mediterraneus  Doumet,  1863: 
222,  by  monotypy.  Sete,  France.  A  synonym 
of  Cuhiceps  gracilis  Lowe,  1843:82.) 

Miilichthys  Lloyd,  1909:1.56.  (Type  species:  Muli- 
chtJiys  s<piatuiceps  Lloyd,  1909:158,  by 
monotypy.    Arabian  Sea.) 

MamlclichtJujs  Nichols  and  Murphy,  1944:247. 
(Subgenus.  Type  species:  Cuhiceps  cari- 
natus  Nichols  and  Murphy,  1944:245,  by 
monotypy.  180  miles  SW  of  Cape  Mala, 
Panama. ) 


^  I^owe  described  his  fish  as  a  species  in  the 
genus  Seriola,  but  noted  (p.  82),  "Still  it  is  not 
unlikely  that  a  comparison  of  the  two  fishes 
[gracilis  and  S.  hijiinnulata  ( Quoy  and  Gaimard)] 
may  warrant  .  .  .  their  separation  from  Seriola 
into    a    genus,    winch    may    be    called    Cuhiceps." 


Stromateoid  Fishes  •  HaedricJi 


79 


The  combination  of  elongate  body,  long 
winglike  pectoral  fins,  insertion  of  pelvics 
behind  pectoral  fin  base,  scales  on  top  of 
head,  cheeks,  and  opercles,  and  a  patch  of 
teeth  on  the  tongue  distinguishes  Cubiceps 
from  all  other  stromateoid  genera.  The 
name,  a  masculine  noun,  is  from  the  Greek 
Kvf^o'i,  cube,  +  K£<^oA?/,  head,  in  reference  to 
the  square  profile  of  the  fish's  head. 

Description.  Body  elongate,  maximum 
depth  25  to  30  per  cent  of  the  standard 
length;  musculature  firm.  Peduncle  short, 
deep,  and  compressed.  Two  dorsal  fins, 
scarcely  divided.  First  dorsal  originating 
behind  insertion  of  pectoral  fins,  with  about 
ten  stiff  spines  folding  into  a  groove,  the 
longest  spine  longer  than  the  longest  ray  of 
the  second  dorsal.  Anterior  rays  of  the 
second  dorsal  the  longest,  those  that  follow 
decreasing  in  length,  14  to  23  finrays  in 
all.  Anal  and  genital  papillae  behind  mid- 
body,  in  a  slit.  Anal  fin  originating  behind 
origin  of  second  dorsal  fin,  one  to  three 
short  spines  preceding  the  rays.  Anterior 
rays  the  longest,  those  that  follow  decreas- 
ing in  length,  14  to  21  finrays  in  all.  Pec- 
toral fin  pointed,  becoming  very  long  and 
winglike,  the  relative  length  increasing 
markedly  with  growth;  base  of  the  fin 
inclined  at  an  angle  of  45°.  Pelvic  fins 
inserting  just  under  end  of  or  behind  pec- 
toral fin  base,  attached  to  the  abdomen  by 
a  small  membrane  and  folding  into  a  deep 
groove.  Expanded  coracoid  often  forming 
a  conspicuous  keel  along  mid-ventral  line 
ahead  of  pelvics.  Caudal  fin  forked,  the 
lobes  often  folding  over  one  another.  Scales 
large,  cycloid,  very  deciduous,  covering 
bases  of  the  median  fins.  Simple  tubed 
scales  of  lateral  line  high,  following  dorsal 
profile  and  ending  under  last  dorsal  finray 
or  extending  onto  peduncle.  Skin  thin;  sub- 
dermal  canals  on  flanks  easily  traced.  Main 
canal  may  be  confused  with  lateral  line. 
Pores  to  surface  small.  Head  around  32 
per  cent  of  the  standard  length.  Top  of 
snout  naked,  minute  pores  in  naked  skin. 
Scales  extending  forward  on  top  of  head 


almost  to  level  of  the  nostrils.  Eye  large, 
bony  supraorbital  ridge  pronounced.  Adi- 
pose tissue  around  eye  well  developed, 
extending  forward  around  the  nostrils.  Nos- 
trils near  tip  of  blunt  snout,  small,  both 
round.  Maxillary  ending  under  anterior 
border  of  eye,  angle  of  gape  well  before 
eye.  Premaxillary  not  protractile.  Lacrimal 
bone  completely  covering  upper  jaw  when 
mouth  is  closed,  ventral  border  of  maxillary 
sometimes  remaining  exposed.  Supramaxil- 
lary  absent.  Jaw  teeth  small,  pointed, 
slightly  recurved,  usually  spaced.  Very 
small  teeth  usually  present  on  vomer  and 
in  a  single  series  on  the  palatines  and  basi- 
branchials.  An  oval  patch  of  low  knoblike 
teeth  on  the  glossohyal.  Opercle  and  pre- 
opercle  thin,  scaled,  margins  entire  or  finely 
denticulate;  opercle  with  two  weak  flat 
spines;  angle  of  preopercle  slightly  rounded, 
not  bulging  backward.  Cheeks  scaled.  Cill- 
rakers  slender,  almost  the  length  of  the 
filaments,  toothed  on  inner  edge,  fairly 
close  set,  14  to  18  on  lower  limb  of  first 
arch;  no  rudimentary  rakers  under  the  large 
pseudobranch.  Scapula  prominent.  Verte- 
brae 13  +  17  to  15  +  18  =  30  to  33.  Scle- 
rotic bones  usually  well  ossified  in  adults. 
Stomach  a  simple  sac;  intestine  of  moderate 
length.  Pyloric  caeca  very  numerous,  in  a 
large  dendritic  mass. 

Color  in  preservative  either  brownish, 
darker  on  the  back  than  on  the  sides,  or 
bluish  above  and  silvery  on  the  sides.  First 
dorsal  blackish,  other  fins  usually  the  same 
color  as  the  body,  caudal  dusky.  Iris  grey- 
ish, often  with  golden  semicircles  at  anterior 
and  posterior  borders.  Inside  of  mouth,  gill 
cavity,  and  peritoneum  dark. 

NofumI  history.  All  the  nomeids  are 
oceanic.  Most  species  are  very  rarely  seen, 
and  little  is  known  of  their  habits. 

Small  Cubiceps  gracilis  are  very  numer- 
ous near  the  Azores,  where  they  are  taken 
in  surface  nets  and  from  under  medusae. 
By  the  time  these  fishes  reach  about  200 
mm  SL,  they  are  mature.  With  the  attain- 
ment of  maturity  growth  does  not  stop,  but 


80 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


=^^''?S 


.-.s^ 


Figure  26.      Cubiceps  gracilis,  drawing  of  a   164-mm  specimen,  from  Gunther,  1889. 


continues  significantly.  Specimens  near  800 
mm  SL  have  been  reported  from  the  Medi- 
terranean (Ariola,  1912).  With  growth,  the 
relative  length  of  the  pectoral  fin  increases 
markedly. 

In  the  Philippines,  Cubiceps  is  caught  by 
using  night-lights  and  lift-nets  ( Herre  and 
Herald,  1950).  Large  specimens  are  taken 
occasionally  by  the  near-shore  winter  long- 
line  fishery  for  swordfish  off  southern 
Japan. 

Relationships.  Cubiceps  occupies  the 
central  position  in  nomeid  evolution.  The 
two  other  nomeid  genera,  Nomeus  and 
Psenes,  are  certainly  derived  from  Cubi- 
ceps. In  both  derived  genera  the  patch  of 
teeth  on  the  tongue  has  been  lost,  and 
there  has  been  a  tendency  towards  an 
increase  in  the  number  of  vertebrae  and 
finrays.  The  papillae  in  the  pharyngeal 
sacs  of  Cubiceps  are  very  similar  to  those 
of  Nomeus  (Fig.  25),  but  those  of  Psenes 
are  in  general  smaller  (Figs.  29,  30).  The 
caudal  skeletons  of  all  three  genera  are 
almost  identical  (Fig.  24). 

The  Ariommidae  have  probably  evolved 
from  a  Cubiceps,  although  the  systematic 
position  of  the  family  is  far  from  clear. 
The  ariommids  do  share  certain  characters, 
however,  with  the  fi.shes  of  the  Cubiceps 
pauciradiatus  group,  as  discussed  on  pages 
77  and  90. 

Cubiceps  and  the  tetragonurids  both  have 
patches  of  teeth  on  the  tongue.  The  jaw 
teeth,  the  pharyngeal  sacs,  and  the  scales, 


however,  are  very  different  in  these  two 
groups.  It  seems  likely  that  the  Tetrago- 
nuridae  branched  off  very  early  from  the 
nomeid  stem,  but  their  ancestor  may  have 
been  a  fish  similar  in  many  respects  to  the 
present-day  Cubiceps. 

Species.  Cubiceps  is  found  in  the  tem- 
perate and  tropical  waters  of  the  Atlantic, 
Pacific,  and  Indian  oceans.  It  also  occurs  in 
the  western  Mediterranean  Sea,  and  the 
Gulf  of  Mexico  and  Caribbean  Sea. 

Much  confusion  surrounds  the  species  of 
Cubiceps.  The  counts  of  all  described  are 
fairly  close.  Lacking  sufficient  comparative 
material,  it  is  difficult  to  evaluate  the  small 
differences  which  do  occur,  for  example  in 
vertebral  number.  There  seems  to  be  a 
great  differential  in  the  size  attained  by  the 
adult.  The  species  of  the  Cubiceps  paucira- 
diatus group  may  never  exceed  200  mm  SL 
(Haedrich,  1965),  whereas  fishes  allied  to 
Cubiceps  gracilis  are  reported  (Smith,  1849; 
Ariola,  1912;  Abe,  1955a)  to  approach  a 
meter  in  length. 

The  nominal  species  in  the  genus  Cubi- 
ceps are: 

Cidjiceps  gracilis  (Lowe,  1843)  =  Seriola 
gracilis  Lowe.  Atlantic  Ocean  and  western 
Mediterranean,  type  locality  Madeira,  per- 
haps a  world-wide  species.  D  IX-XI,  I-II 
20-22.  A  I  I-II  I  20-23.  P  20-24.  Gill-rakers 
8-9  +  1  +  14-17.  Vertebrae  15  +  18.  Syn- 
onyms are:  Navarchus  sulcatus  Filippi  and 
Verany,  1859,  from  the  Mediterranean; 
Trachelocirrhus      mediterraneus     Doumet, 


Stromateoid  Fishes  •  Haedrich 


81 


1863,  from  Sete;  Ciihiceps  lowei  Osorio, 
1909,  from  the  Cape  \'erde  Islands;  and 
Aphareus  ohtusirostrls  Borodin,  1930,  from 
the  Azores. 

Cuhiceps  capensis  (A.  Smith,  1849)  = 
*Atimostoma  capensis  Smith.  Type  locahtv' 
South  Africa.  D  IX-X,  I-III  24-26.  A  I'l 
22-23.  P  16-18.  Gill-rakers  8-9  +  1  +  16- 
17.  Vertebrae  14  +  17.  The  stuffed  type  is 
about  900  mm  SL.  The  margin  of  the  oper- 
cle  between  the  two  flat  opercular  spines 
may  be  strongly  serrate  in  this  species.  A 
probable  synonym  is  Cubiceps  niger  Franca, 
1957,  from  Angola. 

*Cubiceps  pauciradiatiis  Giinther,  1872. 
Central  and  western  Pacific  Ocean,  type 
locality  Misol  Island.  D  X-XII,  I  16-18.  A 
I-II  14-17.  P  18-19.  Gill-rakers  8-9  +  1  + 
16.  This  is  a  diminutive  species,  rarely 
exceeding  160  mm  SL.  Closely  related 
forms  are:  *C.  longimanus  Fowler,  1934a, 
C.  carinatus  Nichols  and  Murj^jhy,  1944,  and 
*C.  athenae  Haedrich,  1965.  '  "^Cubiceps 
nesiotes  Fowler,  1938,  from  Christmas  Is- 
land, Central  Pacific,  is  a  probable  synonym. 

Cubiceps  sqiiomiceps  (Lloyd,  1909)  = 
Midichthys  squamiceps  Lloyd.  South  Africa 
to  Japan,  type  locality  Arabian  Sea.  D  IX- 
XI,  I-II  19-21.  A  II-III  18-21.  P  18-20. 
Gill-rakers  8-9  +  1  +  16-17.  This  is  a 
chunky-looking  fish,  with  a  very  short 
peduncle  and  broad,  winglike  pectoral. 
*Cubiceps  nataJensis  Gilchrist  and  von 
Bonde,    1923,    is   probably   a   synonym. 

^Cubiceps  caendeus  Regan,  1914a.  Tas- 
man  Sea,  tvpe  locality  Three  Kings  Island. 
D  X-XI,  I-II  21-24.  A  II-III  21-24.  P  19- 
21.  Gill-rakers  7  +  1  +  16-17.  Vertebrae  13 
+  18.  A  few  of  the  jaw  teeth  on  the  types 
are  long  and  project  like  fangs.  Cubiceps 
baxteri  McCulloch,  1923,  based  on  a  dam- 
aged specimen  371  mm  long,  may  be  the 
adult. 

"^Cubiceps  longimanus  Fowler,  1934. 
V^estern  Indian  Ocean,  type  locality  Dur- 
ban. D  X-XI,  I  15-16.  a'I-I1  15.  p' 18-20. 
Gill-rakers  9+1+14.  All  specimens  known 
are  less  than  50  mm  SL  long. 

Cubiceps  carinatus  Nichols  and  Murphy, 


1944.  Pacific  Coast  of  Central  America, 
type  locality  Gulf  of  Panama.  D  IX-X,  I 
14-16.  A  II  14-15.  P  17-19.  Gill-rakers 
7-8  +  1  +  14-16.  Vertebrae  13  +  17.  This 
species  has  fewer  median  finrays  than  the 
closely  related  C.  pauciradiatus  Giinther, 
1872,  from  the  Central  and  Western  Pacific. 
*Cubiceps  athenae  Haedrich,  1965.  East 
coast  of  North  America.  D  X-XI,  I  15-16. 
A  II  14-15.  P  18-19.  Gill-rakers  8  +  1  + 
16-17.   Vertebrae  13  +  18. 

Genus  NOMEUS  Cuvier,  1817 

Figure  27 

Nomeus  Cuvier,  1817:315.  (Type  species:  Gobius 
gronovii  Gmelin,  1788:1205,  by  subsequent 
designation  of  Jordan  and  Gilbert,  1882: 449. ^ 
Atlantic  Ocean. ) 

The  combination  of  elongate  body,  black 
fanlike  pelvic  fins  with  the  full  length  of 
the  trailing  edge  attached  to  the  abdomen, 
insertion  of  the  pelvics  (usually)  ahead  of 
the  pectorals,  blotched  and  spotted  pattern, 
and  41  vertebrae  distinguishes  Nomeus 
from  all  other  stromateoid  genera.  The 
name,  a  masculine  noun,  is  from  the  Greek 
vofxeis,  herdsman,  a  translation  of  the  Dutch 
vernacular  "Harder"  (Marcgrave,  1648), 
probably  in  reference  to  the  fishes'  habit 
of  following  Fhijsalia. 

Description.  Body  elongate,  maximum 
depth  around  30  per  cent  of  the  standard 
length;  musculature  firm.  Peduncle  slightly 
tapered,  compressed.  Two  dorsal  fins, 
scarcely  divided.  First  dorsal  fin  originat- 
ing over  or  a  little  behind  insertion  of  the 
pectoral  fin,  with  about  ten  soft  spines 
folding    into    a    deep    groove,    the    longest 


1  Cuvier  and  Valenciennes  (1833:242)  desig- 
nated Nomeus  nuniritii  Cuvier  (  1817:.315)  type 
for  the  genus.  Cu\ier's  species,  howe\er,  based 
on  the  "Harder"  of  Marcgrave  (1648:153),  ap- 
peared in  name  only,  the  description  being  later 
supplied  by  Cuvier  and  Valenciennes  (1833:243). 
Under  the  International  Code,  a  uotncn  uiiduru 
is  unavailable  as  a  type,  and  Nomeus  mauritii 
Cuvier,  1817,  is  thus  rejected.  Nomeus  mauritii 
( non  Cuvier,  1817 )  Cuvier  and  Valenciennes, 
1833,  is  a  synonvm  of  Nomeus  gronovii  ( Gmelin, 
1788). 


82        Bulletin  Museum  of  Comparative  Zoolo<:,ij,  Vol.  135,  No.  2 


Figure  27.     Nomeus    gronov/i,    drawing    of    an    approximately   40-mm   specimen,   courtesy   of  tfie   Smitfisonian    Institution. 


spine  .slightly  longer  than  the  longest  ray 
of  the  second  dorsal.  Anterior  rays  of  the 
second  dorsal  fin  the  longest,  those  that 
follow  shorter,  subequal,  24  to  28  finrays 
in  all.  Anal  and  genital  papilla  at  mid-body, 
in  a  deep  slit.  Anal  fin  originating  under 
or  slightly  behind  origin  of  second  dorsal 
fin,  one  or  two  weak  spines  preceding  the 
rays.  Anterionnost  rays  the  longest,  those 
that  follow  shorter,  subequal.  Pectoral  fin 
rounded  in  the  young,  pointed,  winglike,  in 
the  adult;  the  relative  length  increases 
markedly  with  growth.  Pelvic  fins  inserting 
before,  or  in  large  specimens  under,  pec- 
toral fin  base,  fan-shaped,  innermost  ray 
the  longest,  attached  to  the  abdomen  for  its 
entire  length  by  a  strong  membrane  and 
folding  into  a  deep  groove  which  reaches 
to  the  anus.  Relative  length  of  the  fin 
decreasing  markedly  with  growth.  Caudal 
fin  deeply  forked,  lobes  very  long.  Scales 
small,  cycloid,  very  thin,  deciduous,  covering 
bases  of  median  fins.  Simple  tubed  scales 
of  lateral  line  high,  following  dorsal  profile 
and  not  appearing  to  extend  onto  peduncle. 
End  of  lateral  line  under  last  dorsal  finray 
in  most  specimens  (in  the  single  large 
specimem   known,   the   lateral   line   reaches 


the  caudal  base).  Skin  thin;  subdermal 
canals  on  flanks  easily  traced.  Main  canal 
may  be  confused  with  lateral  line.  Pores  to 
surface  minute.  Head  around  30  per  cent 
of  the  standard  length.  Top  of  snout  naked, 
minute  pores  visible  in  naked  skin.  Scales 
extending  forward  over  nape  to  level  of  an- 
terior border  of  the  eye.  Interorbital  space 
and  top  of  snout  covered  with  adipose 
tissue.  Eye  of  moderate  size,  bony  supra- 
orbital ridge  pronounced.  Adipose  tissue 
around  eye  very  well  developed,  extending 
forward  to  cover  the  lacrimal  and  surround- 
ing the  nostrils.  Nostrils  on  tip  of  truncate 
snout,  small,  the  anterior  round,  the  pos- 
terior a  slit.  Maxillary  ends  under,  or  before 
in  large  specimens,  anterior  border  of  the 
eye,  angle  of  gape  well  before  eye.  Pre- 
maxillary  not  protractile.  Lacrimal  bone 
almost  completely  covering  upper  jaw  when 
mouth  is  closed,  ventral  border  of  maxillary 
remaining  exposed.  Supramaxillary  absent. 
Jaw  teeth  small,  pointed,  slightly  recurved, 
spaced.  Small  recurved  teeth  present  on 
vomer  and  in  a  single  series  on  the  pala- 
tines and  basibranchials.  No  teeth  on  the 
glossohyal.  Opercle  and  preopercle  thin, 
scaled,   margins  very  finely  denticulate  or 


Stromateoid  Fishes  •  Haedrich 


83 


entire;  opercle  with  two  weak  hardly  de- 
fined flat  spines;  angle  of  preopercle 
rounded,  bulging  backward  in  large  speci- 
mens. Cheeks  scaled.  Gill-rakers  slender, 
half  the  length  of  the  filaments,  toothed  on 
inner  edge,  fairly  close-set,  about  16  on 
lower  limb  of  first  arch;  sometimes  a  few 
rudimentary  rakers  under  the  large  pseudo- 
branch.  Six  branchiostegal  rays,  four  on 
the  ceratohyal,  two  on  the  epihyal.  Scapula 
prominent.  Vertebrae  15  +  26  =  41.  In  the 
single  large  specimen  known,  the  sclerotic 
bones  are  well  ossified.  Stomach  a  simple 
sac;  intestine  of  moderate  length.  Pyloric 
caeca  numerous,  in  a  small  dendritic  mass. 

Color  in  life  bright  blue  above,  blotched 
and  spotted  with  blue  on  the  brilliant 
silvery  sides.  In  preservative,  the  base 
color  is  tan,  the  blotches  and  spots  appear- 
ing dark  brown.  Median  fins  \\'ith  about 
three  dark  stripes.  First  dorsal  and  pel  vies 
black,  pectorals  light.  The  single  large 
specimen  known  is  uniform  dark  brown. 
Eye  usually  greyish;  in  the  large  specimen 
the  iris  is  golden,  divided  by  a  dark  ver- 
tical bar.  Inside  of  mouth,  gill  cavity,  and 
peritoneum  light. 

Natural  history.  The  association  between 
Nomeus  and  PhysaJia  is  commonly  cited 
as  an  example  of  commensalism,  but  ac- 
tually very  little  is  known  about  the  true 
nature  of  the  association  or  about  the  life 
histories  of  the  animals  involved.  No77ieus 
has  been  observed  eating  Physalia  (Kato, 
1933)  and  vice  verso  (Carman,  1896).  In 
comparison  to  other  fishes,  however,  No- 
meus is  relatively  immune  to  the  toxin  of 
the  siphonophore   (Lane,   1960). 

The  eggs  and  larvae  of  Nomeus  are  not 
known.  Fishes  of  10-mm  SL  have  been 
taken  from  under  Physalia,  however,  indi- 
cating that  the  association  must  fonn  early 
in  the  fish's  life.  I  have  seen  150-mm  SL 
specimens  also  taken  with  Physalia.  These 
fishes  appeared  to  be  adults,  but  none  were 
ripe.  The  largest  known  specimen  of  No- 
meus was  taken  with  a  bottom  trawl  in  the 
Caribbean  Sea.  This  225-mm  specimen  was 
apparently  not  in  association  with  Physalia, 


was  living  fairly  deep  in  the  water,  and  was 
a  uniform  dark  brown  instead  of  blotched 
like  smaller  specimens.  This  fish  appeared 
to  be  a  mature  male.  Many  young  stromat- 
eoids  live  with  jellyfishes,  have  a  blotched 
or  mottled  color  pattern,  and  both  desert 
their  coelenterate  host  and  become  uni- 
formly colored  with  growth.  The  discovery 
of  this  large,  dark  Nomeus  suggests  that  the 
familiar  small,  blotched  Nomeus  found  un- 
der Physalia  may  only  be  the  young  form 
of  a  bigger  fish  which  lives  in  the  depths. 

Relationships.  Nomeus  is  derived  from 
the  central  Cubiceps  stock.  It  is  very  simi- 
lar in  appearance  to  Cubiceps.  The  differ- 
ences between  the  two  are  slight,  but 
sufficient  to  consider  Nomeus  a  genus  in 
its  own  right.  In  Nomeus  the  pelvic  bones 
have  become  much  shortened,  the  patch  of 
teeth  on  the  glossohyal  has  disappeared, 
the  number  of  vertebrae  and  finrays  has 
increased,  and  there  are  only  two  spines 
preceding  the  anal  finrays. 

Nomeus  lives  in  a  very  specialized  en- 
vironment, under  Physalia.  The  features 
which  distinguish  it  from  Cubiceps  doubt- 
less reflect  the  demands  of  this  unusual 
habitat.  Adapted  to  a  particular  way  of 
life,  Nomeus  has  given  rise  to  no  other 
forms. 

Species.  Nomeus  is  a  wide-ranging  genus, 
found  in  the  temperate  and  tropical  waters 
of  all  the  major  oceans.  It  does  not  occur, 
however,  in  the  eastern  Atlantic  or  the 
Mediterranean.  I  have  examined  specimens 
from  the  Atlantic  Ocean,  the  Culf  of  Mexico 
and  Caribbean  Sea,  the  Indian  Ocean,  and 
the  central  and  western  Pacific  Ocean.  The 
counts  of  all  these  are  essentially  the  same. 
I  cannot  but  conclude  that  in  the  genus 
Nomeus  there  is  but  one  species: 

Nomeus  g,ronovii  (Gmelin,  1788)  =  Gobius 
gronovii  Gmelin.  Temperate  and  tropical 
regions  of  the  Atlantic,  Pacific,  and  Indian 
Oceans,  Gulf  of  Mexico  and  Caribbean  Sea, 
type  locality  "American  Ocean  in  the  Tor- 
rid Zone."  D  IX-XII,  24-28.  A  I-II  24-29. 
P  21-23.  Gill-rakers  8-9  +  1  +  15-18.  Ver- 
tebrae 41.   Synonyms  are:    Eleotris  mauritii 


84 


Bulletin  Museum  of  Comparative  Zoology ,  Vol.  135,  No.  2 


Bloch  and  Schneider,  1801,  from  Mauritius; 
Nomeu.s  maculosus  Bennett,  1831,  from  the 
Atlantic  coast  of  North  Africa;  Nomeus 
maculattis  Valenciennes,  1836;  Nomeu.s 
oxytini.s  Poey,  1861,  from  Cuba;  and  No- 
meu.s dijscritus  Whitley,  1931,  from  New 
South  Wales.  *Nomeus  peronii  Cuvier  and 
Valenciennes,  1833,  from  the  seas  of  Java 
is  known  only  from  three  specimens  in  very 
poor  condition.  It  is  hard  to  be  certain  that 
these  fish  belong  to  Nomeus,  though  they 
are  almost  surely  nomeids.  I  was  able  to 
make  counts  on  only  one.  The  counts  were 
D  IX,  30;  A  31  (total  elements);  P  21;  gill- 
rakers  8  +  1  +  16.  No  subsequent  material 
has  appeared,  and  I  consider  *iV.  peronii  a 
synonym  of  N.  gronovii. 

Genus  PSENES  Cuvier  and  Valenciennes, 
1833 

Figure  28 

Psenes  Cuvier  and  Valenciennes,  1833:259.  (Type 
species:  *P.senes  cyanophrys  Cuvier  and  Val- 
enciennes, 1833:260,  by  original  designation. 
New  Ireland,  western  Pacific  Ocean. ) 

Icticus  Jordan  and  Thompson,  1914:242.  (Type 
species:  *  Icticus  i.schaini.s-  Jordan  and  Thomp- 
son, 1914:242,  by  original  designation.  Oki- 
nawa, western  Pacific  Ocean.  A  synonym  of 
*Pscucs  pcUiicidus   Liitken,    1880:516.) 

Papyrichthys  J.  L.  B.  Smith,  1934:90.  (Type  .spe- 
cies: *Psenes  pcUucidus  Liitken,  1880:516,  by 
original  designation.  Straits  of  Surabaja, 
Java. ) 

Thecopsenes  Fowler,  1944a: 63.  (Type  species: 
*P.'ienes  chapmani  Fowler,  1906:119,  by  orig- 
inal designation.  Cape  Verde  Islands,  Atlantic 
Ocean.  A  synonym  of  *Psene.s  cyanophrys 
Cuvier   and   Valenciennes,    1833:260.) 

Carlstioides  Whitley,  1948:87.  (Type  species: 
Cari^ioides  amplipinni.s-  Whitley,  1948:88,  by 
monotypy.  Lord  Howe  Island,  Tasman  Sea. 
A  svnonym  of  '*P.sencs  pclhicidus  Liitken, 
1880:516.) 

Parapsenes  J.  L.  B.  Smith,  1949a :847.  (Type 
species:  Psenes  rotundus  Smith,  1949:307,  by 
original  designation.  Dassen  Island,  South 
Africa. ) 

The  combination  of  two  dorsal  fins,  the 
first  dorsal  fin  originating  before  or  over 
the  pectoral  insertion,  pelvic  fins  present, 
deep  body,  teeth  on  the  palatines  and  basi- 
branchials,  and  no  teeth  on  the  glossohyal 


distinguishes  Psenes  from  all  other  stromat- 
eoid  genera.  The  name,  a  feminine  noun, 
is  from  the  Greek  i//?;!?/,  the  osprey  Pandion, 
the  allusion  not  evident.  The  authors  of  the 
name,  Cuvier  and  Valenciennes  (1833), 
may  have  been  impressed  by  the  resem- 
blance of  the  "sourcil  bleu"  on  their  little 
fish  to  the  similar  brow  of  the  fish  hawk. 

Description.  Body  deep,  maximum  depth 
usually  greater  than  40  per  cent  of  the  stan- 
dard length,  but  sometimes  less  in  large 
specimens;  musculature  firm  to  soft  and 
flabby.  Regions  at  bases  of  median  fins  may 
be  very  compressed  and  translucent.  Pe- 
duncle short,  compressed,  may  be  fairly 
slender.  Two  dorsal  fins,  scarcely  divided. 
First  dorsal  fin  originating  before  insertion 
of  pectoral  fins,  with  about  ten  soft  spines 
folding  into  a  deeper  groove.  Rays  of  the 
second  dorsal  fin  nearly  as  long  as  the  long- 
est Di  spine,  all  approximately  the  same 
length  or  decreasing  in  length  posteriorly, 
18  to  30  finrays  in  all.  Anal  papilla  a  little 
before  mid-body,  in  a  depression.  Anal  fin 
originating  at  mid-body,  slightly  behind 
origin  of  second  dorsal  fin,  two  or  three 
weak  spines  preceding  the  17  to  30  rays. 
Pectoral  fin  rounded  or  winglike;  relative 
length  of  fin  decreasing  slightly  or  increas- 
ing markedly  with  growth.  Pelvic  fins  in- 
serting under  posterior  portion  of  pectoral 
fin  base,  attached  to  the  abdomen  by  a 
small  membrane  and  folding  into  a  groove. 
Pelvics  very  long  in  the  young  of  some,  the 
relative  length  decreasing  markedly  with 
growth.  Caudal  fin  deeply  forked.  Scales 
small  to  minute,  with  a  few  weak  cteni, 
very  thin,  deciduous,  covering  bases  of  me- 
dian fins.  Simple  tubed  scales  of  lateral 
line  high,  following  dorsal  profile  and  end- 
ing under  last  dorsal  finray  or  extending 
onto  peduncle.  Skin  thin;  main  subdermal 
canals  along  intermuscular  septum  appar- 
ent, may  be  confused  with  lateral  line,  side 
branches  not  visible.  Pores  to  surface  mi- 
nute or  absent.  Head  around  30  per  cent 
of  the  standard  length.  Top  of  snout  naked, 
minute  pores  in  naked  skin.  Scales  extend- 
ing forward  on  top  of  head  almost  to  level 


Stromateoid  Fishes  •  Haedrich        85 


20  mm 


Figure   28.      Psenes  pellucidus,   drawing   of  a   1 30-mm   specimen  by  Margaret  Bradbury. 


of  anterior  border  of  the  eye.  Eye  moderate 
to  large.  Adipose  tissue  around  eye  mod- 
erately developed.  Nostrils  near  tip  of  trun- 
cate snout,  small,  the  anterior  one  round, 
the  posterior  a  slit.  Maxillary  ending  under 
anterior  portion  of  the  eye.  Premaxillary 
not  protractile.  Lacrimal  bone  almost  com- 
pletely covering  upper  jaw  when  mouth  is 
closed,  ventral  border  of  maxillary  remain- 
ing exposed.  Supramaxillary  absent.  Teeth 
in  both  jaws  uniserial,  pointed;  teeth  in 
upper  jaw  small,  slightly  recurved,  spaced; 
teeth  in  lower  jaw  larger,  may  be  long  and 
bladelike  with  small  cusps,  close-set.  A 
few  small  recurved  teeth  in  a  patch  on  the 
head  of  the  vomer  and  in  a  single  series  on 
each  palatine  and  on  the  basibranchials. 
Opercle  and  preopercle  thin,  scaled,  mar- 
gins very  finely  denticulate  or  entire;  oper- 
cle with  two  weak,  hardh"  defined,  flat 
spines;  angle  of  preopercle  rounded,  pro- 
jecting backward  very  slightly.  Gill-rakers 
slender,  a  little  shorter  than  the  filaments, 
toothed  on  inner  edge,  moderately  spaced, 
14  to  19  on  lower  limb  of  first  arch;  some- 
times a  iew  rudimentary  rakers  under  the 
long  pseudobranch.  Six  branchiostegal  rays, 


four  on  the  ceratohyal,  two  on  the  epihyal. 
Scapula  not  prominent.  Vertebrae  13  to  15 
+  18  to  23  =  31  to  38,  or  15  +  26  or  27  =  41 
or  42.  Stomach  a  simple  sac;  intestine  very 
long.  Pyloric  caeca  numerous,  in  a  dendritic 
mass. 

Color  in  preservative  bro\\'n  to  yellowish, 
some  species  with  a  conspicuous,  dark, 
blotched  or  longitudinally  striped  pattern. 
Median  fins  and  pelvics  often  darker  than 
the  body.  Region  at  bases  of  median  fins 
translucent  in  P.  pcUucidus.  Inside  of  mouth 
light  brown,  gill  cavity  usually  darker. 
Peritoneum  dark  or  light. 

Natural  history.  The  young  of  Psenes  are 
fairly  common  in  the  surface  layers  on  the 
high  seas.  They  do  not  seem  to  associate 
with  jellyfishes  to  any  extent,  but  are  \ery 
often  dipnetted  from  under  floating  Sargas- 
sum.  The  larger  adults,  as  with  most  other 
stromateoids,  probably  live  deeper  in  the 
water.  Most  species  in  Psenes  are  strictly 
oceanic,  and  large  specimens  are  rarely 
seen.  Longley  and  Hildebrand  ( 1941 )  re- 
port the  remains  of  120-mm  P.  cijanophnjs 
from  bird  rookeries  in  the  Tortugas.  Other 
species  found  there  included  Monacanthus 


86        BuUcfiu  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure  29.      Branchial    region    of   Psenes    cyanophrys,    drawing    of  a  cleared-and-stoined  preparation  from  a  128-mm  specimen. 
Elements    identified    in    Figure    2. 


lii.spichis,  Trachurops  crumenoptJialmo,  and 
Caranx  ruber,  all  fishes  commonly  associ- 
ated with  floating  Gulf  weed. 

In  the  Caribbean  area,  the  monthly  dis- 
tribution of  post-larval  and  juvenile  Psenes 
cyanophrys  suggested  that  spawning  oc- 
curred from  March  through  October 
( Legaspi,  1956 ) .  The  young  fish  fed  at 
first  on  copepods,  but  later  a  variety  of 
foods — copepods,  amphipods,  chaetognaths, 
fish  eggs  and  larvae — was  taken  (Lloyd, 
1909;  Legaspi,  1956).  In  large  P.  pellucidus, 
a  sombre-colored  probably  mesopelagic  spe- 
cies, I  have  found  gonostomatids  of  the 
genus  Mauroliciis. 


Relationships.  Psenes,  like  N omens,  is  a 
derivative  of  the  central  Cubiceps  stock.  In 
Psenes,  the  number  of  median  finrays  and 
vertebrae  has  tended  to  increase,  the  first 
interhaemal  has  moved  forward  fonning  an 
abrupt  angle  with  the  haemal  process  of  the 
first  precaudal  vertebra,  the  teeth  on  the 
glossohyal  have  disappeared,  the  tooth- 
bases  in  the  pharyngeal  sac  have  become 
smaller,  and  the  body  has  become  deeper. 
In  the  meso-  or  bathypelagic  species  the 
teeth  are  highly  differentiated.  Those  in  the 
lower  jaw  are  long  and  knifelike,  while 
those  in  the  upper  jaw  are  small  and 
strongly    recurved.     In    many    species    of 


Stromateoid  Fishes  •  Hacdrich        87 


mm^^ 


Figure  30.      Branchial    region   of  Psenes  pellucidus,  the   type  species  for  the  nominal  genus  /cficus,  drawing  of  a  cleared-and- 
stained    preparation    from    a    166-mm    specimen,    ABE   60-106.    Elements   identified   in    Figure  2. 


Psenes,  the  teeth  in  the  lower  jaw  are 
cusped. 

Psenes  has  given  rise  to  no  other  forms, 
although  species  such  as  Psenes  pellucidiis, 
which  have  entered  the  bathypelagic  realm, 
seem  to  be  diverging  rapidly  from  the  cen- 
tral bauplan.  This  fish  has  been  described 
as  a  separate  genus,  Icticus.  Were  it  not 
for  the  great  similarity  between  the  struc- 
ture of  the  pharyngeal  sacs  of  this  nominal 
genus  and  Psenes  (Figs.  29,  30),  and  for 
the  presence  of  species  intennediate  in 
other  characters,  Icticus  would  stand  as  a 
genus  derived  from  Psenes. 

Species.  Psenes  is  widespread  in  the  tem- 
perate and  tropical  parts  of  the  Atlantic, 
Pacific,  and  Indian  oceans.  There  are  nu- 
merous records  of  the  genus  from  the  Gulf 


of   Mexico   and   the   Caribbean,   but   none 
from  the  Mediterranean  Sea. 

As  in  Cubiceps,  there  is  some  confusion 
surrounding  the  species  of  Psenes.  Most 
seem  to  be  world-^\ide,  with  minor  differ- 
ences from  ocean  to  ocean.  Larger  collec- 
tions and  more  extensive  series  than  those 
now  available  are  necessary  for  proper  com- 
ment on  the  significance  of  these  differ- 
ences. 

The  nominal  species  in  the  genus  are: 
*Psenes  cyanophiys  Cuvier  and  Valen- 
ciennes, 1833.  Atlantic,  Pacific,  and  Indian 
oceans,  type  locality  New  Ireland.  D  IX- 
XI,  24-28.  A  III  24-28.  P  17-20.  Gill-rakers 
8-9  +  1  +  19.  Vertebrae  13  +  18.  The  counts 
recorded  for  specimens  from  the  Gulf  of 
Mexico  ( Legaspi,  1956 )  are  modally  higher 


88 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


than  those  of  Pacific  specimens.  A  very 
characteristic  feature  of  this  species  is  the 
numerous  longitudinal  streaks  on  the  sides. 
According  to  Le  Danois  ( 1962 ) ,  the  follow- 
ing are  synonyms:  Pscnes  jovanicus  Cuvier 
and  Valenciennes,  1833,  from  Java,  Psenes 
aurafus  Cuvier  and  Valenciennes,  1833, 
from  Guam,  and  Psenes  ftiscus  Guichenot, 
1866,  from  Madagascar.  Psenes  leucurus 
Cuvier  and  Valenciennes,  the  color  of  which 
is  described  by  the  authors  (1833:265)  as 
"jaunatre,  marbre  de  noiratre  et  finement 
raye  de  traits  longitudinaux  noiratres,"  is 
also  probably  a  synonym,  but  Le  Danois 
(1962)  does  not  agree.  Other  synonyms 
are:  *Cubiceps  niultimdiatus  Gimiher,  1871, 
from  Manado,  Philippines;  Psenes  chapmani 
Fowler,  1906,  from  the  Cape  Verde  Islands; 
Psenes  pocifieus  Meek  and  Hildebrand, 
1925,  from  the  Bay  of  Panama;  and  Psenes 
kamoharai  Abe,  Kojima,  and  Kosakai,  1963, 
from  Kyushu. 

*Psenes  pelhieidtis  Liitken,  1S80.  Atlan- 
tic, Pacific,  and  Indian  oceans,  type  locality 
Surabaja  Strait,  Java  Sea.  D  X-XI,  I-II  27- 
32.  A  III  26-31.  P  18-20.  Gill-rakers  8-9 
+  1  +  14-16.  Vertebrae  15  +  26-27.  This 
meso-  or  bathypelagic  species  is  immedi- 
ately recognizable  by  the  soft,  flabby  mus- 
culature, the  long  knifelike  teeth  in  the 
lower  jaw,  the  sombre  coloration,  and  the 
high  vertebral  and  median  finray  counts. 
With  growth,  this  species  becomes  quite 
slender.  Synonyms  are:  *Psenes  edtcardsii 
Eigenmann,  1902,  from  the  Atlantic  Ocean 
south  of  Rhode  Island;  '^Ictieus  ischanus  Jor- 
dan and  Thompson,  1914,  from  Okinawa; 
Caiistioides  ampUpinnis  Whitley,  1948, 
from  Lord  Howe  Island,  and  Cidneeps 
ismaelensis  Dieuzeide  and  Roland,  1955, 
from  the  coast  of  Algeria. 

*Psenes  macuhtns  Liitken  1880.  Atlan- 
tic, Pacific,  and  Indian  oceans,  type  locality 
central  Atlantic  Ocean.  D  IX-XI,  I  22-24. 
A  III  21-23.  P  20-21.  Vertebrae  15  4-  18- 
20.  Psenes  ni^rescens  Lloyd,  1909,  from  the 
Andaman  Sea  is  a  possible  synonym. 

*Psenes  arafurensis  Giinther,  1889.  Atlan- 
tic, Pacific,  and  Indian  oceans,  type  locality 


Arafura  Sea.  D  X-XI,  I-II  19-21.  A  III  20- 
21.  P  18-20.  Gill-rakers  7-9  +  1  +  15-16. 
Vertebrae  13  +  18.  This  species  is  very 
similar  to  *P.  moculatus  Liitken,  1880,  but 
the  body  is  deeper  at  comparable  sizes.  The 
teeth  in  the  lower  jaws  of  both  nominal 
species  are  like  those  of  *P.  pellueidus.  P. 
rotundtis  Smith,  1949,  from  South  Africa  is 
a  probable  synonym. 

Psenes  wlutcle^gii  Waite,  1894.  Coasts  of 
New  South  Wales,  type  locality  Maroubra 
Bay.  D  XI,  I  19.  A  III  18.  P  18.  (From 
Waite,  1894.)  Psenes  hiUii  Ogilby,  1915, 
from  Queensland  is  a  proliable  synonym. 

Psenes  guttatns  Fowler,  1934a.  Coast  of 
Natal.  D  XI,  I  20.  A  III  18.  Gill-rakers  7 
+  18.  (From  Fowler,  1934a).  This  name 
appears  as  a  nomen  nudum  in  Fowler 
(1906).  Psenes  stigmapleuron  Fowler,  1939, 
is  a  synonym.  Perhaps  a  synonym  of  P. 
liliiteleggii  Waite,  1894. 

*Psenes  henardi  Rossignol  and  Blache, 
1961.  Gulf  of  Guinea.  D  XI,  I  19-20.  A  III 
21-22.  P  19.  Gill-rakers  9  +  1  +  16-17.  *P. 
arafurensis  Giinther,  1889,  has  a  much 
deeper  body,  and  is  striped  and  mottled  on 
the  sides.  P.  henardi  is  uniform  light  brown. 
It  is  perhaps  a  svnonym  of  P.  guttatus  Fow- 
ler, 1934a. 

Family  ARIOMMIDAE,  new  family 

Type  genus:    Ariomma  Jordan  and  Snyder, 

1904 

Diagnosis.  Stromateoid  fishes  with  pelvic 
fins  present  in  adults,  two  dorsal  fins,  tooth- 
less palate,  six  branchiostegal  rays,  two  hy- 
pural  bones  in  the  tail,  and  well  ossified 
sclerotic  bones.  Bases  of  papillae  in  the 
pharyngeal  sacs  round;  papillae  not  in  bands 
and  in  upper  halves  of  the  sacs  only. 

Deseription.  Body  slender  or  deep, 
rounded  to  compressed.  Peduncle  slender, 
with  two  low  fleshy  lateral  keels  on  each 
side.  Two  dorsal  fins.  The  first  dorsal  with 
about  ten  slender  spines  folding  into  a 
groove.  The  longest  spine  twice  the  length' 
of  the  longest  ray  of  the  second  dorsal  fin. 
Three  anal  spines,  not  separated  from  the 
rays.    Soft    dorsal    and    anal    fins    approxi- 


Stromateoid  Fishes  •  Haedrich 


89 

1 


Figure  31.  Bronchial  region  of  Anomma  cf.  nigriorgenfeo, 
an  elongate  species,  drawing  of  a  cleared-and-sfained  prep- 
aration from  a  140-nrim  specinnen.  Elements  identified  in 
Figure  2. 


mately  the  same  length,  each  with  14  or  15 
finrays,  the  large  basals  protruding  into  the 
body  profile.  Bases  of  median  fins  not 
sheathed  by  scales.  Pelvic  fins  attached  to 
the  abdomen  by  a  thin  membrane  and  fold- 
ing into  a  groove.  Scales  large,  cycloid, 
thin,  extremely  deciduous.  Lateral  line 
high,  following  dorsal  profile  and  not  ex- 
tending onto  peduncle;  tubes  in  the  lateral 
line  scales  sometimes  branched.  A  branch 
of  the  lateral  line  extending  forward  over 
the  eye  in  a  bony  tract.  Skin  thin;  subder- 
mal  mucous  canal  system  well  developed. 
Opercular  and  preopercular  margins  entire 
or  very  finely  denticulate.  Opercle  very  thin, 
brittle,  with  two  weak,  ill-defined,  flat 
spines.  Six  branchiostegal  rays.  Mouth 
small,  maxillary  barely  extending  to  below 
eye.  Teeth  small,  simple  or  three-cusped, 
uniserial  in  the  jaws.  Vomer,  palatines,  and 
basibranchials  toothless.  Supramaxillary 
bone  absent.  Eye  large,  adipose  tissue  well 
developed  and  covering  the  lacrimal  bone. 
Sclerotic  bones  well  ossified.  Vertebrae  29 
to  32.  Caudal  skeleton  with  tvvo  hypurals 
and  three  epurals.  Pharyngeal  sacs  with 
papillae  in  the  upper  halves  only.  The  pa- 
pillae not  in  bands,  their  bases  rounded 
with  a  stalk  with  teeth  seated  all  along  it 
arising  off-center.  Adults  usually  about  a 
foot  long,  but  in  some  species  exceeding 
two  feet.    Silvery  to  blue-brownish,   some 


/y 


Figure  32.  Brancfiial  region  of  Anomma  indica,  a  deep- 
bodied  species,  drawing  of  a  cleared-and-stained  prepara- 
tion from  a  164-mm  specimen,  NTU  51942.  Elements  iden- 
tified  in    Figure   2. 


species  with  conspicuous  spotted  or  counter- 
shaded  pattern. 

Distri])ntion.  Ariommids  apparently  live 
near  bottom  in  deep  water  of  the  subtropics 
and  tropics.  They  occur  along  the  east 
coast  of  North  America,  in  the  Gulf  of  Mex- 
ico and  Caribbean  Sea,  along  the  coasts  of 
West  and  South  Africa,  along  Asian  coasts 
from  the  Red  Sea  to  Japan,  and  off  Hawaii 
(Fig.  55). 

Relationships.  Ariomma,  the  single  genus 
in  the  family,  superficially  appears  to  be  a 
nomeid.  The  two  dorsal  fins,  persistent  pel- 
vies,  and  six  branchiostegal  rays  have  been 
the  cause  for  placement  in  this  group  close 
to  Ciibiceps  (Psenidae  of  Jordan  and 
Snyder,  1907;  Nomeidae  of  Katayama,  1952). 
Some  authors  (Regan,  1914a;  Jordan,  1923) 
have  even  considered  Ariomma  a  synonym 
of  Cubiceps.  But  the  complete  absence  of 
teeth  on  the  vomer,  palatines,  and  basi- 
branchials in  Ariomma  contrasts  with  the 
situation  in  the  Nomeidae.  The  structure 
of  the  caudal  skeleton  ( Fig.  33 )  and  of  the 
pharyngeal  sacs  (Figs.  31,  32)  in  Ariomma 
is  unique  among  stromateoids,  and  diver- 
gent enough  from  any  others  to  warrant 
separation  at  the  family  level.   The  unique 


90        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


EPURALS 


URONEURAL  2 


HYPURAL  4  +  5  +  6 


HYPURAL  1  +  2  +  3- 


Figure  33.      Caudal   skeleton   of  Ariomma  cf.   n/griorgenfeo,   drawing    of    a    cleared-and-stained    preparation    from    a    28-mm 
specimen.     All    elements   identified    in    Figure    1. 


characters  of  the  family,  far  from  the  con- 
dition of  others  in  the  suborder,  are  prob- 
ably due  to  the  shift  the  ariommids  have 
made  to  an  adaptive  zone  atypical  for  stro- 
mateoids  (see  below:  Natural  history,  p. 
93). 

Ariomma  presents  a  confusing  array  of 
characters  which  could  suggest  affinities 
with  the  centrolophids,  the  nomeids,  or  the 
stromateids.  The  well  ossified  sclerotic 
bones,  minute  body  pores,  and  extremely 
deciduous  scales  are  characters  held  in 
common  with  the  advanced  Seriolella  group 
of  the  centrolophids  and  the  diminutive 
Cuhiccps  pauciradiatus  group  of  the  no- 
meids and  the  stromateids.  The  teeth  on 
the  papillae  occur  all  along  the  bony  stalk, 
and  the  jaw  teeth  of  Ariomma  indica  are 
cusped.  Both  characters  are  typical  of  stro- 
mateids. The  general  body  shape  is  like 
Seriolella.   The  complete  absence  of  palatal 


dentition  suggests  an  affiliation  with  the 
line  connecting  the  advanced  centrolophids 
with  the  stromateids. 

However,  Ariomma  has  two  distinct  dor- 
sal fins  and  very  large  scales,  and  the  bony 
bridge  over  the  anterior  vertical  canal  of 
the  ear  is  either  very  reduced  or  absent, 
all  in  marked  contrast  to  the  situation  in 
centrolophids  and  stromateids.  These  con- 
ditions are  found  in  the  Cuhiceps  pauci- 
radiatus group,  nomeids  which  in  addition 
have  reduced  palatal  dentition.  The  Ariom- 
midae  are  probably  derived  from  some- 
where in  this  line,  and  have  lost  the  teeth 
on  the  palate  and  basibranchials. 

Genus  ARlOtAhAA  Jordan  and  Snyder,  1904 

Figures  34,  35 
Ariomma   Jordan    and    Snyder,    1904:942.     (Type 
.species:    Ariomma  hirida  Jordan  and  Snyder, 
1904:943,  by  original  designation.    Honolulu, 
Hawaii. ) 


Stromateoid  Fishes  •  Haedrich 


91 


Figure  34.      Anomma   ledanoisi,    an    elongate   species,    drawing  of  a   189-mm  specimen,   from  Poll,   1959. 


Paracithiceps  Belloc,  1937:356.  (Type  .species: 
Paracuhiceps  ledanoisi  Belloc,  1937:356,  by 
original   designation.    Coast  of  West  Africa. ) 

The  combination  of  slender  caudal  pe- 
duncle with  lateral  keels,  deeply  forked, 
stiff  caudal  fin,  about  fifteen  dorsal  and  anal 
finrays  with  their  basal  elements  pro- 
nounced and  entering  into  the  body  profile, 
well-developed  adipose  tissue  around  the 
eye,  two  dorsal  fins,  and  toothless  palate 
distinguishes  Ariomma  from  all  other  stro- 
mateoid genera.  The  name  is  a  feminine 
noun  of  unknown  derivation. 

Description.  Body  either  elongate,  maxi- 
mum depth  about  25  per  cent  of  the  stan- 
dard length,  rounded,  or  deep,  maximum 
depth  40  per  cent  of  the  standard  length, 
compressed;  musculature  firm,  often  oily. 
Peduncle  short,  slender,  square  in  cross- 
section,  with  two  ill-defined,  low,  fleshy 
keels  on  each  side  at  base  of  caudal  fin. 
Two  dorsal  fins,  scarcely  separated.  First 
dorsal  originating  directly  over,  or  a  little 
before,  insertion  of  pectoral  fin,  with  about 
ten  long,  brittle  spines,  folding  into  a  deep 
groove.  Second  dorsal  fin  \\'ith  14  or  15 
rays,  each  about  half  as  long  as  the  longest 


Di  spine;  the  anterior  rays  a  little  longer 
than  those  which  follow.  Anal  papilla  a 
little  behind  mid-body,  in  a  slit.  Anal  fin 
originating  behind  middle  of  body  and  be- 
hind origin  of  the  second  dorsal  fin,  two  or 
three  spines  preceding  the  14  to  15  rays; 
rays  short,  the  anterior  ones  the  longest. 
Rays  of  the  median  fins  close-set  anteriorly, 
becoming  more  widely  spaced  posteriorly. 
Basals  of  the  finrays  often  projecting  above 
the  body  profile.  Pectoral  fin  rounded  in 
the  young,  becoming  pointed  with  growth; 
relative  length  decreasing  slightly  with 
growth  of  elongate  form  but  increasing 
markedly  with  growth  of  deep-bodied  form. 
Pelvic  fins  inserting  under  end  or  behind 
base  of  pectoral  fin,  attached  to  abdomen 
with  a  membrane  and  folding  into  a  pro- 
nounced groove  which  reaches  to  the  anus. 
Caudal  fin  stiff,  deeply  forked,  rays  on  the 
leading  edge  stiff  and  spinelike.  Scales 
large,  cycloid,  very  thin,  extremely  decidu- 
ous, not  covering  bases  of  the  median  fins. 
Scales  of  the  lateral  line  \\'ith  branched 
tubes,  located  high  on  the  body,  following 
dorsal  profile  and  not  extending  onto  pe- 
duncle. A  branch  of  the  lateral  line  ext(Mid- 


92        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure   35.      Ariomma  regu/us,    a   deep-bodied    species,    from   McKenney,   1961. 


ing  forward  over  the  eye  in  a  bony  tract 
from  the  head  of  the  hyomandibular.  Skin 
thin;  main  subdennal  canal  along  inter- 
muscular septum  very  apparent,  may  be 
confused  with  lateral  line;  side  branches  not 
as  conspicuous,  pores  to  surface  seem  to  be 
wanting.  Head  around  30  per  cent  of  the 
standard  length.  Top  of  snout  naked,  pores 
and  subdermal  canals  barely  visible.  Scales 
extending  forward  over  nape  to  level  of  an- 
terior border  of  the  eye.  Eye  large,  bony 
supraorbital  ridge  pronounced.  Adipose  tis- 
sue around  eye  well  developed,  extending 
forward  over  the  lacrimal  and  around  the 
nostrils.  Nostrils  near  tip  of  the  obtuse 
snout,  small,  the  anterior  round,  the  pos- 
terior a  slit.  Maxillary  scarcely  reaching  to 
under  eye,  angle  of  gape  well  before  eye 
and  nearer  to  tip  of  snout.  Premaxillary  not 
protractile.  Lacrimal  bone  transparent,  al- 
most completely  covering  upper  jaw  when 
mouth  is  closed,  the  ventral  border  of  the 
maxillary  remaining  exposed.  Supramaxil- 
lary  absent.  Jaw  teeth  minute,  covered 
basally  with  a  membrane,  usually  pointed 
but  three-cusped  in  a  few  deep-bodied 
forms,  uniserial,  close  set  or  slightly  spaced; 


vomer,  palatines,  and  basibranchials  tooth- 
less. Opercle  and  preopercle  thin,  brittle, 
margins  very  finely  denticulate  or  entire; 
opercle  rounded,  with  two  weak,  ill-defined, 
flat  spines;  angle  of  preopercle  rounded 
and  not  bulging  backward.  Gill-rakers 
slender,  half  the  length  of  the  filaments, 
toothed  on  inner  edge,  fairly  close  set, 
about  19  on  lower  limb  of  first  arch;  no 
rudimentary  rakers  under  the  well-devel- 
oped pseudobranch.  Six  branchiostegal 
rays,  four  on  the  ceratohyal,  two  on  the 
epihyal.  Scapula  prominent.  Vertebrae  12 
to  14  +  17  or  18  =  29  to  32.  Two  hypurals 
and  three  epurals.  Sclerotic  bones  well  os- 
sified. Stomach  large,  a  simple  sac;  intes- 
tine very  long.  Pyloric  caeca  numerous,  in 
a  dendritic  mass.  Air  bladder  present,  ex- 
tending the  length  of  the  abdominal  cavity. 
Color  in  preservative  brown,  bluish,  or 
silvery.  First  dorsal  blackish,  pelvic  fins 
dark  or  clear,  other  fins  usually  light.  Color 
pattern  may  be  blue  above,  silvery  below, 
the  shades  not  intergrading,  uniformly  dark, 
or  light  with  dark  spots.  Young  have  three 
to  five  dark  vertical  stripes.  Opercles  sil- 
very or  blackish.    Iris  usually  golden,   di- 


Stromateoid  Fishes  •  Hacdrich 


93 


vided  by  a  dark  vertical  bar.  Mouth  and 
gill  cavity  light  or  dark.  Peritoneum  silvery. 

Natural  history.  Most  stromateoids  are 
pelagic,  but  Ariomma  is  a  bottom  or  near- 
bottom  fish  of  deep  water.  Very  small 
Ariomma  are  pelagic,  for  they  are  taken  in 
surface  collections,  but  all  large  specimens 
reported  have  been  taken  in  bottom  trawls, 
usually  at  depths  in  excess  of  100  meters 
(Lowe,  1962;  McKenney,  1961;  Poll,  1959). 
Ariomma  apparently  schools,  for  single  net 
hauls  contain  numerous  specimens  (Poll, 
1959). 

The  pharyngeal  sacs,  but  not  the  stom- 
achs, of  most  specimens  dissected  contained 
grit  and  mud,  indicative  of  a  bottom-feed- 
ing habit.  This  habit  could  account  for  the 
relatively  high  number  of  gill-rakers,  around 
30,  found  in  Ariomma.  The  stomachs  of 
specimens  examined  by  Poll  ( 1959 )  con- 
tained bits  of  crustaceans  and  unidentifi- 
able meat.  The  thick  adipose  tissue  on  the 
head  probably  protects  the  eyes  and  nos- 
trils as  the  fish  scoops  its  prey  from  the  sea 
floor. 

Most  Ariomma  mature  around  160  to  180 
mm  SL  (A.  regula  [McKenney,  1961];  A. 
Ieda7wisi  [Poll,  1959];  A.  indica;  A.  nigriar- 
gentea),  and  probably  do  not  grow  much 
larger  than  this.  Very  large  specimens  are 
known  from  the  Red  Sea  (800  mm,  Klun- 
zinger,  1884),  Japan  (356  mm  SL,  Abe, 
1954),  and  Hawaii  (635  mm,  Fowler,  1923). 
The  Japanese  species,  at  least,  does  not 
seem  to  mature  before  reaching  this  size. 

Species.  The  problem  of  delineating  the 
species  of  Ariomma  is  one  of  the  most  per- 
plexing in  the  entire  suborder.  Fourteen 
have  been  described,  but  all  of  these,  from 
the  Gulf  of  Mexico  to  Hawaii,  have  almost 
the  same  numbers  of  finrays  and  gill-rakers. 
Reports  of  lateral  line  scale  counts  vary 
slightly,  but  the  scales  are  so  deciduous  in 
Ariomma  that  this  information  must  be 
viewed  with  caution. 

The  species  of  Ariomma,  with  one  inter- 
mediate exception,  are  either  elongate,  with 
the  maximum  depth  less  than  30  per  cent  of 
the  standard  length,  or  deep-bodied,  with 


the  maximum  depth  greater  than  40  per 
cent  of  the  standard  length.  It  is  possible 
that  these  two  groups  warrant  subgeneric 
recognition,  but,  pending  further  study,  this 
action  is  deferred. 

The  nominal  species  in  Ariomma  are: 

Ariomma  regidiis  (Poey,  1868)  =  Psenes 
regidus  Poey.  Gulf  of  Mexico  to  British 
Guiana,  type  locality  Cuba.  Deep-bodied. 
D  XI,  I  15.  A  III  15.  P  21-24.  Gill-rakers 
usually  7  +  1  +  15.  ( From  McKenney, 
1961.)    Spotted.    Teeth  not  cusped. 

Ariomma  indica  (Day,  1870)  =  *P.sene.s 
indicus  Day.  India  to  southern  Japan,  type 
locality  Madras.  Deep-bodied.  D  XI-XIl, 
14-15.  A  III  14-15.  P  21-23.  Gill-rakers 
usually  7  +  1  +  15.  Vertebrae  12  or  13  + 
18.  The  pectoral  fin  of  this  silvery  species 
becomes  produced,  up  to  35  per  cent  of  the 
standard  length,  with  growth.  The  teeth 
in  the  hind  part  of  the  lower  jaw  are  three- 
cusped;  the  rest  are  simple.  Psenes  extra- 
neiis  Herre,  1950,  known  from  a  single 
Philippine  specimen,  is  very  likely  a  syn- 
onym. 

Ariomma  brevimanus  (Klunzinger,  1884) 
=  Ctibiceps  brevimanus  Klunzinger.  Red 
Sea.  Elongate,  known  from  a  single  speci- 
men 800  mm  long.  D  XI,  15.  A  II  15.  P  24. 
( From  Klunzinger,  1884. ) 

"^Ariomma  lurida  Jordan  and  Snyder,  1904. 
Hawaii.  Elongate.  D  XI-XII,  14-15.  A  III 
13-14.  P  20-21.  Gill-rakers  9  +  1  +  20.  Ver- 
tebrae 14  +  18.  This  species  is  distinguished 
from  *A.  cvcrmanni  Jordan  and  Snyder, 
1907,  by  the  large  eye,  greater  than  30  per 
cent  of  the  length  of  the  head,  and  fewer 
pectoral  finrays. 

*  Ariomma  evermanni  Jordan  and  Snyder, 
1907.  Hawaii.  Elongate.  D  XI-XII,  15.  A 
III  14.  P  25.  Gill-rakers  9  +  20.  Vertebrae 
13  +  18.  Eye  is  less  than  28  per  cent  of  the 
length  of  the  head.  Attains  a  large  size; 
Cubiceps  thompsoni  Fowler,  1923  (type 
635  mm  long),  is  a  probable  synonym. 

Ariomma  africami  (Gilchrist  and  von 
Bonde,  1923)  =  '^Psenes  africanus  Gilchrist 
and  von  Bonde.  South  Africa.  Deep-bodied. 
D  IX-X,  15.    A  III  16.    P  22.    Gill-rakers  8 


94 


Bulletin  Muscutn  of  Comparative  Zoology,  Vol.  135,  No.  2 


+  1  +  16.  Vertebrae  13  +  18.  Very  similar 
to  A.  re<iu1m  (Poey,  1868).  Spotted.  Teeth 
not  eusped. 

Ariomma  dollfusi  ( Chabanaud,  1930)  = 
'^Cubicc'ps  doUfiisi  Chabanaud.  Gulf  of 
Suez.  Intermediate,  maximum  depth  of  co- 
type  32  per  cent  of  the  standard  length  (112 
mm).  D  XI-XII,  15.  A  III  15.  P  22.  Gill- 
rakers  7  +  1  +  14.  Vertebrae  12  +  18.  The 
teeth  are  said  to  be  "comprimees  .  .  .  et 
crenelees"  (Chabanaud,  1930:520),  sug- 
gesting close  relationship  or  synonymy  with 
A.  indica  (Day,  1870). 

'■Ariomma  hondi  Fowler,  19.30.  Grenada, 
British  West  Indies.  Elongate,  known  from 
the  holotype,  79  mm  TL.   D  XI-XII,  14.   A 

II  15.  Gill-rakers  8  +  15.  Lateral  line  scales 
43.  Dark  above,  light  on  sides.  (From  Fow- 
ler, 1930.)  Possible  synonyms  are  *A.  nigri- 
argcntca  and  or  *A.  melana,  both  of  Gins- 
burg,  1954. 

Ariomma  ledanoi.si  (Belloc,  1937)  =  Para- 
cuhiceps  ledanoisi  Belloc.  West  equatorial 
Africa.  Elongate.  D  XI-XII,  14-15.  A  III 
14-15.  P  20-22.  Gill-rakers  ?  +  ?  +  16-17. 
(From  Poll,  1959.) 

Ariomnia  niiiriargcntca  (Ginsburg,  1954) 
=  *Cid)icepii  nigriargenteus  Ginsburg.  Gulf 
of  Mexico,  Caribbean,  and  north  to  Cape 
Cod,  type  locality  Cape  Romain,  South 
Carolina.    Elongate.    D  XI-XII,   15-16.    A 

III  15.  P  21-22.  Gill-rakers  9-10  +  17-19. 
(From  Ginsburg,  1954. )  Vertebrae  13  +  17. 
Said  to  differ  from  A.  meJana  ( Ginsburg, 
1954)  by  the  smaller  scales  (62  to  68  in 
lateral  line),  color  (blue  above,  silvery 
below),  less  scalation  on  the  head,  and 
shorter  maxillary.  These  characters,  how- 
ever, seem  to  intergrade. 

Ariomma  melana  (Ginsburg,  1954)  = 
'^Cuhiceps  meJanti.s  Ginsburg.  Gulf  of  Mex- 
ico, Caribbean,  and  north  to  Cape  Hatteras, 
tvpe  locality  Mississippi  Delta.  Elongate. 
li)  XI-XII,  15.  A  III  14-15.  P  21-22.  Gill- 
rakers  9-11  +  18-20.  Lateral  line  scales  39- 
56.  ( From  Ginsburg,  1954. )  Vertebrae  ap- 
parently 15  +  16.    Uniform  dusky  l)rown. 

Ariomma  midti.'iqiiami.s  (Marchal,  1961) 
=  *raracuhice]).'i     m\dtis(piamis      Marchal. 


West  equatorial  Africa.  Elongate.  D  XI- 
XII,  1.5-16.  A  III  14-16.  P  21-23.  (From 
Marchal,  1961.)  Gill-rakers  9  +  1  +  18.  Said 
to  differ  from  A.  ledanoisi  (Belloc,  1937) 
by  having  more  scales  in  the  lateral  line 
(61-63  t;s.  36-40). 

Family  TETRAGONURIDAE 

Type  genus:   Tetragonurus  Risso,  1810 

Tetragonuridiie.  Risso,  1826:382  (def. ).  Liitken, 
1880:437  (disc,  rel.  to  scombroids).  Ramsay 
and  Ogilby,  1888:9  (disc,  rel.  to  Atherinidae). 
Resan,  1902:206  (rel.  to  Stromatcidae).  Boul- 
enger,  1904:642  (popular  account).  Grey, 
1955: 1    ( world-wide  revision  ) . 

Tetragonurina.  Giinther,  1861:407  (def.,  rel.  to 
Atherinidae ) . 

Tetragonuroidei.  Berg,  1940:323  (definition);  1955: 
247   (definition).    Smith,   1953:53   (review). 

Diagnosis.  Elongate  stromateoid  fishes 
with  pelvic  fins  present  in  the  adults,  two 
dorsal  fins,  teeth  on  vomer  and  palatines, 
five  or  six  branchiostegal  rays,  heavy  adher- 
ent keeled  scales,  and  four  hypural  and  two 
epural  bones  in  the  tail.  Papillae  in  the 
pharyngeal  sacs  with  rounded  bases,  not  in 
bands. 

Description.  Body  slender,  rounded.  Pe- 
duncle thick,  square  in  cross-section,  with 
modified  scales  forming  two  prominent 
lateral  keels  on  each  side.  Two  dorsal  fins, 
the  first  with  10  to  20  short  spines,  folding 
into  a  groove;  the  base  of  the  fin  as  long  as 
or  longer  than  the  base  of  the  second  dor- 
sal. One  anal  spine,  not  separated  from  the 
rays.  Soft  dorsal  and  anal  fins  approxi- 
mately the  same  length,  with  10  to  17  fin- 
rays.  Last  ray  of  pelvic  fin  attached  to 
abdomen  for  its  entire  length,  fin  folding 
into  a  depression.  Scales  moderate  in  size, 
ctenoid,  with  heavy  longitudinal  ridges, 
very  adherent,  arranged  in  a  geodesic  pat- 
tern around  the  body.  Lateral  line  slightly 
arched  forward,  descending  to  run  along 
middle  of  side  and  extending  onto  peduncle; 
no  tubed  scales.  Skin  thick;  subdermal 
mucous  canal  system  well  developed,  but 
barely  visible.  Opercular  and  preopercular 
margins  entire  or  finely  denticulate.  Oper- 
cle  thick,  spines  not  apparent.    Five  or  six 


Stromateoid  Fishes  •  Haedricii        95 


EPURALS 


HYPURAL   6 


HYPURAL  4  +  5 


HYPURAL  2+3 


Figure  36.      Caudal   skeleton   of   Tefraganurus   atlanticus,   drawing   of  a   cieored-and-stained    preparation   from   a  66-mm   speci- 
men, MCZ  41791.    All  elements  identified  in  Figure  1. 


branchiostegal  rays.  Mouth  large,  maxil- 
lary extending  below  eye.  Teeth  moderate 
to  large,  simple  and  cusped,  uniserial  in  the 
jaws.  Vomer,  palatines,  basibranchials,  and 
usually  the  tongue  with  teeth.  Supramaxil- 
lary  bone  absent.  Eye  large,  no  adipose 
tissue.  Sclerotic  bones  not  well  ossified. 
Vertebrae  43  to  58.  Caudal  skeleton  with 
four  hypurals  and  two  epurals.  Pharyngeal 
sacs  with  small  papillae  in  upper  and  lower 
halves;  bases  of  papillae  not  in  bands, 
rounded,  central  stalk  with  a  few  teeth. 
Adults  one  to  two  feet  in  length.  Unifonn 
dark  brown,  with  no  pattern  or  counter- 
shading. 

Distribution.  The  distribution  of  the 
Tetragonuridae  largely  parallels  that  of  the 
nomeids  (Fig.  54).  Tetragonurids  are  oceanic 
fishes  of  tropical,  subtropical,  and  temper- 
ate seas.  None  have  been  taken  in  the  east- 
em  Mediterranean  Sea,  the  Red  Sea,  and 
the  South  and  East  China  seas. 


ReIotionsJu})s.  Tetiagomirus,  the  single 
genus  in  the  family,  has  teeth  on  the  vomer, 
palatines,  and  basibranchials.  Thus,  it  is 
affiliated  with  the  nomeid  stock.  The  cau- 
dal skeleton  (Fig.  36)  is  similar  to  that  of 
the  nomeids,  but  is  advanced  an  evolution- 
ary grade  in  having  lost  one  of  the  epurals. 
The  pharyngeal  sacs  (Fig.  37)  and  the 
heavy,  keeled  scales  of  Tetragonuriis,  how- 
ever, are  markedly  different  from  the  sacs 
and  scales  of  the  nomeids.  The  pharyngeal 
sacs  are  exceedingly  elongate.  The  papillae 
are  poorly  ossified  and  are  very  reduced  in 
size.  The  bases  of  the  papillae  are  rounded, 
as  they  are  in  the  Ariommidae,  but  there 
are  less  than  six  teeth  seated  on  top  of  a 
short  stalk.  The  fourth  pharyngobranchial 
is  very  elongate  and  is  fused  to  the  third 
pharyngobranchial.  This  long  bone  is 
studded  with  teeth  and  extends  well  back- 
ward into  the  sac,  where  it  no  doubt  aids 
both  in  shredding  the  food  and  in  support- 
ing the  sacs.   All  nomeids  have  six  branchi- 


96 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


) 


UPPER    PHARYNGEAL   BONE. 


0    \   'T\    O     O 


:  3  1 


r 


r 


r 


^•'    "tt     ,c 


Figure  37.  Branchial  region  of  Tetragonurus  cuvieri,  drawing  of  a  cleared-and-stoined  preparation  from  a  340-mm  specimen, 
CNHM  64218.  Elements  identified  in  Figure  2.  Tfie  upper  pfiaryngeal  bone  is  formed  by  tfie  fusion  of  tfie  third  and  fourth 
pharyngobranchials. 


ostegal  rays;  tetragon iirids  have  either  six 
or  five. 

The  highly  differentiated  jaw  teeth  of 
Tetrogominis  are  very  similar  to  those  of 
certain  species  in  the  genus  Psenes.  But 
Tetragonurus  has  teeth  on  the  tongue,  and 
cannot  be  derived  from  Psenes.  It  is  very 
likely  that  Tetragonurus  branched  off  fairly 
early  from  the  nomeid  stem  and  is  derived 
from  no  living  nomeid  genus.  The  loss  of 
an  epural  and  a  branchiostegal  ray,  the 
heavy  keeled  scales,  the  very  elongate  pha- 
ryngeal sacs  with  the  reduced  papillae,  and 
the  greatly  increased  number  of  vertebrae 
are  quite  divergent  from  the  situation  in 
nomeids,  and  together  imply  that  evolution 
has  proceeded  independently  in  these  two 
groups  for  some  time. 

The  divergent  characters  of  Tetragonurus 
are  part  of  its  adaptation  to  a  very  particu- 
lar mode  of  life.  Tetragonurus  is  certainly 
a  derived  form,  and  is  probably  quite  unlike 
the  ancestral  nomeid.  The  central  nomeid 
genus  Cuhiceps,  like  Tetragonurus,  has 
teeth  on  the  tongue.  The  nomeid  stock 
from  which  Tetragonurus  arose  may  have 
been  in  many  respects  similar  to  the  pres- 
ent-day Cuhiceps. 

Genus  TETRAGONURUS  Risso,  1810 

Figure  38 

Tetragonurus  Ris.so,  1810:347.  (Type  .species: 
Tetragonurus  cuvieri  Ri.sso,  1810:347,  by 
monotypy.    Mediterranean. ) 


Ctenodax  Macleay,  1885:718.  (Type  species: 
Ctenodax  wilkinsoni  Macleay,  1885:718,  by 
monotypy.  Lord  Howe  Island,  Tasman  Sea. 
A  synonym  of  *Tetragonurus  atlantieus  Lowe, 
1839:79.) 

The  combination  of  elongate  body  and 
peduncle,  modified  scales  forming  two  keels 
on  the  peduncle,  origin  of  first  dorsal 
slightly  to  well  behind  pectoral  insertion, 
base  of  first  dorsal  longer  than  base  of  sec- 
ond dorsal,  heavy  keeled  scales,  and  pecu- 
liar lower  jaw  with  heavy  knifelike  teeth 
distinguishes  Tetragonurus  from  all  other 
stromateoid  genera.  The  name,  a  mascu- 
line noun,  is  from  the  Greek  t€t pay ojvo'i,  with 
four  angles,  +  ovpd,  tail,  in  reference  to  the 
shape  of  the  caudal  peduncle. 

Description.  Body  very  elongate,  maxi- 
mum depth  less  than  20  per  cent  of  the 
standard  length,  rounded;  musculature  firm. 
Peduncle  long,  thick,  with  modified  scales 
forming  two  prominent  keels  at  base  of 
caudal  fin.  Two  dorsal  fins,  scarcely  di- 
vided. First  dorsal  originating  slightly  to 
well  behind  insertion  of  pectoral  fins,  with 
10  to  21  stiff  spines  folding  into  a  groove, 
the  longest  spine  half  the  length  of  the 
longest  ray  of  the  second  dorsal.  Anterior 
rays  of  the  second  dorsal  the  longest,  those 
that  follow  decreasing  slightly  in  length,  10 
to  17  finrays  in  all.  Anus  well  behind  mid- 
body,  in  a  depression.  Anal  fin  originating 
shghtly  behind  origin  of  second  dorsal  fin. 


Stromateoid  Fishes  •   Haedricli 


97 


■—■■■■,,"•"  '■■-■-  ■  ■  ■•^~•-■■-■■.^■■■'=^37;i?M9t15^;s<MSsass^ 


Figure  38.      Tetragonurus  cuvieri,  drawing  of  a  129-mm  specimen,  from  Grey,   1955. 


2  cm 


one  short  spine  preceding  the  rays.  Anterior 
rays  the  longest,  those  that  follow  decreas- 
ing slightly  in  length,  9  to  15  finrays  in  all. 
Pectoral  tin  small,  the  central  rays  the  long- 
est. Pelvic  fins  small,  inserting  behind  pec- 
toral fin  base  and  before  origin  of  first  dor- 
sal, innermost  ray  attached  to  abdomen  for 
its  entire  length.  Caudal  fin  forked.  Scales 
moderate  in  size,  with  heavy  longitudinal 
keels,  very  adherent,  following  a  geodesic 
pattern  around  the  body.  Very  small  scales 
on  bases  of  median  fins.  Lateral  line  usu- 
ally slightly  arched  anteriorly,  descending 
to  run  along  mid-lateral  line  of  body  and 
ending  on  peduncle  at  origin  of  keels;  no 
tubed  scales.  Skin  thick;  sulidennal  canals 
cannot  be  traced.  Pores  to  surface  minute. 
Head  30  to  20  per  cent  of  the  standard 
length.  Top  of  head  and  snout  naked,  small 
pores  in  naked  skin.  Scales  extending  for- 
ward over  nape  almost  to  level  of  posterior 
border  of  the  eye.  Eye  large,  with  a  series 
of  grooves  on  the  posterior  rim.  Nostrils 
toward  top  of  blunt  snout,  large,  well  sepa- 
rated, the  anterior  round,  the  posterior  a 
slit.  Maxillary  ending  under  eye,  angle  of 
gape  well  before  eye.  Premaxillary  not  pro- 
tractile. Lacrimal  bone  covering  most  of 
upper  jaw  at  all  times,  ventral  border  of 
maxillary  remaining  exposed.  Lower  jaw 
almost  completely  within  upper  jaw  when 
mouth  is  closed.  Supramaxillary  absent. 
Teeth  in  upper  jaw  small,  pointed,  recurved, 
spaced.  Teeth  in  lower  jaw  large,  knifelike, 
close  set,  with  small  cusps,  deeply  em- 
bedded in  the  gum  with  only  the  tips  show- 
ing. Strong  recurved  teeth  present  on  head 
of  vomer,  and  in  a  single  series  on  shaft  of 


vomer  and  palatines.  Small  teeth  on  basi- 
branchials  and,  usually,  profusely  scattered 
on  tongue.  Tongue  high-sided,  depressed 
in  center.  Opercle  and  preopercle  thin, 
fleshy,  scaled,  margins  entire  in  adult, 
spinulose  in  the  young;  opercle  with  two 
very  weak  spines;  angle  of  preopercle 
slightly  rounded,  not  bulging  backward. 
Cheeks  scaled.  Gill-rakers  broad,  fleshy, 
shorter  than  the  filaments,  toothed  on  inner 
edge,  spaced,  8  to  14  on  lower  limb  of  first 
arch;  rudimentary  rakers  often  present  un- 
der the  large  pseudobranch.  Scapula  not 
visible.  Vertebrae  43  to  58.  Sclerotic  bones 
well  ossified  in  adult.  Stomach  a  simple 
sac;  intestine  long.  Pyloric  caeca  numerous, 
in  a  large  dendritic  mass. 

Color  in  preservative  uniform  brown, 
ranging  from  tan  to  almost  black.  Fins  the 
same  color  as  the  body.  Inside  of  mouth, 
gill  cavity,  and  peritoneum  dark. 

Natural  history.  Young  Tetragonurus  have 
been  found  associated  with  medusae  ( Man- 
sueti,  1963).  There  are  also  reports  (Emery, 
1882;  Lo  Bianco,  1909;  Fitch,  1949)  of 
young  specimens  found  within  salps,  usu- 
ally Pyrosoma.  Mansueti  (1963)  felt  that 
the  association  of  Tetragonurus  with  jelly- 
fishes  was  a  chance  occurrence,  but,  since 
jellyfish  associations  are  commonly  formed 
by  other  stromateoids,  it  is  likely  that  the 
association  is  actively  sought. 

Tetragonurus  probably  feeds  almost  ex- 
clusively on  coelenterates  and  ctenophores 
(Risso,  1826;  Fitch,  1952).  The  large  sHc- 
ing  teeth  of  the  lower  jaw  and  the  peculiar 
boxlike  jaw  seem  admirably  suited  for  such 
a  diet   (Grey,  1955). 


98 


Bulletin  Musciini  of  Comparative  Zoology,  Vol.  135,  No.  2 


Risso  (1826)  reported  that  the  flesh  of 
Tetragonurus  was  poisonous,  attributing 
this  quahty  to  the  fish's  diet  of  venomous 
jell\'fishes  of  the  genus  Stcphanomic.  His 
report  has  been  widely  spread,  but  has  only 
recently  been  reinvestigated.  Fitch  ( 1952 ) 
analyzed  four  California  specimens  and 
found  them  to  be  not  poisonous.  The  pos- 
sibility remains  that  Tetragonunis  is  poi- 
sonous only  during  certain  seasons,  for 
example  at  the  time  of  spawning  ( Petit  and 
Amar,  1946). 

In  the  Atlantic,  Tetragonurus  cuvieri  ap- 
parently spawns  in  spiing  and  summer 
(Grey,  1955).  Guiglia  (1950)  reports  ma- 
ture females  of  T.  cuvieri  taken  throughout 
the  year  in  the  Mediterranean.  T.  atlonticus 
spawns  during  the  fall  in  the  eastern  and 
northern  Atlantic,  but  in  winter  and  spring 
in  the  western  Atlantic  (Grey,  1955). 

Tetragonurus  is  strictly  oceanic  (Grey, 
1955).  Although  young  specimens  occur 
near  the  surface  with  jellyfish,  the  adults, 
judging  from  their  somber  coloration,  prob- 
ably are  members  of  the  meso-  or  ]:)athy- 
pelagic  faunas.  Tetragonurus  has  been 
considered  a  rare  fish  (Fitch,  1949),  but 
recent  surveys  in  the  North  Pacific  (Lark- 
ins,  1964)  indicate  that  it  is  much  more 
common  than  is  usually  thought,  bearing 
out  a  prediction  of  Grey  ( 1955). 

Species.  Tetragonurus  has  been  ably 
treated  by  Marion  Grey  (1955),  who  rec- 
ognized three  species  in  the  genus.  As  she 
has  pointed  out,  each  species  varies  widely, 
and  further  division  may  be  justified  when 
more  specimens  are  available.  The  species 
are: 

T.  cuvieri  Risso,  1810.  Mediterranean  Sea, 
Atlantic,  Pacific  oceans.  D  XV-XXI,  10-17. 
A  I  10-15.  P  14-21  (?).  Lateral  line  scales 
97-114.  Vertebrae  52-58.  (From  Grey, 
1955. ) 

*T.  atlanticus  Lowe,  1839.  Atlantic,  Pa- 
cific, Indian  oceans.  D  XIV-XVII,  10-13. 
A  I  9-12.  P  14-18.  Lateral  line  scales  83- 
95.  Vertebrae  45-51.  (From  Grey,  1955.) 
Ctenodax  icilkinsoni  Macleay,  1885,  is  a 
synonym. 


T.  pacificus  Abe,  1953.  Pacific,  Indian 
oceans.  D  X-XI,  10-12.  A  I  10-12.  P  15- 
17.  Lateral  line  scales  73-78.  Vertebrae 
40(?)-43.    (From  Grey,  1955.) 

Family  STROMATEIDAE 

Type  genus:    Sfromateus  Linnaeus,   1758 

Stromatees.     Cuvier   and    Valenciennes,    1833:372 

( in  part,  descr. ) . 
Stromateina.     Giinther,   1860:397    (in   part,   def. ). 

Gill,   1862:126   (genera  listed). 
Stroniateinae.     Gill,    1884:669    (def.,   gen.).    Bi\h- 

ler,   1930:62   (digestive  system). 
Stromateidae.      Jordan      and      Gilbert,      1882:449 

(descr.).     Jordan    and    Evermann,     1896:964 

(descr.,    North    America).     Jordan,    1923:182 

(list,  +  Pampidae).    Berg,  1940:323   (dist.); 

19.5.5:248   (dist.). 

Diagnosis.  Deep-bodied  stromateoid 
fishes  with  pelvic  fins  absent  in  the  adults, 
continuous  dorsal  fin,  toothless  palate,  four 
hypural  and  two  or  three  epural  bones  in 
the  tail,  and  well  ossified  sclerotic  bones. 
The  papillae  in  the  pharyngeal  sacs  not  in 
bands,  in  both  the  upper  and  lower  halves 
of  the  sac;  bases  of  the  papillae  stellate. 

Description.  Body  deep,  compressed. 
Single  dorsal  and  anal  fins,  with  none  or 
one  to  ten  flat,  bladelike  spines  and  three 
to  five  slender,  graduated  spines  preceding 
the  rays.  Median  fins  about  the  same  length, 
usually  falcate;  caudal  fin  deeply  forked. 
Pectoral  fin  long  and  pointed.  Pelvic  fins 
present  only  in  young  Stromateus;  absent 
in  all  others.  Scales  small,  cycloid,  ex- 
tremely deciduous.  Lateral  line  high,  fol- 
lowing dorsal  profile,  and  extending  onto 
the  short  peduncle.  Opercular  and  preoper- 
cular  margins  entire.  Opercle  very  thin, 
with  two  short,  flat,  weak  spines.  Gill  mem- 
branes usually  free  from  isthmus,  but  united 
in  Pampus.  Five  or  six  branchiostegal  rays. 
Mouth  temiinal  to  sub-terminal,  small,  an- 
gle of  gape  rarely  reaching  below  eye. 
Teeth  very  small,  laterally  flattened,  with 
three  minute  cusps,  and  uniserial  in  the 
jaws.  Vomer,  palatines,  and  basibranchials 
toothless.  Supramaxillary  absent.  Eye  fairly 
small;  adipose  tissue  usually  not  well  de- 
veloped.  Sclerotic  bones  well  ossified.  Ver- 


Stromateoid  Fishes  •  Haedrich 


99 


tebrae  30  to  48.  Caudal  skeleton  with  four 
hxpurals  and  t\\o  epurals,  except  three 
epurals  in  Stromafeus  fiatola.  Pharyngeal 
sacs  with  papillae  in  upper  and  lower  halves, 
not  in  bands;  bases  of  the  papillae  stellate, 
with  teeth  seated  all  along  a  central  stalk. 
Adults  usually  about  a  foot  in  length.  Sil- 
very to  blue,  some  with  spots. 

Distribution.  Stromateids  live  over  the 
continental  shelves  and  in  the  bays  of  tropi- 
cal, subtropical,  and  temperate  waters. 
They  are  found  on  the  east  and  west  coasts 
of  North  and  South  America,  from  the 
Mediterranean  Sea  to  South  Africa,  and 
from  the  Iranian  Gulf  to  Japan  (Fig.  56). 
None  occur  near  oceanic  islands,  and  none 
have  reached  Australia.  Stromateus,  in 
southern  South  America  and  western  Africa, 
is  the  only  genus  that  has  representatives 
on  both  sides  of  an  ocean.  The  genera  are 
allopatric  except  for  a  small  area  of  overlap 
between  Stromateus  and  PepriJus  in  South 
America.  All  stromateids  school,  and  many 
are  important  commercially. 

Relationships.  The  Stromateidae  are  the 
current  zenith  in  stromateoid  evolution. 
The  reduced  number  of  branchiostegals  and 
elements  in  the  caudal  skeleton  (Figs.  42, 
47),  the  absence  of  pelvic  fins,  the  small 
mouth  with  cusped  teeth,  the  broad  stellate 
bases  of  the  papillae  in  the  pharyngeal  sacs 
(Figs.  43,  46),  and  the  deepened  body  are 
all  advanced  conditions. 

The  stromateids  were  derived  from  some- 
where near  the  Seriolella  group  of  the  Cen- 
trolophidae,  perhaps  from  a  fish  very  like 
the  deep-bodied  Psenopsis.  In  Stromateus 
fiatola  the  presence  of  peKic  fins  in  the 
young  and  the  three  epurals  recalls  the 
centrolophid  heritage. 

The  stromateids  may  have  evolved  only 
recently.  Though  the  three  genera  are  dis- 
tinct, little  speciation  has  occurred,  but,  in 
some  cases,  seems  to  be  in  an  incipient 
stage.  Almost-sibling  species  occur  on 
either  side  of  the  Isthmus  of  Panama. 

The  stromateids  are  an  end-point,  and  no 
other  groups  have  been  deri\ed  from  them. 
But  they  are  by  no  means  an  evolutionar>' 


deadend.  Pampus,  with  its  restricted  gill- 
opening,  elongate  pharyngeal  sac,  and  pe- 
culiar spines  in  some  species,  is  diverging 
rapidly  from  the  central  stromateid  bauplan 
and  is  widely  successful  along  the  coasts  of 
southern  Asia. 

Key  to  Stromateid  Genera 

1  (4).    Inter-  and  subopercles  not  united  to 

the  isthmus.  End  of  maxillary  before 
or  at  anterior  border  of  eye.  Cusps 
on  teeth  in  lower  jaw  subequal,  the 
teeth  appearing  truncate  to  the  naked 
eye.  Spine  on  end  of  pelvic  bone 
present  or  absent.  In  small  specimens 
( less  than  80  mm  SL )  pelvic  fins 
present  or  absent.  Six  branchiostegal 
rays.    2 

2  (3).    One    to    three    flat,    bladelike    spines 

ahead  of  median  fins.  A  small  spine 
projecting  posteroventrally  from  end 
of  pelvic  bone.  Median  fins  falcate 
or  not.    Pelvic  fins  never  present.    30 

to  35  vertebrae.   

Peprilus,  p.  103.  Figs.  40,  41 

3  (2).    No    flat,    bladelike    spines    ahead    of 

median  fins.  No  spine  at  end  of 
pelvic  bone.  Median  fins  never  fal- 
cate. Pelvic  fins  absent  in  adult,  but 
present  in  some  small  specimens.    40 

to  48  vertebrae.  

Stromateus,  p.  99.  Fig.  39 

4  (1  ).    Inter-  and  subopercles  broadly  united 

to  isthmus.  End  of  maxillar>'  under 
eye.  Central  cusp  on  teeth  of  lower 
jaw  much  larger  than  the  other  two 
cusps,  which  can  hardly  be  seen  with- 
out extreme  magnification.  No  spine 
at  end  of  pelvic  bone.  Pelvic  fins 
never  present.  Five  branchiostegal 
rays.  Pampiis,  p.  108.  Figs.  44,  45 

Genus  STROMATEUS  Linnaeus,  1758 

Figure  39 

Stromateus  Linnaeus,  1758:248.  (Type  species: 
Stromateus  fiatola  Linnaeus,  1758:248,  by 
monotypy.     Mediterranean. ) 

Chrysostromus  Lacepede,  1802:697.  (Type  spe- 
cies: Chrysostromus  fiatoloides  Lacepede, 
1802:697,  by  monotypy.  Mediterranean.  A 
synon>Tn  of  Stromateus  fiatola  Linnaeus,  1758: 
248.) 

Fiatola  Cuvier,  1817:342.  (Tvpe  species:  Stroma- 
teus fiatola  Linnaeus,  1758:248,  by  monotypy. 
Mediterranean. ) 

Seserinus  Cuvier,  1817:342.  (Type  species:  "Seser- 
imts  Rondelet"  [Seserinus  rondeleti]  Cuvier, 
1817:343,  by  subsequent  designation  of  Jor- 


100         Bulletin  Museum  of  Comparatwe  Zoology,  Vol.  135,  No.  2 


•"^'^n/.-f^m 


"'^ 


'M 


^vc^fes 


>■?»' 

V 


Figure  39.     S/romofeus  fiatola,  drawing  of  a  228-mm  specimen,   from   Poll,    1959 


^iJSfeh- 


dan,  1923:106.  Mediterranean.  A  synonym 
of  Stwmatcus  fiatola  Linnaeus,  1758:248.) 
Ptcrorhomhus  Fowler,  1906:118.  (Subgenus.  Type 
species:  Fiatola  fasciata  Risso,  1826:289,  by 
original  designation.  Mediterranean.  A  syn- 
onym of  Stwmatcus  fiatola  Linnaeus,  1758: 
248.) 

The  combination  of  deep  body,  small 
eye,  moderate  pectoral  fin,  no  bladelike 
spines  ahead  of  the  median  fins,  no  ventral 
spine  on  the  pelvic  bone,  spotted  body  pat- 
tern, and  (sometimes)  pelvic  fins  in  the 
young  distinguishes  Stromateus  from  all 
other  stromateoid  genera.  The  name,  a  mas- 
culine noun,  is  from  the  Greek  (TTpwfiaTev;, 
a  brightly  colored  quilt  or  bedding,  prob- 
ably in  reference  to  the  shape  and  pattern 
of  the  fish. 

Description.  Body  deep,  maximum  depth 
generally  greater  than  40  per  cent  of  the 
standard  length,  compressed;  musculature 
firm.  Peduncle  very  short,  compressed.  Dor- 
sal fin  continuous,  originating  over  base  of 
the  pectoral  fins,  the  anteriormost  elements 


usually  very  small  and  buried  in  the  skin, 
44  to  53  elements  in  all.  Anal  papilla  before 
mid-body,  in  a  deep  slit.  Anal  fin  origi- 
nating slightly  before  middle  of  body  but 
behind  dorsal  origin,  the  small  anteriormost 
elements  buried  in  the  skin,  35  to  47  ele- 
ments in  all.  Anterior  rays  of  the  median 
fins  produced,  two  to  three  times  longer 
than  the  subequal  rays  of  the  posterior  two- 
thirds  of  the  fin,  the  lobes  rounded  and  not 
falcate.  No  bladelike  spines  preceding  me- 
dian fins.  Pectoral  fin  moderate  in  length, 
broad.  Pelvic  fins  present  in  the  young  of 
at  least  some  species,  inserting  under  end 
of  pectoral  fin  base,  the  fins  lost  in  adult 
but  with  two  dark  flaps  of  skin  sometimes 
indicating  their  former  presence.  Pelvic 
bone  usually  not  visible  on  mid-line  and 
lacking  a  ventral  spine.  Caudal  fin  stiff, 
deeply  forked,  the  lobes  very  long.  Scales 
small,  cycloid,  deciduous,  minute  scales 
covering  all  fins.  Simple  tubed  scales  of  the 
lateral  line  moderately  high,  following  dor- 


Strom ATEOiD  Fishes  •   Hacdrich         101 


sal  profile  and  extending  onto  peduncle  but 
not  to  caudal  base.  Skin  moderately  thick, 
subdermal  canals  not  visible,  pores  to  sur- 
face seem  to  be  wanting.  Head  around  25 
per  cent  of  the  standard  length,  very  deep 
and  broad.  Top  of  head  naked,  small  pores 
easily  seen,  naked  skin  underlain  with  nu- 
merous parallel  canals  projecting  slightly 
backward  over  the  nape.  Eye  small,  adi- 
pose tissue  around  eye  well  developed  and 
extending  forward  surrounding  the  nostrils. 
Nostrils  moderate  in  size,  the  anterior 
round,  the  posterior  a  slit,  located  nearer 
to  tip  of  blunt  snout  than  to  eye.  Mouth 
broad.  Maxillary  scarcely  reaching  anterior 
border  of  the  eye,  angle  of  gape  well  before 
eye.  Premaxillary  not  protractile.  Lacrimal 
bone  reduced,  scarcely  covering  top  of  up- 
per jaw  when  mouth  is  closed,  end  of  maxil- 
lary exposed.  Supramaxillary  absent.  Jaw 
teeth  minute,  uniserial,  laterallv  flattened, 
with  three  subequal  cusps,  close  set,  cov- 
ered laterally  by  a  membrane;  vomer,  pala- 
tines, and  basibranchials  toothless.  Gill 
membranes  broadly  united  across  the  isth- 
mus, divided  from  level  of  the  back  part  of 
the  eye.  Opercle  and  preopercle  thin, 
scaled,  margins  entire;  opercle  rounded, 
with  two  ill-defined,  weak  spines;  angle  of 
preopercle  broadly  rounded,  projecting 
backward  slightly.  Cheek  scaled.  Gill- 
rakers  a  little  less  than  half  the  length  of 
the  filaments,  diminishing  in  size  anteriorly, 
not  toothed,  fairly  close-set,  about  12  on  the 
lower  limb  of  the  first  arch;  no  rudimentary 
rakers  under  the  small  pseudolnanch.  Six 
branchiostegal  rays,  four  on  the  ceratohyal, 
two  on  the  epihyal.  Scapula  not  visible. 
Vertebrae  slightly  variable,  usually  16  +  26 
=  42  to  19  +  26  =  45.  Sclerotic  bones  well 
ossified.  Stomach  a  simple  sac;  intestine 
very  long.  Pyloric  caeca  numerous,  in  a 
long  dendritic  mass. 

Color  in  preservative  brown  or  bluish 
with  a  silvery  or  \\4iitish  overlay,  dark 
above,  lighter  below.  Back  and  sides  with 
numerous  dark  spots.  The  young  may  have 
four  or  five  dark  vertical  bands.  Fins 
darker  or  lighter  than  the  body;  pectoral  in 


some  species  blackish.  Gill  cavity,  inside 
of  mouth,  and  peritoneum  light. 

Natural  history.  The  young  of  Stromatetis 
commonly  associate  with  medusae  (Padoa, 
1956),  and  Lo  Bianco  (1909)  observed  them 
eating  jellyfish.  Fish  up  to  five  inches  in 
length  have  been  reported  in  association 
(Smith,  1949a),  but  the  majority  probably 
desert  their  coelenterate  host  before  reach- 
ing this  size. 

Poll  ( 1959 )  found  the  adult  common  in 
depths  from  12  to  50  meters  off  the  coast 
of  West  Africa.  I  have  examined  adults 
taken  by  the  Guinean  Trawling  Survey  in 
30  meters  of  water,  and  adults  from  Chile 
captured  with  a  trammel  net.  Adult  Stro- 
mateus  may  rarely  descend  to  deeper  water. 

Small  Stromatetis  fiatolo  have  a  vertically 
barred  pattern  and  small  pelvic  fins.  The 
bars  and  the  pelvics  are  lost  usually  before 
the  fish  reaches  100  mm  standard  length. 
At  this  point,  the  young  fish  probably 
moves  into  the  adult  habitat.  Whether  or 
not  the  young  of  South  American  Stroma- 
tetis have  pelvic  fins  is  unknown.  Poll 
(1959)  reports  a  Stromateus  fiatola  500  mm 
long  weighing  151  grams;  this  is  probably 
near  the  maximum  size  attained. 

In  the  Patagonian  region,  Stromatetis 
moves  shoreward  to  spawn  in  earh'  summer 
( Hart,  1946 ) .  At  this  time  the  fishes  form 
their  maximum  concentrations.  Following 
spawning,  they  move  offshore  during  the 
fall  and  winter,  and  become  widely  dis- 
persed. In  addition  to  inshore  and  offshore 
movement,  Hart  ( 1946 )  found  evidence 
that  Stromatetis  moves  from  lower  to  higher 
latitudes  in  the  summer,  and  back  in  the 
winter. 

Said  to  be  a  fish  with  "delicate  flesh  and 
fine  flavour"  (Gilchrist  and  von  Bonde, 
1923:11),  Stromateus  apparently  does  not 
receive  the  attention  it  deserves.  Once  de- 
scribed as  numerous  in  the  markets  at  Lima 
(Cuvier  and  Valenciennes,  1833),  it  is  to- 
day the  subject  of  only  small  local  fisheries 
in  parts  of  Chile  and  along  the  African 
coast. 

Relationships.    Stromateus    contains    the 


102         BiiUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


most  primitive  species  of  the  family  Stro- 
mateidae.  Stwinatcus  fiafola,  the  African 
species,  has  three  epural  elements  in  the 
caudal  skeleton  and,  when  young,  has  pel- 
vic fins.  These  characters  indicate  the  cen- 
trolophid  heritage  of  the  stromateids.  In 
the  South  American  species  of  Stromoteus, 
there  seem  to  be  only  two  epural  elements 
in  the  tail,  and  the  young  may  lack  pelvic 
fins.^  This  situation  is  typical  of  the  stro- 
mateid  grade. 

Stromateiis  has  a  very  high  number  of 
vertebrae,  more  than  forty,  an  advanced 
condition.  This  high  number  may  be  a  re- 
cently acquired  characteristic  of  the  genus. 
The  number  itself  is  variable,  and,  in  a 
large  proportion  of  cases,  fusions  of  the 
centra  occur  in  the  caudal  series.  Centra 
\\'ith  two  or  three  neural  and  haemal  spines 
appeared  in  four  of  the  fourteen  specimens 
radiographed.  The  variability  and  high  in- 
cidence of  fusions  suggests  a  genetic  in- 
stability perhaps  correlated  with  recent 
acquisition. 

Because  of  its  high  vertebral  count,  Stro- 
mateus  cannot  be  the  direct  ancestor  of 
either  of  the  other  two  stromateid  genera, 
both  of  which  have  lower  counts.  Rather, 
all  three  must  share  a  common  ancestor, 
a  fish  most  like  Stromatetis  but  with  a  ver- 
tebral count  somewhere  near  13  +  17. 

Species.  Stromateiis  lives  in  temperate  to 
tropical  waters  along  the  coasts  of  Medi- 
terranean countries,  West  Africa,  Argentina, 
and  Chile.  Though  numerous  species  have 
been  described,  there  seem  to  be  no  more 
than  two,  or  possibly  three,  valid  ones.  Spe- 
cies are  widespread  along  a  coastline,  but 
none  jump  ocean  barriers.  The  species  in 
the  genus  are: 

Stromateiis  fiatola  Linnaeus,  1758.  Medi- 
terranean, coast  of  West  Africa  south  to 
Capetown,  type  locality  Mediterranean  Sea. 
D  48-51  (total  elements).  A  35-38  (total 
elements).    P  22-24.    Cill-rakers  usually  3 


■■•  I  have  seen  no  small  Stromatetis  from  South 
America,  but  the  adults  lack  the  two  dark  flaps 
of  skin  wlu'ch  bespeak  the  former  presence  of  pel- 
vies  in  adult  African  Stromateiis. 


+  1  +  11.  Vertebrae  18-19  +  24-26.  Speci- 
mens less  than  100  mm  SL  have  pelvic  fins 
and  vertical  bars  on  the  sides.  The  fins  and 
the  bars  are  gone  in  adults.  The  names  of 
the  Mediterranean  Fiatola  fasciata  Risso, 
1826,  and  Seseriniis  microchiriis  Cuvier  and 
Valenciennes,  1833,  are  synonyms  based  on 
juveniles.  Other  synonyms  are  Chrijsostro- 
miis  fiatoloides  Lacepede,  1802,  from  the 
Mediterranean  and  Stromateiis  capensis 
Pappe,  1866,  from  South  Africa. 

Stromateiis  steUatus  Cuvier,  1829.  Pacific 
coast  of  South  America,  Chile  and  rarely 
north  to  Lima,  type  locality  coasts  of  Peru. 
D  44-53  (total  elements).  A  39-44  (total 
elements).  P  19-24.  Gill-rakers  around  4 
+  1  +  12.  Vertebrae  16  +  26-27.  *Stroma- 
teus  maculatus  Cuvier  and  Valenciennes, 
1833  (=S.  advectitius  Whitley,  1935),  a 
junior  synonym,  is  the  name  most  often 
used  for  this  fish.  *S.  maculatus  is  also  gen- 
erally applied  to  the  species  of  Stromateus 
which  occurs  along  the  Atlantic  coast  of 
South  America.  The  two  forms  are  very 
close  in  appearance  but  the  Chilean  form 
is  a  slenderer  fish  with  a  slightly  longer 
head.  The  counts  of  the  two  overlap  almost 
completely.  The  Atlantic  fonn  seems  to 
breed  near  the  northern  limit  of  its  range 
(Hart,  1946).  Nothing  is  known  concern- 
ing the  breeding  habits  of  the  Pacific  fonn, 
but  the  spawning  area  is  probably  well 
north  of  Tierra  del  Fuego.  It  seems  un- 
likely that  there  is  any  gene  exchange  be- 
tween the  two  populations,  and  I  suspect 
that  future  study  will  show  that  sufficient 
difference  exists  to  warrant  recognition  of 
both  at  the  species  level.  The  available 
name  for  the  Atlantic  population  is: 

*Stromateus  ])rasilicnsis  Fowler,  1906.  At- 
lantic coast  of  South  America,  Tierra  del 
Fuego  north  to  Uruguay,  type  locality  Rio 
Grande  do  Sul,  Brazil.  D  47-53  ( total  ele- 
ments). A  44-47  (total  elements).  P  19- 
20.  Gill-rakers  around  3+1  +  12.  Verte- 
brae 16-17  +  27-30.  The  bionomics  and 
potential  fishery  for  this  fish  are  the  sub- 
ject of  an  excellent  discussion  by  Hart 
(1946). 


Stromateoid  Fishes  •  HacdricJi        103 


Genus  PEPRILUS  Cuvier,  1829 

Figures  40,  -U 

Rliumhus  Lacepede,  1800:60.  (Type  species: 
Chaetodon  alepidotus  Linnaeus,  1766:460,  by 
monotypy.  Charleston,  South  CaroHna.  Pre- 
occupied by  Rhombus  Humphrey,  1797,  Mol- 
lusca. ) 

Pcprihis  Cuvier.  1829:213.  (Type  species:  Sfcr- 
noptyx  gardcnii  Bloch  and  Schneider,  1801: 
494,  by  subsequent  designation  of  Gill,  1862: 
126.  Charleston,  South  Carolina.  A  synonym 
of  Chaetodon  alepidotus  Linnaeus,  1766:460.) 

Powrwtus  Gill,  1861:35.  (Type  species:  Stro- 
juatcus  triacanthus  Peck,  1804:51,  by  mono- 
typy.    Piscataqua  River,   New  Hampshire. ) 

Palometa  Jordan  and  Evennann,  1896:966.  (Sul> 
genus.  Type  species:  *Stromateus  palometa 
Jordan  and  Bollman,  1889:156,  by  original 
designation.    Bay  of  Panama,  Pacific  Ocean. ) 

Simohrama  Fowler,  1944b:2.  (Type  species:  *Seser- 
inus  xanthurus  Quoy  and  Gaimard,  1824:384, 
by  original  designation.  Rio  de  Janeiro.  A 
synonym  of  Stromateus  paru  Linnaeus,  1758: 
248.) 

The  combination  of  deep  body,  large  eye, 
long  pectoral  fin,  one  to  three  bladeHke 
spines  ahead  of  the  median  fins,  a  ventral 
spine  on  the  pelvic  bone,  and  no  pelvic  fins, 
distinguishes  Pcprihis  from  all  other  stro- 
mateoid genera.  The  name,  a  masculine 
noun,  is  from  the  Greek  TrcTrptAos,  one  of 
Hesychian's  many  l^dv^  ttoios,  unknown  fish. 

Description.  Body  deep,  maximum  depth 
35  to  70  per  cent  of  the  standard  length, 
highly  compressed;  musculature  firm.  Pe- 
duncle ver\'  short,  compressed.  Dorsal  fin 
continuous,  originating  just  behind  insertion 
of  the  pectoral  fin;  one  to  three  flat,  blade- 
like spines,  the  first  spine  pointed  on  both 
ends,  preceding  the  30  to  40  finrays.  An- 
teriormost  rays  of  the  median  fins  pro- 
duced, the  fins  often  falcate,  the  rays  which 
follow  much  shorter,  diminishing  very 
slightly  in  length  to  the  last  ray,  the  shortest. 
Pectoral  fins  long,  winglike,  their  bases 
slightly  inclined.  No  pelvic  fins.  Pelvic 
bone  visible  on  midline  of  bod\'  under  the 
end  of  the  pectoral  fin  base;  a  small  spine 
on  the  end  of  the  bone  projecting  postero- 
ventrally  through  the  skin.  Tip  of  coracoid 
sometimes  projecting  slightly  underneath 
the  head  at  about  level  of  margin  of  the 


preopercle.  Caudal  fin  stiff,  deeply  forked, 
the  lobes  long  and  equal.  Scales  very  small, 
cycloid,  thin,  very  deciduous,  extending 
onto  all  fins.  Simple  tubed  scales  of  the 
lateral  line  moderately  high,  following  dor- 
sal profile  and  extending  onto  peduncle  but 
not  to  caudal  base.  A  branch  of  the  lateral 
line  extending  upward  from  the  head  of  the 
hvomandibular  in  a  short,  wide,  bonv  tract. 
Skin  very  thin;  main  subdermal  canal  along 
intermuscular  septum  and  side  branches 
conspicuous,  pores  to  surface  very  small. 
In  some  species,  a  row  of  large  conspicuous 
pores  in  the  back  under  the  first  half  of  the 
dorsal  fin.  Head  around  25  per  cent  of  the 
standard  length.  Top  of  head  naked,  pores 
clearly  \dsible,  naked  skin  underlain  with 
numerous  parallel  canals  projecting  back- 
ward over  the  nape.  Eye  large.  Adipose 
tissue  around  the  eye  developed,  extending 
forward  and  surrounding  the  nostrils.  Nos- 
trils small,  the  anterior  round,  the  posterior 
a  slit,  located  near  tip  of  the  obtuse  snout 
at  level  of  top  of  the  eye.  End  of  maxillary 
barely  reaching  to  below  eye,  angle  of  gape 
well  before  eye.  Premaxillary  not  protrac- 
tile. Lacrimal  bone  reduced,  scarcely  cov- 
ering top  of  upper  ]a.\y  when  mouth  is 
closed,  end  of  maxillary  exposed.  Supra- 
maxillary  absent.  Jaw  teeth  minute,  uni- 
serial,  laterally  compressed,  with  three  sub- 
equal  cusps,  close  set,  covered  laterally  by 
a  membrane.  Vomer,  palatines,  and  basi- 
branchials  toothless.  Gill  membranes  united 
across  the  isthmus,  divided  from  about  le\el 
of  the  forward  part  of  the  eye.  Opercle  and 
preopercle  thin,  not  scaled,  preopercle  finely 
striated,  opercle  smooth,  margins  entire; 
opercle  with  two  ill-defined  flat  spines; 
angle  of  preopercle  rounded,  not  projecting 
backward.  Cheek  not  scaled.  Gill-rakers  a 
little  more  than  half  the  length  of  the  fila- 
ments, with  fine  teeth  on  the  inner  edge, 
the  rakers  close  set,  about  18  on  the  lower 
limb  of  the  first  arch;  no  rudimentary 
rakers  under  the  small  pseudobranch.  Six 
branchiostegal  rays,  four  on  the  ceratohyal, 
two  on  the  epihyal.  Scapula  visible.  Ver- 
tebrae variable,  13  +  17  =  30  to  12  4-  23  = 


104        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure    40.      Pepnlus    triacanthus,    drawing    of    a    7.5-inch    specimen,    courtesy    of    the    Smithsonian    Institution. 


35.  Sclerotic  l)oncs  well  ossified.  Stomach 
a  simple  sac;  intestine  very  long.  Pyloric 
caeca  verv  numerous,  in  a  dendritic  mass. 

Color  in  preservative  brownish,  often 
with  a  silvery  overlay,  dark  above,  lighter 
below.  Back  and  sides  sometimes  with 
spots.  Median  fins  darker  or  lighter  than 
the  body;  pectoral  light.  Gill  cavity,  inside 
of  mouth,  and  peritoneum  light. 

Naturol  history.  The  species  of  Pcprihis 
are  prized  for  food  wherever  they  occur. 
The  biggest  fishery  is  for  P.  triaconthus, 
which  is  found  along  the  east  coast  of  the 
United  States.  In  some  years,  more  than 
two  million  pounds  of  this  species  are  landed 
in  Massachusetts  alone  (Bigelow  and 
Schroeder,  1953).  Considering  its  commer- 
cial importance,  surprisingly  little  is  known 
of  the  habits  of  Peprilus.  Almost  all  our 
knowledge  of  the  natural  history  of  the 
genus  comes  from  a  few  general  studies  on 
P.  triacanthus.  The  discussion  here  is  based 
largely  on  the  excellent  review  of  Bigelow 
and  Schroeder  (1953). 

Though  young  P.  triacanthus  are  some- 
times found  with  Cyanea,  they  do  not  seem 


to  associate  with  medusae  as  actively  as 
some  other  stromateoids.  The  small  fishes 
are  just  as  often  observed  swimming  inde- 
pendently at  the  surface  or  clustered  under 
floating  Gulf  weed.  A  much  stronger  as- 
sociation is  formed  by  P.  alepidofus  with 
the  sea  nettle  Chrysaora  in  Chesapeake  Bay. 
P.  alepidotus  feeds  actively  on  the  medusa 
(Mansueti,  1963). 

Pcprihis  triacanthus  spawn  in  Massachu- 
setts waters  from  summer  into  early  fall. 
Spawning  takes  place  a  few  miles  offshore, 
but,  except  at  this  time,  the  adults  are  usu- 
ally closer  to  shore  in  water  less  than  30 
fathoms  deep.  The  eggs  are  pelagic  and, 
at  a  temperature  of  65 °F,  hatch  in  less  than 
two  days.  During  their  first  summer,  the 
young  fish  may  grow  to  a  length  of  three 
or  four  inches.  They  probably  mature  when 
about  two  years  old  at  a  length  of  seven 
inches.  A  large  adult  is  no  more  than  a  foot 
long. 

During  th(>  late  fall,  schools  of  Pcprihis 
triacanthus  apparently  move  offshore  where 
they  winter  near  bottom  in  about  100  fath- 
oins.    Though  sometimes  occurring  as  far 


Stromateoid  Fishes  •  Hacdrich 


105 


Figure    41.      Peprilus    alepidotus,    drawing    of    a    7.5-inch    specimen,   courtesy   of  the   Smithsonian    Institution. 


north  as  Newfoundland,  P.  triacanthus,  like 
all  members  of  the  genus,  is  basically  a  wami 
water  fish. 

Relationships.  Peprilus  is  more  advanced 
than  Stromatcus  in  that  all  species  lack  pel- 
vic fins  and  have  only  two  epural  elements 
in  the  caudal  skeleton  (Fig.  42).  But  it  is 
more  primitive  than  Pampus  in  having  six 
instead  of  five  branchiostegal  rays  and  a 
shorter  pharyngeal  sac  (Fig.  43;  cf.  Fig.  46). 
The  few  spines  before  the  median  fins  in 
Peprilus  are  very  similar  in  fonu  to  the 
more  numerous  spines  of  some  Pampus,  but 
the  similarity  is  probably  due  to  common 
heritage  rather  than  to  direct  ancestry.  It 
is  unlikely  that  Pampus  is  derived  from 
Peprilus.    The    most    primitive    species    in 


Pampus  has  no  spines  before  the  median 
fins  and  shows  no  trace  of  the  pelvic  spine 
so  characteristic  of  Peprilus.  Peprilus  is  de- 
rived from  a  fish  somenhat  like  Stroma- 
teus,  but  probably  with  fewer  vertebrae. 

Species.  Peprilus  is  widespread,  with  a 
number  of  species  along  both  the  Atlantic 
and  Pacific  coasts  of  the  New  World  north 
of  the  equator.  One  species  is  found  as  far 
south  as  Montevideo  on  the  east  coast  of 
South  America.  Peprilus  has  been  divided 
into  several  nominal  genera,  but  the  dif- 
ferences on  which  these  are  based — depth 
of  body,  fins  falcate  or  not,  certain  pores 
well  developed  or  not — are  the  differences 
between    species,    not    genera.    Osteologi- 


106        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


EPURALS 


HYPURAL  6 


HYPURAL  4  +  5  — 


HYPURAL  2  +  3 


Figure  42.      Caudal    skeleton   of  Peprilus    triacanthus,   drawing    of  a  cleored-and-stained   preparation   from   a  36-mm  specinnen. 
All   elements  identified  in    Figure   1. 


cally,  all  members  of  the  nominal  genera 
are  very  similar. 

The  species  differ  but  little  from  one  an- 
other. There  is  some  question  as  to  whether 
or  not  certain  populations  are  to  be  re- 
garded as  full  species  or  only  as  subspecies. 
The  marked  similarity  between  species,  and 
therefore  the  doubt  as  to  the  rank  to  be 
accorded  certain  forms,  is  probably  due  to 
the  fact  that  active  speciation  is  occurring 
now  in  the  genus.  More  variational  studies 
and  increased  knowledge  of  the  natural  his- 
tory are  needed  to  resolve  these  questions. 

The  species  in  Peprilus  are: 

Pcprilus  parti  (Linnaeus,  1758)=S/ro- 
mateus  paru  Linnaeus.  West  Indies  to  Uru- 
guay, type  locality  Jamaica.  D  III  38-44. 
A  II  3,5-41.  P  20-22.  Vertebrae  13  +  17. 
This  species  is  distinguished  from  the  very 
closely  related  P.  alcpidotus  ( Linnaeus, 
1766)  by  the  lower  counts  and  narrower 
pectoral  fin  (Hildebrand,  MS).  Synonyms 
are:       '^'Seserinus     xcinthuriis     Quoy     and 


Gaimard,  1824,  from  Brazil;  Rhombus 
crenulotus  Cuvier  and  Valenciennes,  1833, 
from  Cayenne;  and  Rliombus  orbicularis 
Guichenot,  1866a,  also  from  Cayenne. 

Pcprilus  alcpidotus  ( Linnaeus,  1766 )  = 
Chaetodon  alcpidotus  Linnaeus.  East  coast 
of  North  America  from  Massachusetts  to 
Florida  and  Gulf  of  Mexico,  type  locality 
Charleston,  South  Carolina.  D  III  43-49. 
A  II  39-43.  P  21.  Vertebrae  13  +  17.  This 
species  is  considered  distinct  from  P.  paru 
by  Hildebrand  (MS),  although  both  are 
often  lumped  under  this  name.  This  is  the 
"harvestfish"  of  the  North  American  fish- 
eries literature.  Stromateus  longipinnis 
Mitchill,  1815,  from  New  York  Bay,  is  a 
synonym. 

Pcprilus  triacanthus  (Peck,  1804)  =  Stro- 
mateus triacanthus  Peck.  East  coast  of 
North  America  from  Newfoundland  to  Flor- 
ida, type  locality  Piscataqua  River,  New 
Hampshire.  D  III  43-46.  A  II  37-43.  P 
19-21.  Vertebrae  around  13  4-  19.  This  spe- 


Stromateoid  Fishes  •  Haedrich        107 


Figure  43.      Branchial    region   of  Pepnius   triacanthus,  drawing   of  a  cleored-and-sfained  preparation  from  a  120-mm  specimen. 
Elements  identified  in    Figure  2. 


cies  is  very  close  to  P.  biiiti  Fowler,  from 
which  it  is  distinguished  by  a  slightly 
higher  vertebral  count  (Collette,  1963). 
These  two  fonns  have  only  recently  di- 
verged. This  fish  is  usually  known  as  Poro- 
notus  triacanthits,  the  "butterfish"  of  North 
American  fisheries  literature.  Stromatcus 
ciyptoms  Mitchill,  1815,  from  New  York 
Bay,  is  a  synonym. 

Peprilus  simillimiis  (Ayres,  1860)  =  Poro- 
notus  similliiims  Ayres.  West  coast  of  North 
America,  British  Columbia  to  Baja  Califor- 
nia, type  locality  San  Francisco.  D  III  45- 
47.  A  III  39-44.  P  20-22.  Vertebrae  13  + 
17.  This  species  is  a  member  of  the  P.  tiia- 
canthiis-burti  complex. 

Peprilus  medius  (Peters,  1869)=S7ro- 
matcus  medius  Peters.    Known  onlv  from 


Mazatlan,  Mexico,  Pacific  Coast.  D  III  42. 
A  III  32.  (From  Fordice,  1884.)  Possibly 
a  synonym  of  P.  simiUimus  (Ayres,  1860). 

Peprilus  polometa  (Jordan  and  Bollman, 
1889)  =  "^Stromateus  palometa  Jordan  and 
Bollman.  Pacific  coasts  of  Panama  and 
Colombia.  D  III  44-48.  A  II  43-46.  P  22- 
23.  Gill-rakers  5-6  +  1  +  15-16.  Vertebrae 
13  +  20-21.  This  deep-bodied  fish  with  fal- 
cate median  fins  is  a  member  of  the  Atlan- 
tic P.  alepidotus-paru  group. 

*P.  smjderi  Gilbert  and  Starks,  1904. 
Known  only  from  Panama  Bay.  D  III  41- 
47.  A  III  41-42.  P  22-23.  Gill-rakers  4  +  1 
+  14.  Vertebrae  13  +  23.  This  rarely  seen 
species  is  distinguished  from  P.  palometa 
(Jordan  and  Bollman,  1889)  in  having  more 
vertebrae,  a  longer  snout,   and  very  short 


108        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure  44.      Pampas   chinensii,   a    species   lacking   spines   before    the    median    fins,    drawing    of   a    4-incfi    specimen,    from    Day, 
1875. 


lobes  on  the  median  fins.  P.  snijderi  ap- 
proaches Stromateiis  in  the  increased  num- 
ber of  vertebrae  and  in  the  very  reduced 
spines  preceding  the  rays  in  the  median 
fins.  It  may  be  very  Hke  the  stromateid  an- 
cestral to  Pcprilus  and  Stromatcus.  Inves- 
tigations of  its  systematic  position,  geo- 
graphic distribution,  and  natural  history 
should  be  very  instructive. 

Peprihts  biirti  Fowler,  1944b.  Gulf  of 
Mexico,  type  locality  Breton  Island,  Loui- 
siana.  D  III  43-45.   A  III  40-41.   P  20-21. 


(From  Fowler,  1944b.)  Vertebrae  around 
13+17  (Collette,  1963).  This  species  is 
very  close  to  P.  triacanthus  (Peck,  1804). 
Caldwell  (1961)  and  Collette  (1963)  differ 
in  their  interpretation  of  its  systematic 
status. 

Genus  PAMPUS  Bonaparte,  1837 

Figures  44,  45 

P(iin})iis  Bonaparte,  1837:48.  ( Subtrenus.  Type 
species:  Strotmitcus  cfinilidiis  Cuvier  and 
Valenciennes,  1833:391,  by  subsequent  desig- 
nation  of  Jordan,    1923:187.     Malabar  Coast. 


Stromateoid  Fishes  •  Haedrich 


109 


Figure  45.     Pampus    argenteus,    a    species    with    spines    before    the  median  fins,  drawing  of  a  10.5-inch  specimen,  from  Jordan 
and  Metz,  1913. 


A  .synonym  of  Stronmtcus  argenteus  Euphra- 
,sen,   1788:53.) 

Stromateoides  Bleeker,  1851:368.  (Type  .species: 
Stromateus  cinereus  Bloch,  1793:90,  by  sub- 
sequent designation  of  Gill,  1862:126.  A 
synonym  of  Stromateus  argenteus  Euphrasen, 
1788:53.) 

Chondroplites  Gill,  1862:126.  (Type  species: 
Stromateus  atous  Cu\  ier  and  Valenciennes, 
1833:389,  by  original  designation.  After  Rus- 
sell's "atoo-'koia"  (1803:  plate  21),  Viza- 
gapatam.  A  synonym  of  Stromateus  chinensis 
Euphrasen,   1788:54.) 

The  combination  of  deep  body,  no  pelvic 
fins,  fixed  maxillary,  and  gill  membranes 
broadly  united  to  the  isthmus  distinguishes 
Pampus  from  all  other  stromateoid  genera. 
The  name,  a  masculine  noun,  is  from  the 
vernacular  of  the  19th  century  East  Indian 
Spanish  and  Portuguese  colonials,  who  gen- 
erally used  the  term  "pampus"  (ultimately 
from  "pampano")  for  any  silvery,  com- 
pressed fish. 


Description.  Body  very  deep,  maximum 
depth  greater  than  60  per  cent  of  the  stan- 
dard length,  highly  compressed;  muscula- 
ture finn.  Peduncle  very  short,  compressed. 
A  continuous  dorsal  fin;  both  median  fins 
preceded  by  either  none  or  five  to  ten  flat, 
bladelike  spines,  pointed  on  both  ends,  pro- 
truding but  slightly  and  resembling  the 
ends  of  free  interneurals.  In  forms  with 
bladelike  spines,  dorsal  fin  originating 
slightly  behind  end  of  pectoral  fin  base,  the 
first  spine  generally  over  or  slightly  before 
the  pectoral  insertion;  in  forms  lacking 
spines,  dorsal  fin  originating  over  the  pec- 
toral fin  base.  Anal  papilla  well  before  mid- 
body,  in  a  slit.  Anal  fin  originating  at  or 
before  middle  of  body  and  only  slightly 
behind  origin  of  the  soft  dorsal.  Anterior- 
most  rays  of  the  median  fins  produced,  the 
fins  often  falcate,  rays  which  follow  shorter; 
in  forms  with  bladelike  spines,  rays  of  the 


110        Bulletin  Museum  of  Cotnparative  Zoology,  Vol.  135,  No.  2 


posterior  two-thirds  of  the  fin  short  and 
subequal,  the  anal  fin  lobe  often  extremely 
produced;  in  forms  lacking  spines,  rays  of 
the  posterior  two-thirds  of  the  fin  decreas- 
ing in  length  evenly  to  the  last  ray,  the 
shortest.  Pectoral  fin  long,  winglike,  the 
base  of  the  fin  inclined  about  45°.  No  pel- 
vic fins.  Pelvic  bone  not  visible  on  midline 
and  lacking  a  ventral  spine.  Tip  of  coracoid 
often  projecting  slightly  underneath  head 
at  about  level  of  margin  of  preopercle.  Cau- 
dal fin  stiff,  deeply  forked,  in  forms  with 
bladelike  spines  the  ventral  lobe  often  ex- 
tremely produced.  Scales  very  small,  cy- 
cloid, thin,  deciduous,  extending  onto  bases 
of  all  fins.  Simple  tubed  scales  of  the  lat- 
eral line  fairly  high,  following  dorsal  pro- 
file, and  extending  onto  peduncle.  Skin 
thin;  main  subdermal  canal  along  intermus- 
cular septum  and  side  branches  usually 
quite  apparent,  pores  to  surface  seem  want- 
ing. Head  around  25  per  cent  of  the  stan- 
dard length,  \'ery  deep  and  broad.  Top  of 
head  naked,  subdermal  canals  visible  under 
naked  skin  but  pores  not  visible,  naked  skin 
underlain  with  numerous  parallel  canals 
projecting  backwards  over  the  nape  and 
along  first  part  of  lateral  line.  Eye  small. 
Adipose  tissue  around  eye  developed  and 
extending  forward  around  the  nostrils.  Nos- 
trils large,  the  anterior  round,  the  posterior 
a  long  slit,  located  near  tip  of  the  inflated 
snout  at  level  of  the  top  of  the  eye,  nasal 
capsules  greatly  expanded.  Mouth  subter- 
minal,  curved  downward,  small,  maxillary 
scarcely  reaching  to  below  eye  and  angle 
of  gape  before  eye.  Premaxillary  not  pro- 
tractile. Maxillary  immobile,  covered  with 
skin  and  united  to  cheek.  Lacrimal  bone 
very  much  reduced.  Supramaxillary  absent. 
Jaw  teeth  minute,  uniserial,  flattened,  with 
a  large  rounded  central  cusp  and  two 
shorter  auxiliary  cusps,  close  set,  covered 
laterally  by  a  membrane.  Vomer,  palatines, 
and  basibranchials  toothless.  Gill  mem- 
branes broadly  united  to  the  isthmus.  Gill 
opening  a  straight  slit,  covered  with  a  flap 
of  skin.  Gill-rakers  small,  about  one-quar- 
ter the  length  of  the  filaments,  not  toothed. 


widely  spaced.  Pseudobranch  absent.  Five 
branchiostegal  rays,  three  on  the  ceratohyal, 
two  on  the  epihyal.  Scapula  not  prominent. 
Vertebrae  variable  in  species  with  blade- 
like spines,  14  +  20  =  34  to  16  +  25  =  41; 
in  species  without  bladelike  spines,  verte- 
brae 14  +  19  =  33.  Dermal  skeleton  soft 
and  spongy,  but  sclerotic  bones  well  ossi- 
fied; skeleton  in  general  fibrous.  Stomach 
a  simple  sac;  intestine  very  long.  Pyloric 
caeca  numerous,  in  a  small  dendritic  mass. 

Color  in  life  very  silvery  with  a  bluish 
cast  on  the  back.  Color  in  preservative 
brown  or  bluish  with  a  silvery  or  whitish 
overlay.  Median  fins  and  caudal  yellowish 
with  dark  borders.  Plead  a  little  darker  than 
the  body,  with  fine  speckling.  Gill  mem- 
branes and  inside  of  mouth  dark.  Perito- 
neum silvery  with  black  speckles. 

Natural  liistonj.  Pampii.s  is  the  most 
sought  after  of  all  the  stromateoid  fishes. 
Throughout  the  Orient,  it  commands  a  good 
price  wherever  it  appears.  In  India,  where 
it  is  known  as  "pomfret,"  the  1962  landings 
totaled  25.7  thousand  metric  tons,  more 
than  four  per  cent  of  the  total  marine  catch 
( FAO  1964 ) .  However,  despite  its  com- 
mercial importance,  virtually  nothing  is 
known  of  the  life  history  of  Pampus. 

The  young  occur  in  shallow  water  along 
the  coasts,  and  may  even  ascend  estuaries 
( Day,  1875 ) .  The  small  mouths  with  cut- 
ting teeth  and  the  long  pharyngeal  sacs  sug- 
gest that  soft-bodied  coelenterates  may 
figure  largely  in  the  diet.  Most  stomachs 
examined  seemed  to  contain  the  shredded 
remains  of  these  animals,  but  bits  of  fish 
were  also  found.  Chopra  (1960)  found  that 
a  sudden  appearance  of  numerous  cteno- 
phores  and  medusae  in  the  waters  off  Bom- 
bay was  accompanied  by  a  marked  increase 
in  the  local  catch  of  Pampus. 

Rehitionsliips.  Pampus  is  the  most  ad- 
vanced stromateid  genus.  The  advanced 
characters  are  the  reduction  in  the  number 
of  branchiostegal  rays  to  five,  the  lengthen- 
ing of  the  pharyngeal  sac  (Fig.  46),  the 
restriction  of  the  gill  opening,  the  loss  of 
th(^  pseudobranch,  and  the  development,  in 


Stromateoid  Fishes  •  Hacdrich         111 


Figure  46.      Branchial    region  of  Pampus  echinogosfer,  drawing    of   a    cleared-and-stained    preparation    from    a    180-mm    speci 
men,  ABE  1743.     Elements  identified  in   Figure  2. 


some  species,  of  flat,  bladelike  spines  ahead 
of  the  median  fins.  The  genus  is  derived 
from  a  fish  very  Hke  Stromateus,  but  with 
fewer  vertebrae.  The  most  primitive  spe- 
cies in  Fampus  has  14  +  19  vertebrae,  but 
the  more  advanced  may  have  as  many  as 
16  +  25.  All  members  of  the  genus  have 
the  typical  stromateid  caudal  skeleton  (Fig. 
47). 

Species.  Pampus  is  widely  distributed  in 
tropical  waters  over  the  continental  shelves 
from  the  Iranian  Gulf  to  Japan.  There  are 
reports  of  specimens  from  Hawaii  ( Fowler, 
1938)  and  from  the  Adriatic  (Soljan,  1948). 
No  subsequent  records  have  appeared  from 
either  place.  The  two  localities  are  so  far 
out  of  the  established  range  of  the  genus 
that  the  records  can  only  have  been  based 
on  specimens  brought  from  elsewhere. 

Gill  (1884)  divided  the  genus  Stromat- 
eoides  {=  Pampus)  into  tsvo  groups,  which 
he  apparently  regarded  as  subgenera.  The 
group  Stromateoides  had  falcate  fins  and 


prominent  dorsal  and  anal  spines;  the  other 
group,  CJwndwplitcs,  had  neither.  This 
dichotomy  does  exist  in  Pampus,  but  more 
work  is  needed  to  decide  whether  or  not 
the  distinction  merits  subgeneric  recogni- 
tion. 

Numerous  species  have  been  described 
in  the  genus.  The  majority  are  probably 
synonyms.  Published  descriptions  provide 
for  the  most  part  no  clear-cut  means  of  dis- 
tinguishing species.  At  the  present  state  of 
knowledge,  only  three  can  be  recognized: 

Pampus  cliincnsis  (Euphrasen,  1788)  = 
Stromateus  cJiinensis  Euphrasen.  India  to 
China,  type  locality  "Castellum  Chinense 
Bocca  Tigris."  D  43-50  (total  elements). 
A  39-42  (total  elements).  P  24-27.  Ver- 
tebrae 14  +  19.  This  species  lacks  the  pe- 
culiar flat  spines  before  the  median  fins 
which  are  found  in  the  other  two  species. 
The  median  fins  are  not  falcate.  The  fin- 
rays  gradually  diminish  in  length  posteri- 
orly  (Fig.  44).    P.  chincnsis,  the  type  for 


112        Bulletin  Museum  of  Comparative  Zoology,  Vol  135,  No.  2 


EPURALS 


HYPURAL   6 


HYPURAL  4  +  5  — 


HYPURAL   2+3  — 


Figure  47.     Caudal   skeleton  of  Pampus  argenteus,   drawing  of    a    cleared-and-stained    preparation    from    a    48-mm    specimen, 
ABE    1937.     All    elements    identified    in    Figure    1. 


Gill's  (1884)  genus  Chondroplites,  is  cer- 
tainly the  most  primitive  species  in  Pompus. 
Synonyms  are:  Stromateus  alhus  Cuvier 
and  Valenciennes,  1833,  from  Pondichery; 
Strot7uiteus  atous  Cuvier  and  Valenciennes, 
1833,  from  Vizagapatam;  and  Stromateoides 
atokoia  Bleeker,  1852,  from  Malaysia.  The 
trivial  name  is  commonly  written  incorrectly 
sinensis. 

Pampiis  argenteus  (Euphrasen,  1788)  = 
Stromateus  argenteus  Euphrasen.  Iranian 
Gulf  to  Japan,  type  locality  "Castellum 
Chinense  Bocca  Tigris."  D  V-X  38^3.  A 
V-VII  34-43.  P  24-27.  Vertebrae  14-16  + 
20-25.  Falcate  median  fins,  preceded  by 
flat  bicuspid  spines,  are  characteristic  of 
this  species  (Fig.  45).  The  species  is  ap- 
parently very  wide  ranging.  Further  study 
will  no  doubt  show  it  to  be  composed  of 
numerous  subspecies  or  even  species.  This 
fish   is  the  "pomfret"  of  Eastern   fisheries 


literature,  and  is  important  commercially  in 
India,  China,  and  Japan.  Probable  syn- 
onyms are:  Stromateus  cinereus  Bloch, 
1793;  Stromateus  candidus  Cuvier  and 
Valenciennes,  1833,  from  Pondichery;  Stro- 
mateus securifer  Cuvier  and  Valenciennes, 
1833,  from  Bombay;  Stromateus  griseus 
Cuvier  and  Valenciennes,  1833,  from  Pondi- 
chery; Stromateus  punetatissimus  Temminck 
and  Schlegel,  1850,  from  Japan;  and  Pampus 
simoprosopus  Fowler,  1934b,  from  Siam. 

Pampus  echinogaster  (Basilewsky,  1855) 
=  Stromateus  echinogaster  Basilewsky. 
China,  Korea,  and  Japan,  tvpe  locality 
China.  D  VIII-X  42-49.  A  V-VII  42-47. 
P  24-25.  Vertebrae  14-15  +  24-26.  This 
species  has  more  median  finrays  than  P. 
argenteus.  Abe  and  Kosakai  (1964)  report 
that  P.  echinogaster  has  fewer,  thicker 
pyloric  caeca  than  P.  argenteus,  and  3-6  + 
12-15  gill-rakers  as  opposed  to  2-3  +  8-10 


Stromateoid  Fishes  •  Haedrich         113 


in  P.  orgcnteus.  Pompus  lighti  Evermann 
and  Shaw,  1927,  from  Nanking  is  a  prob- 
able synonym. 

EVOLUTIONARY  TRENDS  IN  THE 
STROMATEOIDEI 

Gosline  (1959)  and  Liem  (1963)  have 
recently  stressed  the  need  for  an  under- 
standing of  functional  moiphology  in  con- 
nection with  phylogenetic  studies.  Natural 
selection  acts  on  efficiencies  and  abilities, 
and  evolutionary  change  results.  If  func- 
tion is  understood,  or  at  least  taken  cogni- 
zance of,  an  attempt  can  be  made  to  evalu- 
ate observed  differences  in  terms  of  effi- 
ciences  and  abilities.  Within  this  framework, 
evolutionary  trends  can  be  discussed.  Anat- 
omy alone  is  insufficient;  the  way  of  life 
must  also  be  taken  into  account. 

The  functional  significance  of  many  char- 
acters, however,  is  not  known.  It  is  hard 
to  understand,  for  example,  why  selection 
should  favor  a  fish  with  15  principal 
branched  rays  in  the  caudal  fin  or  with  25 
vertebrae,  yet  these  numbers  have  appeared 
in  many  independent  phyletic  lines.  Studies 
of  characters  of  this  sort  are,  nonetheless, 
of  much  importance.  The  argument  that  the 
overall  trends  observed  are  real  is  made 
stronger  when  these  characters  change  in 
step  with  characters  for  which  the  function 
is  at  least  partially  understood. 

The  trends  in  the  evolution  of  the  stro- 
mateoid fishes  are  discussed  below,  treat- 
ing separately,  as  much  as  possible,  groups 
of  characters  which  can  be  considered  ana- 
tomical units,  i.e.,  caudal  skeleton,  branchial 
region,  etc.  Drawing  on  what  little  is  known 
of  the  way  of  life  of  stromateoids,  func- 
tional significance  can  be  attached  to 
changes  in  the  teeth  and  jaws,  the  pharvn- 
geal  sacs,  and  the  caudal  skeleton.  But  the 
meaning  of  the  changes  in  the  size  and 
shape  of  the  body,  the  fin  pattern,  the  bran- 
chial region,  and  the  number  of  vertebrae 
cannot  be  easily  interpreted.  By  and  large, 
the  discusion  under  each  of  these  is  neces- 
sarily brief  and  loaded  with  conjecture. 
Though  each  unit  is  treated  separately,  it 


must  be  remembered  that  they  have  evolved 
together.  The  organism  responds  as  a  whole 
to  the  environment,  and  the  products  of 
natural  selection  are  all  interdependent. 

Attention  should  be  drawn  to  the  distinc- 
tion between  characters  typical  of  a  taxon 
and  those  typical  of  the  grade  of  a  taxon. 
In  the  first  case,  the  characters  are  found 
in  all  members  of  the  taxon.  In  essence, 
they  are  a  part  of  the  definition  of  that 
taxon.  Grade  refers  to  the  "average"  evo- 
lutionary status  of  the  taxon  vis-a-vis  other 
taxa.  The  characters  typical  of  a  grade  need 
not  be  found  in  all  members  of  the  taxon. 
In  fact,  because  different  parts  of  the  ani- 
mal respond  to  the  environment  at  differ- 
ing evolutionary  rates,  it  is  unlikely  that 
these  characters  will  be  found  in  all.  The 
use  of  grades  pro\'ides  a  convenient  way  of 
discussing  evolutionary  trends  without  con- 
stantly itemizing  the  exceptions  to  the  gen- 
eral picture.  Thus,  while  25  vertebrae  char- 
acterize the  centrolophid  grade,  not  all 
members  of  the  family  Centrolophidae  have 
25  vertebrae.  Most  do  have  25  but  some 
have  26  and  others  have  near  60. 

Before  proceeding  to  the  discussion,  let 
us  briefly  recall  the  more  salient  features 
of  each  of  the  five  stromateoid  families. 
The  characters  of  the  individual  genera  are 
presented  in  Table  2. 

The  Centrolophidae  are  one  to  four  feet 
in  length  with  moderately  deep  to  elongate 
bodies.  All  have  pelvic  fins.  Their  mouths 
and  the  teeth  in  their  jaws  are  fairly  large. 
There  are  no  teeth  on  the  palate.  The  pa- 
pillae in  the  round  pharyngeal  sacs  have 
irregularly  shaped  bases.  There  are  seven 
branchiostegal  rays,  a  pseudobranch,  and 
25,  26,  29,  30,  or  50  to  60  xertebrae.  The 
caudal  skeleton  has  six  hypural  and  usually 
three  epural  elements.  There  are  six  genera: 
Hyperoghjphe,  Schedophilus,  CentroJophus, 
Icichthys,  SerioIcUa,  and  Psenopsis. 

The  Nomeidae  are  six  inches  to  over  two 
feet  in  length  with  deep  to  elongate  bodies. 
All  have  pelvic  fins.  Their  mouths  and  the 
teeth  in  their  jaws  are  small.  The  palatines 
and  the  vomer  bear  teeth.   The  papillae  in 


114        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Table  2.     Characters  of  the  stromateoid  genera.     +  denotes   presence;  —  absence. 


Pal- 

Bran- 

Body 

Maximum 

Pelvic 

Form   of 

atal 

CHIOS- 

Verte- 

Epurals -\- 

Shape 

Length 

Fins 

Teeth 

Denti- 
tion 

TEGAL 

Rays 

brae 

Hypurals 

Hijperoghjphe 

moderate 

4  feet 

+ 

simple 

— 

7 

25 

3  +  6 

SchedopJiihi.s 

deep 

3  feet 

+ 

simple 

— 

7 

25,  26,  29,  30 

3  +  6 

CcntroJoplui.s 

elongate 

4  feet 

+ 

simple 

— 

7 

25 

3  +  6 

Icichthijs 

elongate 

3  feet 

+ 

simple 

— 

7 

50  to  60 

2  +  6 

Seriolella 

moderate  to 
elongate 

3  feet 

+ 

simple 

— 

7 

25 

3  +  6 

Psenopsis- 

deep 

1   foot 

+ 

simple 

— 

7 

25 

3  +  6 

Stromateus 

deep 

1  foot 

—  in 
adult 

cusped 

— 

6 

42-48 

3  +  4 

Peprihis 

deep 

1  foot 

— 

cusped 

— 

6 

30-33 

2  +  4 

Pampiis 

very  deep 

18  inches 

— 

cusped 

— 

5 

33  41 

2  +  4 

Cubiceps 

elongate 

3  feet 

+ 

simple 

+ 

6 

30-33 

3  +  4 

Nomeus 

elongate 

1   foot 

+ 

simple 

+ 

6 

41 

3  +  4 

Psenes 

moderate 
to  deep 

1   foot 

+ 

simple  or 
cusped 

+ 

6 

30-42 

3  +  4 

Ariomiiia 

deep 

to 

elongate 

3  feet 

most 

<1  foot 

+ 

simple  or 
cusped 

6 

30-33 

3  +  2 

Tetragonurus 

very 
elongate 

2  feet 

+ 

simple  and 
cusped 

+ 

5-6 

43-58 

2  +  4 

the  round  pharyngeal  sacs  have  stellate 
bases.  There  are  six  branchiostegal  rays,  a 
pseudoliranch,  and  30  to  38,  41  or  42  ver- 
tebrae. The  caudal  skeleton  has  four  hypu- 
ral  and  three  epural  elements.  There  are 
three  genera:  Cubiceps,  Nomeus,  and 
Psenes. 

Most  species  of  the  Ariommidae  are  less 
than  a  foot  long,  although  a  few  species 
attain  more  than  twice  that  length.  Their 
bodies  are  either  deep  or  elongate.  All  have 
pelvic  fins.  Their  mouths  and  the  teeth  in 
their  jaws  are  very  small.  There  are  no 
teeth  on  the  palate.  The  papillae  in  the 
elongate  pharyngeal  sacs  have  round  bases. 
There  are  six  branchiostegal  rays,  a  small 
pseudobranch,  and  30  to  33  vertebrae.  The 
caudal  skeleton  has  two  hypural  and  three 
epural  elements.  There  is  one  genus: 
Ariomma. 

The   Tetragonuridae    are   less    than    two 


feet  long  with  very  elongate  bodies.  All 
have  small  pelvic  fins.  Their  mouths  are 
fairly  large.  In  the  upper  jaw,  their  teeth 
are  small  and  recurved;  in  the  lower  jaw, 
they  are  large  and  knifelike.  The  palatines 
and  the  vomer  bear  teeth.  The  papillae  in 
the  very  elongate  pharyngeal  sacs  are  much 
reduced  and  have  small  round  bases.  There 
are  five  or  six  branchiostegal  rays,  a  pseu- 
dobranch, and  43  to  58  vertebrae.  The 
caudal  skeleton  has  four  hypural  and  two 
epural  elements.  There  is  one  genus:   Tetra- 


gonurus. 


The  Stromateidae  are  usually  no  more 
than  a  foot  long  with  deep  bodies.  None 
have  pelvic  fins  when  adult.  Their  mouths 
and  the  teeth  in  their  jaws  are  very  small. 
There  are  no  teeth  on  the  palate.  The  pa- 
pillae in  the  round-to-elongate  pharyngeal 
sacs  have  stellate  bases.  There  are  five  or 
six  branchiostegal  rays,  either  a  small  pseu- 


Stromateoid  Fishes  •  Hacdrich        115 


dobranch  or  none  at  all,  and  30  to  48  verte-  gonurid  is  600  mm  long  (Fitch,  1951).  These 

brae.  The  caudal  skeleton  has  four  hypural  highly    modified    fishes    are    very    slender, 

and  usualh'  two  epural  elements.  There  are  with  the  maximum  depth  usually  less  than 

three    genera:     Stromateus,    Peprilus,    and  20  per  cent  of  the  standard  length.    The 

Pampus.  stromateids  rarely  exceed  450  mm  in  length, 

There  are  two  main  lineages  in  the  stro-  and  mature  when  less  than  200  mm  long, 

mateoids  (p.  51).    One  is  composed  of  the  These  fishes  are  very  deep  bodied,  the  max- 

Centrolophidae    and    their    derivative,    the  imum  depth  ranging  from  35  to  over  70  per 

Stromateidae.    The  other,  a  looser  assem-  cent  of  the  standard  length, 
blage,  is  composed  of  the  Nomeidae  and         The  course  of  evolution  in  the  form  of 

their  two  derivatives,  the  Ariommidae  and  the  body  has  been  one  of  diminution  in  size 

the  Tetragonuridae  ( Fig.  7 ) .    The  Centro-  and  of  increase  in  depth.   These  t\\'o  evolu- 

lophidae  and  the  Nomeidae  are  the  basal  tionary    tendencies    are    also    displayed    in 

stocks.    Of  these  two,  the  Centrolophidae  other  teleostean  groups  (Myers,  1958;  Liem, 

are   in   most   respects   the   most   primitive.  1963). 

Familial   and   generic   relationships,   to   be         Fins  (Fig.  48).    Only  one  major  change 

touched  upon  only  briefly  here,  have  been  has  occurred  in  the  fin  pattern  of  stromat- 

discussed    in    the    individual    accounts    of  eoids — the   loss   of  the   pelvic   fins    at   the 

family  and  genus.  stromateid  grade.    The  presence  of  pelvics 

Figure  48  summarizes  some  of  the  major  in  young  Stromotctis  fiafolo  and  their  sub- 
evolutionary  trends  in  the  stromateoid  sequent  loss  in  the  adult  are  important  clues 
fishes.  Each  group  is  treated  more  or  less  in  understanding  the  phylogeny  of  stroma- 
as  a  grade  in  the  diagram.  The  characters  teids.  There  is  a  difference  between  the 
sho\\'n  are  relative  size  and  shape  of  the  basic  fin  patterns  of  the  t\\'o  stromateoid 
bod>-,  fin  pattern,  presence  or  absence  of  lineages.  Members  of  the  nomeid  line  have 
palatal  dentition,  number  of  branchiostegal  two  dorsal  fins;  members  of  the  centro- 
rays,  vertebrae,  epural  plus  hypural  ele-  lophid  line  usually  have  but  one. 
ments  in  the  tail,  and  the  shape  of  the  pa-  The  thoracic  pelvic  fins  of  perciform 
pillae  in  the  pharyngeal  sacs.  These  are  dis-  fishes  are  used  in  braking  and  turning 
cussed  in  detail  in  the  following  accounts.  (Harris,  1938).  In  deep-bodied  fishes,  how- 
The  width  of  the  arrow  leading  to  each  ever,  the  effecti\eness  of  the  fins  for  these 
grade  is  proportional  to  the  number  of  purposes  is  probably  greatly  decreased  due 
genera  in  that  family.  to  the  change  in  the  hydrodynamic  profile 

Body  (Fig.  48).  The  largest  stromateoids  of  the  fish.    In  this  situation,  selective  pres- 

are    members     of    primitive    centrolophid  sure  may  favor  loss  of  the  fins.    This  has 

genera.    McCulloch  ( 1914)  reports  a  1,072-  apparently  been   the   case   in   stromateids, 

mm    Hyperoghjphe    porosa    weighing    41  and    is    also    observed    in    Parasfromatcus, 

pounds   from   the  Great  Australian   Bight,  Monodactyhis,   Psetfus,   and   a   number   of 

and  I  have  seen  an  1,195-mm  Centrolophus  other  unrelated  deep-bodied  teleosts. 
niger  taken   south  of  New  England  on   a         In  the  great  majority  of  stromateoids,  the 

long-line.     The   maximum    depth    in    most  anterior  rays  of  the  median  fins  are  no  more 

centrolophids  is  within  25  to  30  per  cent  than  two  or  three  times  the  length  of  the 

of  the  standard  length  and  never  exceeds  posterior  rays.   In  the  stromateids,  however, 

50  per  cent.    In  the  nomeids   and   in   the  the  median  fins  have  become  falcate  and 

ariommids,  there  are  a  few  species  whose  the  anterior  rays  are  very  much  produced, 

maximum  length  approaches  a  meter  but  In   Stromateus  this   tendency  is   but   little 

most  are  smaller.    The  maximum  depth  in  pronounced.  The  deeper-bodied  species  in 

these  families  is  from  about  25  to  40  per  Peprilus  have  very  falcate  fins.  The  anterior 

cent  of  the  standard  length.    A  large  tetra-  finrays  of  the  anal  fin  in  these  species  are 


116        BiiUrtin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


CENTROLOPHID 


Figure  48.  Evolutionary  trends  in  the  Stromoteoldel,  showing  relative  size  and  shape,  fin  pattern,  and  (within  the  outline 
from  left  to  right)  presence  or  absence  of  palatal  dentition,  and  numbers  of  branchlostegals,  vertebrae,  and  epurals  + 
hypurals.     Inset  shows  a   papilla.     Width  of  arrows   proportional   to  number  of  genera   in  the  family.     See  text. 


seven  or  eight  times  longer  than  the  pos-  fin  may  be  ten  to  20  times  longer  than  the 

terior  rays.    In  Pampu.s  both  the  anterior  posterior  finrays.    Lacking  observations  on 

rays  of  the  anal  fin  and  of  the  lower  caudal  living  Pampus,  it  is  difficult  to  know  what 

lobe  are  produced,  and  those  in  the  anal  advantage  these  elongate  fins  confer. 


Stromateoid  Fishes  •  Hacdrich        117 


Teeth  and  jaws  (Fig.  48).    In  most  cen-  flattened,  cusped  teeth  are  more  suited  to 

trolophids  the  angle  of  the  gape  may  he  slicing  the  tissues  of  coelenterates.    With 

helow  the  eye  hut  in  all  other  stromateoids  such  animals,  there  is  little  need  for  catch- 

the  angle  of  the  gape  is  hefore  the  eye.  The  ing   and    holding.    Nonetheless,    the   tetra- 

ariommids  and  stromateids  have  the  small-  gonurids,  existing  almost  entirely  on  salps 

est   mouths   of  all.    In    these   two   families  and   coelenterates,   have   conical,   recurved 

even  the  end  of  the  maxillary  is  before  the  teeth  in  the  upper  jaw  and  on  the  palate 

eye.    Centrolophids,  with  the  exception  of  which  must  hold  the  prey  firmly  while  the 

Psenopsis,  ha\e  a  small  supramaxillary  bone,  long,  knifelike  teeth  of  the  lower  jaw  slice 

but  this  is  gone  in  all  other  stromateoids.  off  mouthfuls. 

The  presence  of  teeth  on   the  palate  is  In  the  course  of  evolution,  the  jaws  of 

usually  considered  primitive  (Liem,  1963).  stromateoids  have  become  shorter  and  the 

The   palatine    and   vomerine    teeth   in    the  supramaxillary   bone   is   lost.    The   conical 

nomeids  indicate  that  this  group  branched  teeth     have     become     flattened,     cusped, 

off  early  from  the  ancestral  stock.   Both  the  smaller,  and  more  closely  set.    In  the  no- 

centrolophids,    in    most   respects   the   most  meid  lineage,  the  palatine  dentition  is  lost, 

primitive  of  stromateoids,  and  their  deriva-  Changes  in  the  dentition  by  and  large  re- 

tive,  the  stromateids,  lack  these  teeth.  Pala-  fleet  increasing  specialization   in  the  food 

tal  dentition  is  well  developed  in  the  tetra-  habits. 

gonurids,  which  are  derived  from  fish  an-  Pharyngeal  sacs  (Figs.  49,  .50).  The  sacs 

cestral    to    the    nomeids.    The    ariommids,  of  centrolophids  are  higher  than  they  are 

which  are  probably  derived  directly  from  a  long,  and  the  papillae  are  arranged  in  ten 

nomeid,  have  lost  these  teeth.  or  more  elongate  patches  (Fig.  49A).    The 

In  the  majority  of  primitive  centrolophids  bases  of  the  papillae  ( Fig.  50A )  are  irregu- 
and  in  many  nomeids,  the  jaw  teeth  are  lar  in  shape,  with  the  teeth  seated  all  over 
relatively  large,  strong,  spaced,  and  often  the  inner  face;  the  base  is  often  humped  up 
slightly  recurved.  The  advanced  centro-  to  fit  over  a  ridge  of  muscle  in  the  sac.  In 
lophids,  stromateids,  and  ariommids  have  the  nomeids,  the  sac  is  not  so  high  in  re- 
much  smaller,  close-set,  straight  teeth.  The  ^pect  to  its  length,  and  the  papillae  are  in 
teeth  of  all  stromateids  and  of  some  ariom-  ^bout  five  longitudinal  patches  (Fig.  49B). 
mids  are  laterally  flattened  and  bear  minute  ^he  papillae  are  verv  different  from  those 
cusps.  Tetragonurids  and  some  species  in  ^^  ^j^^  centrolophids.  'The  bases  ( Fig.  SOB  ) 
the  nomeid  genus  Psenes  have  two  sorts  of  ^^^  ^^^^^^^^  ^^^^  ^j^^  ^^^^j^  ^^^  concentrated 
teeth  m  the  jaws.    Tliose  m  the  upper  ,aw  ^^^^^  ^^^  ^^^^  ^^  ^  ^^^j^    ^^^  ^^^  .^^  ^^^^^^_ 

are  conical,  spaced,  and  recurved;  those  m  .   . ,     ,  t--      Ar,r^  \   •      i.  i      i.        i               -^  • 

^,     ,           .     ^        ,         ,    .r  ,.,                ,  teids  (Fig.  49C)  is  at  least  as  long  as  it  is 

the  lower  jaw  are  long,  knitelike,  very  close  r,-  i,    •     p              /t^-      4f-\   i.  • .         .j   j 

set,  and  often  bear  verv  minute  cusps.  ,°  '         „    '       v     &•      I                          h    ■ 

The  structure  of  the  teeth  and  jaws  is  ^he   papillae  are   m    only   two   ill-defmed 

certainlv  a  function  of  the  diet.   In  stromal-  P^tdies  in  the  top  and  bottom  halves  of  the 

eoids,   which   possess   a   masticatory   organ  sac.  As  in  the  nomeids,  the  bases  (Fig.  50C) 

in  the  pharyngeal  sacs,  the  jaw  teeth  are  are  stellate,  but  they  are  in  general  larger, 

primarily  for  catching  and  holding  prey.  and  the  teeth  are  seated  all  along  the  long 

The  diet  of  most  centrolophids  and  no-  central  stalk  instead  of  only  near  the  end. 

meids  is  fairly  diverse  and  often  includes  The  sacs  in  both  the  ariommids  ( Fig.  49D ) 

rather  large  animals.    The  strong,  conical,  and  the  tetragonurids  are  longer  than  high, 

slightly  recurved  teeth  are  ably  suited  to  markedly  so  in  the  latter  (Fig.  36).    The 

the  catching  and  holding  of  fairly  vigorous  large  papillae  of  ariommids  (Fig.  SOD)  have 

prey.    The  stromateids   feed  rather  exten-  round  bases,  and  the  small  teeth  are  seated 

sively  on  jellyfishes.  Their  smaller  jaws  and  all  along  the  central  stalk.   The  papillae  are 


118        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


A 


B 


.^ri1?«^?;'#'^*^ 


5?aki     ..       ;.  I     ^  ,■•,'- 73^"t  »•!  'J'. ^ 


Figure  49.  Comparison  of  bronchial  regions  in  four  stromateoid  families.  A.  Centrolophidoe,  Hyperoglyphe,  from  Figure  9. 
B.  Nomeidoe,  Nomeus,  from  Figure  25.  C.  Stromateidae,  Peprilus,  from  Figure  43.  D.  Ariommidae,  Ariommo,  from  Figure 
31 .    See  text. 


in  a  single  patch,  in  the  upper  half  of  the 
sac  only.  In  tetragonurids,  the  small  papil- 
lae are  widely  separated  and  are  not  in 
bands.  They  are  rounded,  and  there  are  a 
few  weak  teeth  on  the  end  of  a  short  stalk. 
The  nomeids  and  the  stromateids  do  not 
share  a  direct  common  ancestor.  The  no- 
meids are  derived  from  a  pre-centrolophid 
form  and  the  stromateids  are  derived  from 
an    advanced    centrolophid.     Nonetheless, 


there  is  a  great  similarity  in  the  stellate 
papillae  found  in  both  families  (Fig.  SOB, 
C),  but  this  similarity  is  due  to  parallelism. 
The  centrolophid  fishes  are  unspecialized 
in  their  diets.  They  feed  on  other  fishes, 
on  squids,  on  crustaceans,  on  jellyfishes 
and,  sometimes  but  certainly  not  customar- 
ily, on  garbage.  The  large  sacs  are  capable 
of  admitting  fairly  large  objects.  The  crude 
papillae  do  shred  the  prey  to  some  extent. 


Stbomateoid  Fishes  •  Haedrich 


119 


Al 


D 


A,  A,,    B,  C 


bin 


Figure  50.  Comparison  of  papillae  in  the  pharyngeal  sacs  of  four  stromateoid  families.  A.  Centrolophidae,  Hyperoglyphe, 
from  preparation  shown  in  Figure  9,  large  papilla.  A).  Same,  small  papilla.  B.  Nomeidae,  Nomeus,  from  preparation 
shown  in  Figure  25.  C.  Stromateidae,  Pepri/us,  from  preparation  shown  in  Figure  43.  D.  Ariommidae,  Ariomma,  from  prep- 
aration shown  in  Figure  31.    See  text. 


but  never  so  much  as  to  render  stomach 
contents  completely  unrecognizable. 

Little  is  known  of  the  feeding  habits  of 
nomeids.  Fish  and  jelhfish  remains  have 
been  found  in  their  stomachs.  The  fairly 
small  sacs  and  the  papillae  with  their  stel- 
late bases  firmly  seated  in  the  muscular 
wall  of  the  sac  combine  to  make  a  good 
shredding  organ,  and  stomach  contents  are 
often  difficult  to  identify. 

Stromateids    may    feed    very    largely   on 


small  crustaceans  and  medusae.  The  rela- 
tively smaller  and  more  elongate  sacs,  the 
papillae  with  greatly  extended  bases,  and 
the  teeth  ranged  all  along  the  central  stalk 
of  the  papillae  make  a  very  efficient  shred- 
ding organ,  ably  suited  to  rendering  the 
rubbery  tissues  of  medusae.  The  shredded 
stomach  contents  of  stromateids  are  almost 
impossible  to  identify. 

Too  little  is  known  of  the  natural  history 
of  ariommids  to  be  able  to  understand  the 


120        Bulletin  Musctim  of  Comparative  Zoology,  Vol.  135,  No.  2 


structure  of  the  peculiar  pharyngeal  sacs 
found  m  this  group.  In  almost  every  speci- 
men examined,  the  sacs  were  filled  with 
mud  and  silt.  Is  it  possible  they  perform 
some  sort  of  filtering  function? 

Tetragonurids  may  live  largely  on  jelly- 
fishes.  The  sacs  of  these  fishes  are  very 
elongate,  as  might  be  expected,  but  the 
papillae  are  very  reduced  and  are  probably 
not  very  efficient  shredders.  The  upper 
pharyngeal  bones,  houever,  are  studded 
with  teeth  and  extend  verv  far  backward 
into  the  sacs  (Fig.  46).  The  pharyngeal 
bones  are  capable  of  considerable  back-and- 
forth  motion  (Grey,  1955)  and,  in  tetra- 
gonurids, may  perform  the  shredding  action 
for  which  the  papillae  do  not  seem  suited. 

The  main  changes  that  have  occurred  in 
the  pharyngeal  sacs  of  stromateoids  have 
been  elongation  of  the  sac,  reduction  in  the 
number  of  bands  of  papillae,  and  increase 
in  complexity  of  the  papillae.  These  changes 
are  correlated  with  a  change  from  more  or 
less  omnivorous  feeding  habits  to  increasing 
utilization  of  jelly  fishes  for  food. 

Branchial  rciiion  (Figs.  4<S,  49).  Seven, 
blunt-ended  branchiostegals  and  a  large 
ceratohyal  fenestra  are  found  at  the  centro- 
lophid  grade  (Fig.  49A).  The  advanced 
centrolophids  ScriolcUa  and  Pscnop.sis  have 
pointed  branchiostegals,  and  the  first  one  is 
reduced  in  size  (Figs.  20,  23).  In  nomeids 
(Fig.  49B)  and  ariommids  (Fig.  49D)  there 
are  six  tapering  branchiostegals,  and  the 
ceratohyal  fenestra  is  much  smaller,  or,  in 
some  species,  closed.  At  the  stromateid 
grade  ( Fig.  49C )  there  are  six  tapered 
branchiostegals  and  the  ceratohyal  fenestra 
is  closed.  The  stromateid  genus  Potnpus, 
perhaps  the  most  advanced  of  stromateoids, 
has  but  five  branchiostegal  rays.  Within  the 
tetragonurids,  the  number  of  branchioste- 
gals is  either  six  or  five. 

A  pseudobranch  is  present  in  all  stromat- 
eoids with  the  exception  of  Pampas.  Its 
loss  may  be  correlated  with  the  unification 
of  the  gill-covers  to  the  isthmus.  In  most 
stromateoids,  the  pseudobranch  is  very  well 
developed  and  the  gill-covers  are  cleft  well 


fonvard.  In  the  stromateid  genera  Stronia- 
teus  and  Pcprilus  the  pseudobranch  is  small 
and  the  gill-covers  are  united  across,  but 
not  to,  the  isthmus.  Finally,  in  Pampiis  the 
pseudobranch  is  gone  and  the  gill-covers 
have  become  broadly  united  to  the  isthmus. 

In  the  hyal  series  of  stromateoids,  two 
changes  have  occurred.  The  branchiostegal 
rays  ha\'e  become  more  slender  and  one 
ray  is  lost,  and  the  ceratohyal  fenestra  be- 
comes closed.  The  pseudobranch,  well  de- 
veloped in  most  stromateoids,  is  lost  in 
Pampiis. 

Axial  skeleton  (Fig.  48).  Most  centro- 
lophids have  10  -I-  15  vertebrae,  the  well- 
know  n  basic  perciform  number.  The  excep- 
tions are  in  some  Schedophilu.s  with  10  4- 
16,  12  +  17,  or  10  +  20,  and  Icichthys  with 
a  total  of  50  to  60.  In  the  nomeids  both 
numbers  have  increased;  there  are  13  to  15 
precaudal,  and  17  to  23,  26,  or  27  caudal  ver- 
tebrae. The  stromateids  have  12  to  IS  pre- 
caudal, and  19  to  27  caudal  vertebrae,  and 
within  any  one  species  the  number  may  be 
quite  variable.  Ariommids,  derived  from  a 
nomeid  stock,  usually  have  12  or  13  +  17 
or  IS  vertebrae.  The  tetragonurids,  also 
derived  from  the  nomeids,  have  continued 
the  increase,  to  a  total  of  43  to  5S  vertebrae. 

It  is  difficult  to  understand  the  selective 
pressures  responsible  for  an  increase  in  ver- 
tebral niunber.  In  Nomeu.s,  however,  the 
situation  is  reasonably  clear.  This  genus  has 
41  vertebrae  and  is  elongate  in  fomi.  The 
high  number  of  vertebrae  allows  the  fish 
to  move  in  a  very  sinuous  manner,  and  to 
turn  in  a  very  small  radius.  The  pelvic  fins 
are  also  much  enlarged  and  aid  in  the  turn- 
ing. This  ability  enables  the  fish  to  avoid 
more  easily  the  stinging  tentacles  of  the 
PJujsalia  under  which  it  lives. 

In  general,  as  the  number  of  vertebrae 
has  increased,  the  relative  length  of  each 
individual  vertebra  has  decreased.  Fishes 
with  an  increased  number  of  vertebrae  have 
more  neural  and  haemal  spines  than  fishes 
with  fewer  vertebrae,  and  these  are  closer 
together.    Hence  there  is  a  stronger  frame 


Stromateoid  Fishes  •  Hacdrich 


121 


Figure  51.  Comparison  of  caudal  skeletons  of  four  stromateoid  families.  A.  Centrolopfiidae,  Hyperoglyphe,  from  Figure  10. 
B.  Nomeidae,  Nomeus,  from  Figure  24.  C.  Stromoteidae,  Peprilus,  from  Figure  42.  D.  Ariommidae,  Ariomma,  from  Figure 
33.    See  text. 


for  attachment  of  the  muscles.  This  prob- 
ably has  permitted,  or  even  encouraged, 
the  evolution  of  the  deep,  firm  body  char- 
acteristic of  the  stromateid  grade. 

With  the  increase  in  vertebral  number, 
the  number  of  median  finravs  increases  as 
well.  This,  of  course,  might  be  expected, 
but  need  not  necessarilv  follo\\ .  The  ariom- 
mids  and  the  tetragonurids,  both  with  in- 
creased numbers  of  vertebrae,  have  very 
decreased  numbers  of  median  finrays.  In 
Psenes  and  in  the  stromateids,  the  increased 
number  of  anal  finrays  has  resulted  in  a 
forward  swing  of  the  first  interhaemal  so 
that  it  forms  an  abrupt  angle  with  the 
haemal  spine  of  the  first  precaudal  verte- 


bra. This  tends  to  support  and  protect  the 
belly  of  the  fish.  In  the  stromateids,  the 
enlargement  and  extension  of  the  pelvic 
bones,  which  bear  no  fins,  and  of  the  post- 
cleithrum  almost  complete  this  trend  to- 
wards support  and,  possibly,  protection. 

There  has  been  a  general  tendency  to- 
^^'ards  increase  in  the  number  of  \ertebrae 
in  the  evolution  of  the  stromateoids.  The 
number  of  caudal  vertebrae  has  tended  to 
increase  the  most,  but  the  number  of  pre- 
caudal vertebrae  has  been  affected  as  well. 
Secondary  increase  in  the  number  of  ver- 
tebrae from  a  basic  number  near  10  +  15  is 
of  common  occurrence  in  teleosts  ( Gregory, 
1951;  Liem,  1963).    In  many  stromateoids. 


122        Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  2 


there  has  been  a  concomitant  increase  in 
the  number  of  median  finrays. 

Caudal  skeleton  (Figs.  48,  51).  The  gen- 
erahzed  percifonn  type  of  caudal  skeleton 
with  six  hypurals  and  three  epurals  is  t\  pi- 
cal  of  the  centrolophid  grade  (Fig.  51A). 
With  the  fusion  of  hypurals  2  +  3  and  3  + 
4,  the  number  is  reduced  to  fovu"  in  the  no- 
meids  (Fig.  51B),  stromateids  (Fig.  51C), 
and  tetragonurids.  Three  epurals  are  pres- 
ent in  nomeids,  but  one  of  these  is  lost  at 
the  stromateid  grade  (Fig.  51C).  In  the 
ariommids  (Fig.  51D),  the  fusion  of  hy- 
purals 1  +  2  +  3  and  4  +  5  +  6  fonns  two 
solid  blocks.  In  both  the  ariommids  and 
some  stromateids  there  is  a  tendency  to- 
wards further  fusion  of  hypural  elements 
with  the  urostylar  vertebrae.  The  ariom- 
mids have  three  epurals  but  the  second  one 
is  very  reduced  in  size,  and  is  probably  on 
its  way  to  becoming  lost.  The  tetragonurids 
have  two  epurals.  Fusion  of  the  uroneurals, 
both  with  each  other  and  with  the  urostylar 
vertebrae,  has  occurred  in  several  stromat- 
eoid  genera. 

Most  of  the  centrolophid  fishes  spend  at 
least  the  first  part  of  their  lives  hovering 
quietly  under  floating  objects.  These  fishes 
are  able  to  hang  almost  motionless  with  a 
slight  fanning  of  the  pectorals  and  strong 
rotary  motion  of  the  caudal  fin.  This  rotary 
motion  is  possible  because  of  the  nimierous 
elements  in  the  caudal  skeleton.  In  the 
advanced  centrolophids — fast-swimming, 
schooling  fishes  such  as  Seriolella  violaceo — 
partial  fusions  in  the  hypural  series  result 
in  a  more  rigid  tail. 

Many  of  the  nomeids  are  hovering  fishes. 
At  this  grade,  fusions  in  the  hypural  series 
tend  to  make  the  tail  fairly  stiff.  This  may 
be  counteracted  by  the  long,  well-developed 
autogenous  haemal  spines,  which  may  be 
moved  laterally  to  produce  a  rotary  motion 
in  the  fin.  The  long  pectoral  fins  of  no- 
meids, too,  may  aid  their  hovering.  Obser- 
vations on  living  fishes  are,  however,  lack- 
ing, and  are  sorely  needed. 

The  consolidated  tail  of  stromateids  al- 
lows for  little  rotarv  motion.   These  school- 


ing fishes  probably  do  not  hover  as  much 
as  nomeids  or  centrolophids  but  may  swim 
fairly  constantly.  Specimens  of  Feprilus 
triacanthus  observed  in  the  Woods  Hole 
Aquarium  never  remained  still,  but  moved 
slowly  forward,  bouncing  up  and  down  with 
beats  of  their  long,  broad  pectoral  fins. 

There  are  no  observations  of  living  ariom- 
mids or  tetragonurids.  From  the  structure 
of  their  caudal  skeleton  and  fin,  it  can  be 
assumed  that  the  former  at  least  are  very 
strong,  fast  swimmers.  Living  near  the  bot- 
tom, they  may  not  need  to  hover,  but  may 
cruise  over  the  sea  floor  buoyed  up  by  their 
well-developed  air  bladder. 

The  evolution  of  the  caudal  skeleton  in 
stromateoids  is  marked  by  a  reduction  in 
the  number  of  elements.  In  the  hypural 
series,  this  reduction  is  accomplished  by  a 
series  of  fusions;  in  the  epural  series,  an 
element  is  lost.  The  tendency  toward  con- 
solidation and  reduction  of  elements,  ulti- 
mately resulting  in  a  fused  hypural  plate,  is 
a  general  phenomenon  found  in  numerous 
percifonn  lineages  (Gosline,  1961a).  The 
changes  in  the  stromateoid  tail  coincide  ap- 
proximately with  a  change  from  hovering 
to  swimming  fairly  constantly  in  schools. 

DISTRIBUTION  OF  THE  STROMATEOIDEI 

Distributional  data  for  stromateoid  fishes 
are  at  best  scanty.  Nonetheless,  all  avail- 
able data  tend  to  support  the  conclusions 
based  on  anatomical  data,  that  is,  that  the 
centrolophids  arose  first,  followed  by  the 
nomeids,  tetragonurids,  and,  most  recently, 
the  stromateids  and  ariommids.  Because  of 
the  scantiness  of  the  data,  the  map  figures 
accompanying  this  section  must  be  con- 
sidered approximate  only.  In  general,  the 
distributions  have  been  extrapolated  from  a 
few  records.  Although  I  am  fairly  sure  of 
the  general  picture  presented,  fine  details 
of  the  distribution  of  stromateoids  are  lack- 
ing. 

CentroJophidae.  The  major  features  of 
the  centrolophid  distribution  are  disconti- 
nuity, bipolarity,  endemism,  and  sympatry 
of  genera.   The  first  three  are  found  in  the 


Stromateoid  Fishes  •  Haedrich 


123 


Figure  52.      Distribution  of  ttie  soft-spined   Centrolophidae. 


more  primitive  members  of  the  family,  the 
soft-spined  centrolophids  (Fig.  52).  Schedo- 
philus  is  found  mainly  in  the  Atlantic 
Ocean.  The  presence  of  two  isolated  popu- 
lations, in  the  China  seas  and  in  the  en- 
virons of  the  Tasman  Sea,  indicates  that  the 
former  range  of  the  genus  was  once  much 
wider.  The  distribution  of  the  two  Pacific 
area  populations  of  SchedopJiihis  and  the 
distribution  of  Centrolophus  are  bipolar, 
again  indicative  of  a  shrinking  range. 
IcichtJujs,  fonnerly  considered  an  endemic 
element  of  the  North  Pacific  fauna,  has  re- 
cently been  found  off  New  Zealand  ( Haed- 
rich, in  press),  and  is  thus  bipolar.  Icichthys 
and  Centrolophus  probably  share  a  com- 
mon ancestor;  the  characteristics  of  each 
genus  may  well  have  developed  in  the  iso- 
lation provided  by  an  ancestral  relict  dis- 
tribution. In  the  hard-spined  centrolophids 
(Fig.  53),  Hyperoglyphe,  like  SchcdopJiihts, 
is  bipolar  in  the  Pacific  but  widespread  in 
the  Atlantic.  The  most  primitive  species  in 
this  genus,  H.  antarctica,  is  found  only 
south  of  30°  S.  Seriolella,  a  relatively  ad- 
vanced genus,  is  widespread  in  the  higher 
latitudes  of  the  Southern  Hemisphere, 
where  it  is  no  doubt  endemic.  The  most 
recently  evolved  centrolophid  genus,  Pse- 


nopsis,  may  be  spreading  out  from  the  wa- 
ters of  the  East  Indian  region.  By  and  large, 
centrolophid  species  are  oceanic  or  found 
near  the  edge  of  the  continental  shelf.  Some 
species  of  the  soft-spined  centrolophids  may 
even  be  meso-  or  bathypelagic.  The  ad- 
vanced genera  Seriolella  and  Psenopsis, 
however,  commonly  occur  in  shallow  wa- 
ter, and  some  species  may  even  enter  estu- 
aries. Some  overlap  with  at  least  one  other 
genus  occurs  within  the  ranges  of  all  cen- 
trolophid genera.  Four  of  the  six  genera 
occur  in  Australia  and  New  Zealand. 

Nomeidae.  In  the  distribution  of  the  no- 
meids  (Fig.  54)  there  are  no  relicts,  no  bi- 
polar species,  and  no  regional  endemism. 
For  the  most  part,  the  three  genera  seem 
broadly  sympatric,  but  records  are  too  few 
to  discuss  the  limits  of  each  genus  with 
precision.  In  the  North  Atlantic,  however, 
Nomeus  is  found  in  the  western  parts,  but 
has  never  been  reported  from  Madeira, 
where  its  companion  Physalia  is  common. 
Cuhiceps,  though  it  occurs  in  the  western 
Atlantic,  is  much  more  common  in  the  east- 
em  portions  and  the  Mediterranean.  Most 
nomeid  species  are  oceanic;  a  few  species 
in  Psenes  seem  to  be  mesopelagic.  In  gen- 
eral  confined   to   more   tropical   waters,    a 


124        Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  2 


Figure    53.      Distribution    of    the   hard-spined    Centrolophidae. 


number  of  species,  such  as  Nomeus  ^ronovii 
and  Psenes  cyonoplinjs,  are  found  in  all 
oceans.  All  genera  are  found  in  Australia 
and  New  Zealand. 

Tctruiiomuidae.  The  distribution  of  the 
tetragonurids  is  very  poorly  known.  In  gen- 
eral it  seems  to  approximate  the  distribu- 
tion of  the  nomeids  ( Fig.  54 ) . 

Ariommidae.  The  family  Ariommidae,  a 
nomeid  derivative,  is  found  mainly  in  tropi- 
cal waters  (Fig.  55).  One  deep-bodied  spe- 
cies occurs  off  South  Africa.  All  members 
of  the  single  genus  Arionima  seem  to  be 
engybenthic  in  deep  water  over  continental 
shelves  or  near  islands.  The  deep-bodied 
and  elongate  forms  of  Ario77rma  occiu-  to- 
gether in  the  New  World,  but  tend  to  be 
allopatric  elsewhere.  Apparently  there  are 
no  representatives  on  the  west  coast  of  Cen- 
tral America,  although  the  genus  is  wide- 
spread throughout  the  Gulf  of  Mexico  and 
the  Caribbean,  and  elongate  species  occur 
in  Hawaii.  The  latter  are  undoubtedly  de- 
rived from  Japanese  forms.  The  most  ad- 
vanced species  in  the  genus,  A.  indica,  is  a 
deep  bodied  silvery  species  with  cusped 
teeth,  found  from  the  Gulf  of  Iran  through- 
out the  East  Indian  region  to  the  East  China 


Sea.    No  ariommids  occur  in  Australia  or 
New  Zealand. 

Sticmmteidae.  The  stromateid  distribu- 
tion is  characterized  by  continuity,  wide- 
spread species,  restriction  to  continental 
shelves,  a  trans-Isthmian  genus  in  the  New 
World,  and  allopatry  of  genera.  The  dis- 
tributions of  each  of  the  three  genera  ( Fig. 
56)  are  more  or  less  continuous.  In  S^ro- 
matciis,  one  species  is  found  from  the  Medi- 
terranean to  South  Africa.  Stromatctis  is 
the  only  genus  that  has  managed  to  cross 
an  ocean.  This  has  been  accomplished 
across  the  shortest  possible  gap,  from  Africa 
to  South  America,  and  in  the  direction  of 
the  prevailing  winds  and  currents.  The  ad- 
vanced StromateiLS  of  the  east  and  west 
coasts  of  southern  South  America  are  very 
little  differentiated  from  one  another,  and 
may  be  speciating  at  the  present  time.  The 
genus  Feprilus,  apparently  derived  from 
Stromateu.s  through  a  species  such  as  the 
west  coast  P.  snijderi.  has  spread  on  both 
coasts  of  North  America  and  southward 
along  the  east  coast  of  South  America  to 
Uruguay,  where  it  occurs  sympatrically 
with  Stwmatetis.  This  is  the  only  place 
where  two  stromateid  genera  are  found  to- 


Stromateoid  Fishes  •  Haedrich        125 


Figure  54.      Distribution  of  the  Nomeidae:    Cubiceps,  Nomeus,  and  Psenes. 


gether.  The  most  advanced  stromateid 
genus,  Pampiis,  occurs  from  the  Gulf  of 
Iran  to  Japan.  Both  an  advanced  species, 
P.  orgenteits,  and  a  more  primitive  species, 
P.  chincnsis,  occur  through  most  of  the  East 
Indian  region.  No  stromateids  have  crossed 
Wallace's  line  into  Australia  or  New  Zea- 
land. 

Discussion.  The  two  most  recently 
evolved  families  are  the  ariommids  and  the 
stromateids.  In  the  ariommids,  a  single 
genus  is  widespread.  Containing  two  in- 
cipient genera,  one  elongate  and  one  deep- 
bodied,  AriotJima  has  apparently  had  insuf- 
ficient time  for  characteristics  worthy  of 
generic  division  to  develop.  Three  genera 
have  evolved  in  the  stromateids,  but  in 
general  each  genus  is  restricted  to  a  sepa- 
rate continental  area.  Peprihis  has  spread 
in  a  classical  circular  pattern,  from  Pacific 
South  America  across  the  Isthmus  of  Pan- 
ama and  south  to  Uruguay,  to  re-encounter 
the  ancestral  Strornatcus  stock.  This  small 
region  in  the  western  South  Atlantic  is  the 
only  area  where  stromateid  genera  occur 
sympatrically.  Speciation  is  currently  active 
in  both  families. 

The  ariommids  and  the  stromateids,  in 
contrast  to  the  other  stromateoid  families. 


are  restricted  to  near  land.  The  ariommids 
live  in  deep  water  over  the  shelves  and  in 
the  vicinity  of  Hawaii.  The  stromateids  may 
prefer  quite  shallow  water,  and  occur  in 
large  schools  in  wide  embayments.  Be- 
cause of  this  relationship  with  the  land,  it 
is  possible  to  examine  and  possibly  date  the 
emergence  of  the  two  families  in  the  light 
of  past  tectonic  activity. 

The  present  stromateid  distribution 
throughout  Asia  (but  not  the  Red  Sea), 
the  Mediterranean,  West  Africa,  and  the 
New  World  is  strongly  suggestive  of  an 
ancestral  Tethyan  distribution.  The  stro- 
mateid ancestor  could  have  been  a  member 
of  the  warm  water  shelf  fauna  which  ex- 
tended uninterrupted  across  this  region  in 
Tertiary  times.  In  the  Upper  Eocene  or 
Oligocene,  the  emergence  of  land  in  the 
Near  East  divided  this  fauna  in  two.  The 
ancestral  stromateid  isolated  in  the  East 
gave  rise  to  Pampus.,  the  form  in  the  West 
was  the  central  Stwmafetis  stock.  In  the 
Pacific  Panamanian  region,  separated  from 
the  southern  stock  of  Strornatcus  by  long 
coastlines  unsuitable  for  stromateids,  Pep- 
rihis evolved.  Found  today  in  both  oceans, 
this  genus  must  have  been  established  be- 


126        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  2 


Figure  55.      Distribution  of  the  Ariommidoe. 


fore  the  emergence  of  the  Isthmus  of  Pan- 
ama in  the  lower  PHocene. 

The  ariommids  are  not  so  tightly  bound 
to  the  coasts  as  the  stromateids.  Oceanic 
dispersal  may  be  facilitated  by  pelagic 
juveniles,  a  few  of  which  have  been  taken 
at  Bermuda  and  in  the  tropical  Central  At- 
lantic. Widespread  in  the  Gulf  of  Mexico 
and  the  Caribbean,  no  ariommids  occur  on 
the  west  coast  of  the  New  World.  This 
suggests  at  least  a  late  Pliocene  dispersal. 
This  fact,  plus  the  remarkable  homogeneity 
of  the  group  and  the  structure  of  the  pha- 
ryngeal sacs  and  the  fused  hypural  fan,  are 
strong  evidence  for  considering  the  ariom- 
mids the  most  recently  evolved  stromateoid 
fishes. 

The  remaining  stromateoid  families,  the 
centrolophids,  nomeids,  and  tetragonurids, 
are  by  and  large  all  oceanic.  Changes  in 
the  configuration  of  the  land  would  not 
have  affected  these  fishes  as  they  did  the 
ariommids  and  stromateids.  Since  the  major 
ocean  basins  have  probably  been  a  stable 
feature  since  well  before  the  Cretaceous, 
the  period  of  the  great  flowering  of  the 
teleosts,  it  is  unlikely  that  tectonic  activity 
has  been  an  important  isolating  mechanism 
in  the  evolution  of  these  groups. 


The  centrolophid  distribution  bears  all 
the  earmarks  of  an  older  group.  Disjunct 
distributions,  including  bipolarity,  are  char- 
acteristic of  an  old  group  which  has  passed 
its  peak.  Another  indication  of  the  age  of 
the  centrolophids  is  their  diversity.  There 
are  six  genera  in  the  family,  and  the  spe- 
cies inhabit  a  wide  range  of  en\'ironments. 
In  Australia  and  New  Zealand,  where  no 
stromateids  occur,  the  advanced  centro- 
lophid genus  Seriolella  lives  in  shallow  wa- 
ters near  the  coast,  the  typical  stromateid 
habitat.  Numerous  ebbs  and  flows  have 
occurred  in  the  distribution  of  the  centro- 
lophids, for  numerous  genera  are  found  to- 
gether. Lacking  fossils,  it  is  impossible  to 
date  the  emergence  of  the  centrolophids. 
Nonetheless,  they  certainly  antedate  the 
stromateids  which  had  their  beginnings  in 
the  mid-Tertiary.  The  centrolophids,  then, 
probably  arose  in  the  early  Tertiary,  or  per- 
haps even  in  the  late  Cretaceous. 

The  nomeids  probably  arose  concurrently 
with,  or  perhaps  a  little  after,  the  centro- 
lophids. The  genera  occur  together  through- 
out the  range  of  the  family.  Little  specific 
differentiation  seems  to  have  developed,  al- 
though the  apparent  commonness  of  cir- 
cumtropical  species  in  this  group  may  only 


Strom ATEOiD  Fishes  •  Haedrich        127 


Figure  56.      Distribution   of  the  Stromateidae. 


reflect  the  premium  placed  on  a  particular 
phenotype  in  the  rigorous  oceanic  environ- 
ment. 

The  great  divergence  from  the  nomeid 
condition  of  a  derived  family,  the  tetra- 
gonurids,  suggests  that  they  branched  off 
at  an  early  stage.  The  tetragonurids  have 
become  very  specialized,  and  may  be  an 
evolutionary  dead-end.  There  are  appar- 
ently only  three  species  in  the  single  genus. 

Two  general  features  of  the  stromateoid 
distributions  are  of  interest.  The  first  is  the 
tendency  for  the  more  primitive  taxa  to  be 
found  in  higher  latitudes.  Included  in  this 
group  are:  CentroJophus\  Icichthys,  Hijpero- 
ghjphe  antarctica,  Stromateus  in  the  New 
World,  and,  perhaps,  the  giant  nomeid 
Cubiceps  capensis.  The  Ariommidae  and 
Stromateidae,  both  advanced,  have  not 
reached  the  Australian  region.  The  second 
feature  is  that  the  most  advanced  or  most 
recently  evolved  taxa  have  their  centers  of 
distribution  in  the  East  Indian  region.  Ex- 
amples are  the  centrolophid  genus  Psenop- 
sis,  the  stromateid  genus  Tampus,  and  the 
ariommid  Ariomma  indica. 

ACKNOWLEDGEMENTS 

In  the  course  of  this  studv  I  have  been 


supported  by  an  Emerson  Fellowship  from 
Harvard  University  and  a  Summer  Fellow- 
ship and  Predoctoral  Fellowship  from  the 
Woods  Hole  Oceanographic  Institution. 
The  National  Science  Foundation  spon- 
sored my  participation  on  Cruise  6  of  the 
ANTON  BRUUN,  as  a  part  of  the  U.  S. 
Program  in  Biology,  International  Indian 
Ocean  Expedition,  and  also  permitted  me 
to  visit  Japan  under  the  joint  U.  S. -Japan 
Cooperative  Program  in  Science  (OF- 147 
to  Harvard  University).  Funds  from  the 
Harvard  Committee  on  Evolutionary  Biol- 
ogy covered  my  visits  to  natural  history 
museums  in  Washington,  London,  Paris, 
and  Copenhagen.  Incidental  support  has 
been  derived  from  National  Science  Foun- 
dation Grant  GB-543  to  the  Woods  Hole 
Oceanographic  Institution. 

This  study  would  not  have  been  possible 
without  the  kind  cooperation  of  numerous 
curators  who  have  freel\-  sent  me  specimens 
from  far-flung  collections.  For  these  ser- 
vices my  thanks  go  to  Frank  J.  Talbot  of 
the  Australian  Museum,  Sydney;  William 
J.  Richards  of  the  Bureau  of  Commercial 
Fisheries  Biological  Laboratory,  Washing- 
ton, D.  C;  Alfred  W.  Ebeling  of  the  Bing- 
ham Oceanographic  Laboratory,  Yale  Uni- 


128        Bulletin  Muscuttt  of  Comparative  Zoology,  Vol.  135,  No.  2 


versitv;  Frank  Williams  of  the  Guinean 
Trawling  Survey,  Lagos;  M.  J.  Penrith  of 
the  South  African  Museum,  Capetown; 
George  S.  Myers  of  the  Natural  History 
Museum,  Stanford  University;  Richard  H. 
Rosenblatt  and  Joseph  F.  Copp  of  the 
Scripps  Institution  of  Oceanography,  La 
Jolla;  Werner  Ladiges,  Zoologisches  Mu- 
seum, Hamburg;  and  J0rgen  G.  Nielsen  of 
the  Zoological  Museum,  Copenhagen.  On 
my  visits  to  various  institutions,  which  also 
supplied  specimens  for  the  study,  I  received 
the  very  best  of  treatment  from  James  Tyler 
of  the  Academy  of  Natural  Sciences  of 
Philadelphia;  C.  C.  Lindsey  and  Norman 
J.  Wilimovsky  of  the  University  of  British 
Columbia,  Vancouver;  P.  H.  Greenwood 
and  A.  C.  Wheeler  of  the  British  Museum 
(Natural  History),  London;  Loren  P.  Woods, 
Marion  Grey,  and  Pearl  Sonoda  of  the  Chi- 
cago Museum  of  Natural  History;  M. 
Bauchot  and  M.  Blanc  of  the  Museum  Na- 
tional d'Histoire  Naturelle,  Paris;  and 
Daniel  M.  Cohen,  Bruce  B.  Collette,  Robert 
H.  Gibbs,  Jr.,  Nathaniel  Gramblin,  and 
Leonard  P.  Schultz  of  the  United  States 
National  Museum,  Washington,  D.  C. 
Frank  J.  Mather,  HI  of  the  Woods  Hole 
Oceanographic  Institution  allowed  me  free 
access  to  collections  made  in  connection 
with  his  pelagic  fish  studies  and  also  ex- 
amined the  types  of  SerivIeUa  in  Paris. 
Especial  thanks  go  to  Erik  Bertelsen  of  the 
Danish  Carlsberg  Foundation  who  provided 
the  nucleus  of  specimens  which  initiated 
this  study  and  who  has  encouraged  it 
throughout,  and  to  Tokiharu  Abe  who  pro- 
vided numerous  Japanese  specimens,  who 
has  increased  my  understanding  through 
discussions,  and  who  was  to  me  a  most 
gracious  host  during  my  stay  in  his  country. 
Especial  thanks  also  go  to  the  staff  of  the 
Museum  of  Comparative  Zoology — particu- 
larly Myvanwy  M.  Dick,  Josie  DeFalla, 
Elaine  Kelley,  and  Charles  Karnella — and 
to  the  staff  of  the  Marine  Biological  Labo- 
ratory Library — particularly  Jane  Fes- 
senden — for  help  cheerfully  given. 

I   gratefully  acknowledge   the  particular 


services  rendered  by  a  number  of  people, 
all  friends  and  colleagues  at  the  Woods 
Hole  Oceanographic  Institution.  Louise 
Russell  made  the  drawings  of  the  gill  arches 
and  pharyngeal  sacs.  Clifford  Hinton,  Jr. 
made  the  final  plates,  and  through  discus- 
sion gave  me  much  insight  into  the  prob- 
lems of  photographic  reproduction  and  of 
technical  illustration.  Barbara  Jones,  Mar- 
tin Bartlett,  and  George  Grice  have  been 
good  listeners,  and  have  lent  assistance  at 
many  points.  James  Craddock  has  criticized 
parts  of  the  manuscript,  in  particular  the 
keys.  And  Jane  Peterson  has  never  faltered 
in  her  cheerful  typing,  retyping,  and  typing 
again  of  the  manuscript. 

Finally,  I  owe  a  special  debt  of  thanks 
to  two  people  who  have  discussed,  criti- 
cized, and  encouraged  this  work  in  all  of 
its  stages.  The  first  of  these  is  Richard  H. 
Backus,  Senior  Scientist  at  the  Woods  Hole 
Oceanographic  Institution.  The  second  is 
Giles  W.  Mead,  Curator  of  Fishes  at  the 
Museum  of  Comparative  Zoology,  my  ad- 
viser at  Harvard  University,  and  the  person 
who  first  introduced  me  to  the  stromateoid 
fishes. 

SUMMARY 

Known  to  the  ancients,  and  investigated 
by  such  able  ichthyologists  as  Giinther,  Gill, 
and  Regan,  the  development  of  the  stromat- 
eoid classification  has  a  long  history.  Only 
recently,  however,  has  sufficient  material 
become  available  to  clarify  the  confusion 
surrounding  the  systematics  of  these  fishes. 

The  percifonn  suborder  Stromateoidei 
is  diagnosed  by  the  possession  of  toothed 
pharyngeal  sacs  and  small  uniserial  teeth 
in  the  jaws.  Comparative  study  of  the  na- 
ture of  the  pelvic  and  dorsal  fins,  the  tooth 
pattern,  the  number  of  vertebrae  and  bran- 
chiostegal  rays,  and,  in  particular,  the 
structure  of  the  caudal  skeleton  and  the 
pharyngeal  sacs  suggests  a  separation  of  the 
suborder  into  five  families  and  fourteen 
genera.  These  are:  Centrolophidae — 
Ihjpcroiijyphc,  Schedophilus,  Centrolophus, 
Icichthijs,  ScrioIcJla,  Psc)ioi).sis-,  Nomeidae — 


Stromateoid  Fishes  •  Haedrich        129 


Cuhiceps,  Noirwus,  Psenes;  Ariommidae 
(fam.  nov. ) — Aiiommo;  Tetragoniiridae — 
Tetmg,onunis-,  and  Stromateidae — Stroma- 
teiis,  Pcprilus,  Fiimpus.  The  Centrolophidae 
are  the  most  primitive  in  the  suborder,  and 
have  given  rise  directly  to  the  Stromateidae. 
The  Nomeidae  have  evolved  parallel  to  the 
centrolophid-stromateid  line,  and  have 
gi\en  rise  to  the  Tetragonuridae  and  the 
Ariommidae. 

Within  the  suborder,  evolutionary  trends 
from  the  generalized  to  the  highly  evolved 
condition  are  marked.  The  maximum  size 
attained  becomes  smaller,  and,  in  the  stro- 
mateids,  the  relative  depth  of  the  body  in- 
creases. The  pelvic  fins  are  lost.  The  mouth 
becomes  smaller,  the  jaw  teeth  become 
cusped,  and  the  palatine  dentition  may  be 
lost.  The  phar)ngeal  sacs  become  more 
elongate  and  the  structure  of  the  papillae 
within  them  becomes  more  complex.  The 
number  of  branchiostegal  rays  is  reduced 
from  seven  to  five.  The  pseudobranch  is 
lost.  The  vertebrae  increase  in  number  from 
a  basic  25.  The  number  of  elements  in  the 
caudal  skeleton  is  reduced  through  losses 
and  fusions. 

The  major  features  of  the  centrolophid 
distribution  are  discontinuity,  bipolarity, 
endemism,  and  sympatry  of  genera.  Four 
of  the  six  genera  occur  in  Australia  and 
New  Zealand.  The  three  nomeid  genera  are 
broadly  sympatric  in  temperate  and  tropi- 
cal oceans,  and  there  are  no  relicts,  no  bi- 
polarity, and  no  regional  endemism.  The 
tetragonurid  distribution  is  very  poorly 
known,  but  is  similar  to  that  of  the  nomeids. 
The  ariommids  are  found  in  deep  water 
over  the  edge  of  the  continental  shelves 
from  the  east  coast  of  the  New  World  to 
Japan,  and  near  Hawaii.  The  stromateid 
distribution  is  characterized  by  discontinu- 
ity, widespread  species,  restriction  to  con- 
tinental shelves,  and  allopatry  of  genera. 
None  occur  in  Australia  and  New  Zealand. 
The  distributional  data  support  the  conclu- 
sions based  on  comparative  morphology. 

The  relationships  and  natural  history  of 
the   stromateoid   taxa   are   discussed.    Svn- 


onymies,  keys,  and,  under  each  genus,  lists 
of  nominal  species  are  included. 

LITERATURE  CITED 

Abe,  ToKraARU.  1953.  New,  rare  or  uncommon 
fishes  from  Japanese  waters.  II.  Records  of 
rare  fishes  of  the  famiHes  Diretniidae,  Luvari- 
dae,  and  Tetragonuridae,  with  an  appendix 
(description  of  a  new  species,  Tetragonurns 
pacificu.s,  from  off  the  Solomon  Islands).  Jap. 
J.    Ichthyol.,    3(l):.39-47,   7   figs. 

.      1954.     New,   rare,   or   uncommon   fishes 

from  Japanese  waters.  V.  Notes  on  the  rare 
fishes  of  the  suborders  Stromateoidei  and 
Tetragonuroidei  (Berg).  Jap.  I.  Ichthyol., 
3:222,  246,  255. 

.      1955.      New,   rare,    or   uncommon   fishes 

from  Japanese  waters.  V.  Notes  on  the  rare 
fishes  of  the  suborders  Stromateoidei  and 
Tetragonuroidei  (Berg).  Jap.  J.  Ichthyol., 
4:113-118. 

.      1955a.     Notes  on  the  adult  of  Cuhiceps 

gracilis  from  the  western  Pacific.  J.  Oceanogr. 
Soc.  Japan,   ll(2):75-80. 

.      1963.     Unusual    occurrences    of    several 


species  of  boreal,  amphipacific  and  bathy- 
pelagic  fishes  in  Sagami  Bay  and  adjoining 
waters  during  the  first  half  of  1963,  a  cold- 
water  season  in  southern  Japan.  Bull.  Tokai 
Reg.  Fish.  Res.  Lab.,  No.  37:27-35. 

Abe,  T.,  S.  Kojima,  and  T.  Kosakai.  1963.  De- 
scription of  a  new  nomeid  fish  from  Japan. 
Jap.  J.  Ichthyol.,  9(  I,  2)  :31-35. 

Abe,  T.,  and  Toshio  Kosakai.  1964.  Notes  on 
an  economically  important  but  scientifically 
little-known  silver  ponifret,  Pampus  echino- 
gaster  (Pampidae,  Teleostei).  Jap.  J.  Ichthyol., 
12(l/2):29-31. 

Agassiz,  Louis.  1846.  Nomenclatoris  Zoologici. 
Index  universalis  continens  nomina  systematica 
classium,  ordinum,  familiariun  et  generum 
animalium  omnium,  tam  viventium  quam 
fossilium.  Soloduri,  Jent  and  Cassman,  393 
pp. 

Alcock,  Alfred  William.  1890.  On  the  bathyb- 
ial  fishes  collected  in  the  Bay  of  Bengal 
during  the  season  1889-1890.  Ann.  Mag.  Nat. 
Hist.,   Ser.   6,   6:197-222. 

.      1892.     Illustrations    of    the    zoolog>-    of 

H.  NL  Indian  marine  surveying  steamer  "In- 
vestigator," under  the  command  of  Comm. 
A.  Carpenter  and  Comm.  R.  F.  Hoskyn.  Part 
I,  Fishes,  7  pis.  Calcutta. 

Aldrovandi,  Ulisse.  1613.  De  Piscibus  libri 
V,  et  de  Cetis  liber  unus.  Bononife,  fol.,  765 
pp. 

Apsaxgikar,  D.  K.  1953.  The  systematic  posi- 
tion of  Stromatetis  niger.  J.  L^niv.  Bombay, 
N.  S.,  2IB(5):41-50,  32  figs. 


130         Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  2 


Arambourg,  Camille.  1954.  Les  poissons  cre- 
taces  du  Jebel  Tselfat  (Maroc).  Not.  Mem. 
Serv.  Geoi.  Maroc,  No.  118:1-185. 

Ariola,  V.  1912.  Nuovo  pesce  abissale  del 
Golfo  di  Genova  (  Cuhiccps  capcnsi.'i  (  Smith )  ) . 
Riv.   Mens.   Pesca,  Pavia,   7(  14) :  185-192. 

Ayres,  William  O.  1860.  [On  new  fishes  of 
the  Californian  coast.]  Proc.  Calif.  Acad.  Nat. 
Sci.,   1st  Ser.,  2:81-86. 

B.\HX.\RD,  K.  H.  1927.  A  monograph  of  the 
marine  fishes  of  South  Africa,  Part  11.  Ann. 
S.   Afr.   Mns.,   21(  2)  :419-1065. 

.      1948.     Further  notes   on   South   African 

marine  fishes.  Ann.  S.  Afr.  Mus.,  36:341- 
406,  4  pis. 

Basilewsky,  Stephanus.  1855.  Ichthyographia 
Chinae  Boreahs.  Nouv.  Mem.  Soc.  Nat. 
Moscou,   10:215-264,  9  pis. 

Bean,  Tarleton  H.  1912.  Description  of  new 
fishes  of  Bermuda.  Proc.  Biol.  Soc.  Washing- 
ton,  25:121-126. 

Beaufort,  L.  F.  de  and  W.  M.  Chapman.  1951. 
The  fishes  of  the  Indo-Australian  Archipelago. 
IX.  Percomorphi  (concluded),  Bleimoidea. 
Leiden,  E.  J.  Brill,  xi  +  484  pp.,  89  figs. 

Belloc,  Gerard.  1937.  Note  sur  un  poisson 
comestible  nouveau  de  la  cote  occidentale 
d'Afrique.  {Paraciibiceps  ledenoisi  nov.  gen., 
nov.  sp.).  Rev.  Trav.  I'Office  Peches  Marit., 
10(3):353-356.    4  figs. 

Belon,  Pierre.  1553.  De  Aquatilibus  Libri  ii. 
Cum  eiconibus  ad  vivam  ipsornm  effigiem, 
quoad  eius  fieri  potuit,  expressis.  Paris,  448 
pp. 

Bennett,  Edward  Turner.  1831.  [Characters 
of  new  genera  and  species  of  fishes  of  the 
Atlantic  coast  of  northern  Africa  presented  by 
Capt.  Belcher.]  Proc.  Zool.  Soc.  London, 
1831:146-148. 

Berg,  Lev  Semenovich.  1940.  Classification  of 
fishes  both  recent  and  fossil.  Trav.  Inst.  Zool. 
Acad.  Sci.  U.R.S.S.,  Leningrad,  5:87-517, 
190  figs.  [In  Russian  with  complete  English 
translation.  Reprint  1947,  Edwards  Brothers, 
Ann   Arbor.] 

.  1955.  Classification  of  fislies  and  fish- 
like vertebrates,  both  living  and  fossil.  Second 
edition.  Trudy  Zool.  Inst.  Akad.  Sci.  SSSR, 
20:1-286,  figs.  1-263.    [In  Russian.] 

Bertin,  Louis,  and  Camille  Arambourg.  1958. 
Super-ordre  des  Teleosteens  in  P.  P.  Grasse 
et  at.,  Traite  de  Zoologie,  vol.  13.  Paris. 
Masson  et  Cie.    2758  pp. 

Besednov,  L.  N.  1960.  Some  data  on  the  ich- 
thyofaima  of  the  Pacific  Ocean  driftwood. 
Trudy  Inst.  Okean.  Akad.  Nat.  Sci.  SSSR,  41 : 
192-197.    [In  Russian.] 

BiGELOW,  Henry  Bryant  and  William  C. 
ScHROEDER.      1953.     Fishes    of    the    Gulf    of 


Maine.    Fish.  Bull.  74.    Fish.  Bull.  U.  S.  Fish 
and  Wildlife  Serv.,  53:1-577. 

Blache,  J.  1962.  Liste  des  poissons  signales 
dans  TAtlanticjue  Tropico-Oriental  Sud — du 
Cap  des  Palmes  (4°  Lat.  N)  a  Mossamedes 
(15°  Lat.  S)  (Province  Guineo-Equatoriale). 
Trav.  Centre  Ocean,  de  Point-Noire, 
O.R.S.T.O.M.   1962.    Pp.   13-106. 

Blacker,  R.  W.  1962.  Rare  fishes  from  the 
Atlantic  Slope  fishing  grounds.  Ann.  Mag. 
Nat.  Hist.,  Ser.  13,  5:261-271. 

Bleeker,  Pieteh.  1851.  Over  einige  nieuwe 
geslachten  en  soorten  van  Makreelachtige 
visschen  van  den  Indischen  Archipel.  Nat. 
Tijdschr.    Neder.-Indie,    1:341-372. 

.  1852.  Bijdrage  tot  de  kennis  der  Mak- 
reelachtige visschen  van  den  Soenda-Moluk- 
schen  Archipel.  Verb.  Batav.  Genootsch.,  24: 
1-93. 

1859.     Enumeratio     specierum     piscium 


hucusque  in  Archipelago  Indico  observatarum. 
Act.  Soc.  Sci.  Indo-Neerl.,  6:1-276. 

Blegvad,  J.  and  B.  L0PPENTHIX.  1944.  Fishes 
of  the  Iranian  Gulf.  Danish  scientific  investi- 
gations in  Iran,  Part  111:1-247,   10  pis. 

Bloch,  Marc  Elieser.  1793.  Naturgeschichte 
der  Ausliindischen  Fische,  Siebenter  Theil. 
Berlin,  J.  Morino  &  Comp.,  144  pp. 

.  1795.  Naturgeschichte  der  Ausliindis- 
chen Fische,  Neunter  Theil.  Berlin,  J.  Morino 
&  Comp.,   192  pp. 

Bloch,  M.  E.  and  1.  G.  Schneider.  1801. 
Systema  ichthyologiae  iconibus  ex  illustratum. 
Post  obitum  auctoris  opus  inchoatum  absolvit, 
correxit,  interpolavit  Jo.  Gottlob  Schneider, 
Sa.xo.    Berolini,   584  pp.,   110  pis. 

Bonaparte,  Charles  Lucien.  1837.  Icono- 
grafia  della  fauna  italica,  per  le  cjuattro  classi 
degli  animali  vertebrati.  Tome  III,  part  2, 
Pesce.    Roma. 

.      1846.     Catalogo      metodico      dei      pesci 

Europei.     Napoli,   97   pp. 

Bonde,  Niels.  1966.  The  fishes  of  the  Mo-Clay 
Formation  (Lower  Eocene).  Medd.  Dansk 
Geol.  Foren.,  16:198-202. 

Borodin,  Nicholas.  1930.  Some  more  new 
deep-sea  fishes.  Proc.  New  England  Zool. 
Club,    11:87-92. 

Boulenger,  George  Albert.  1904.  Teleostei 
(Systematic  Part).  The  Cambridge  Natural 
History,  vol.  7:539-727,  figs.  32.5-440.  Lon- 
don, Macmillan  &  Co. 

BiJHLER,  Hermann.  1930.  Die  Verdauungsor- 
gane  der  Stromateidae  ( Teleost ) .  Zeitschr. 
Morphol.  Okolog.  Tiere,  19:59-115,  29  figs. 

Caldwell,  David  K.  1961.  Populations  of  the 
butterfish,  Poronotus  triacanthus  (Peck), 
with  systematic  comments.  Bull.  S.  Calif. 
Acad.  Sci.,  60(1):  19-31,  4  pis. 


Stromateoid  Fishes  •   Hocdrich 


131 


Caxestrixi,  C.iovaxxi.  1865.  Sopra  alcuni  pesci 
poco  noti  o  nuovi  del  Mediterraneo  nota. 
Mem.  R.  Accad.  Sci.  Torino,  Ser.  2,  21:.359- 
367,  2  pis. 

Carmichael,  Capt.  Dugald.  1818.  Some  ac- 
count of  the  Island  of  Tristan  da  Cunha  and 
of  its  natural  productions.  Trans.  Linn.  Soc. 
London,    12:483-513. 

Ca.stelnau,  Fraxcis  L.  1872.  Contributions  to 
the  ichthyologv'  of  Australia.  Proc.  Zool.  Ac- 
clini.  Soc.  Victoria,  1:29-247. 

Cetti,  Fraxcesco.  1777.  Historia  naturale  di 
Sardagni.   Pesci,  pp.  71-208.    Sassari. 

Chabaxaud,  p.  1930.  Description  d'lm  nouveau 
Cubiceps  (Pisces  Stromateidae)  de  la  Mer 
Rouge.  Bull.  Mus.  Hist.  Nat.  Paris,  2"  Ser., 
2:519-523. 

Chabanaud,  p.  and  G.  Tregouboff.  1930.  Ob- 
servations morphologiques  et  biologiques  sur 
un  Cenirolophus  niger  Gmelin,  ayant  \equ.  dans 
Faquarium  de  Villefranche-sur-mer.  Bull.  Soc. 
Zool.  France,  55:479-484. 

Chopra,  Shashi.  1960.  A  note  on  the  sudden 
outburst  of  ctenophores  and  medusae  in  the 
waters  off  Bomba\'.  Current  Science,  Octo- 
ber 1960,  29:392-393. 

Chu,  Y.  T.  et  al.  1962.  The  fishes  of  South 
China  Sea.  Peking,  xii  -(-  1184  pp.  [In 
Chinese.] 

.      1963.     The    fishes    of    East    China    Sea. 

Peking,   xxviii  +  642  pp.     [In  Chinese.] 

Clemexs,  W.  a.  axd  G.  V.  Wilby.  1961. 
Fishes  of  the  Pacific  Coast  of  Canada.  Second 
Edition.  Fish.  Res.  Bd.  Canada,  Bull.  No. 
68.    443  pp. 

Clothier,  Charles  R.  1950.  A  key  to  some 
southern  California  fishes  based  on  vertebral 
characters.  Calif.  Fish.  Bull.,  No.  79:1-83, 
23  pis. 

Cocco,  Axastasio.  1838.  Su  di  alcuni  sal- 
monidi  del  mare  di  Messina;  lettera  al  Ch. 
D.  Carlo  Luciano  Bonaparte.  Nuovi  Ann. 
Sci.   Nat.,   Bologna,   2:161-194. 

.      1839.     Sopra  un  nuovo  genere  di  pesci 

della  famiglia  de'  Centrolofini  e  di  ima  nuova 
specie  di  Trachurus.  Innom.  Messina,  3(7): 
56-59. 

1840.     Su  di  alcuni  nuovi  pesci  del  mare 


di   Messina.     Maurolico,   Messina,   4:236-244. 
CoLLETT,     Robert.      1896.     Poissons     pro\enants 

des  campagnes  du  yacht  I'Hirondelle   ( 1885- 

88).    Res.  Camp.  Sci.  Monaco,  Fasc.  X,  \iii  + 

198  pp. 
CoLLETTE,  Bruce  Badex.     1963.     The  systematic 

status  of  the  Gulf  of  Mexico  butterfish,  Poro- 

notus  burti  (Fowler).    Copeia,  1963(3)  :582- 

583. 
Corxish,    Thomas.      1874.     Pimelepterus    Cornu- 

biensis,    a    supposed    new   fish,    at    Penzance. 

The  Zoologist,  Ser.  2,  9:4255-4258. 


Costa,  Oroxzio  Gabriele.  1866.  Illustrazione 
di  un  centrolofo  pescato  nel  Golfo  di  Napoli. 
Ann.  Mus.  Zool.  Napoli,  6:84-87. 

Couch,  Joxathax.  1863.  A  history  of  the 
fishes  of  the  British  Islands.  Vol.  II.  London, 
Groombridge   &    Sons,    iv  -f  265   pp.,    63   pis. 

CowpER,  T.  R.  1960.  Occurrence  of  Pijrosoma 
on  the  continental  slope.  Nature,  187:878- 
879. 

Cuxxixgham,  J.  T.  1910.  On  the  marine  fishes 
and  invertebrates  of  St.  Helena.  With  descrip- 
tions of  new  species  of  Hydrozoa  and  Porifera 
by  R.  Kirkpatrick.  Proc.  Zool.  Soc.  London, 
1910:86-1.30.    3  pis. 

CuviER,  Georges.  1817.  Le  regne  animal  dis- 
tribue  d'apres  son  organisation,  pour  servir  de 
base  a  I'histoire  naturelle  des  animaux  et 
d'introduction  a  I'anatomie  comparee.    Edition 

I.  Tome  II,  Poissons.    Paris,  532  pp. 
.      1829.     Le       regne      animal,       distribue 

d'apres  son  organisation,  pour  servir  de 
base  a  I'histoire  naturelle  des  animaux  et 
d'introduction  a  I'anatomie  comparee.    Edition 

II.  Tome  II,  Poissons.    Paris,  406  pp. 
CuviER,    Georges    axd    Achille    Valexciexxes. 

1833.  Histoire  naturelle  des  poissons.  Tome 
IX.    Paris,   512  pp. 

Dahl,  George.  1823.  Coleoptera  und  Hy- 
menoptera.  Ein  systematisches  Verzeichniss, 
etc.    Wien,   105  pp. 

Day,  Fraxcis.  1870.  On  the  fishes  of  the  An- 
daman Islands.  Proc.  Zool.  Soc.  London, 
1870:677-705. 

.      1875.     The    fishes    of    India,    being    a 

natural  history  of  the  fishes  known  to  inhabit 
the  seas  and  fresh  waters  of  India,  Burma, 
and  Ceylon.  Vol.  I,  Text,  xx  +  778  pp.  Vol. 
II,  Atlas,   198  pis. 

DeKay',  James  Ellsworth.  1842.  Zoology  of 
New  York,  or  the  New  York  fauna.  Compris- 
ing detailed  descriptions  of  all  the  animals 
hitherto  observed  within  the  state  borders. 
Class  V.  Fishes.  Natural  History  of  New 
York  Geological  Survey,  Part  I.  Zoology. 
Albany,  415  pp.,   79  pis. 

Dieuzeide,  R.  axd  I.  Rolaxd.  1955.  Sur  im 
Stromateidae  nou\eau  du  genre  Cubiceps. 
Bull.  Sta.  Aquic.  Peche,  Castiglione  (N.S.), 
7:  .341-368,   1  pi. 

DouMET,  Napoleox.  1863.  Description  d'un 
nouveau  genre  de  poissons  de  la  Mediterranee. 
Rev.  Mag.  Zool.,  2-  serie,   15:212-223,   1  pi. 

Duarte-Bello,  Pedro  Pablo.  1959.  Catalogo 
de  peces  cubanos.  Univ.  Villaneuva  Mono- 
grafia  6:1-208. 

Eigexmaxx,  C.  H.  1902.  Description  of  a  new 
oceanic  fish  foimd  off  southern  New  England. 
Bull.  U.  S.  Fish  Comm.,  21:35-37. 

Emery,  Carlo.  1882.  Contribuzioni  all'Ittio- 
logia.    I\'.    Sulle  candizimi  di  \ita  di  giovani 


132        Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  2 


iiidi\idiii   de   Tetrafioniirus   ciivicri.     Mitdieil. 
Zool.  Stat.  Neapel.,  3:281-282. 

EuPHRASEN,  Bengt  Anders.  1788.  Beskrifning 
pa  trenne  fiskar.  Vetensk.  Akad.  Nya  Handl. 
Stockholm,  9:51-55. 

EvERMAXN,  B.  W.  AND  T.  H.  Shaw.  1927. 
Fishes  from  eastern  China,  with  descriptions 
of  new  species.  Proc.  Calif.  Acad.  Sci.,  Ser. 
4,   16(4):97-122. 

Fabricius,  Johaxx  Christian.  1798.  Supple- 
mentum,  J.  C.  Fabricii  .  .  .  Entomologia  sys- 
tematica emendata  et  aucta  .  .  .  adjectis 
synonymis,  locis,  observationibus,  descriptioni- 
bus.    Hafnife,  572  pp. 

FAO.  1964.  Yearlwok  of  fishery  statistics. 
Catches  and  landings.  Vol.  16.  Food  and 
Agriculture  Organisation  of  the  United  Na- 
tions.    Rome. 

FiLiPPi,  FiLippo  DE  and  J.  B.  Verany.  1859. 
Sopra  alciini  pesci  nuovi  o  poco  noti  del 
Mediterraneo.  Mem.  R.  Accad.  Sci.  Torino, 
Ser.  2,   18:187-199. 

Fitch,  John  E.  1949.  Some  unusual  occur- 
rences of  fish  on  the  Pacific  Coast.  Calif. 
Fish  and  Game,  35(l):41-49. 

.      1951.     Notes   on   the   squaretail,    Tetra- 

gontiru.s  cuvieri.    Calif.   Fish   and    Came,    37 
(l):55-59. 

1952.     Toxicity  and  taxonomic  notes  on 


the  sfiuaretail,  Tctragomiru.s  cuvieri.  Calif. 
Fish  and  Game,   38(2) :25l-252. 

Fordice,  Morton  W.  1884.  A  review  of  the 
American  species  of  Stromateidae.  Proc. 
Acad.   Nat.   Sci.   Philadelphia,    1884:311-317. 

Fowler,  Henry  Weed.  1906.  New,  rare  or 
little  known  scombroids,  no.  .3.  Proc.  Acad. 
Nat.    Sci.    Philadelphia,    58:114-122,    3    figs. 

.     1923.     New    or    little   known    Hawaiian 

fishes.  Occ.  Pap.  Bernice  P.  Bishop  Mus. 
Honolulu,   8:37.5-392. 

.      1928.     The    fishes    of    Oceania.     Mem. 

Bernice  P.  Bishop  Mus.  Honolulu,  10:1-540, 
49  pis. 

.     1930.     The     fishes     obtained     by     Mr. 

James  Bond  at  Grenada,  British  West  Indies, 
in  1929.  Proc.  Acad.  Nat.  Sci.  Philadelphia, 
82:269-277,   2  figs. 

.  1931.  The  fishes  of  Oceania — Supple- 
ment 1.  Mem.  Bernice  P.  Bishop  Mus.  Hono- 
lulu,   11(5):311-381. 

.  1934.  The  fishes  of  Oceania — Supple- 
ment 2.  Mem.  Bernice  P.  Bishop  Mus.  Hono- 
lulu, ll(6):383-466. 

.      1934a.     Fishes  obtained  by  Mr.   H.  W. 

Bell-Marley  chiefly  in  Natal  and  Zululand  in 
1929  to  1932.  Proc.  Acad.  Nat.  Sci.  Phila- 
delphia, 86:405-514,  53  figs. 

.      1934b.     Zoological   results   of   the   third 

de    Schauensee    Siamese    Expedition.     Part    1, 


Fishes.  Proc.  Acad.  Nat.  Sci.  Philadelphia, 
86:67-163. 

— .  1936.  The  marine  fishes  of  West 
Africa,  based  on  the  collection  of  the  Ameri- 
can Museum  Congo  Expedition,  1909-1915. 
Part  II.  Bull.  Amer.  Mus.  Nat.  Hist.,  80(2): 
609-1493. 

.      1938.     The   fishes   of   the   George   Van- 

derbilt  South  Pacific  Expedition,  1937. 
Monogr.  Acad.  Nat.  Sci.  Philadelphia,  2:1- 
349,   12  pis. 

.      1939.     Ichthyological      notes.      No.      1. 

Notulae  Naturae,  No.  3:1-2. 

.     1941.     A  list  of  the  fishes  known  from 

the  coast  of  Brazil.  Arquiv.  Zool.  Estad.  Sao 
Paulo,  3(6):  115-184. 

.  1944.  Fishes  of  Chile.  Systematic  cata- 
log.   Rev.  Chil.  Hist.  Nat.,  46:15-116. 

.      1944a.     Results  of  the  fifth  George  Van- 

derbilt  Expedition,  1941.  The  fishes.  Monogr. 
Acad.  Nat.  Sci.  Philadelphia,  6:57-529,  20 
pis. 

.      1944b.     Description     of    a    new     genus 

and  a  new  species  of  American  stromateid 
fishes.  Notulae  Naturae,  No.  142:1-4,  2  figs. 
1949.     The  fishes   of   Oceania — Supple- 


ment 3.  Mem.  Bernice  P.  Bishop  Mus.  Hono- 
hdu,   12(2):35-186. 

Franca,  P.  da.  1957.  Contribugao  para  o  con- 
hecimento  dos  Stromateidae  de  Angola. 
Anais,  Jta.  Invest.  Ultramar,  Lisboa,  12(2): 
7-26,  2  pis. 

Fraser-Brunneb,  a.  1935.  New  or  rare  fishes 
from  the  Irish  Atlantic  slope.  Proc.  Roy. 
Irish  Acad.,  426:319-326,  5  figs. 

Freihofer,  Warren  C.  1963.  Patterns  of  the 
ramus  lateralis  accessorius  and  their  system- 
atic significance  in  teleostean  fishes.  Stanford 
Ichthyol.    Bull.,    8(2):81-189. 

Carman,  Samuel  F.  1896.  Report  on  the  fishes 
collected  by  the  Bahama  Expedition  of  the 
State  University  of  Iowa,  under  Prof.  C.  C. 
Nutting  in  1893.  Bull.  Lab.  Nat.  Sci.  Univ. 
Iowa,  4(l):76-93. 

CIeimtz,  Hans  Bruno.  1868.  Die  fossilen 
Fischschuppen  aus  dem  Pliiner-Kalke  von 
Strehlen.  Denkschr.  Ges.  Natur-  imd  Heilk. 
Dresden,   1868:31-56,   4   pis. 

Gesner,  Conrad.  1560.  Nomenclator  aquati- 
lium  animantium.  Icones  animalium  aquati- 
limii  in  mari  &  dulcibus  acjuis  dcgentiinu, 
plus  ([uam  DCC  cum  nomenclaturis  singu- 
lorum  Latinis,  Grecis,  Italicis,  Hi.spanicis, 
Gallicis,  Germanicis,  Anglicis,  alij'sq;  inter- 
dum,  per  certos  ordines  digestae.    374  pp. 

Gilbert,  Charles  Henry.  1904.  Notes  on 
fishes  from  the  Pacific  Coast  of  North  Amer- 
ica. Proc.  Calif.  Acad.  Sci.,  Ser.  3,  Zool., 
3:255-271. 

CliLRERT,  C.   H.   AND  E.   C.   Stabks.      1904.     The 


Stromateoid  Fishes  •   Hacdrich 


133 


fishes  of  Panama  Bay.  Mem.  Calif.  Acad. 
Sci.,  4:1-403,  33  pis. 

Gilchrist,  J.  D.  F.  1922.  Note  on  the  oesopha- 
geal teeth  of  the  Stromateidac.  Ann.  Mag. 
Nat.  Hist.,  Ser.  9,  9:249-255. 

Gilchrist,  J.  D.  F.  and  C.  von  Bonde.  1923. 
The  Stromateidae  ( Butter  Fishes )  collected 
by  the  S.  S.  "Pickle."  Rep.  Fish.  Mar.  Biol. 
Survey  Union  of  S.  Afr.  Capetown.  Rep.  No. 
3,  Spec.  Rep.  No.  4:1-12. 

C:iLL,  Theodore  Nicholas.  1860.  Notes  on  the 
nomenclature  of  North  American  fishes.  Proc. 
Acad.  Nat.  Sci.  Philadelphia,  11:20-21. 

.      1861.     Catalogue    of    the   fishes    of   the 

eastern  coast  of  North  America,  from  Green- 
land to  Georgia.  Proc.  Acad.  Nat.  Sci.  Phila- 
delphia, 13  ( supplement ):  1-63. 

.      1862.     On  the  limits   and   arrangements 

of  the  family  of  scombroids.  Proc.  Acad. 
Nat.   Sci.  Philadelphia,   14:125-127. 

1884.     Notes      on      the      Stromateidae. 


Proc.   Amer.   Phil.    Soc,   21  ( 116) :  664-672. 

Ginsburg,  Isaac.  1954.  Four  new  fishes  and 
one  little  known  species  from  the  east  coast 
of  the  United  States  including  the  Gulf  of 
Mexico,  y.  Wash.  Acad.  Sci.,  44:256-264, 
6  figs. 

Gmelin,  Johann  Frederick.  1788.  Caroli  a 
Linne  .  .  .  Systema  naturas  per  regne  triae 
naturae,  secundum  classes,  ordines,  genera, 
species,  cum  characteribus,  differentis,  syn- 
onymis,  locis.    Lipsia\    Vol.   1:1-1516. 

GooDE,  George  Browne  and  Tarleton  H.  Bean. 
1896.  Oceanic  ichthyology,  a  treatise  on  the 
deep-sea  and  pelagic  fishes  of  the  world, 
based  chiefly  upon  the  collections  made  by 
the  steamers  BLAKE,  ALBATROSS,  and 
FISH  HAWK  in  the  northwestern  Atlantic, 
with  an  atlas,  containing  417  figures.  Smith- 
sonian Institution,  United  States  National 
Museum,  Special  Bull.  No.  2.  Washington, 
Government   Printing  Office,   xxxv  +  553   pp. 

GosLiNE,  William  A.  1959.  Mode  of  life,  func- 
tional morphology,  and  the  classification  of 
modern  teleostean  fishes.  Syst.  Zool.,  8:160- 
164. 

.  1960.  Contributions  toward  a  classifica- 
tion of  modern  isospondylous  fishes.  Bull. 
Brit.  Mus.  Nat.  Hist.,  (Zoology),  6(6):325- 
365,  15  figs. 

.      1961.     Some     osteological     features     of 

modern   lower  teleostean  fishes.     Smithsonian 
Misc.  Coll.,   142(3):  1-42,   6  figs. 

1961a.     The  perciform   caudal  skeleton. 


Copeia,   1961(3):265-270,  3  figs. 

GosLiNE,  W.  A.  AND  Vernon  E.  Brock.  1960. 
Handbook  of  Hawaiian  fishes.  Honolulu, 
University  of  Hawaii  Press,   ix  +  372  pp. 

Gregory,    William    K.     1907.     The    orders    of 


teleostoTiious    fishes.     Ann.    New    York    Acad. 
Sci.,   17:437-508. 

.  1933.  Fish  skulls:  a  study  of  the  evo- 
lution of  natural  mechanisms.  Trans.  Amer. 
Phil.  .Soc,  23:75-481,  302  figs. 

1951.     Evolution  Emerging.    New  York, 


Macmillan,  2  vols.    1013  pp. 
Grey,   Marion.     1955.     The   fishes   of  the   genus 

Tetrogoniinis    Risso.     DANA-Report   No.    41: 

1-75,  16  figs. 
Griffin,   L.   T.      1928.     Studies   in   New   Zealand 

fishes.    Trans.  Proc.  New  Zealand  Inst.  Well- 
ington,  59:374-388,  9  pis. 
GiJNTHER,  Albert.      1859.     Catalogue  of  the  acan- 

thopterygian   fishes    in    the   collection   of   the 

British     Museum.      Volume     First.      London, 

Taylor  and  Francis,  xxxi  -|-  524  i^p. 
.      1860.     Catalogue  of  the  acanthopterygian 

fishes  in  the  collection  of  the  British  Museum. 

Volume  Second.    London,  Taylor  and  Francis, 

xxi  -|-  548  pp. 
.      1860a.     On  Centrolophus  hritannicus,  a 

new  British  fish.    Ann.  Mag.  Nat.  Hist.,  Ser. 

6,  3:46-48. 
.      1861.     Catalogue  of  the  acanthopterygian 

fishes  in  the  collection  of  the  British  Museum. 

Vohnne  Third.    London,  Taylor  and  Francis, 

XXV  -|-  586  pp. 
.      1869.     Contribution   to   the   ichthyology 

of  Tasmania.    Proc.  Zool.  Soc.  London,  1869: 

429. 
.      1871.     Report   on   several  collections   of 

fishes  recently   obtained   for  the   British   Mu- 
seum.    Proc.    Zool.    Soc.    London,    1871:652- 

675. 
.      1872.     On  some  new  species  of  reptiles 

and  fishes  collected  by  J.  Brenchley,  Esq.  Ann. 

Mag.   Nat.   Hist.,   Ser.  4,    10:418-426. 
.      1876.     Andrew     Garrett's     Fische     der 

Siidsee,  beschreiben  und  redigirt  von  A.  C.  L. 

G.  Giinther.    Band  II.    J.  Mus.  Godeffrey,  4: 

129-260,  67  pis. 

1880.     An   introduction  to  the  study  of 


fishes.  Edinburgh,  Adam  and  Charles  Black, 
720  pp. 

.      1882.     Description    of    a    specimen    of 

Scliedophihis  medtisophagiis,  a  fish  new  to 
the  British  fauna.  Trans.  Zool.  Soc.  London, 
11:223-224. 

1889.     Report  on  the  pelagic  fishes  col- 


lected by  H.M.S.  Challenger,  during  the  years 
1873-1876.  Report  on  the  scientific  results 
of  the  voyage  of  H.M.S.  Challenger.  Zoology, 
31:1-47,  plates  I-VI. 

GuiCHENOT,  Alphonse.  1848.  Peces  de  Chile. 
In:  Claudio  Gay,  Historia  fisica  y  politica  de 
Chile.  Zoologica.  Tomo  Segundo.  Paris  and 
Santiago,  pp.   137-370. 

.  1866.  Catalogue  des  poissons  de  Mada- 
gascar  de   la   collection   du   Musee   de   Paris, 


134         Bulletin  Museum  of  Coviparnfive  Znnlnfiy.  Vol.  135,  No.  2 


avec     la     description     de     plusieiirs     especes 
nouvelles.     Mem.    Soc.    Sci.    Nat.    Cherbourg, 
12:129-148. 
— .      1866a.     Notice   sur   un   nouvelle   espece 


de  poissons  appartenant  au  genre  des  rhombes 
du  Museum  de  Paris.  Mem.  Soc.  Sci.  Nat. 
Cherbourg,   12:243-247. 

Gui(;lia,  D.  1950.  II  Tcimgonuni.'i  cuvieri  Risso 
1810  nel  Golfo  di  Genova.  Ann.  Mus.  Civ. 
Stor.  Nat.  Genova,  64:158-169. 

Haedrich,  Richard  L.  1965.  Ciibiceps  athenae, 
a  new  nomeid  fisli  from  the  western  North 
Athintic,  and  its  systematic  position  among 
stromateoids.   Copeia,  1965(4)  :50l-505. 

.     In    press.     The    stromateoid    fish    genus 

Icichthijs:  notes  and  a  new  species.  Vidensk. 
Medd.  bansk  Naturh.  Foren.,  129. 

Harris,  J.  E.  1938.  The  role  of  the  fins  in  the 
eciuilibrium  of  the  swimming  fish.  II.  The 
role  of  the  pelvic  fins.  J.  Exper.  Biol.,  15 
(l):32-47. 

Hart,  T.  John.  1946.  Report  on  trawling  sur- 
veys on  the  Patagonian  continental  shelf.  Dis- 
covery Reports,  23:223-408,  1  pi. 

Heckel,  Johaxn  Jakob.  1856.  Neue  Beitriige 
zur  Kenntnis  der  fossilen  Fische  Oesterreichs. 
Denkschr.  Akad.  Wiss.  Wien,  11:187-274,  15 
pis. 

Herald,  Earl  S.  1961.  Living  fishes  of  the 
world.  Garden  City,  New  York,  Doubleday 
and  Co.,  304  pp. 

Herre,  a.  W.  C.  T.  1950.  Six  additions  to  the 
Philippine  fish  fauna,  including  two  new 
species.    Philipp.  J.  Sci.,  79:341-346. 

.      1953.     Check    list    of    Philippine    fishes. 

Res.  Rep.  U.  S.  Fish  and  Wildlife  Service,  No. 
20:1-977. 

Herre,  A.  W.  C.  T.  and  E.  S.  Herald.  1950. 
Noteworthy  additions  to  the  Philippine  fish 
fauna,  including  two  new  species.  Philipp.  J. 
Sci.,  79:309-340,  12  figs. 

Hildebrand,  Samuel  Frederick.  1946.  De- 
scriptive catalog  of  the  shore  fishes  of  Peru. 
Bull  U.  S.  Nat.  Mus.,  No.  189:1-530,  95  figs. 

.     MS.     The     marine    fishes     of     Panama. 

[Partially  completed  revision  of  the  earlier 
work  by  Meek  and  Hildebrand,  1925.  Stro- 
mateidae  section  obtained  from  CNHM 
through  Ira  Rubinoff.] 

HoBBS,  Kenneth  L.  1929.  A  new  species  of 
CcntroIo))hu.s  from  Monterey  Bay,  California. 
J.  Wash.  Acad.  Sci.,  19(20) : 460-462. 

Hollister,  Gloria.  1934.  Clearing  and  dyeing 
fish  for  bone  study.  Zoologica,  12:89-101, 
4  figs. 

Holt,  E.  W.  L.  and  L.  W.  Byrne.  J 903.  On 
the  British  and  Irish  species  of  the  family 
Stromateidae.  Rep.  Fish.  Ireland  for  1901, 
Part  2:70-76. 


HuBBS,  Carl  Leavitt  and  Karl  F.  Lagler. 
1958.  Fishes  of  the  Cireat  Lakes  region.  Re- 
vised Edition.  Cranbrook  Institute  of  Science, 
Bulletin  No.  26.  Bloomfield  Hills,  Michigan, 
xi  +  135  pp. 

Humphrey,  George.  1797.  Museum  Calon- 
nianum.  Specification  of  the  various  articles 
which  compose  the  .  .  .  Museum  of  Natural 
History  collected  by  M.  de  Calonne  in  France, 
etc.   Part  I,  84  pp. 

HuTTON,  Capt.  Frederick  Wollaston.  1872. 
Contributions  to  the  ichthyology  of  New  Zea- 
land. Trans.  Proc.  New  Zealand  Inst.,  5:259- 
272. 

Isokawa,  Sohiti,  K.  Kubota,  T.  Kosakai,  I. 
Satomura,  M.  Tsukouchi  and  a.  Sera. 
1965.  Some  contributions  to  study  of  eso- 
phageal sacs  and  teeth  of  fishes.  J.  Nihon 
Univ.   Sch.   Dent.,   7(3)  :103-111. 

Johnson,  James  Yate.  1862.  Description  of 
some  new  genera  and  species  of  fishes  ob- 
tained at  Madeira.  Proc.  Zool.  Soc.  London, 
1862:167-180. 

Jordan,  David  Starr.  1923.  A  classification  of 
fishes  including  families  and  genera  as  far  as 
known.  Stanford  Univ.  Pub.,  Univ.  Ser.,  Biol. 
Sci.,  3(2):77-243.  Reprint  1963,  Stanford 
University  Press. 

.      1923a.     Note    on    Icichthys    lockingtoni 

Jordan  and  Gilbert,  a  pelagic  fish  from  Cali- 
fornia. Proc.  U.  S.  Nat.  Mus.,  63(2472):1- 
3,  1  pi. 

Jordan,  D.  S.  and  Charles  Harvey  Bollman. 
1889.  Descriptions  of  new  species  of  fishes 
collected  at  the  Galapagos  Islands  and  along 
the  coast  of  the  United  States  of  Colombia, 
1887-1888.  Proc.  U.  S.  Nat.  Mus.,  12:149- 
183. 

Jordan,  D.  S.  and  Barton  Warren  Evermann. 
1896.  The  fishes  of  North  and  Middle  Amer- 
ica; a  descriptive  catalogue  of  the  species 
of  fish-like  vertebrates  found  in  the  waters 
of  North  America,  north  of  the  Isthmus  of 
Panama.  Part  I.  Bull.  U.  S.  Nat.  Mus.,  No. 
47:1-1340.  Reprint  1963,  Smithsonian  In- 
stitution. 

Jordan,  D.  S.,  B.  W.  Evermann  and  H.  W.  Clark. 
1930.  Check  list  of  the  fishes  and  fishlike 
vertebrates  of  North  and  Middle  America 
north  of  the  northern  boundary  of  Venezuela 
and  Colombia.  Appendix  X  to  the  Report  of 
the  United  States  Commissioner  of  Fisheries 
for  the  fiscal  year  1928.  Reprint  1955,  Wash- 
ington,  U.  S.  Govt.  Printing  Office,  670  pp. 

Jordan,  D.  S.  and  Charles  Henry  Gilbert. 
1880.  Description  of  a  new  species  of  deep- 
water  fish  (Icichihy.s  lockingtoni)  from  the 
coast  of  California.  Proc.  U.  S.  Nat.  Mus., 
3:305-308. 


Strom ATEOiD  Fishes  •  Hacdrich        135 


.      1882.     Synopsis   of  the  fishes   of   North 

America.    Bull.    U.    S.   Nat.   Mus.,   No.    16:1- 

1018. 
JoHDAx,  D.  S.  AND  C.  L.  HuBBs.      1925.     Records 

of  fishes  obtained  by  D.  S.  Jordan  in  Japan, 

1922.     Mem.    Carnegie    Mus.    Pittsburgh,    10 

(2):93-346,  7  pis. 
Jordan-,  D.  S.  and  C.  W.  Metz.      1913.     A  cata- 
logue of  the  fishes  known  from  the  waters  of 

Korea.     Mem.    Carnegie    Mus.    Pittsburgh,    6 

(l):l-65,   10  pis. 
Jordan,  D.  S.  and  J.  O.  Snyder.     1904.     Notes 

on  the  collections  of  fishes  from  Oahu  Island 

and  Laysan  Island,  Hawaii,  with  descriptions 

of  four  new  species.   Proc.  U.  S.  Nat   Mus..  27 

(1377): 939-948. 
.      1907.     Notes  on  fishes  of  Hawaii,  with 

descriptions    of    new    species.     Dept.    Conim. 

Lab.,  Bull.  Bur.  Fish.,  26:205-218,  2  pis. 
Jordan,    D.    S.    and    W.    F.    Thompson.      1914. 

Record   of   the    fishes    obtained    in    Japan    in 

1911.     Mem.    Carnegie    Mus.    Pittsburgh,    6: 

205-313. 
Kamohara,     Toshiji.      1940.     Fauna     Nipponica. 

Scombroidei     ( exclusive    of    Carangiformes ) . 

Vol.  15,  fasc.  2,  no.  5,  225  pp.    [In  Japanese.] 
Katayama,  Masao.      1952.     A  record  of  Ariomma 

lurida    Jordan    et    Snyder    from    Japan,    with 

notes     on     its     systematic    position.      Jap.     J. 

IchthyoL,  2:31-34,  2  figs. 
.      1959.     Studies  on  the  serranid  fishes  of 

Japan  (I).   Bull.  Fac.  Educ,  Yamaguchi  Univ., 

8(2): 103-181. 
Kato,  K.      1933.     Is  Nomeus  a  harmless  inquilinus 

of  Plujsalia?    Proc.  Imp.  Acad.  Japan,  9(9): 

537-538. 
Klunzinger,  Carl  Benjamin.      1884.     Die  Fische 

des  Roten  Meeres.   Fine  kritische  Revision  mit 

Bestinimungs-Tabellen.     Teil   I.    Acanthopteri 

veri  Owen.    Stuttgart,   133  pp. 
Kner,  Rudolf  and  Franz  Steindachner.     1866. 

Neue    Fische   aus   dem    Museum   der   Herren 

Joh.     C.     Godeffroy    &     Sohn    in     Hamburg. 

Sitzungsber.     Akad.   Wiss.   Wien,   Abt.    1,   54 

(8):356-.395. 
Lacepede,   Bernhard  Germain.      1800.     Histoire 

naturelle  des  poissons.    Vol.  II.    372  pp. 
.      1802.     Histoire    naturelle    des    poissons. 

Vol.  III.    378  pp. 
.      1803.     Histoire    naturelle    des    poissons. 

Vol.  IV.    360  pp. 
Lane,  C.  E.      1960.     The  Portuguese  man-of-war. 

Sci.   Amer.,   202(  3)  :158-168,  9  figs. 
Larkins,     Herbert    A.      1964.     Some    epipelagic 

fishes  of  the  North  Pacific  Ocean,  Bering  Sea, 

and  Gulf  of  Alaska.    Trans.  Amer.  Fish.  Soc, 

93(3):286-290. 
Le  Danois,  Yseult.      1962.     Catalogue  des  types 

de   poissons   du   Mustnun    National   d'Histoire 

Naturelle.    ( Nomeidae,  Stromateidae,  Apolecti- 


dae,   Kurtidae. )     Bull.   Mus.   Hist.   Nat.   Paris, 
2"  Sen,  35(3):228-234. 

Le  Gall,  Jean.  1925.  Un  poisson  peu  commun: 
le  Centrolophus  hrifanniciis  Giinther.  Bull. 
Mus.  Hist.  Nat.  Paris,  2'  Ser.,  31(1):288- 
295,  2  figs. 

Legaspi,  V.  A.  1956.  A  contribution  to  the  life 
history  of  the  nomeid  fish  Psenes  ct/anophrys. 
Bull.  Mar.  Sci.  Gulf  &  Carib.,  6:179-199. 

Liem,  Karel  F.  1963.  The  comparative  os- 
teology and  phylogeny  of  the  Anabantoidei 
(Teleostei,  Pisces).  Illinois  Biological  Mono- 
graphs: No.  30.  Urbana,  University  of  Illinois 
Press,  V  -|-  149  pp. 

Linnaeus,  Carl.  1758.  Systema  Naturae,  ed.  X, 
vol.  1,  824  pp. 

.      1766.     Svstema    Naturae,    ed.    XII,    vol. 

1(1),  532  pp. 

Lloyd,  R.  E.  1909.  A  description  of  the  deep- 
sea  fish  caught  by  the  R.I. M.S.  ship  "In- 
vestigator" since  the  year  1900,  with  supposed 
evidence  of  mutation  in  Mcdthopsls.  Mem. 
Indian  Mus.  Calcutta,  2:1.39-180,  7  pis. 

Lo  Bianco,  Salvatore.  1909.  Notizie  biologiche 
riguardanti  specialmente  il  period  di  maturita 
sessuale  degli  animali  del  Golfo  di  Napoli. 
Mittheil.    Zool.  Stat.  Neapel,   19(4)  :51.3-761. 

Longley,  William  H.  and  Samuel  F.  Hilde- 
brand.  1941.  Systematic  catalogue  of  the 
fishes  of  Tortugas,  Florida,  with  observations 
on  color,  habits,  and  local  distribution.  Car- 
negie Inst.  Washington  Pub.,  No.  535:1-331, 
34  pis. 

Lowe,  Richard  Thomas.  1833.  Characters  of 
a  new  genus  Leirtis  and  of  several  species  of 
fishes  from  Madeira.  Proc.  Zool.  Soc.  London, 
1:142-144. 

.      1839.     A   supplement    to   a   synopsis   of 

the  fishes  of  Madeira.    Proc.  Zool.  Soc.   Lon- 
don, 7:76-92. 

1843.     Notices  of  fishes  newly  observed 


or  discovered  in  Madeira  during  the  years 
1840,  1841,  and  1842.  Proc.  Zool.  Soc.  Lon- 
don, 11:81-95. 

Lowe,  Rosemary  H.  1962.  The  fishes  of  the 
British  Guiana  continental  shelf,  Atlantic 
coast  of  South  America,  with  notes  on  their 
natural  historv.  T-  Linn.  Soc.  London,  Zool., 
44  (.301):  669-700. 

LozANO  Y  Rey,  Luis.  1952.  Feces  fisoclistos. 
Subserie  toracicos.  Segimda  parte.  Ordenes 
Escombriformes.  Mem.  Real  Acad.  Cien. 
Madrid.   Ser.   Cien.   Nat.,    14:387-703. 

LuTKEN,  Chr.  1880.  Spolia  Atlantica.  Bidrag  til 
Kundskab  om  Formforandringer  hos  Fiske  un- 
der deres  Vitxt  og  Udvikling,  sterligt  hos 
nogle  af  Atlanterhavets  H0js0fiske.  Dansk. 
Vidensk.  Selsk.  Skr.,  (5)Nat.  Math.  Afd.,  12 
(6):409-613,  5  pis. 

Macleay,    William.     1885.     A    remarkable    fish 


136        Bulletin  Museum  of  Comparaiive  Zoologtj,  Vol.  135,  No.  2 


from  Lord  Howt-  Island.  Proc.  Linn.  Soc.  New 
South  Wales,  10:718-720. 

Mann,  Cuillermo.  1953.  Sobre  le  identidad 
de  la  "cojinoha"  del  norte  v  centro  Chile. 
Invest.  Zool.  Chil.,  2(1) :3. 

Man.sueti,  Romeo  L  1963.  Symbiotic  behavior 
between  small  fishes  and  jelKfishes,  with  new 
data  on  that  between  the  stroniateid,  Pcprihis 
alepidottis,  and  the  scyphomedusa,  Chrt/saora 
quinquecirrha.    Copeia,    1963(  1 )  :40-80. 

Marcgrave,  George.  1648.  Historiae  rernm 
naturalinm  Brasilia'  libri  viii.  Liber  iv,  de 
Piscibns.  Lugduni  Batavorum  et  Amstelodami, 
293  pp. 

Marchal,  E.  1961.  Description  d'une  nouvelle 
espece  de  Faracuhiceps  (Belloc),  Paraciibiccps 
mtiltisqiia77}i.'>  n.  sp.  (Poisson  Teleosteen). 
Bull.  Mns.  Hist.  Nat.  Paris.  2"  Ser.,  33(5): 
487-491,  2  figs. 

Marshall,  Tom  C.  1964.  Fishes  of  the  Great 
Barrier  Reef  and  coastal  waters  of  Queensland. 
Sydney,  Angus  &  Robertson,  566  pp. 

Maul,  G.  E.  1964.  Observations  on  young  live 
Miipiis  niaculatus  (Giinthcr)  and  Mtiptis 
ovalis  (Valenciennes).  Copeia,  1964(1):93- 
97. 

McCuLLOCH,  Allen  Riverstone.  1911.  Report 
on  the  fishes  obtained  by  the  F.I.S.  "En- 
deavor" on  the  coasts  of  New  South  Wales, 
Victoria,  South  Australia,  and  Tasmania.  Part 
1.  Zoological  results  of  the  fishing  experi- 
ments carried  out  by  F.I.S.  "Endeavor"  1909- 
10  under  H.  C.  Dannevig,  Commonwealth 
Director  of  Fisheries.    Vol.   I,  Part  1:1-87. 

.     1914.     Report  on   some   fishes   obtained 

by  the  F.I.S.  "Endeavor"  on  the  coasts  of 
Queensland,  New  South  Wales,  Victoria, 
Tasmania,  South  and  South-Western  Australia. 
Part  2.  Zoological  results  of  the  fishing  ex- 
periments carried  on  by  F.I.S.  "Endeavor" 
1909-10  under  H.  C.  Dannevig,  Common- 
wealth of  Fisheries.    Vol.  II,  Part  111:75-165. 

.      1923.     Notes    on    fishes    from    Australia 


and  Lord  Howe  Island.  Rec.  Aust.  Mus., 
Sydney,  14:1-7,  3  pis. 

McKenney,  Thomas  W.  1961.  Larval  and  adult 
stages  of  the  stromateoid  fish  Psenes  regulus, 
with  comments  on  its  classification.  Bull.  Mar. 
Sci.  Gulf  &  Carib.,   11(2)  :210-236. 

Mead,  Giles  W.  and  Margaret  G.  Bradbury. 
1963.  Names  of  bones.  In:  H.  B.  Bigelow 
and  W.  C.  Schroeder,  Fishes  of  the  western 
North  Atlantic.  Mem.  Sears  Found.  Mar.  Res., 
l(3):20-23. 

Meek,  Seth  Eugene  and  Samuel  Frederick 
Hilderrand.  1925.  The  marine  fishes  of 
Panama.  Part  II.  Publ.  Field  Mus.  Zool.  Ser. 
(Chicago),  15(226)  :331-707,  54  pis. 

Mees,  G.   F.      1962.      Additions  to  the  fish   fauna 


of  Western  Australia — 3.  Fish.  Dept.  W.  Aust., 
Fish.  Bull.  No.  9,  part  3:23-29. 

Merriman,  D.  1945.  Notes  on  Palimiricluhys 
perciformis  and  the  evidence  for  its  bath- 
ypelagic  habitat.  Trans.  Conn.  Acad.  Arts 
Sci.,  36:837-848,  1  pi. 

Miller,  R.  R.  1946.  Distributional  records  for 
North  American  fishes,  with  nomenclatorial 
notes  on  the  genus  Psenes.  ].  Wash.  Acad. 
Sci.,  36(6)  :206-212. 

Miller,  Robert  V.  1964.  The  morphology  and 
function  of  the  pharyngeal  organs  in  the 
clupeid  Dorosoma  petense  (  Giinther ) .  Ches- 
apeake Science,  5(4)  :194-199. 

Minding,  Julius.  1832.  Lehrbuch  der  Naturges- 
chichte  der  Fische.    Berlin,   xii  -f  131   pp.,    1 

pl. 

Miraxda-Ribeiro,  Alipio  de.  1915.  Fauna 
brasiliense,  peixes.  Tomo  V.  ( Eleutherobran- 
chios  aspirophores. )  Arch.  Mus.  Nac.  Rio  de 
Janeiro,  17,  about  600  pages,  no  continuous 
pagination. 

Mitchill,  Samuel  Latham.  1815.  The  fishes 
of  New  York,  described  and  arranged.  Trans. 
Lit.  Phil.  Soc.  New  York,  1:355-492. 

.      1818.      [Supplement    to    "The    fishes    of 

New  York  .  .  ."]  Amer.  Month.  Mag.,  2: 
241-248. 

Moore,  Donald.  1962.  Development,  distribu- 
tion, and  comparison  of  rudderfishcs  Kijphosus 
seciatrix  (  Linnaeus )  and  K.  incisor  ( Cuvier ) 
in  the  western  North  Atlantic.  Fish.  Bull. 
196,  Fish.  Bull.  Fish  &  Wildlife  Service,  61: 
451-480. 

MoREAU,  Emile.  1881.  Histoire  naturelle  des 
pois.sons  de  la  France.  Vol.  2.  Paris,  572  pp., 
220  figs. 

Mori,  Tamezo.  1952.  Check  list  of  the  fishes 
of  Korea.  Mem.  Hyogo  Univ.  Agric,  Sa- 
sayama.    Biol.  Ser.,   l(3):l-228. 

Morton,  Alexander.  1888.  Description  of  two 
new  fishes.  Pap.  Proc.  Roy.  Soc.  Tasmania 
for  1887:77-78,  1  fig. 

MuNRO,  Ian  S.  R.  1958.  Handbook  of  Austrahan 
fishes.  No.  29:117-120.  Families  Lepidopi- 
dae,  Luvaridae,  Centrolophidae,  Nomeidae, 
Tetragonuridae.  Australian  Fisheries  News- 
letter, Nov.   1958:17-20. 

Myers,  George  Sprague.  1958.  Trends  in  the 
evolution  of  teleostean  fishes.  Stanford 
Ichthyol.    Bull.,   7(3):27-30. 

Nardo,  Giovanni  Domenico.  1827.  Prodromus 
observatiomun  et  disquisitionum,  Adriaticae 
ichthyologiu'.  Giorn.  Fis.  Chim.  Stor.  Nat. 
Pavia,  10:22-40. 

Nichols,  John  Treadwell  and  Robert  Cush- 
MAN  Murphy.  1922.  On  a  collection  of 
marine  fishes  from  Peru.  Bull.  Amer.  Mus. 
Nat.  Hist.,  46(9):501-516. 

.      1944.     A    collection    of   fishes    from    the 


Strom ATEOiD  Fishes  •  Haedrich        137 


Panama  Bight,  Pacific  Ocean.  Bull.  Anier. 
Mus.   Nat.   Hist.,   83(  4  )  ;221-260. 

Nielsen,  J0rgen  G.  1963.  Marine  fishes  new 
or  rare  to  the  Danish  fauna  ( from  the  period 
1937-1961).  Vidensk.  Medd.  Dansk  Naturh. 
Foren.,  125:147-165. 

NoBRE,  AuGUSTO.  1935.  Fauna  marinha  de 
Portugal.  Vertebrados  I.  Descricao  do  peixes 
de  Portugal.   Porto,  547  pp.,  77  pis. 

Norman,  John  Roxbrough.  1937.  Coast  fishes. 
Part  II.  The  Patagonian  region.  Discovery 
Reports,  16:1-150,  5  pis. 

.      1957.     A   draft   synopsis   of   the   orders, 

families  and  genera  of  the  recent  fishes  and 
fish-like  vertebrates  ( excluding  Ostariophysi, 
Scleroparei,  Ammodytidae  and  a  few  other 
families,  notalily  Centrarchidae,  Percidae,  and 
Cichlidae)  covering  literature  up  to  1938, 
and,  as  far  as  it  was  available  to  the  author, 
from  1939  to  1944.  Brit.  Mus.  (Nat.  Hist.), 
649  pp.   Unpublished  manuscript. 

OcHiAi,  Akira  and  Koichiro  Mori.  1965. 
Studies  on  the  Japanese  butter  fish  referable 
to  the  genus  Psenopsis.  Bull.  Misaki  Mar. 
Biol.   Inst.,   Kyoto   Univ.,   No.    8:1-7. 

Ogilby,  James  Douglas.  1893.  Description  of 
a  new  pelagic  fish  from  New  Zealand.  Rec. 
Aust.  Mus.  Sydney,  2(5):64-65. 

.      1915.     On    some    new    or    little    known 

Australian  fishes.  Mem.  Queensland  Mus. 
Brisbane,  3:117-129. 

Okada,  Y.,  K.  Uchida  and  K.  Matsubar.\.  1935. 
Figures  and  descriptions  of  Japanese  fishes. 
Tokyo,  425  pp.,  166  pis.    [In  Japanese.] 

OsoRio,  Balthazar.  1909.  Contribuyao  para  o 
conhecimento  da  fauna  bathypelagica  visinha 
das  costas  de  Portugal.  Mem.  Mus.  Bocage, 
Lisboa,  No.   1. 

Padoa,  Emanuele.  1956.  Centrolophidae,  No- 
nieidae.  In:  Fauna  e  Flora  Golfo  di  Napoli, 
Monogr.  38,  part  111:538-545. 

Pappe,  Carl  Wilhelm  Ludwig.  1854.  Syn- 
opsis of  the  edible  fishes  of  the  Cape  of  Good 
Hope.    Capetown,  34  pp. 

Parin,  N.  V.  1958.  Rare  pelagic  fishes  of  the 
northwest  part  of  the  Pacific  Ocean.  (  Taractes 
stcindachncri,  Palinuiichthijs  japoniciis,  and 
Centrolophiis  lockingtoni. )  Voprosy  Ikhtiol., 
12(11):162-170.    [In  Russian.] 

Parr,  Albert  Eide.  1956.  On  the  original  vari- 
ates  of  taxonomy  and  their  regressions  upon 
size  in  fishes.  Bull.  Amer.  Mus.  Nat.  Hist., 
110(5):369-398,  14  figs. 

Patterson,  Colin.  1964.  A  review  of  Mesozoic 
acanthopterygian  fishes,  with  special  reference 
to  those  of  the  English  Chalk.  Phil.  Trans. 
Roy.  Soc.  London.  Ser.  B.  Biol.  Sci.  247 
(739):213-482. 

Peck,  William  D.  1804.  Description  of  four 
remarkable  fishes,   taken  near  the  Piscataqua 


in  New  Hampshire.  Mem.  Amer.  Acad.  Arts 
Sci.,  2(2):46-57. 

Pellegrin,  Jacques.  1937.  Le  centrolophe  oval, 
poisson  comesti])le.  Bull.  Soc.  Aq.  Peche, 
Paris,  44:33-36. 

Peters,  Wilhelm  Carl  Hartwig.  1869.  Uber 
neue  oder  weniger  bekannte  Fische  des  Ber- 
liner Zoologischen  Museums.  Monatsber. 
Akad.   Wiss.    Berlin,    1869:703-712. 

Petit,  G.  and  R.  Amar.  1946.  Capture  d'un  Tctra- 
gonurus  cuvieri  Risso  a  Saint-Tropez  (Var. ). 
Bull.    Mus.    Hist.    Nat.    Marseille,   6:34-39. 

Poey,  Felipe.  1861.  Meniorias  sobre  la  historia 
natural  de  la  Isla  de  Cuba  .  .  .  Hal:)ana,  vol. 
2:337^42. 

.      1868.     Synopsis  piscium  Cubensium.  In: 

Repertorio  fisico-natural  de  la  Isla  de  Cuba 
.  .  .  vol.  2:279-484.  Habana.  Barcina  y  Comp. 

Poll,  M.  1959.  Poissons  V — Teleosteens  acan- 
thopterygiens  (deuxieme  partie).  Res.  Sci. 
Exped.  Oceanogr.  Beige  Cote  Afr.  Atlant. 
Sud.,  vol.  4,  3B:  1-417,   127  figs. 

QuoY,  J.  R.  C.  AND  J.  p.  Gaimard.  1824.  Voy- 
age autour  du  monde  des  corvettes  1'  "Uranie" 
et  le  "Physicienne"  pendant  les  annees  1817— 
20.  Zool.,  part  II,  Poissons.  Paris,  pp.  192-401. 

Rafinesque-Schmaltz,  C.  S.  1810.  Caratteri  di 
alcuni  nuovi  generi  e  nuove  specie  de  animale 
e  piante  della  Sicilia.    Palenno,  105  pp.,  20  pis. 

Ramsay,  E.  P.  and  J.  D.  Ogilby.  1888.  On  the 
genus  Teiragonurus  of  Risso.  Proc.  Linn.  Soc. 
New  South  Wales,  Ser.  2,  3(1):9-16. 

Regan,  Charles  Tate.  1902.  A  revision  of  the 
fishes  of  the  family  Stromateidae.  Ann.  Mag. 
Nat.  Hist.,  Ser.  7,  10:115-131,  194-207. 

.      1913.     The  classification  of  the  percoid 

fishes.  Ann.  Mag.  Nat.  Hist.,  Ser.  8,  12: 
111-145. 

.      1914.     British  Antarctic  ("Terra  Nova") 


Expedition,  1910.  Zoology  I,  Fishes,  pp.  1- 
54. 

.      1914a.     Diagnoses  of  new  marine  fishes 

collected  by  the  British  Antarctic  ("Terra 
Nova")  Expedition.  Ann.  Mag.  Nat.  Hist., 
Ser.  8,   13:11-17. 

Richardson,  Sir  John.  1845.  Ichthyology.  In: 
John  Richardson  and  John  E.  Gray,  The  zool- 
ogy' of  the  voyage  of  H.M.S.  "Erebus  and 
Terror,"  under  the  command  of  Capt.  Sir. 
J.  C.  Ross  .  .  .  during  .  .  .  1839-43.  London, 
viii  -f-  139  pp.,  60  pis. 

Risso,  AxNastase.  1810.  Ichthyologie  de  Nice. 
Paris,  F.  Schoell,  xxxvi  +  388  pp.,  51  pis. 

.      1826.     Histoire  naturelle  des  principales 

productions  de  I'Europe  Meridionale,  vol.  III. 
Paris,  480  pp. 

Rondelet,  Gulielmus.  1554.  Libri  de  piscibus 
marinis,  in  quibus  vent  piscium  effigies  ex- 
pressie  sunt.    Lugduni,  fob,  620  pp. 


138 


Bulletin  Musctiiu  of  Comparative  Zoology,  Vol.  135,  No.  2 


RossiGNOL,  M.  AND  J.  Blache.  1961.  Sur  im 
poisson  Stromateidae  notiveaii  dn  Colfe  dc 
(uiimV,  Pscncs  hcnardi  nov.  sp.  Bull.  Mu.s. 
Hist.    Nat.    Paris,   2'    Ser.,    33(4)  :384-386,    1 

fig. 

Russell,  Patrick.  1(S()3.  Descriptions  and  fij^- 
ures  of  two  hundred  fishes  collected  at 
\'izagapatani  on  the  coast  of  Coromandel. 
\'ol.    1.    London,  vii  +  78  pp.,   100  pis. 

Sanz-Echeverria,  J.  1949.  Identificacion  de  los 
peces  de  la  faniilia  Centrolophidae  de  Espana 
por  medio  de  los  otolitos.  Real  Soc.  Esp. 
Hist.  Nat.,   Tomo  Extraordinaire:  151-156. 

Sauvace,  Hexri  Emile.  1879.  Mcmoire  sur  la 
faune  ichthyologique  de  I'ile  Saint-Paul.  Arch. 
Zool.  Exped.  Gen.  1879-1880,  8:1-46. 

Schneider,  Johaxx  Cottloh.  1784.  Sammlunji 
\erniischter  Abhandlungen  ziir  Aiifkliirung  der 
Zoologie  und  der  Handlungsgeschichte.  Ber- 
lin, 348  pp.,   1  pi. 

Sc:hwartz,  Frank  J.  1963.  The  barrelfish  from 
Chesapeake  Bay  and  the  Middle  Atlantic  Bight, 
with  comments  on  its  zoogeography.  Chesa- 
peake Science,  4{  3  ):  147-149. 

Scott,  Trevor  D.  1962.  The  marine  and  fresh- 
water fishes  of  South  Australia.  Adelaide, 
W.    L.    Hawes,    Government   Printer,    338   pp. 

Shojima,  Yoichi.  1961.  On  the  postlarvae  and 
juveniles  of  a  kind  of  butterfish,  Fscnopsis 
sp.,  accompanying  the  jellyfishes.  Bull.  Seikai 
Reg.  Fish.  Res.  Lab.,  No.  21:67-74. 

Simpson,  George  Gaylord.  1953.  The  major 
features  of  evolution.  New  York,  Columbia 
University  Press,  434  pp. 

SivERTsEN,  E.  1945.  Fishes  of  Tristan  da  Cunha, 
with  remarks  on  age  and  growth  based  on 
scale  readings.  Res.  Norwegian  Expedition 
to  Tristan  da  Cunha  1937-38,  Oslo,  No.  12: 
1-44,  8  pis. 

Smith,  Andrew.  1849.  Illustrations  from  the 
zoology  of  South  Africa;  consisting  chiefly  of 
figures  and  descriptions  of  objects  of  natural 
history  collected  during  an  expedition  into 
the  interior  of  South  Africa  in  1834-36.  Vol. 
IV,  Fishes.    31  pis. 

Smith,  C.  Lavett  and  Reeve  M.  Bailey.  1962. 
The  subocular  shelf  of  fishes.  J.  Morph., 
110(1):1-18. 

Smith,  Hugh  M.  1907.  The  fishes  of  North 
Carolina.  North  Carolina  Geological  and 
Economic  Survey,  Vol.  11:1-453,  188  figs. 
Raleigh,  E.  M.  Uzzell  &  Co. 

Smith,  J.  L.  B.  1934.  Marine  fishes  of  seven 
genera  new  to  South  Africa.  Trans.  Roy.  Soc. 
South   Africa,   22:89-100. 

.      1949.     The  .sea  fishes  of  southern  Africa. 

Capetown,  Central  News  Agency,  Ltd.,  South 
Africa,  550  pp.    Rev.  ed.  1953,  564  pp. 

.      1949a.     The  stromatcid   fishes   of   South 


Africa.     Ann.    Mag.    Nat.    Hist.,    Ser.    12,    2: 

839-851. 
— .      1953.     The    genus    Tetrafiontirtis    Risso, 

1810.    Ann.  Mag.  Nat.  Hist.  Ser.  12,  6:53-66, 

1   pi. 
.      1960.     Two  interesting  fishes  from  South 

Africa.    S.  Afr.  J.  Sci.,  56:91-92,  1  fig. 
— .      1961.     A     new     stromateid     fish     from 


South  Africa,  and  another  new  to  that  area. 
S.  Afr.  J.  Sci.,  57:158-160. 

SoLjAN,  ToNKO.  1948.  Fauna  et  flora  Adriatica. 
Vol.  1,  Pisces.  Split,  Jugoslavia,  437  pp., 
1350  figs. 

Starks,  Edwin  Chapin.  1906.  A  collection  of 
fishes  made  by  P.  O.  Simons  in  Ecuador  and 
Peru.  Proc.  U.  S.  Nat.  Mus.,  30:761-800, 
2  pis. 

Steindachner,  Franz.  1868.  tJbersicht  der 
Meeresfische  an  den  Kiisten  Spanien's  und 
Portugal's.  Sitzungsber.  Akad.  Wiss.  Wien, 
Abt.  1,  57(3):351-424. 

.      1874.     Ichthyologische     Beitriige.      Sitz- 

ung.sber.  Akad.  Wiss.  Wien,  Abt.  1,  70(32): 
37.5-390,    1   pi. 

Steindachner,  F.  and  L.  Doderlein.  1885. 
Beitriige  zur  Kenntniss  der  Fische  Japans 
(III).  Denkschr.  K.  Akad.  Wiss.  Wien,  49: 
171-212,  7  pis. 

Suzuki,  Kijoshi.  1962.  Anatomical  and  taxo- 
nomical  studies  on  the  carangid  fishes  of 
Japan.  Rep.  Fac.  Fish.,  Pref.  Univ.  Mie,  4 
(2):4.3-233. 

SwAiNsoN,  William.  1839.  The  natural  history 
and  classification  of  fishes,  amphibians,  and 
reptiles  or  monocardian  animals.  Vol.  II. 
London,  vi  +  452  pp.,  135  figs. 

Tchang,  T.  L.  ct  al.  1955.  The  fishes  of  the 
Gulf  of  Pechili  and  Yellow  Sea.  Peking,  360 
pp.    [In  Chinese.] 

Temminck,  Coenraad  Jacob  and  Hermann 
Schlegel.  1850.  Pisces  (last  part).  In: 
C.  T.  E.  von  Siebold,  Fauna  Japonica,  sive 
descriptio  animalium  quae  in  itinere  per 
Japoniam  suscepto  annis  1823-30  collegit, 
notis  observationibus  et  odumbrationibus  illus- 
travit  P.  F.  de  Siebold.  Lugduni  Batavorum, 
pp.  270-324. 

Templeman,  Wilfred  and  Richard  L.  Haedrich. 
1966.  Distribution  and  comparisons  of  Ccii- 
frolophits  nip,er  ( Gmelin )  and  Cciitrolophits 
hritannicus  Giinther,  family  Centrolophidae, 
from  the  North  Atlantic.  J.  Fish.  Res.  Bd., 
Canada,  23(8)  :1161-1185. 

Thompson,  D'Arcy  Wentworth.  1947.  A  glos- 
sary of  Greek  fishes.  London,  Oxford  Uni- 
versity Press,  vi  -f  302  pp. 

ToRTONESE,  E.  1959.  Revisione  dei  Centrolophi- 
dae ( Pisces  Perciformes )  del  Mare  Ligure 
(1).    Ann.  Mus.  Stor.  Nat.  Genova,  71:57-82. 


Stromateoid  Fishes  •  Uacdrich 


139 


ToTTEX,  A.  K.  1960.  Studies  on  PhysaJia  plujsa- 
lis  (L.).  Part  1.  Natural  history  and  mor- 
phology. Discovery  Reports,  30:301-367,  18 
pis. 

Ueno,  T.  1954.  First  record  of  a  strange  bathy- 
pelagic  species  referable  to  the  genus  Cen- 
frolopJuis  from  Japanese  waters,  with  remarks 
on  the  specific  differentiation.  Bull.  Fac.  Fish. 
Hokkaido,  5:240-247. 

Valenciennes,  Achille.  1836.  Les  poissons. 
In:  G.  Cuvier,  Le  regne  animal  .  .  .  Edition 
III,  Vol.  4,  392  pp. 

.      1848.     Ichth\'ologie    des    iles    Canaries, 

ou  histoire  natnrelle  des  poissons,  rapportes 
par  MM.  Webb  et  Berthelot.  In:  P.  B.  Webb 
and  S.  Berthelot,  Histoire  natnrelle  des  iles 
Canaries,  1835-1850.    Vol.  2,  part  2,  109  pp. 

Waite,  Edgar  Ravenswood.  1894.  New  or  rare 
fishes  from  Maroubra,  N.  S.  W.  Proc.  Linn. 
Soc.  New  South  Wales,  Ser.  2,  9:215-227. 

.      1910.     Notes    on    New    Zealand    fishes. 


Trans.   Proc.   New  Zealand   Inst.,   Wellington, 
1909,   42:384-391. 

Whitley,  Gilbert  P.  1931.  New  names  for 
Australian  fishes.  Aust.  Zool.  Sydney,  6:310- 
334. 

.      1933.     Studies    in    ichthyology.     No.    7. 

Rec.  Aust.  Mus.  Sydney,  19:60-112,  4  pis. 

.      1935.     Studies    in    ichthyology.     No.    9. 

Rec.  Aust.  Mus.   Sydney,   19(4)  :215-250. 

.  1943.  Ichthyological  notes  and  illus- 
trations (  part  2 ) .  Australian  Zoologist,  Syd- 
ney, 10:167-187,   10  figs. 

.      1948.     Studies   in   ichthyology.     No.    13. 

Rec.   Aust.  Mus.  Sydney,  22(l):70-94. 

1958.     Descriptions      and      records      of 


fishes.     Proc.    Roy.    Zool.    Soc.    New    South 

Wales,  1956-57:28-51,  12  figs. 
WiLLUGHBY,  Francis.      1686.     De  historia  piscium. 

Libri   quatuor,  jussu   &  sumptibus,   Soc.   Reg. 

Lond.  ed.  etc.,  Oxon.,  fob,  373  pp.,   186  pis. 
(Received  26  July  1965.) 


u.':ri'i>.€  V?'-'JV 


^uitetin  OF  THE 

Museum   of 

Comparative 

Zoology 


Morphology  and  Relationships  of  the  Holocephali 
with  Special  Reference  to  the  Venous  System 


BARBARA  J.  STAHL 


MwiiOMm  of  rnmparflljye  Zoology.  Cambridge.  Massachusetts 
Jstituto  di  Btotogter-Generak^UDraexsUa-df-PtsurlTaty — 


HARVARD  UNIVERSITY 

CAMBRIDGE,  MASSACHUSETTS,  U.S.A. 


VOLUME  135,  NUMBER  3 
JANUARY  27,  1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

MEMoms  1864-1938 

JoHNSONLA,  Department  of  Mollusks,  1941- 

OccASiONAL  Papers  on  Mollusks,  1945- 


Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    ( Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).   $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.  $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.   ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


@  The  President  and  Fellows  of  Harvard  College  1967. 


MORPHOLOGY  AND  RELATIONSHIPS  OF  THE  HOLOCEPHALI 
WITH  SPECIAL  REFERENCE  TO  THE  VENOUS  SYSTEM 


BARBARA  J.   STAHL' 


CONTENTS 

Introduction  __,  141 

Materials   and   Methods  143 

Description  of  the  Venous  System  144 

The  Subcutaneous  Veins  145 

The  Deep  Veins 

Precardiac  Group  147 

Postcardiac   Group   ._  151 

The  Hepatic  Portal  System  154 

Discussion  of  the  Significance  of  the  Circula- 
tory  System   to   the   Problem   of   Holoce- 

phalian  Relationships  155 


Discussion  of  Phylogenetic  Clues  from  Other 

Organ    Systems    162 

The  Nervous   System  162 

The   Skeletal   System   _   167 

The  Muscular  System  176 

The  Urogenital  System   181 

The   Digestive  System  186 

Conclusion   190 

Acknowledgments   ._ 193 

References  Cited  193 


INTRODUCTION 

Although  the  evolutionary  relationships 
of  the  Holocephali  have  been  under  con- 
sideration for  years,  no  one  theory  of  their 
descent  has  appeared  so  satisfactory  that 
the  question  may  be  laid  to  rest.  This  paper 
is  the  result  of  the  continuing  search  for 
progress  in  clarifying  the  position  of  these 
fishes.  To  this  end  an  investigation  of  the 
venous  system  of  Chimacra  colliei  Lay  and 
Bennett  (Hydrolo^tis  colliei )  has  been  un- 
dertaken, and  a  reassessment  of  the  anatomy 
of  the  Holocephali  has  been  made,  taking 
into  consideration  both  the  structural  ar- 
rangements revealed  by  the  new  dissection 
and  current  paleontological  knowledge. 

The  problem  of  classifying  the  holo- 
cephalian  fishes  (the  living  genera  of  which 
are:  Chimacra,  CaUorhijnchus,  Rhinochi- 
maera,  and  Harriotta)  has  become  more 
and  more  difficult  as  the  understanding  of 

^  St.  Anselm  College,  Manchester,  N.H. 

Bull.  Mus.  Comp. 


the  history  of  fishes  has  grown.  In  contrast 
to  Linnaeus  who  set  the  Holocephali  down 
in  the  same  group  with  the  sharks,  rays, 
sturgeons,  and  lampreys  because  of  their 
cartilaginous  skeleton,  modem  scholars  are 
giving  much  thought  to  the  propriety  of 
even  including  them  with  the  elasmo- 
branchs. 

The  day  has  passed,  too,  \\'hen  an  inves- 
tigator could  seek  to  solve  the  problem  by 
focussing  upon  a  single  structure  and  sug- 
gesting that  it  indicates  a  probable  evolu- 
tionary relationship.  This  point  bears  men- 
tion because  the  spotting  of  isolated  simi- 
larities has  generated  several  hypotheses, 
concerning  the  evolution  of  the  Holocephali, 
which  have  proven  untenable  \\'hen  an  in- 
tensive examination  is  pressed.  Based  upon 
reasoning  of  this  sort  is  the  idea  that  the 
Holocephali  might  possibly  be  allied  to  the 
lungfishes  through  the  common  possession 
of  autostyly  and  cutting  toothplates.  When 
the  idea  was  tested  by  further  study,  it  was 

Zool.,  135(3):   141-213,  January,   1967         141 


142        BiiUctiu  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


shown  that  the  nature  of  the  palatoquadratc 
fusion  was  different,  that  tlie  toothphites 
were  surely  not  homologous  structures,  and 
that  other  anatomical  characteristics  were 
not  alike.  When  paleontological  evidence 
is  considered,  the  probability  of  a  relation- 
ship between  the  Holocephali  and  Dipnoi 
recedes  still  further.  Despite  the  large 
amount  of  cartilage  in  the  skeleton,  lung- 
fishes  have  definitely  sprung  from  ances- 
tral bony  fishes,  which  sets  them  far  from 
the  holocephalians.  The  latter  fishes  arose 
probably  from  forms  more  nearly,  though 
not  necessarily  very  closely,  allied  with  the 
ancestors  of  sharks  than  with  the  predeces- 
sors of  the  Osteichthyes.  Assuming  the 
truth  of  this  statement,  one  can  cast  aside 
the  hypotheses  which  link  the  Holocephali 
to  fishes  like  Latimcrio  and  Pohjptenis 
whose  position  as  bony  fish  is  well  estab- 
lished, and  also  those  which  embed  the 
holocephalians  in  the  line  of  fishes  leading 
to  tetrapods. 

Currently  only  two  possibilities  of  holo- 
cephalian  origin  are  receiving  serious  atten- 
tion. One  opinion  holds  that  holocephalians 
are  aberrant  off-shoots  from  ancient  carti- 
laginous fishes.  Specifically,  this  school 
favors  the  idea  that  the  Holocephali  are 
descendants  of  the  bradyodonts,  an  extinct 
group  of  presumed  shark  relatives  distin- 
guished by  nonreplaceable  teeth  of  a  pecu- 
liar histological  structure.  Although  no 
wealth  of  fossil  material  exists,  paleontol- 
ogists have  speculated  that  at  least  some 
of  the  bradyodonts  were  autostylic,  as  are 
the  Holocephali.  A  leading  advocate  of  the 
bradyodont  origin  of  the  Holocephali,  Moy- 
Thomas  ( 1936 )  has  studied  one  of  the  rare 
bradyodont  fossils  which  consist  of  more 
than  teeth  and  spines  and  found  in  it  many 
resemblances  to  holocephalian  design.  This 
specimen,  the  cochliodont  Ilclodus  simplex, 
dates  from  Carboniferous  times.  If  it  is 
ancestral  to  the  Jurassic  chimaerids,  one 
must  assume  that  all  the  distinctive  holo- 
cephalian characteristics  which  Helodus 
does  not  possess  were  evolved  in  the  inter- 
vening years.    Other  bradyodonts  such  as 


Menaspis  and  Omcantlius  have  been  dis- 
cussed in  relationship  to  the  problems  of 
holocephalian  origin,  but  they  have  either 
possessed  structures  like  the  spines  on  the 
head  of  the  former  which  makes  one  hesi- 
tate to  place  them  in  the  direct  ancestral 
line  or  they  have  been,  like  the  latter,  in  too 
fragmentary  a  condition  to  allow  a  thorough 
comparison.  In  a  recent  paper,  Patterson 
( 1965 )  concludes  that  the  bradyodonts  are 
closely  enough  related  to  the  Holocephali 
to  be  grouped  with  them  in  the  class  Holo- 
cephali, but  abandons  the  idea  that  Helodus 
or  any  other  bradyodont  is  ancestral  to  the 
holocephalian  line. 

The  second  possible  source  of  the  Holo- 
cephali is  an  older  one.  Amongst  the 
ptyctodonts,  a  placodenu  group,  have  been 
found  several  fossil  forms  that  show  charac- 
teristics which  could  be  ancestral  to  those 
of  holocephalians.  The  resemblances  were 
recognized  early  (Pander,  1858),  but  ne- 
glected after  the  ptyctodonts  were  alhed 
with  the  arthrodires,  and  after  Moy-Thomas 
offered,  in  Helodus,  a  bradyodont  ancestor 
for  the  holocephalians  which  had  long  been 
classified  in  a  general  category  with  sharks. 
The  idea  of  a  ptyctodont  ancestor  has  re- 
turned to  favor,  however,  as  the  magnitude 
of  the  differences  between  holocephalians 
and  sharks  has  been  revealed.  It  seems  now 
most  attractive  to  find  a  stock,  traceable 
far  back  into  the  Devonian,  which  could  be 
ancestral  to  the  Holocephali.  There  are  a 
number  of  fossils  ( in  a  more  complete  state 
than  many  of  the  cochliodont  forms )  which 
have  been  used  as  a  basis  for  comparison 
with  extinct  and  Recent  chimaerids.  Of 
these  forms,  students  of  holocephalian  evo- 
lution cite  most  often  RliampJwdopsis, 
Pttjetodus,  and  Ctenuiella.  The  last  is 
considered  by  0rvig  (1962)  to  show  a  re- 
markable number  of  similarities  to  the  Holo- 
cephali. However,  the  ptyctodont-holo- 
cephalian  relationship,  while  possible,  is 
far  from  proved.  There  are  still  serious 
questions  to  be  solved.  One  must  suppose, 
for  instance,  if  the  relationship  is  a  fact, 
that  over  the  countless  generations  which 


Morphology  and  Relationships  of  Holocephali  •  Stahl        143 


separated  the  Devonian  ptyctodonts  from 
the  Jurassic  chimaerids  the  animals  lost 
their  distinctixe  pattern  of  dermal  armor, 
their  pectoral  spines,  and  their  internal 
bone.  While  changes  of  this  nature  are  not 
impossible,  there  is  no  fossil  evidence  to 
prove  that  they  did  take  place. 

In  trying  to  decide  whether  it  is  more 
likely  that  holocephalians  originated  from 
ptyctodonts  than  from  a  group  closer  to  the 
shark  line,  one  tiuns  normally  to  the  data 
available  from  embryological  studies.  In 
the  case  of  the  Holocephali,  very  little 
embryological  work  has  been  done.  Since 
the  holocephahan  fishes  lay  their  eggs,  al- 
ready fertilized  and  enclosed  in  a  case,  in 
deep  water,  the  embryos  are  not  often  ob- 
tained. There  have  been  only  two  studies 
made  of  embryonic  forms:  that  of  Schauins- 
land  on  CoUorhynchus  (1903)  and  that  of 
Dean  on  Chimacia  (1906).  Although  both 
studies  were  elegant  pieces  of  work,  a  lack 
of  certain  stages  resulted  in  the  absence  of 
observations  of  the  fusion  of  the  upper  jaw 
and  the  step-by-step  fomiation  of  the  hyoid 
arch,  for  example.  An  understanding  of 
these  two  points  would  shed  great  light 
upon  the  evolutionary  question. 

Besides  the  paleontological  and  embryo- 
logical approaches,  there  is  a  third  useful 
avenue  of  investigation.  The  contribution 
from  the  area  of  comparative  anatomy  can- 
not be  omitted  in  assembling  e\'idence 
which  bears  upon  the  problem.  Although 
holocephahan  fishes  have  been  dissected 
numerous  times,  the  work  upon  the  anat- 
omy of  these  fishes  is  not  completed.  In 
early  anatomical  inxestigations  the  dissector 
often  placed  his  emphasis  upon  structures 
which  are  not  the  best  keys  to  the  evolu- 
tionary problem.  There  is  no  information 
available  concerning  some  of  the  areas 
which  are  of  great  interest  from  the  com- 
parative point  of  view. 

One  such  area,  that  of  the  venous  system, 
has  been  completely  untouched.  Although 
there  have  been  publications  concerning 
the  distinctive  portions  of  the  arterial  path- 
way, there  is  nothing  in  the  literature  about 


the  pattern  of  vessels  returning  blood  to  the 
heart.  It  was  in  the  hope  that  the  venous 
system  would  show  special  features  which 
might  serve  as  clues  to  a  better  understand- 
ing of  holocephahan  evolution  that  this 
study  was  undertaken. 

MATERIALS  AND  METHODS 

Since  it  was  desirable  to  avoid  describ- 
ing as  the  general  occurrence  an  anomalous 
vessel  in  a  single  fish,  dissections  were  re- 
peated until  it  appeared  certain  that  a  par- 
ticular pattern  was   a  normal  and  not   an 
abnormal  feature.  The  relatively  large  num- 
ber of  specimens  available  made  this  method 
possible.  The  first  specimen  to  be  dissected 
was  a  female  Chimaera  coUiei,  uninjected, 
which  had  been  preserved  in  formalin  and 
transferred  to  alcohol.    A  group  of  twelve 
specimens  of  Chimaera  colliei  were  obtained 
fresh-frozen  from  Vancouver,  B.  C,  through 
the  kindness  of  Dr.  Norman  J.  Wilimovsky. 
The    procedure    used    with    these    animals 
was  to  defrost  them,  inject  immediately  with 
latex,  preserve  first  in  formalin,  and  after 
five  days  to  transfer  them  in  several  steps 
to  70  per  cent  alcohol.    The  last  six  speci- 
mens of  Chimaera  coUiei,  four  females  and 
two  males,  were  received  already  injected 
with  latex  through  the  kind  efforts  of  Dr. 
Richard  Snyder.   For  comparative  purposes 
one    specimen    of   CaUorhynchus   and   one 
specimen  of  Rhinoehimaera  were  examined. 
Several  methods  of  injection  \\'ere  tried. 
Because  of  the  delicate  nature  of  the  vein- 
walls    and    the    consequent    similarity    be- 
tween veins  and  strands  of  connective  tis- 
sue in  some  areas,  nothing  was  interpreted 
as  a  vein  unless  it  was  observed  filled  with 
an  injecting  material  or  remnants  of  brown- 
colored    agglutinated    blood.    In    the    first 
uninjected  specimen  described  above,  a  car- 
mine suspension  was  injected  in  area  after 
area  as  the  dissection  proceeded.   India  ink 
was  also  tried.    In  the  defrosted  specimens, 
as  noted,  latex  was  used.    Finallv,   to  fill 
certain   empty   areas   in   the   professionally 
injected   specimens,   ordinary   poster  paint 


144         Bulletin  Mu.scinii  of  Comparative  Zoology,  Vol.  135,  No.  3 


was  employed  in  its  regular  concentration 
and  also  in  a  slightly  diluted  form. 

All  these  media  were  introduced  through 
a  glass-barreled  syringe  fitted  with  a  num- 
ber 23  needle  inserted  into  an  inch-and-a- 
half-long  piece  of  polyethylene  tubing,  size 
50.  The  tubing  was  tied  into  the  vessel 
through  which  the  injection  was  made. 

The  routes  that  were  available  for  injec- 
tion were  limited.  Injection  via  a  sinus 
proved  impractical,  because  the  injecting 
apparatus  could  not  be  tied  tightly  to  the 
delicate  sinus- wall.  Very  fine  veins  disin- 
tegrated under  the  most  careful  handling. 
Only  large  veins  of  well-defined  cylindrical 
shape  were  useful.  Injecting  through  them 
was  hampered  only  by  the  presence  of 
valves  which  restricted  the  amount  of  in- 
jection material  able  to  pass  beyond  into 
tributary  vessels.  In  particular,  this  diffi- 
culty arose  in  getting  material  to  pass  from 
the  common  cardinal  vein  forward  into  the 
anterior  cardinal  sinus  and  also  in  filling 
the  deep  veins  of  the  fins.  Although  various 
vessels  were  tried  as  the  dissection  ad- 
vanced, for  the  initial  attempt  to  fill  as 
much  of  the  venous  system  as  possible,  two 
veins  were  relied  upon.  To  inject  the  he- 
patic portal  system,  the  posterior  dorsal  in- 
testinal vein  was  employed.  To  fill  the 
systemic  vessels,  injection  was  made  into 
the  lateral  cutaneous  vein  immediately  pos- 
terior to  the  scapula.  This  vein  could  be 
uncovered  easily  over  a  considerable  dis- 
tance by  removing  the  skin  just  below  the 
lateral  line.  Injection  was  made  through 
this  vessel  first  in  an  anterior  and  then  in  a 
posterior  direction. 

DESCRIPTION  OF  THE  VENOUS  SYSTEM 

The  veins  return  blood  to  the  two  com- 
mon cardinal  vessels  which  lie  medial  to 
the  anterior  edge  of  the  scapular  process 
of  the  pectoral  girdle  and  empty  into  the 
lateral  corners  of  the  sinus  venosus.  On 
either  side,  three  veins  meet  at  the  point  at 
which  the  last  pharyngobranchial  cartilage 
articulates  with  a  facet  of  the  scapula,  to 
create  the  common   cardinal   of  that   side. 


These  three  are  the  anterior  cardinal,  the 
posterior  cardinal,  and  the  lateral  cutaneous 
veins.  The  first  comes  from  a  fonvard  di- 
rection, the  second  comes  from  the  pos- 
terior region,  and  the  third  runs  downward 
and  slightly  caudad  to  meet  the  other  two. 
Into  the  upper  end  of  the  common  cardinal, 
the  inferior  jugular  vein  opens.  More  ven- 
trally,  the  brachial  sinus  opens  into  the 
common  cardinal  from  the  posterior  side 
(PI.  5,  B,  C). 

Each  of  the  major  venous  trunks  will  be 
described  with  its  tributaries  and  the  areas 
which  they  drain.  So  that  the  description 
may  be  more  easily  understood,  the  pattern 
of  the  venous  system  is  presented  first  in 
concise,  outline,  form: 

I.     The  Subcutaneous  System 
Lateral  eiitaneous  vein 
Caudal  tributary 
Axial  tributaries 

Pelvic  anastomotic  area 
Clasper  veins 
Ventro-lateral  tributary 
Postscapular  tributary 

Dorso-lateral  axial  branch 
Dorsal  fin  branch 
Prescapular  tributary 

Dorsal  ceplialic  branch 
Anterior  subcutaneous  tributary 
Ventral  cephalic  branch 
Opercular  branches 
Subscapular  tributary 
II.      The  Deep  Veins 

A.  Precardiac  vessels 
Anterior  cardinal  sinus 

Inferior  jugular  vein 
Posterior  cerebral  vein 
Postorbital  vein 
Hyoid  tributary 
Orbital  sinus 

Maxillo-facial  vein 
Preorbital  branch 
Deep  labial  branch 
Orbito-nasal  vein 
Posterior  palatal  vein 
Superior  adductor  mandibidar 

vein 
Anterior  cerebral  \  ein 

Posterior  cerebral  tributary 
Anterior  cerebral  tributary 
Ethmoidal  \'ein 

B.  Postcardiac  vessels 
Brachial  sinus 

Posterior  brachial  vein 


Morphology  and  Relationships  of   Holocephali  •  Stahl        145 


Anterior  brachial  \  e in 
Posterior  cardinal  sinus 

Ventro-anterior  parietal  vein 
Deep  epaxial  veins 
Dorsal  fin  sinus 

Medial  dorsal  vein 
Anterior  epaxial  \'ein 
Spino-basal  vein 
Esophageal  veins 
Anterior  parietal  veins 
Veins  of  the  reproductive  tract 
Renal  veins 
Femoral  vein 

Rectal  tributary 

Dorsal  fin  tributaries 

Ventral  fin  tributaries 
Hepatic  veins 
Renal  portal  veins 
Caudal  vein 
Parietal  veins 

Ventro-posterior  parietal  \  ein 
Hepatic  portal  vein 
Intra-intestinal  vein 

Anterior  dorsal  intestinal  tribu- 
taries 
Anterior  ventral  intestinal  vein 
Mesenteric  vein 

Dorsal  posterior  intestinal  vein 

Ventral  posterior  intestinal  vein 

Auxiliary  splenic  veins 

Lieno-pancreatic  vein 

Auxiliary  pancreatic  veins 

The  subcutaneous  system  is  shown  in 
Plate  1.  The  deeper  veins  are  represented 
diagrammatically  in  Plates  2  and  3. 

THE  SUBCUTANEOUS  VEINS 

There  is  an  extensive  system  of  sub- 
cutaneous drainage  (PI.  1).  The  vessels 
which  form  it  lie  in  the  loose  connective 
tissue  under  the  skin.  Although  their  path- 
ways vary  somewhat  in  different  specimens, 
the  basic  pattern  of  flow  is  generally  the 
same.  Assigning  names  to  the  vessels  of 
this  system  is  a  hazardous  business  because 
of  the  numerous  anastomotic  connections 
which  are  present,  but  there  are  several 
principal  trunks  \\'hich  can  be  specifically 
distinguished. 

The  chief  collecting  trunk  deserves  the 
name  lateral  cutaneous  vein,  for  it  courses 
anteriorly,  paralleling  the  lateral  line.  In 
the  caudal  region  it  is  located  about  a  half 
inch  belo\\'  the  lateral  line  canal,  but  at  the 


level  of  the  base  of  the  pelvic  fin  it  bends 
dorsally  somewhat  and  can  be  followed  for- 
ward into  the  trunk  region  where  it  is  to  be 
found  just  ventral  to  the  lateral  line.  At  its 
anterior  end  it  continues  forward  lateral  to 
the  muscle-covered  dorsal  extension  of  the 
scapular  cartilage,  bends  medially  around 
the  anterior  edge  of  this  cartilage,  and  then 
runs  ventrally  for  a  short  distance  to  form, 
with  the  anterior  and  posterior  cardinals, 
the  common  cardinal  vein.  As  it  passes  ven- 
trally on  the  medial  side  of  the  scapula,  it 
enlarges  sufficiently  to  merit  the  name  of 
subscapular  sinus.  Where  it  approaches  the 
upper  end  of  the  common  cardinal  it  is 
flanked  by  passing  nerves,  the  anterior 
nerve  trunk  containing  fibers  of  the  cervical 
plexus  which  innervate  the  hypobranchial 
muscles,  and  the  posterior  trunk  containing 
branches  of  the  first  through  third  spinal 
nerves.  ( There  is  also  in  the  anterior  trunk 
a  small  group  of  visceral  vagus  fibers. ) 

The  lateral  cutaneous  vein,  as  the  prin- 
cipal trunk  of  the  subcutaneous  system,  has 
the  firmest  wall  of  any  vein  involved  in  the 
superficial  drainage.  The  toughness  of  the 
wall  is  due  primarily  to  an  ensheathing 
layer  of  dense  connective  tissue.  This  vein 
receives  many  tributaries  which  will  be  de- 
scribed below,  beginning  with  those  bring- 
ing blood  from  the  most  posterior  regions. 

Although,  in  the  caudal  region,  the  lateral 
cutaneous  runs  forward  a  short  distance 
ventral  to  the  lateral  line,  there  is  another, 
smaller  vein  which  accompanies  the  sensory 
canal.  This  caudal  tributary  turns  ventrally 
to  empty  into  the  lateral  cutaneous  vein  at 
the  point  along  the  length  of  the  body 
which  is  on  a  level  with  the  posterior  limit 
of  the  pelvic  fin  attachment. 

As  the  lateral  cutaneous  vein  courses  for- 
ward, it  collects  blood  returning  from  the 
superficial  regions  of  the  axial  musculature 
dorsal  and  ventral  to  it.  The  axial  tributaries 
are  arranged  in  an  orderly  but  not  a  rigidly 
segmental  pattern.  The  dorsal  tributaries 
are  relatively  short  and  in  the  region  of  the 
trunk  posterior  to  the  dorsal  fin  spine  have 
as   their   source   a   network   of  little   veins 


146         Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


which  forms  a  narrow  band  dorsal  to  the 
lateral  line  and  parallel  to  it.  The  ventral 
tributaries  collect  blood  from  a  much 
greater  area  and  in  the  pelvic  region  are 
considerably  enlarged.  There,  they  draw 
from  an  anastomotic  network  of  veins.  As 
part  of  that  network,  a  vein  can  be  seen 
running  along  the  line  of  origin  of  the  su- 
perficial levator  muscle  of  the  fin.  The 
location  of  this  line  may  be  described  as 
being  about  halfway  between  the  lateral 
line  above  and  the  base  of  the  pelvic  fin 
below^  Into  this  vein  run  tributaries  from 
the  levator  muscle,  from  the  axial  muscle 
medial  to  the  levator,  and  from  the  axial 
muscles  which  are  posterior  and  ventral  to 
the  pelvic  region.  These  tributaries  have 
connections,  also,  with  two  veins  which  to- 
gether encircle  the  base  of  the  fin.  One 
runs  around  the  base  laterally;  the  other 
runs  around  it  medially,  thus  edging  the 
anal  region.  These  two  vessels  receive  veins 
draining  the  fin  web  and  the  superficial 
muscles  of  the  fin  itself.  In  the  male  Chi- 
maera,  the  veins  of  the  clasper,  which  re- 
ceive blood  from  the  erectile  tissue  in  the 
clasper  tips,  become  superficial  as  they 
course  proximally  and  empty  into  the  ve- 
nous ring  at  the  fin  base  (Pi.  4,  A).  The 
chief  clasper  veins  are  two  which  appear  on 
the  ventral  side  of  the  clasper.  One  drains 
each  prong,  and  they  merge  shortly  before 
emptying  at  the  posterior  edge  of  the  fin 
base. 

From  the  anterior  corner  of  the  venous 
network  in  the  pelvic  region  there  flows 
forward  a  vessel  of  rather  large  size  which 
gathers  blood  from  the  skin  and  superficial 
axial  musculature  ventral  to  the  field  served 
by  the  axial  tributaries  to  the  lateral  cuta- 
neous vein.  This  vessel  meanders  forward 
over  a  slightly  wavy  pathway,  finally  curv- 
ing dorsally  behind  the  pectoral  region  to 
empty  into  the  lateral  cutaneous  trunk  just 
before  the  latter  turns  inward  around  the 
anterior  edge  of  the  scapula.  The  name 
ventro-lateral  tributary  seems  appropriate 
for  this  vein.  In  one  specimen  which  had 
been  injected  with   India   ink,   small   veins 


were  seen  entering  it  from  the  posterior 
edge  of  the  operculum  dorsal  to  the  gill 
opening  and  from  the  ventral  part  of  the 
trunk  immediately  behind  the  opening  from 
the  gill  chamber.  The  veins  in  this  area 
were  not  injected  successfully  in  any  other 
specimen. 

The  lateral  cutaneous  trunk  receives  two 
sizable  tributaries  bearing  blood  from  dor- 
sal regions.  The  first  one  to  be  described 
begins  lateral  to  the  muscle-covered  pos- 
terior tip  of  the  scapular  cartilage  which  is 
bound  against  the  epaxial  muscles  at  the 
base  of  the  dorsal  spine.  This  vein,  called 
the  post.scapular  tril)iitani,  receives  blood 
from  the  dorsal  fin  branch,  draining  the 
web  and  muscles  of  the  dorsal  fin.  Halfway 
along  its  course  to  the  lateral  cutaneous 
vein,  the  postscapular  tributary  receives  the 
(lorso-Iateral  axial  branch.  The  latter  vessel 
is  a  long  one,  running  parallel  but  dorsal  to 
the  lateral  cutaneous  vein.  It  collects  blood 
returning  from  the  superficial  epaxial  mus- 
cles which  lie  dorsal  to  those  drained  by  the 
axial  tributaries  of  the  main  lateral  trunk. 
Some  of  the  branches  which  join  the  dorso- 
lateral axial  branch  can  be  seen  to  connect 
also  with  a  median  dorsal  vessel  whose 
blood  flows  eventually  into  the  posterior 
cardinal  sinus.  These  connections  represent 
one  of  the  few  anastomoses  between  the 
subcutaneous  and  the  deep  venous  drain- 
age systems. 

Far  dorsally,  near  the  base  of  the  dorsal 
fin  spine,  there  are  prominent  vessels  which 
form  an  anastomosis  between  the  postscap- 
ular vein  and  the  second  of  the  two  sizable 
tributaries  from  the  dorsal  region.  The 
second  one,  the  prescapular  tributary, 
courses  ventrally  just  in  front  of  the  an- 
terior edge  of  the  scapula  to  join  the  lateral 
cutaneous  vein  at  the  point  at  which  it 
turns  medially  to  meet  the  common  cardi- 
nal. Shortly  before  emptying  into  the  lat- 
eral cutaneous,  the  prescapular  tributary 
receives  the  dorsal  cephalic  brancJi  carrying 
blood  from  the  flattened  triangular-shaped 
dorsal  surface  of  the  head.  Atop  the  head, 
the  dorsal  cephalic  branches  of  the  left  and 


Morphology  and  Relationships  of  Holocephali  •  StaJil        147 


right  sides  are  connected  through  anasto- 
mosing venules.  As  it  runs  toward  its  meet- 
ing with  the  prescapular,  the  dorsal 
cephalic  follows  the  posterior  portion  of  the 
supraorbital  sensory  canal,  collecting  blood 
from  fine  venules  which  parallel  the  mucous 
canals  above  the  eye.  A  small  vein  draining 
the  skin  immediately  above  the  orbit  may 
empty  into  the  dorsal  cephalic  branch  or 
may  be  connected  to  the  tributary  next  to 
be  described. 

This  tributary,  called  the  anterior  sub- 
cutaneous, empties  into  the  lateral  cutane- 
ous vein  at  the  same  point  at  which  the 
prescapular  enters  it.  Approaching  that 
point,  it  courses  dorso-posteriorly,  approxi- 
mately paralleling  the  posterior  quarter  of 
the  suborbital  sensory  canal.  This  vessel 
receives  several  opercular  branches  (some 
of  which  may  anastomose  with  the  ventro- 
lateral tributary  near  its  anterior  end).  It 
receives  also  a  ventral  cephalic  branch 
which  drains  veins  collecting  forward  and 
ventral  of  the  orbit  and  fine  venules  which 
parallel  the  group  of  mucous  canals  an- 
terior and  ventral  to  the  eye.  The  ventral 
cephalic  branch  may  also  receive  blood 
from  the  region  just  posterior  to  the  lower 
jaw,  but  in  no  specimen  could  the  injection 
medium  be  made  to  penetrate  that  far  for- 
ward. 

The  last  tributary  to  the  lateral  cutane- 
ous vein  \\'hich  remains  to  be  mentioned  is 
the  subscapular.  This  one  is  really  a  small 
sinus,  lying  against  the  medial  surface  of 
the  scapular  cartilage.  It  receives  venules 
from  the  cartilage  itself  and  from  two  fine 
veins  which  follo\A'  the  posterior  border  of 
the  cartilage,  one  coming  from  a  xentral 
and  the  other  from  a  dorsal  direction.  The 
subscapular  tributary  is  the  last  one  to  join 
the  lateral  cutaneous  vein  before  it  empties 
into  the  common  cardinal  vein. 

THE   DEEP  VEINS 

Precardiac  Group 

The  return  of  blood  from  the  deep  por- 
tion of  the  bodv  anterior  to  the  heart  takes 


place  through  the  anterior  cardinal  sinus. 
This  sinus  is  exposed  by  lifting  the  dorsal 
constrictor  muscle  which  covers  the  gill 
area.  As  the  connective  tissue  beneath  the 
muscle  is  cleared  away  dorsal  to  the  oper- 
culum, the  scalpel  falls  into  the  sinus.  The 
blood-space  lies  lateral  to  a  muscle  origi- 
nating under  the  subocular  shelf  and  insert- 
ing posteriorly  upon  the  last  pharyngobran- 
chial  cartilages.  This  muscle,  the  trapezius 
internus  of  Vetter  (Vetter,  1878),  covers  a 
portion  of  the  branchial  branches  of  the 
vagus  nerve.  The  latter  are  visible  through 
the  medial  wall  of  the  anterior  cardinal 
sinus  for  a  short  part  of  their  pathway  ven- 
tral to  the  muscle-band.  The  sinus  is  situ- 
ated dorso-laterally  with  respect  to  the 
efferent  branchial  arteries  and  entirely  dor- 
sal to  the  branchial  skeleton. 

Just  as  the  anterior  cardinal  sinus,  at  its 
posterior  end,  curves  slightly  ventrad  to  join 
the  common  cardinal,  it  receives  the  inferior 
jufj,ular  vein.  This  vein,  which  enters  the 
sinus  from  the  ventral  side,  has  so  broad  a 
mouth  that  it  might  be  interpreted  as  open- 
ing partially  into  the  common  cardinal  it- 
self. The  inferior  jugular  originates  far 
anteriorly  behind  the  lower  jaw  (Pi.  5,  A ) . 
Although  its  main  branch  comes  from  within 
the  hyoid  "tongue"  which  protrudes  from 
the  floor  of  the  mouth,  branches  also  reach 
it  from  the  thyroid  gland,  the  ventro-medial 
fibers  of  the  ventral  constrictor  muscle,  and 
the  anterior  portion  of  the  coracomandibu- 
laris.  Veins  from  these  sources  were  actu- 
ally seen,  but  it  is  also  possible  that  there 
exist  venules  which  failed  to  be  injected, 
draining  all  the  tissues  located  posterior  to 
the  mid-ventral  portion  of  the  mandible. 

About  a  centimeter  behind  the  mandible, 
the  inferior  jugular  vein  turns  medially  and, 
running  dorsal  to  the  coracohyoideus  mus- 
cle, almost  meets  its  fellow  of  the  opposite 
side.  Without  actually  doing  so,  however, 
the  vein  turns  posteriorly  and  takes  a  path 
lateral  to  the  insertion  points  of  the  coraco- 
branchial  muscle  fibers  upon  the  branchial 
cartilages.  The  vein  follows  the  coraco- 
branchial  insertion  line,  flaring  widely  from 


148         BiiUctin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


the  ventral  midline  and  curving  dorsally  as 
it  does.  This  route  leads  the  inferior  jugular 
to  the  postero-ventral  corner  of  the  anterior 
cardinal  sinus  as  described  above.  In  its 
course  along  the  inserting  border  of  the 
coracobranchial,  it  receives  blood  from  the 
lateral  and  medial  sides  of  that  muscle- 
sheet. 

The  drainage  of  the  coracomandibularis 
and  coracohyoideus  muscles  is  only  partly 
accounted  for  by  the  inferior  jugular  vein. 
Although  no  other  veins  in  this  area  were 
injected,  dissections  suggest  that  there  may 
be  a  deep  vein  immediately  ventral  to  the 
ventral  aorta  which  provides  additional 
drainage  ( PL  4,  B ) .  From  it  blood  may  re- 
turn through  small  veins  in  the  dorsal  peri- 
cardial wall  to  the  common  cardinal  or 
possibly  over  a  more  ventral  course  to  a 
pair  of  veins,  to  be  described  below,  which 
run  through  a  channel  in  each  side  of  the 
pectoral  girdle. 

Farther  forward  than  the  entry-point  of 
the  inferior  jugular  the  anterior  cardinal 
sinus  receives  into  its  dorsal  side  the  po.s- 
terior  cerelyraJ  vein  (PI.  6,  A).  This  vessel 
collects  l)lood  from  fine  veins  over  the  cere- 
bellum and  from  membranes  in  the  dorsal 
part  of  the  cranial  cavity.  Since  there  is 
little  likelihood,  from  the  position  of  this 
vein,  that  it  returns  blood  from  any  part 
of  the  cerebrum  of  Chimacra,  the  use  of 
the  teiTH  "cerebral"  in  naming  the  vessel  is 
technically  incorrect.  The  adjective  has 
been  retained  merely  as  a  convenience  to 
indicate  that  this  vessel  is  the  posterior  of 
two  draining  the  brain  region.  There  is  a 
possibility  that  fine  veins  which  connect 
with  the  posterior  cerebral  may  also  con- 
nect with  the  orbital  sinus  via  an  anasto- 
mosing vein  that  passes  through  the  wall  of 
the  orbit  with  the  trochlear  nerve.  The 
existence  of  a  vein  traveling  with  that  nerve 
was  not  clearly  demonstrable,  however,  and 
so  is  best  left  in  question. 

The  posterior  cerebral  vein  is  formed  as 
a  median  vessel  in  the  dorsal  portion  of 
the  cranial  cavity  over  the  medulla  oblon- 
gata between  the  endolymphatic  ducts.    In 


addition  to  the  blood  from  the  brain  and 
associated  membranes,  the  posterior  cere- 
bral receives  tributaries  from  the  inner  ear 
of  each  side.  These  veins  pass  through  the 
wide  opening  by  which  the  cavity  of  the 
inner  ear  communicates  with  that  of  the 
brain  and  hence  do  not  pierce  cartilage. 
Immediately  posterior  to  the  point  of  its 
formation,  by  the  confluence  of  the  small 
vessels  described,  the  posterior  cerebral 
vein  widens,  over  the  rear  part  of  the  me- 
dulla, to  fonn  a  small  sinus.  Into  the  pos- 
terior end  of  this  sinus  run  several  little 
tributaries  carrying  blood  forward  from  the 
spinal  cord.  The  blood  collected  in  the 
sinus  leaves  it  through  two  veins  which  may 
be  considered  as  paired  posterior  cerebrals, 
continuing  from  the  median  vessel.  Each 
passes  directly  into  a  long,  ventrally  directed 
channel  in  the  cartilage  on  its  own  side  of 
the  chondrocranium.  Each  channel,  occu- 
pied solely  by  the  paired  portion  of  the 
posterior  cerebral  vein,  tenninates  by  pass- 
ing dorsal  to  the  vagus  nerve  (which  is 
also  traversing  the  cartilage  at  that  point) 
and  opening  ventro-laterally,  anterior  to 
the  foramen  of  the  latter.  The  posterior 
cerebral  vein  runs  forward  close  under  the 
otic  region  of  the  chondrocranium  and  then 
tiuns  laterally  at  the  level  of  the  posterior 
limit  of  the  semicircular  canals  to  join  the 
anterior  cardinal  sinus. 

At  the  anterior  end  of  the  anterior  cardi- 
nal sinus,  lies  the  opening  of  the  postorbital 
vein  (PI.  6,  A).  This  vein,  which  travels 
through  the  posterior  wall  of  the  subocular 
shelf,  in  a  ventral  direction,  with  the  hyo- 
mandibular  branch  of  the  seventh  nerve, 
forms  a  bridge  between  the  orbital  sinus 
and  the  anterior  cardinal.  As  it  enters  the 
latter,  dorsal  nutrient  veins  from  the  gill 
septa  were  seen,  in  one  specimen,  to  send 
a  common  stem  dorsally  to  this  blood  chan- 
nel. In  no  other  specimen  were  these  little 
veins  detected. 

The  postorbital  vein  was  examined  with 
care,  for  it  was  expected  that  the  hyoid 
sinus  should  open  into  it  or  nearby.  How- 
ever, no  evidence  was  found  in  any  speci- 


Morphology  and  Relationships  of  Holocephali  •  Stahl        149 


men  of  the  existence  of  a  shark-like  hyoid 
sinus.  There  was  only  a  small  vein,  the 
hyoid  tributary  (PI.  7,  B),  which  could  be 
traced  ventrally  to  the  dorsal  tip  of  the 
ceratohyal  cartilage  and  no  further  as  a  dis- 
sectable  vessel.  Posterior  to  the  ceratohyal 
cartilage  and  anterior  to  the  afferent  bran- 
chial artery,  in  the  position  of  the  selachian 
hyoid  sinus,  it  was  possible  to  trace  an  un- 
injected  vein  in  specimens  with  a  favorable 
distribution  of  agglutinated  blood.  Al- 
though a  connection  \\'ith  the  above-de- 
scribed small  vein  was  not  clearly  seen,  it 
is  possible  that  there  was  one  and  that  this 
entire  blood  pathway  is  homologous  to  the 
selachian  hyoid  vessel. 

In  each  dissection  of  the  region  ventro- 
anterior  to  the  postorbital  vein,  the  subocu- 
lar  shelf  and  the  cartilaginous  bar  which 
runs  to  the  mandibular  articulation  were 
removed  after  examination  of  the  bordering 
tissues.  Beneath  the  cartilage  and  immedi- 
ately dorsal  to  the  skin  of  the  roof  of  the 
mouth  was  a  layer  of  loose  connective  tis- 
sue. The  veins  running  through  it  were 
visible  only  when  they  remained  filled  with 
blood,  as  the  injection  mass  never  pene- 
trated to  them.  They  anastomosed  with 
each  other  and  one  ran  to  join  the  post- 
orbital  as  it  emerged  from  beneath  the  sub- 
ocular  shelf  (PI.  7,  B ) .  It  seems  possible 
that  the  vessel  traced  from  the  postorbital 
vein  to  the  tip  of  the  ceratohyal  may  have 
connections  with  the  veins  of  the  connec- 
tive tissue  layer  via  a  fine  vessel  which 
passes  forward,  dorsal  to  the  ceratohyal,  in 
company  with  the  efferent  pseudobranchial 
artery.  This  artery  passes  dorsally,  pierces 
the  chondrocranium,  and  splits  into  the 
cerebral  and  optic  arteries.  ( In  its  dorsal 
course  it  runs  along  the  posterior  edge  of 
the  lymphomyeloid  mass  [Kolmer,  1923] 
dorsal  to  the  skin  of  the  palate. )  Although 
it  is  difficult  to  discern,  it  is  probable  that 
a  vein  travels  with  the  artery.  Judging 
from  the  pathway  of  the  artery,  this  vein 
might  have  connections  to  the  network  of 
veins  in  the  connective  tissue  just  described 
and  to  the  orbital  sinus  as  well.    It  is  also 


possible  that  some  drainage  from  the  base 
of  the  brain  might  be  carried  to  the  orbital 
sinus  or  to  the  postorbital  vein  via  the  path- 
ways which  exist  through  the  connective 
tissue. 

The  orbital  sinus  receives  all  the  blood 
returning  from  the  head  except  that  which 
passes  through  the  subcutaneous  vessels, 
the  inferior  jugular,  and  the  posterior  cere- 
bral veins.  The  sinus  encircles  the  orbit 
medial  to  the  nerves  running  through  it. 
Intimately  connected  with  this  sinus  is 
lymphomyeloid  tissue.  This  tissue,  which 
seems  to  be  situated  in  the  lateral  edges  of 
the  blood-space,  is  present  in  such  quantity 
at  the  ventro-anterior  corner  of  the  orbit 
that  it  bulges  laterally  in  two  sizable  masses 
which  are  visible  as  soon  as  the  skin  is  re- 
moved from  that  area.  When  the  skin,  con- 
nective tissues,  and  mucous  canals^  are 
removed  from  the  head  in  front  of  and  be- 
low the  orbit,  the  largest  tributary  to  the 
orbital  sinus  can  be  seen.  This  vessel,  the 
maxillo-facial  vein,  coursing  dorsally  over 
the  posterior  palatoquadrate  region  and  up 
over  the  subocular  shelf,  enters  the  ventral 
side  of  the  orbital  sinus  anterior  to  the  point 
at  which  the  postorbital  vein  leaves  it  (Pi. 
6,  A ) .  As  the  maxillo-facial  vein  approaches 
the  sinus  it  assumes  a  position  medial  to 
the  nerves  which  run  out  of  the  orbit.  The 
vein  carries  blood  from  the  deep  portions 
of  the  overlying  mucous  canals,  from  the 
dermis  of  the  upper  lip  and  the  area  above 
it  lateral  to  the  labial  cartilages,  and  from 
the  muscles  of  the  facial  region  which  in- 
sert upon  the  labial  cartilages  and  the  lower 
jaw.  As  the  maxillo-facial  vein  approaches 
the  orbital  sinus,  it  is  joined  by  the  pre- 
orbitaJ  branch,  draining  the  muscle  tissue 
anterior   to   the    eye,   and   by   small   veins 


1  In  the  head  region  of  Chimaera  there  are  two 
groups  of  tubules  lying  under  the  skin  which  exude 
mucus  through  pores  at  their  posterior  ends.  The 
dorsal  set  of  six  parallel  tubules  lies  above  the  orbit 
and  extends  behind  it.  The  ventral  tubules,  ap- 
proximately the  same  in  number  and  arrangement, 
cover  an  area  of  the  face  ventral  to  the  eye  and 
anterior  to  it. 


150         Bulletin  Museum  of  Coinjxiidtive  Zoology,  Vol.  135,  No.  3 


which  come  from  tissues  lying  just  posterior  because  its  course  Hes  over  that  area.    Its 

to  the  maxillo-facial  vein  itself.  relation  to  the  suprapalatal  lymphomyeloid 

Since  the  maxillo-facial  vein  was  nearly  tissue  suggests  that  it  plays  a  part  in  drain- 
empty  of  blood  in  the  frozen-and-thawed  ing  it.  This  vessel  is  never  filled  with  the 
specimens  and  did  not  proxe  amenable  to  injection  mass.  Its  presence  is  demonstrable 
injection,  its  anatomy  was  studied  in  the  only  because  of  the  blood  left  in  it. 
fish  which  had  been  injected  with  latex  and  There  are  also  veins  which  enter  the 
preserved  immediately  in  formalin.  In  these  orbital  sinus  in  its  antero-dorsal  comer.  To 
animals  the  maxillo-facial  vein  was  ob-  reach  the  orbital  sinus  at  this  point,  the 
served  only  upon  the  left  side.  The  right  veins  must  traverse  the  posterior  part  of 
side  showed  what  appeared  to  be  a  large  the  ethmoid  canal.  The  canal  is  a  large, 
sub-surface  pool  of  agglutinated  blood —  cartilage-roofed,  median  space  dorsal  to 
surely  an  artifact.  Since  the  veins  on  the  the  portion  of  the  cranial  cavity  occupied 
left  were  entirely  empty,  it  is  probable  that  by  the  elongated  telencephalon.  It  is  sep- 
the  fresh-caught  fish  were  stored  right-side-  arated  from  the  brain  cavity  by  a  cartilag- 
downward,  causing  the  blood  to  accumu-  inous  partition.  The  ethmoid  canal,  which 
late  and  to  obliterate  the  vessels  on  that  is  filled  with  lymphomyeloid  material,  en- 
side.  Although  the  vessels  of  the  left  side  closes  the  ophthalmic  nerves  as  they  pass 
of  the  face  were  not  filled  with  latex,  it  was  from  the  orbital  region  towards  the  snout, 
possible  to  trace  them  by  injecting  poster-  One  of  the  veins  which  passes  through  the 
paint  into  the  orbital  sinus  and  expressing  ethmoid  canal  on  its  way  to  the  orbital  sinus 
it  into  the  facial  veins  by  pressing  gently  is  the  small  .superior  adductor  mandibidor 
upon  the  eye.  vein.    It  drains  the  most  dorsal  portion  of 

A  small  deep  labial  branch  of  the  maxillo-  the    deep    adductor   muscle.     Leaving   the 

facial  vein  brings  blood  from  the  lower  lip  muscle,  the  vein  passes  inward  through  the 

and  jaw,  the  upper  jaw  region  medial  to  wall   of  the   ethmoid   canal   and   joins   the 

the  labial  cartilages,  and  the  nasal  capsule  path   of  the   superficial   ophthalmic  trunk, 

(PI.  7,  A).  traveling  with  it  through  its  foramen  into 

There  are  two  other  veins  which  enter  the  orbit, 
the  orbital  sinus  from  the  ventral  side,  but  A  second  vein,  the  anterior  cerebral,  en- 
both  of  them  pierce  the  subocular  cartilage  ters  the  orbital  sinus  by  passing  through  a 
to  do  it.  The  orbito-na.sal  vein  passes  through  foramen  in  the  cartilage  between  the  back 
its  own  foramen.  As  it  travels  toward  the  of  the  ethmoid  canal  and  the  front  edge  of 
foramen  from  the  nasal  region,  it  lies  the  orbit  (PI.  5,  D).  The  foramen,  which 
against  the  dorsal  surface  of  the  supra-  is  medial  to  the  departure-point  of  the 
palatal  lymphomyeloid  mass.  Tracing  this  ophthalmic  profundus  from  the  orbit,  trans- 
vein  anteriorly,  one  finds  that  it  can  no  mits  only  this  vessel.  The  anterior  cerebral 
longer  be  separated  from  the  lymphomye-  vein  brings  blood  back  from  the  anterior 
loid  tissue  where  the  anterior  tip  of  the  end  of  the  brain.  It  is  formed  as  a  median 
mass  abuts  the  posterior  side  of  the  nasal  vessel  within  the  cranial  cavity  by  the  union 
capsule.  of  a  posterior  and  an  anterior  cerebral  trib- 

The  second  vein  which  reaches  the  or-  utary.  The  posterior  tributary  runs  from  the 
bital  sinus  by  piercing  the  subocular  shelf  tip  of  the  long  epiphysis  (which  extends 
has  already  been  mentioned.  This  is  the  forward  to  a  position  above  the  inter- 
vessel  which  accompanies  the  efferent  pseu-  orbital  area )  ventrad  in  a  course  which 
dobranchial  artery  along  the  posterior  edge  follows  the  curving  posterior  edge  of  the 
of  the  lymphomyeloid  mass  and  thence  interorbital  septum.  The  anterior  cerebral 
through  its  subocular  foramen.  The  vein  tributary,  which  drains  the  telencephalic 
has  been  called  the  posterior  pcdatal  vein  lobes,  follows  a  dorsal  pathway  posteriorly 


Morphology  and  Relationships  of  Holocephali  •  Stahl        151 


through  the  cranial  cavity  to  meet  the  pos- 
terior tributary  just  below  the  ventral  edge 
of  the  interorbital  septum.  The  anterior 
cerebral  vein  produced  by  the  union  of 
the  two  tributaries  passes  antero-dorsally 
through  the  edge  of  the  interorbital  parti- 
tion and  then  through  a  short  channel  in 
the  cartilage  to  enter  the  posterior  end  of 
the  ethmoid  canal.  There  it  bifurcates. 
Each  branch  turns  posteriorly  to  enter  the 
orbital  sinus  on  its  own  side.  Just  before 
it  leaves  the  ethmoidal  canal  each  portion 
of  the  bifurcated  anterior  cerebral  vein 
receives  an  ethmoidal  vein.  The  ethmoidal 
veins  bring  blood  back  through  the  ethmoid 
canal  from  the  most  rostral  part  of  the 
snout.  These  vessels  enter  the  anterior  end 
of  the  canal  through  the  same  pair  of 
foramina  through  which  the  superficial 
ophthalmic  nerves  issue. 

THE  DEEP  VEINS 
Postcardiac  Group 

Ventral  to  the  confluence  of  the  anterior 
cardinal,  lateral  cutaneous,  and  posterior 
cardinal  trunks,  there  is  an  opening  into 
the  posterior  side  of  the  common  cardinal 
vein  from  the  brachial  sinus.  That  sinus, 
which  receives  all  the  blood  returning  from 
the  pectoral  fin,  lies  behind  the  base  of  the 
fin  in  the  angle  between  it  and  the  body 
wall.  There  is  an  extension  of  the  sinus 
ventrally  along  the  posterior  side  of  the 
pectoral  girdle  which  meets  its  pair  in  the 
midline.  The  entrance  of  the  brachial 
sinus  into  the  common  cardinal  is  edged 
by  a  shaip  fold  which  acts  as  a  valve.  It 
is  this  valve,  apparently,  which  prevents 
good  injection  of  the  pectoral  veins. 

The  brachial  sinus  receives  blood  from 
two  sources.  The  larger  contributor  is  the 
posterior  brachial  vein.  It  borders  the 
posterior  edge  of  the  muscle  mass  of  the 
fin.  In  dorsal  view  it  can  be  seen  running 
along  the  posterior  side  of  a  deep  levator 
of  the  fin  (Pi.  5,  B).  Although  neither 
Vetter  (1878)  nor  Shann  (1919)  gives  a 
specific   name    to   this    muscle,    it    can   be 


recognized  easily  through  its  origin  from 
the  postero-medial  surface  of  the  scapula, 
its  strap-hke  shape,  and  its  insertion  upon 
the  metapterygial  cartilage.  The  brachial 
nerves  run  from  the  body  wall  through  the 
axial  region  toward  the  posterior  brachial 
vein.  Upon  reaching  it,  they  divide  into 
dorsal  and  ventral  branches,  the  former 
passing  over  the  vein  and  the  latter  under 
it.  The  posterior  brachial  vein  receives 
blood  from  the  deep  portions  of  the  posterior 
half  of  the  pectoral  fin. 

The  second  source  of  the  blood  collected 
by  each  brachial  sinus  is  the  anterior 
brachial  vein.  This  vein  emerges  from  a 
channel  in  the  cartilage  of  the  pectoral 
girdle  to  pour  its  contents  into  the  sinus. 
The  channel,  which  for  most  of  its  length 
contains  the  brachial  artery  as  well  as  the 
anterior  brachial  vein,  is  a  long  one, 
piercing  the  cartilage  in  the  coracoid  re- 
gion at  a  point  close  to  the  ventral  midline 
and  running  dorsally  through  the  girdle  to 
open  on  the  medial  edge  of  the  scapular 
process  near  the  brachial  sinus.  Between 
its  beginning  in  the  coracoid  area  and  its 
temiination  adjacent  to  the  sinus,  the  chan- 
nel opens  to  the  surface  twice  more:  there 
is  a  foramen  facing  ventro-laterally  anterior 
to  the  articulation  of  the  fin  and  another 
facing  posteriorly  dorsal  to  the  base  of  the 
fin.  Although  the  most  ventral  opening  of 
the  channel  is  sizable  and  set  in  the  anterior 
side  of  the  coracoid  bar,  no  veins  could  be 
seen  entering  it  from  the  coracomandibular 
muscle  which  originates  from  that  surface 
of  the  girdle.  Since  the  veins  draining  the 
muscle  fibers  in  that  area  remained  un- 
injected  in  every  specimen,  it  is  possible 
that  such  veins  do  exist  but  \\'ere  not 
observed. 

Between  the  entrance  to  the  channel  in 
the  coracoid  area  and  the  ventro-lateral 
foramen  mentioned  above,  the  channel  is 
filled  with  lymphomyeloid  tissue  like  that 
in  the  head  region.  If  the  passage  does 
carry  a  \'ein  from  the  area  of  the  hypo- 
branchial  musculature,  the  vessel  would 
undoubtedlv    have    connections    with    the 


152        Bullet  hi  Museum  of  Comparatwe  Zoology,  Vol.  135,  No.  3 


vascular  network  of  the  lymphomyeloid 
substance.  The  first  vein  which  appears 
certainly  in  the  channel,  however,  is  the 
tributary  draining  the  deep,  anterior  ven- 
tral part  of  the  fin.  This  vessel  enters  the 
passage  through  the  ventro-lateral  opening 
and  follows  the  path  of  the  channel  dorsad. 
A  tributary  from  the  deep  part  of  the  an- 
terior dorsal  half  of  the  fin  enters  the  chan- 
nel next,  through  the  posterior  foramen, 
and  merges  with  the  tributary  from  the 
ventral  part  of  the  fin  to  form  the  anterior 
brachial  vein.  It  is  this  vein  which  leaves 
the  channel  at  its  dorsal  termination  to 
enter  the  brachial  sinus. 

Of  the  major  trunks  which  empty  into 
the  common  cardinal  vein,  the  only  one 
which  remains  to  be  described  is  the  pos- 
terior cardinal  .sinii.^.  Although  this  vessel 
is  paired,  there  are  numerous,  sizable  com- 
munications between  the  left  and  right 
sides,  and  posteriorly,  at  the  origin  of  the 
trunk  between  the  kidneys,  there  is  a  single 
median  portion.  There  are  four  constant 
features  concerning  the  anterior  portion  of 
this  sinus  which  should  be  noted.  Firstly, 
the  entrance  into  the  common  cardinal  of 
each  side  is  cavernous.  An  injection  mass 
introduced  into  the  lateral  cutaneous  vein 
always  descends  and  turns  posteriorly  into 
the  posterior  cardinal  sinus  rather  than  en- 
tering the  smaller  opening  of  the  anterior 
cardinal.  Just  as  the  sinus  approaches  the 
common  cardinal,  the  subclavian  artery  and 
two  spinal  nerve  branches  cross  through  it. 
Secondly,  there  seems  to  be  a  connection 
between  the  posterior  cardinal  and  the 
brachial  sinus.  The  connecting  passage 
runs  from  the  ventro-lateral  edge  of  the 
posterior  cardinal  to  the  brachial  sinus 
dorsal  and  posterior  to  its  opening  into  the 
common  cardinal.  Thirdly,  the  left  and 
right  posterior  cardinal  sinuses  extend 
ventro-laterally  to  meet  each  other  in  the 
ventral  midline.  This  midline  communica- 
tion parallels  that  of  the  brachial  sinuses 
and  is  separated  from  it  by  a  sheet  of  con- 
nective tissue.  It  is  to  this  part  of  the 
posterior  cardinal   sinus   that  the  left   and 


right  ventro-anterior  parietal  veins  bring 
blood  from  the  deep  anterior  ventral  and 
anterior  ventro-lateral  axial  musculature. 
Fourthly,  the  sinus  of  each  side  extends 
dorso-medially  as  a  blind  pouch  forward 
of  its  point  of  union  with  the  anterior 
cardinal  sinus.  Thus,  a  cross-section  made 
just  in  front  of  the  anterior  edge  of  the 
scapula  shows  the  left  and  right  pouches 
close  to  the  midline  above  the  branchial 
region  and  the  anterior  cardinal  sinus  of 
each  side  lying  in  a  more  ventro-lateral 
position. 

Since  the  posterior  cardinal  sinus  runs 
retroperitoneally  against  the  dorsal  body 
wall  between  the  dorsal  aorta  and  the  more 
laterally  placed  kidney,  the  veins  from  the 
deep  epaxial  muscles  surely  empty  into  it. 
These  fine  deep  epaxial  veins  were  not 
injected  and  so  remained  invisible,  but 
their  presence  may  be  predicated  with 
safety. 

Besides  this  drainage  and  that  from 
superficial  regions  of  the  dorsal  muscula- 
ture via  the  subcutaneous  svstem,  there  is 
one  other  route  to  be  mentioned:  between 
the  left  and  right  epaxial  muscle  groups 
in  the  trunk  region  can  be  found  a  median 
dorsal  vein  (PI.  6,  B).  This  vessel  was 
injected  successfully  and  seen  to  collect 
from  the  most  dorsal  parts  of  the  muscula- 
ture. Anastomoses  existed  between  its  trib- 
utary veins  and  those  of  the  subcutaneous 
system.  The  median  dorsal  vein  carries  its 
blood  forward  to  a  median  dor.ml  fin  sinus 
set  behind  the  base  of  the  dorsal  fin.  This 
sinus  has  a  single  anterior  opening  on  each 
side  through  which  blood  leaves  it.  These 
openings  lead  to  the  posterior  cardinal 
sinuses.  The  blood  returned  from  the 
median  sinus  by  this  route  enters  the  pos- 
terior cardinal  far  forward,  flowing  into 
the  blind  pouch  which  extends  anteriorly 
and  dorsomedially  into  the  anterior  tip  of 
the  coelom. 

Posterior  to  the  opening  of  this  sinus,  on 
a  line  with  the  base  of  the  dorsal  spine,  an 
anterior  epaxial  vein  enters  the  posterior 
cardinal  on  each  side,  bearing  blood  from 


Morphology  and  Relationships  of   Holocephali  •   Stahl 


153 


deep  muscles  forward  of  the  dorsal  fin. 
Medial  to  its  point  of  entry  is  the  cartilage 
plate  which  supports  the  dorsal  spine. 
Again.st  the  side  of  this  plate  runs  a  vessel, 
the  spino-bosal  vein,  which  connects  the 
subcutaneous  veins  at  the  base  of  the  fin- 
spine  with  the  posterior  cardinal  sinus  deep 
below. 

The  remaining  tributaries  to  the  posterior 
cardinal  enter  it  more  ventrally.  There  are 
several  which  come  from  the  esophageal 
wall,  leaving  it  as  the  gut  tulie  makes  its 
entry  into  the  anterior  end  of  the  body 
cavity.  In  the  region  of  tlie  trunk  anterior 
to  the  kidneys,  anterior  parietal  veins  on 
each  side  contribute  l)lood  from  the  most 
dorsal  portions  of  the  hypaxial  musculature. 
The  gonads  and  the  ducts  of  the  reproduc- 
tive tract  send  their  blood  to  the  posterior 
cardinal,  too.  The  blood  from  these  struc- 
tures seems  to  collect  in  sinuses  between 
the  double  walls  of  the  suspending  dorsal 
mesentery.  The  sinus  parallelling  the  ovi- 
duct of  the  mature  female  is  quite  spacious. 
There  are  veins  running  medially  from  it 
over  the  short  distance  to  the  posterior 
cardinal  sinus.  Around  the  anterior  end  of 
the  functional  kidne\  there  is  a  wider  com- 
munication between  the  two  sinuses.  The 
male  fishes  available  for  dissection  were 
small  and  apparently  not  fully  mature.  The 
vas  deferens  was  very  fine  and  bound 
closely  to  the  lateral  border  of  the  gland 
of  Leydig  (the  transformed  anterior  end 
of  the  kidney ) .  The  venous  drainage  of  the 
duct  was  invisible. 

The  renal  veins  run  their  usual  short 
course,  leaving  the  kidneys  ventro-medially 
and  entering  the  posterior  cardinal  sinus. 
Posterior  to  the  entrance  of  all  but  a  few  of 
the  renal  veins  the  posterior  cardinal  sinus 
in  Chimaera  receives  a  vein  which  does  not 
usually  empty  into  it  in  cartilaginous  fishes. 
That  vessel,  the  large,  firm-walled  femoral 
vein  (PI.  7,  C),  is  situated  posterior  to  the 
femoral  artery  and,  at  the  fin-base,  is 
formed  from  dorsal  and  ventral  fin  tribu- 
taries, branches  which  come  from  the  deep 
dorsal  and  ventral  surfaces  of  the  fin.    Al- 


though the  femoral  vein  is  strong  enough  to 
withstand  the  injecting  process,  it  proved 
impractical  to  use  it  as  a  route  for  injecting 
the  deep  veins  of  the  pelvic  fin  because 
of  a  valve  at  its  distal  end  which  prevents 
backflow. 

A  short  distance  proximal  to  the  fin-loase, 
the  femoral  vein  receives  the  rectal  trib- 
utary. This  vessel  collects  blood  from  a 
capillary  network  in  the  wall  of  the  rectum 
and  also,  in  the  female,  from  the  problem- 
atical glandular  "seminal  receptacle." 

As  the  femoral  vein  runs  toward  the  pos- 
terior cardinal  sinus  it  passes  the  lateral 
edge  of  the  kidney  and  turns  ventral  to  it. 
There  is,  however,  a  small  branch  which 
leaves  the  femoral,  passes  over  the  lateral 
edge  of  the  kidney  to  the  dorsal  side  and 
connects  with  the  renal  portal  vein.  The 
connection,  the  iliac  vein  (PI.  3),  is  a 
delicate  one  and  possibly  not  uniformly 
present.  Because  of  the  small  size  of  the 
vessel  its  functional  significance  is  doubtful. 

The  last  contributor  to  the  posterior  car- 
dinal sinus  is  a  strange  one  for  any  verte- 
brate. Without  a  doubt,  the  hepatic  veins 
( PI.  9,  A  and  B )  empty  into  this  dorsal 
channel  instead  of  passing  forward  through 
the  transverse  septum  to  enter  the  sinus 
venosus.  Inspection  of  the  posterior  wall 
of  the  sinus  venosus  showed  a  pair  of  open- 
ings so  reduced  as  to  be  incapable  of  carry- 
ing the  entire  bloocbflow  from  the  liver. 
When  the  liver  was  cut  transversely  through 
the  small  area  which  is  bound  to  the  back 
of  the  transverse  septum,  there  were  no 
sinuses  to  be  seen  in  the  interior  of  the 
tissue.  In  Chimaera,  however,  the  anterior 
attachment  of  the  liver  to  the  back  of  the 
transverse  septum  is  not  the  only  bridge 
between  that  gland  and  surrounding  tissues 
across  which  a  hepatic  vein  might  travel. 
The  front  portion  of  each  lobe  is  attached 
dorsally  to  the  underside  of  the  posterior 
cardinal  sinus  and  thus  to  the  roof  of  the 
coelom.  The  line  of  attachment  extends 
along  the  dorsal  edge  of  the  right  lobe  of 
the  liver  for  about  a  third  of  its  length  and 
along  that  of  the  shorter  left  lobe  for  half 


154         BuUeiin  Mu.scmn  of  Comparative  Zoology,  Vol.  135,  No.  3 


of  its  length.  When  these  areas  of  at- 
tachment were  explored,  it  was  found 
that  hepatic  veins  of  varying  sizes  passed 
through  them,  carrying  blood  from  the  liver 
directly  to  the  posterior  cardinal  sinuses. 
The  largest  of  these  veins  was  one  which 
drained  the  posterior  two-thirds  of  the 
right  lobe  and  entered  the  right  posterior 
cardinal  at  a  point  immediately  anterior 
to  a  mesentery  strap  extending  from  the 
midline,  between  the  sinuses,  to  the  tip 
of  the  pancreas.  If  the  sinus  wall  is  cut 
and  deflected  at  that  point,  the  dissector 
can  look  through  a  fenestrated  membrane 
separating  the  sinus  from  the  interior  of 
the  hepatic  vein.  This  is  the  largest  of  all 
the  hepatic  veins  and  drains  the  entire  free 
end  of  the  large  right  lobe  of  the  liver.  For 
the  most  part,  the  branches  feeding  this 
hepatic  vein  lie  dorsal  to  those  from  the 
hepatic  portal  vein  which  carry  blood  into 
the  right  lobe.  Anterior  to  the  point  at 
which  this  large  hepatic  vein  enters  the 
posterior  cardinal  sinus,  smaller  hepatic 
veins  enter  the  sinus  separately.  Although 
the  left  lobe  of  the  liver  is  smaller,  it  is 
similarly  drained.  There  are  two  main 
hepatic  veins  that  leave  it  to  enter  the 
posterior  cardinal  sinus  on  the  left  side. 

To  conclude  the  description  of  the  sys- 
temic venous  drainage,  one  turns  to  the 
rcnml  portal  veins  (Pi.  8,  A).  The  blood 
entering  the  kidneys  from  the  renal  portal 
veins  comes  largely  from  the  deep  portions 
of  the  axial  musculature.  The  muscles  of 
the  tail  region  send  ])lood  to  the  caudal 
vein  which  bifurcates  behind  the  body 
cavity  to  become  the  left  and  right  renal 
portal  veins.  These  run  forward  along  the 
dorsal  surface  of  each  kidney,  medial  to 
the  mesonephric  duct.  At  segmental  inter- 
vals the  renal  portals  receive  veins  from 
the  body  wall.  The  renal  portals  extend 
far  forward,  even  beyond  the  region  of  the 
functional  kidney,  collecting  the  segmental 
parietal  veins  and  finally  dwindling  to 
nothing  about  a  centimeter  behind  the  pec- 
toral fin.  At  the  level  of  the  anterior  edge 
of  the  pelvic  girdle  the  renal  portal  receives 


the  ventro-posterior  parietal  vein.  This  ves- 
sel can  be  seen  through  the  peritoneum, 
lying  parallel  to  the  long  axis  of  the  body. 
It  begins  in  the  middle  region  of  the  trunk 
and  carries  blood  posteriorly  from  the  deep 
hypaxial  muscles.  Immediately  anterior  to 
the  pelvic  girdle  it  curves  dorsally  to  join 
the  renal  portal.  The  only  tributaries  to  the 
renal  portal  which  do  not  return  blood  from 
axial  musculature  are  the  small  veins  from 
the  posterior  end  of  the  mesovarium.  For 
a  short  distance  beyond  the  end  of  the 
oviducal  sinus  in  the  crowded  posterior  part 
of  the  body  cavity,  these  vessels  from  the 
lower  end  of  the  oviduct  enter  segmental 
parietal  veins  as  they  are  about  to  join  the 
renal  portal. 

The   Hepatic   Portal   System 

Since  the  nature  of  the  digestive  tract 
and  the  arrangement  in  the  body  cavity 
of  the  associated  glands  are  distinctive  in 
chimaerid  fishes,  the  pattern  of  the  veins 
draining  the  system  is  also  singular.  Before 
trying  to  visualize  the  path  of  the  vessels, 
one  must  understand  several  anatomical 
points.  Firstly,  there  is  no  stomach  in  these 
animals.  Instead  of  a  long  J-shaped  struc- 
ture there  is  a  short  continuation  of  the 
esophagus  which  reaches  the  beginning  of 
the  spiral  intestine.  Secondly,  the  spleen 
is  not  attached  to  the  gut  tube  by  mesen- 
teries. The  mesenteries  are  exceedingly 
reduced  and  the  spleen  is  fairly  free,  bound 
only  to  the  posterior  end  of  the  pancreas. 
When  a  fresh  fish  is  opened,  the  pancreas 
and  spleen  appear  to  lie  ventrally  in  the 
body  cavity.  Because  the  spleen  has  no 
relation  to  a  stomach,  the  term  "gastro- 
splenic"  or  "lieno-gastric"  is  not  applicable 
to  any  vessel  in  the  hepatic  portal  system. 
It  is  well  to  keep  these  facts  in  mind  when 
pondering  possible  homologies  between  the 
vessels  of  chimaerid  fishes  and  any  others. 

The  he]Hitie  portal  vein  is  formed  against 
the  surface  of  the  liver  posterior  to  the  base 
of  the  gall  bladder  by  the  confluence  of 
the  intra-intestinal,  the  anterior  ventral  in- 
testinal, and  the  mesenteric  veins  (Pi.  (S,  B). 


Morphology  and  Relationships  of  Holocephali  •  Siahl        155 


It  runs  immediately  into  the  right  lobe  of 
the  liver  in  one  direction  and,  in  the  other, 
sends  a  large  division  along  the  posterior 
edge  of  the  midventral  portion  of  the  liver. 
The  blood  carried  in  this  vessel  is  distrib- 
uted to  the  left  lobe  of  the  liver. 

The  first  of  the  vessels  which  deliver 
blood  to  the  hepatic  portal,  the  intra- 
intestinol  vein,  drains  the  spiral  valve  and, 
as  it  emerges  from  the  intestine  wall,  re- 
ceives anterior  dorsal  intestinal  trilmtaries. 
These  vessels  collect  blood  not  only  from 
the  anterior  wall  of  the  spiral  intestine  but 
also  from  the  posterior  region  of  the  esoph- 
agus which  connects  with  it.  These  trib- 
utaries anastomose  \\'ith  others  which  con- 
verge to  form  the  anterior  ventral  intestinal 
vein  (PI.  10,  A).  The  intra-intestinal  and 
the  anterior  ventral  intestinal  leave  the 
intestine  wall  from  points  lying  close 
against  opposite  sides  of  the  bile  duct. 
Running  closely  apposed  to  the  duct,  these 
vessels  finally  reach  the  hepatic  portal  vein. 

The  last  of  the  vessels  which  contribute 
blood  to  the  hepatic  portal,  the  mesenteric 
vein,  is  fonned  by  the  confluence  of  the 
two  posterior  intestinal  veins.  The  pos- 
terior dorsal  intestinal  vein,  which  drains 
the  lower  end  of  the  intestine  as  far  as 
the  beginning  of  the  rectum,  receives  trib- 
utaries which  can  be  seen  on  the  surface 
of  the  intestine  wall.  These  tributaries 
anastomose  with  others  which  converge  on 
the  opposite  side  of  the  intestine  to  form 
the  posterior  ventral  intestinal  vein  ( PI.  10, 
A).  Both  posterior  intestinal  veins  leave 
the  surface  of  the  intestine  to  run  free  to 
a  position  against  the  side  of  the  pancreas 
where  they  merge  and  are  bound  down. 
In  their  free  portions,  the  dorsal  and  ventral 
veins  have  a  different  appearance:  the 
ventral  one  is  narro\\'er  and  runs  through  a 
band  of  mesentery;  the  dorsal  one  is  very 
wide  in  diameter  and  absolutelv  uncon- 
fined.  At  the  point  at  which  they  merge 
and  are  tied  to  the  pancreas,  there  is  a 
thin  mesentery  strap  which  leaves  to  reach 
the  dorsal  midline  behind  the  dorsal  attach- 
ment of  the  liver. 


The  mesenteric  vein,  thus  formed,  re- 
ceives several  auxiliary  splenic  veins  (the 
spleen  is  tied  to  the  posterior  end  of  the 
pancreas  just  behind  the  point  where  the 
two  intestinals  reach  it)  and  then  receives 
the  relatively  large  lieno-pancreatic  vein. 
The  latter  vessel  travels  through  the  length 
of  the  spleen  and  the  portion  of  the  pan- 
creas which  lies  posterior  to  the  origin  of 
the  mesenteric.  After  receiving  the  lieno- 
pancreatic  vein,  the  mesenteric  turns  an- 
teriorly and  runs  in  company  with  the 
pancreatico-splenic  artery,  collecting  from 
auxiliary  pancreatic  veins  in  its  course.  The 
mesenteric  enters  the  hepatic  portal  in  con- 
junction with  the  intra-intestinal  vein  (Pi. 
10,  B). 

Additional   Observations 

One  specimen  of  Callorhynchus,  a  small 
female,  was  examined  after  the  dissections 
of  Chimaera  colliei  were  completed.  It  was 
found  that  the  pattern  of  the  confluence  of 
the  major  venous  trunks  to  form  the  com- 
mon cardinal  agreed  with  the  findings  in 
Chimaera.  The  hepatic  veins  were  also 
found  to  enter  the  posterior  cardinal  sinus. 
An  inspection  of  the  posterior  wall  of  the 
sinus  venosus  showed  extremely  small  aper- 
tures that  were  similar  to  the  reduced 
hepatic  openings  in  Chimaera. 

DISCUSSION  OF  THE  SIGNIFICANCE 
OF  THE  CIRCULATORY  SYSTEM  TO 
THE    PROBLEM   OF    HOLOCEPHALIAN 
RELATIONSHIPS 

The  venous  system  of  the  chimaerids 
bears  a  greater  resemblance  to  that  of 
sharks  both  in  the  structure  of  its  vessels 
and  their  arrangement  than  it  does  to  the 
system  of  any  other  group  of  extant  fishes. 
There  are  certain  deviations  from  the 
selachian  plan,  however,  which  are  cer- 
tainly clues  to  the  separate  evolution  of 
the  holocephalian  line.  A  consideration  of 
the  significance  of  the  venous  system  to 
the  question  of  holocephalian  relationships 
necessitates  first,  recognition  of  the  resem- 


156         Bulletin  Museum  of  Conjparativr  Zoology,  Vol.  135,  No.  3 


blances,  and  then  evaluation  of  the  differ- 
ences which  exist. 

The  veins  in  both  groups  of  animals,  like 
those  of  all  fishes,  possess  little  muscular 
tissue.  Their  walls  are  therefore  exceed- 
ingl\  delicate  and  difficult  to  differentiate 
from  connective  tissue  in  gross  dissection. 
Even  the  main  vessels  may  be  opened  by 
a  chance  touch  of  a  scalpel  tip.  It  is  the 
combination  of  fragile  walls  and  the  pres- 
ence of  valves  which  gives  rise  in  both 
holocephalians  and  selachians  to  the  diffi- 
culties experienced  in  injection  procedures. 
The  lack  of  detailed  description  of  the 
drainage  of  the  head  region  stems  directly 
from  the  inability  of  investigators  to  intro- 
duce substances  into  the  veins  which  empty 
into  the  orbital  sinuses.  The  valves,  which 
are  mere  folds  of  the  lining  of  the  vein 
wall,  seem  to  have  a  like  distribution  in 
chimaerids  and  sharks  if  non-penetration 
of  injection  media  can  be  taken  as  a  guide 
to  their  location.  The  dissector  can  see 
that  the  entrance  into  the  common  cardinal 
from  the  anterior  cardinal  is  valved  but 
that  the  opening  from  the  posterior  cardinal 
is  not.  This  arrangement  is  also  shown  for 
Hcptanchus  by  Daniel   (1934). 

The  main  venous  channels  in  holoce- 
phalians as  in  sharks  are  sinuses.  In  both 
types  of  fishes  the  largest  ones  are  held 
together  from  within  by  a  network  of  con- 
nective tissue  trabeculae.  The  position  of 
these  sinuses  relative  to  each  other  is  not 
distinctive  in  the  Holocephali. 

A  great  part  of  the  basic  arrangement  of 
the  venous  system  of  selachians  is  dupli- 
cated in  Chimaera  colliei.  The  orbital  sinus 
is  the  major  collecting  point  for  blood  re- 
tinning  from  the  tissues  of  the  head.  As 
in  Ueptanchus,  Mu.stcJus,  and  ScylUtim,  it 
receives  the  orbito-nasal  and  anterior  cere- 
bral veins.  Although  Daniel  (1934),  Parker 
(18S6),  and  O'Donoghue  (1914),  who  in- 
vestigated the  three  sharks  named,  respec- 
tively, do  not  describe  in  detail  the  specific 
structures  drained  by  the  orbito-nasal  vein, 
the  vessel  in  Chhiuwra  is  probably  exactly 
comparable  except   that   it   receives   blood 


from  the  palatal  lymphomyeloid  mass 
which  the  sharks  do  not  possess.  The  an- 
terior cerebral  vein  of  Chimaera  drains  the 
same  regions  of  the  brain  as  the  selachian 
vessel,  although  there  is  no  reception  of 
an  ethmoidal  vein  in  sharks,  which  lack  an 
ethmoidal  canal.  The  dissection  of  Chi- 
maera revealed  a  maxillo-facial  vein  and 
two  smaller  vessels  which  also  empty  into 
the  orbital  sinus.  That  there  are  no  com- 
parable vessels  shown  for  sharks  is  prob- 
ably due  to  their  having  been  uninjected 
and  unreported  rather  than  to  their  absence. 

A  postorbital  channel  exists  in  both  holo- 
cephalians and  sharks  to  carry  blood  from 
the  orbital  sinus  to  the  anterior  cardinal. 
Although  its  location  relative  to  the  two 
blood  spaces  it  connects  is  the  same  in  both 
types  of  fishes,  its  associations  with  skel- 
etal and  nervous  elements  are  not  identical. 
In  Chimaera  the  postorbital  vein  passes 
through  a  foramen  in  the  posterior  comer 
of  the  orbit  accompanied  by  the  hyoman- 
dibular  branch  of  the  seventh  nerve.  In 
sharks,  the  vessel  merely  lies  in  a  post- 
orbital groove  and  the  hyomandibular 
nerve,  which  never  enters  the  orbit,  does 
not  share  this  anterior  pathway.  It  seems 
likely  that  the  postorbital  veins  of  the  two 
types  of  fish  are  homologous  and  that  the 
different  nervous  and  skeletal  arrangements 
are  due  to  the  autostyly  and  forward  com- 
pression of  the  cephalic  structures  in  Chi- 
maera. The  presence  of  a  foramen  rather 
than  a  groove,  for  example,  is  due  to  the 
fusion  with  the  cranium  proper  of  an  otic 
process  extending  from  the  jaw  joint  to  the 
ear  region.  This  cartilage  provides  the  en- 
tire lateral  wall  of  the  foramen.  The  for- 
ward course  of  the  hyomandibular  nerve 
can  be  explained  by  the  anterior  displace- 
ment of  the  tissues  it  serves  and  by  the 
absence  of  a  spiracle  and  hyomandibular 
cartilage  behind  which  it  \\'Ould  normally 
pass. 

The  chimaerid  anterior  cardinal  sinus 
carries  blood  over  the  gill  region  to  the 
common  cardinal  in  the  shark-like  manner, 
receiving  in  its  course  the  posterior  cerebral 


Morphology  and  Relationships  of  Holocephali  •  Stahl        157 


vein.  The  way  in  which  the  anterior  car- 
dinal and  the  other  large  veins  of  sharks 
meet  to  fomi  the  common  cardinal  vessel 
shows  a  degree  of  variation  sufficient  to 
preclude  the  interpretation  of  the  holo- 
cephalian  arrangement  as  worthy  of  special 
note.  Even  the  connection  of  the  lateral 
cutaneous  vein  via  the  subscapular  sinus 
to  this  confluence  of  vessels  has  its  proto- 
type amongst  the  selachians. 

In  the  postcardiac  region  the  similarity 
of  the  selachian  and  holocephalian  plan  is 
still  evident.  The  renal  portal  and  posterior 
cardinal  vessels  are  substantially  the  same 
in  both  groups.  Although  the  hepatic  portal 
system  will  receive  special  consideration 
below,  it  may  also  be  generally  described 
as  more  like  that  of  sharks  than  that  of 
other  fishes.  It  is  in  the  drainage  of  the 
pelvic  fins,  the  ventral  body  wall,  and  the 
liver  that  significant  differences  do  appear 
in  the  chimaerids. 

The  possession  of  a  subcutaneous  system 
of  veins  which  run  unaccompanied  by 
arteries  is  a  final  point  of  likeness  which 
should  be  mentioned.  In  sharks  as  well 
as  chimaerids,  the  chief  vessels  in  the  net- 
work, the  lateral  cutaneous  veins,  receive 
segmental  tributaries  and  finally  lead  to 
the  subscapular  sinuses.  There  are  connec- 
tions elsewhere  with  deeper  vessels  in  the 
region  of  the  dorsal  and  pelvic  fins. 

That  the  similarities  between  the  venous 
systems  of  holocephalians  and  selachians 
do  signify  an  evolutionary  relationship  is 
strongly  suggested  by  the  fact  that  their 
common  pattern  sets  them  off  distinctly 
from  the  bony  fishes.  Neither  ray-finned 
forms  nor  lungfishes  show  the  development 
of  spacious  sinuses.  Allen's  (1905)  excel- 
lent description  and  beautiful  drawings  of 
the  circulatory  system  of  the  teleost,  Ophio- 
don,  reveal  a  complex  arrangement  of  veins 
of  small  bore  whose  homology  to  vessels  in 
cartilaginous  fishes  would  be  difficult  to 
prove.  In  Ophiodon,  fine  facial  vessels 
empty  into  a  pair  of  jugular  veins  directly, 
since  there  are  no  orbital  sinuses.  The 
jugular  veins,  which  carry  blood  posteriorly 


over  the  gill  region,  are  thought  by  Van 
Gelderen  (193S)  not  to  be  homologous  to 
the  anterior  cardinals  in  the  Chondrich- 
thyes.  The  posterior  cardinal  vessels  of 
Ophiodon,  like  those  of  cartilaginous  fishes, 
receive  blood  from  the  renal  capillaries, 
but  in  other  bony  fishes  there  may  be 
direct  connections  with  the  caudal  vein 
which  never  persist  in  sharks  or  chimaerids. 
No  actinopterygians  that  have  been  investi- 
gated have  subcutaneous  veins  except  the 
Thunnidae  (and  here  the  veins  are  ac- 
companied by  arteries).  As  an  adjunct  to 
the  venous  system,  teleosts  have  a  well- 
developed  set  of  lymphatics  which  all 
cartilaginous  fishes  (except  possibly  Tor- 
pedo) lack.  A  glance  at  the  pattern  of 
veins  in  dipnoans  shows  an  even  greater 
departure  from  the  arrangement  found  in 
selachians  and  holocephalians,  since  there 
appear  in  the  lungfishes  vessels  which  are 
similar  to  those  of  amphibians. 

Despite  the  broad  similarities  between 
the  selachian  and  holocephalian  venous 
systems  there  are  points  of  apparent  dif- 
ference whose  significance  must  be  con- 
sidered. A  dissection  of  the  hypobranchial 
region,  for  instance,  does  not  reveal  a  pair 
of  easily  recognizable,  shark-like  inferior 
jugular  veins.  The  area  is  laterally  com- 
pressed and  occupied  by  the  massively 
developed  hypobranchial  muscles  whose 
arrangement  leaves  no  straight  channel  for 
these  veins  to  follow.  Beneath  the  muscles, 
and  ventral  to  the  ventral  aorta,  there  exists 
a  mass  of  loose  connective  tissue  which 
probably  does  have  fine  veins  draining 
l^lood  posteriorly  toward  the  sinus  venosus. 
Allis,  who  made  a  preliminary  sketch  of 
this  area  in  1916,-  drew  a  pair  of  veins  here 
which  he  interpreted  as  inferior  jugulars. 
Not  finding  a  hyoid  sinus  in  the  usual  loca- 
tion, he  ga\e  that  name  to  a  large  vein 
which    begins    behind    the    mandible    and 


-  This  and  other  unpubHshed  sketches  of  the 
anatomy  of  Chimaera  coUiei  were  given  by  the  son 
of  the  hite  E.  P.  AHis  to  Dr.  A.  S.  Romer  and  are 
in  his  hbrary  in  the  Museum  of  Comparative  Zool- 
ogy at  Harvard  University. 


158        Bulletin  Museitm  of  Comparative  Zoology,  Vol.  135,  No.  3 


curves  dorsally  to  enter  the  back  corner  of 
the  anterior  cardinal  sinus.  It  seems  more 
Hkely,  ho\\'e\'er,  that  any  veins  lying  deep 
and  near  the  midline  represent  auxiliary 
drainage  and  that  the  more  lateral,  dorsally 
cur\ing  vessel  is  the  inferior  jugular  vein  of 
CJumoera.  Although  the  curious  position  of 
this  vein  suggests  that  it  might  be  a  dif- 
ferent vessel  from  that  of  sharks,  its  rela- 
tion to  the  base  of  the  gill  arches  is  quite 
similar  and  its  entry  point  into  the  anterior 
cardinal  sinus  near  the  union  with  the  com- 
mon cardinal  is  not  very  different  from 
that  shown  by  Parker  ( 1886 )  for  Mustchis. 
It  seems  that  the  inferior  jugular  vein  has 
been  shifted  dorsally  as  the  head  became 
laterally  compressed,  and  is  not  so  different 
after  all  from  that  of  sharks.  It  contrasts 
markedly  with  the  inferior  jugular  of  tele- 
osts  which  is  squeezed  medially  and  in 
some  forms,  like  Ophiodon,  exists  as  a 
single  median  ventral  vessel  for  most  of 
its  length. 

The  absence  of  a  hyoid  sinus  in  Chimacra 
may  also  be  due  to  the  relative  reduction 
of  the  branchial  apparatus.  In  contrast  to 
the  large  channel  which  connects  the  an- 
terior cardinal  sinus  with  the  inferior  jug- 
ular vein  in  selachians,  no  major  vein 
appears  on  either  side  of  the  ceratohyal 
cartilage.  Examination  with  a  dissecting 
scope  showed  in  two  specimens  a  fine 
vessel  which  lay  anterior  to  the  afferent 
branchial  artery,  but  the  vein  was  more 
easily  traced  along  a  pathway  with  the 
artery  out  upon  the  opercular  flap  than  to 
a  connection  with  the  postorbital  or  an- 
terior cardinal  channels.  If  it  is  correct 
to  assume  that  main  vessels  develop  in  the 
embryo  as  the  result  of  dominance  of  par- 
ticular pathways  through  the  initial  capil- 
lary net,  then  it  is  not  surprising  to  find 
that  chimaerids  lack  a  shark-like  hyoid 
sinus.  In  sharks,  where  the  epihyal  cartilage 
enlarges  and  develops  a  close  association 
with  the  cranium,  it  is  possible  that  a  vein 
would  appear  behind  it  with  a  connection 
to  the  lateral  head  vein  above.  In  Chi- 
maera,    where    the    epihyal   never   departs 


from  its  serial  alignment  with  other  parts 
of  the  visceral  skeleton,  the  absence  of  a 
special  hyoidean  vein  or  sinus  might  be 
expected.  If  one  follows  this  inteipretation 
and  accepts  this  assumption  that  the  non- 
suspensory  hyoid  is  primitive,  it  follows 
logically  that  the  absence  of  a  hyoid  sinus 
may  represent  an  original  character  rather 
than  a  secondary  loss.  If  the  holocephalians 
sprang  from  an  ancestral  stock  which 
lacked  a  suspensory  hyoid  and  a  distinctive 
hyoid  vein  and  then  evolved  the  laterally 
compressed,  short  head  of  the  extant  ani- 
mals, neither  the  space  nor  the  stimulus 
for  the  development  of  a  hyoid  sinus  would 
have  existed. 

In  pointing  out  singularities  in  the  pre- 
cardiac drainage  of  Chimacra,  one  should 
not  omit  mention  of  the  situation  of  the 
anterior  cardinal  sinus.  Although  it  has 
generally  the  same  location  as  the  sinus 
in  selachians,  its  route  does  show  one  varia- 
tion: the  vein  passes  lateral  to  a  muscle 
which  Vetter  ( 1878 )  calls  the  trapezius 
intenius.  In  sharks,  the  sinus  lies  medial 
to  the  entire  levator  (trapezius)  series. 
Again,  this  difference  could  arise  from  a 
shift  of  the  musculature  rather  than  the 
development  of  a  new  vein,  but  it  is  also 
possible  that  there  has  been  emphasis  upon 
an  alternate  embryonic  blood  pathway  in 
Chimacra.  Leaving  to  one  side  the  possi- 
bility of  a  mistake  in  the  identification  of 
the  muscle  (the  fibers  run  posteriorly  from 
the  subocular  shelf  to  insert  upon  the  an- 
terior edge  of  the  scapula  just  beyond  the 
last  pharyngobranchials),  it  seems  that 
either  a  shift  in  the  arrangement  of  the 
muscles  or  the  vein  must  be  admitted.  How 
great  a  significance  should  be  attached  to 
such  an  alteration  remains  an  open  question. 

In  the  postcardiac  part  of  the  venous 
system  of  Chimacra  there  are  differences 
from  the  selachian  pattern  that  are  more 
clear-cut  if  not  easier  to  inteipret.  The 
lateral  abdominal  veins  are  absent,  and 
the  blood  which  they  would  have  collected 
is  differently  distributed.  From  the  pelvic 
fins  and  also  from  a  rectal  capillary  net- 


Morphology  and  Relationships  of  Holocephali  •  Stahl        159 


work  it  is  sent  directly  into  the  posterior 
end  of  the  posterior  cardinal  sinus.  A  small 
part  of  the  returning  blood  may  even  find 
its  way  to  the  renal  portal  by  a  fine  con- 
nection from  the  femoral  vein.  Further 
forward,  the  blood  from  the  deep  part  of 
the  ventral  body  wall  is  collected  by  the 
ventro-posterior  parietal  vein  which  flows 
posteriorly  to  a  confluence  with  the  renal 
portal.  Finally,  the  blood  from  the  pectoral 
fins,  instead  of  joining  the  flow  from  a 
lateral  abdominal,  enters  the  common  car- 
dinal vein  alone.  The  drainage  from  the 
posterior  half  of  the  pectoral  fin  enters 
what  may  be  more  exactly  described  as 
a  brachial  sinus  than  a  brachial  vein  at  the 
base  of  the  fin  and  is  joined  there  by  the 
blood  from  the  anterior  half  of  the  fin 
which  returns  by  a  vein  that  passes  through 
a  channel  in  the  pectoral  girdle.  There  are 
two  possible  explanations  for  the  derivation 
of  this  peculiar  venous  pattern.  The  first, 
which  cannot  be  flatly  dismissed,  is  that 
the  ancestral  stock  of  the  Holocephali  pos- 
sessed this  arrangement  of  vessels.  The 
second  and  perhaps  more  probable  sug- 
gestion is  that  the  lack  of  lateral  abdominals 
is  secondary  in  chimaerids  as  it  seems  to  be 
in  the  teleost  fishes.  The  lateral  abdominals 
and  their  homologues,  the  ventral  abdom- 
inal and  the  umbilical  veins,  play  too  large 
a  part  in  vertebrate  development  to  allow 
one  to  dispose  of  them  lightly.  Their  dis- 
appearance might  be  imagined  to  have 
required  the  longest  possible  evolutionary 
course.  That  these  veins  have  not  been 
described  in  cyclostomes  complicates  rather 
than  solves  the  problem.  If  their  absence 
represents  a  secondary  loss,  one  has  still 
no  clue  to  the  reason  for  their  absence  in 
holocephalians.  If,  on  the  other  hand,  the 
lack  of  lateral  abdominals  is  a  primitive 
vertebrate  characteristic,  which  holocepha- 
lians are  presumed  to  have  retained,  one 
must  then  question  the  homology  between 
the  lateral  abdominals  which  selachians 
have  developed  and  the  ventral  abdominal 
vein  which  appears  in  the  first  tetrapods. 
The  most  distinctive  feature  of  the  ve- 


nous drainage  which  might  merit  the  same 
interpretation  is  the  lack  of  hepatic  sinuses 
opening  into  the  sinus  venosus.  The  an- 
terior portion  of  the  liver,  just  behind  the 
transverse  septum,  is  very  thin,  contains  no 
sinuses,  and  seems  to  send  no  blood  forward 
into  the  heart.  The  back  wall  of  the  sinus 
venosus,  on  the  other  hand,  seems  entire, 
but  may  have  vestiges  of  hepatic  openings. 
Since  no  injection  material  can  be  made 
to  pass  through,  one  might  suppose  that 
no  passages  exist.  Slight  indentations  are 
visible,  however,  in  a  likely  location,  and 
a  needle  can  be  made  to  pass,  after  some 
probing,  without  seeming  to  pierce  tissue. 
If  traces  of  old  entrances  into  the  sinus 
venosus  are  present,  they  prove  that  the 
lack  of  hepatic  sinuses  in  the  usual  location 
is  secondary.  Even  if  it  is  not  certain  that 
such  openings  are  there,  it  would  seem  from 
the  design  of  the  hepatic  drainage  that  a 
secondary  arrangement  has  arisen.  The 
liver  is  bound  to  the  underside  of  the  pos- 
terior cardinal  sinus,  in  its  forward  portion, 
and  sends  to  that  channel  one  main  vein, 
two  or  three  other  large  ones,  and  several 
minor  auxiliary  vessels.  In  contrast  to  this 
pattern,  the  forward  flow  from  liver  to 
heart  is  characteristic  of  every  other  jawed 
vertebrate,  embryo  and  adult.  In  the  most 
deviant  pattern,  the  hepatic  sinuses  of  rays 
open  into  the  common  cardinal  veins  rather 
than  into  the  more  medial  sinus  venosus. 
If  the  holocephahan  pattern  were  to  be 
judged  primitive,  it  would  have  to  be  sup- 
posed that  it  was  the  sole  remaining  ex- 
ample of  a  distinctive  circulatory  arrange- 
ment which  existed  in  ancient  times  among 
ancestral  vertebrates — a  not  too  likely  pos- 
sibility. As  in  the  case  of  the  absence  of 
the  lateral  abdominals,  it  seems  more  sen- 
sible to  suggest  that  the  liver  drainage  rep- 
resents a  great  deviation  from  the  usual 
vertebrate  condition  and  may  well  have 
been  the  product  of  a  long  independent 
evolution. 

The  search  for  differences  between  se- 
lachians and  holocephalians  should  be  ex- 
tended to  the  hepatic  portal  system,   too, 


160 


BiiUetiii  Museum  of  Comporative  Zoology,  Vol.  135,  No.  3 


because  the  homologies  between  these  ves- 
sels of  sharks  and  chimaerids  are  not  at 
once  apparent.  Upon  close  scrutiny  it  seems 
that  the  chimaerid  veins  are  more  simpli- 
fied and  abbreviated  than  those  of  sharks. 
All  of  the  gastric  veins  are  absent,  of  course, 
as  are  those  which  drain  the  spleen  and  run 
through  mesentery.  A  short  lienopancreatic 
vein  and  some  small  splenics  join  the  vessel 
returning  blood  from  the  posterior  regions 
of  the  intestine.  Consequently,  only  one 
vein  runs  forward  to  join  the  intra-intestinal. 
The  union  of  these  two  vessels  and  a  smaller 
one  from  the  anterior  intestine  and  neigh- 
boring esophagus  creates  the  hepatic  portal. 
Although  it  is  probable  that  the  lack  of 
mesenteries  and  the  removal  of  the  spleen 
from  its  usual  place  are  secondary  changes, 
it  is  not  necessary  to  assume  that  the  entire 
pattern  of  portal  tributaries  has  been  only 
recently  evolved.  The  existence  of  two 
sizable  branches  draining  the  posterior  in- 
testine, an  arrangement  which  has  no  coun- 
terpart in  sharks,  may  not  have  arisen  from 
any  selachian  forerunner.  It  would  be 
tempting  to  add  the  lack  of  gastric  veins 
as  another  possible  preselachian  character, 
because  the  absence  of  a  stomach  has  been 
supposed  to  be  a  primitive  arrangement; 
but  it  cannot  be  assumed  definitely  that 
the  lack  of  a  stomach  and  the  minimal 
development  of  the  spiral  valve  are  prim- 
itive rather  than  degenerate  developments. 
The  hepatic  portal  system,  one  must  con- 
clude, does  bear  a  resemblance  to  the 
selachian  pattern  but  may  be  somewhat 
secondarily  modified.  Despite  resemblances, 
on  the  one  hand,  and  late  modifications,  on 
the  other,  however,  it  is  not  impossible  that 
there  might  be  some  elements  of  an  older 
independent  pattern  still  included  in  the 
system. 

The  association  of  lymphomyeloid  tissue 
with  the  venous  system  in  the  Holocephali 
is  another  characteristic  which  distinguishes 
these  fishes  from  the  shark  group.  Kolmer 
( 1923),  who  studied  this  tissue  in  Chimaera 
monstro.sa,  describes  it  as  consisting  of  a 
mass   of   lymphatic   cells   of   varying   sizes 


mixed  with  red  blood  cells  in  all  stages  of 
formation.  These  cells  are  supported  by 
a  network  of  fine  connective  tissue  fibers 
which  merge  with  the  adventitia  of  the 
wide  veins  and  small  arteries  that  ramify 
within  the  mass.  If  the  tissue  found  in  the 
esophageal  wall  of  sharks  and  rays  is  com- 
parable to  that  which  exists  in  a  much 
expanded  state  in  the  chimaerids,  one  might 
predicate  the  presence  of  this  material  in 
the  early  placoderms.  It  may  have  been 
carried  in  several  lines  as  a  hemopoietic 
organ,  its  different  location  and  extent  in 
holocephalians  and  selachians  indicating 
separate  evolutionary  pathways.  It  is 
harder  to  imagine  that  this  tissue,  which 
is  Vvddespread  in  sharks  and  batoids  in  its 
esophageal  location,  should  disappear  from 
that  place  and  appear  in  the  head  and 
girdle  regions  of  the  Holocephali  as  they 
branched  off  from  a  shark  stem.  It  is  also 
possible  that  the  tissue  of  the  two  groups, 
although  it  looks  similar  under  the  micro- 
scope, may  not  have  a  common  origin.  If 
that  be  the  case,  the  hypothesis  that  chi- 
maerids have  been  derived  from  early 
sharks  would  have  another  point  against  it. 
A  study  of  the  remaining  portions  of  the 
circulatory  system  reinforces  the  idea  that 
holocephalians  and  sharks  probably  arose 
from  placodenn  stocks  possessing  some 
characters  which  both  groups  of  fishes  have 
carried  to  the  present  day.  One  has  only  to 
dissect  the  heart  in  each  animal  ( cf.,  for 
example,  Lankester,  1878,  and  Hyman, 
1942:  329)  to  be  convinced  that  chimaerids 
and  sharks,  while  distinct  lines,  cannot  be 
widely  separated  from  each  other  on  the 
evolutionary  tree.  The  hearts  of  the  two 
types  of  fishes  are  identical  in  their  gross 
anatomy,  and  markedly  distinct  from  the 
heart  of  lungfishes  or  that  of  ray-finned 
forms. "^  The  only  characteristic  which  dis- 
tinguishes the  chimaerid  heart  from  the 
shark  structure  is   its  relatively  small  size 


^  Lankester  (1878),  besides  describinn;  the  heart 
of  Chiiiiacni,  makes  a  visil)ly  futile  attempt  to  point 
out  homoloj^ies  between  its  arrangement  of  valves 
and  tliat  in  the  heart  of  dipnoans. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        161 


(Fig.  4B).  Although  no  measurements 
were  made,  it  seems  that  the  heart  of 
Chimaera  would  have  a  smaller  capacity 
than  that  of  a  shark  of  the  same  size.  If 
one  wishes  to  suppose  that  sharks  and 
holocephalians  have  long  traveled  upon 
separate  evolutionary  paths,  one  must  pos- 
tulate that  their  type  of  heart  represents 
the  primitive  gnathostome  structure  which 
has  been  retained  in  all  cartilaginous  fishes 
and  highly  modified  in  bony  ones. 

The  one  salient  difference  in  the  arterial 
pattern  occurs  in  the  head  region.  The 
Holocephali  show  a  type  of  blood  supply 
to  the  brain  which  differs  from  both 
selachians  and  bony  fishes  in  that  the 
pseudobranchial  efferent  alone  reaches  the 
cranial  cavity  (AlHs,  1912).  The  hyoidean 
efferent,  which  feeds  the  internal  carotid 
in  elasmobranchs,  forms  in  holocephalians 
only  a  commissure  with  no  continuation 
running  forward  to  the  brain.  Even  if  the 
investigator  assumes  that  there  were  in  the 
primitive  state  two  pairs  of  efferent  arteries 
which  sent  blood  forward  to  the  brain,  and 
that  in  the  Holocephali  the  more  posterior 
pair  has  degenerated,  he  has  not  disposed 
of  the  entire  problem  posed  by  the  cephalic 
arteries  of  the  Holocephali.  There  is  also 
an  unusual  mandibular  artery  for  which  to 
account.  In  sharks  the  lower  jaw  is  sup- 
plied by  a  vessel  which  leaves  the  ventral 
end  of  the  first  efferent  arterial  loop  ( Hy- 
man,  1942:  324);  in  holocephalians  the 
mandibular  artery  runs  ventrally  from  the 
efferent  pseudobranchial,  itself  a  more  dor- 
sal branch  from  the  loop.  The  suggestion 
has  been  made  by  Allis  ( 1912 )  that  the 
holocephalian  vessel,  which  follows  the  line 
of  the  jaw,  represents  the  \entral  portion 
of  the  afferent  mandibular  artery  and  that 
the  retention  of  this  vessel,  which  has  gen- 
erally disappeared  in  other  vertebrates,  is 
a  primitive  feature.  If  this  supposed  homol- 
ogy is  correct,  it  would  not  be  possible  to 
derive  the  chimaerids  from  a  shark  group 
in  which  the  mandibular  afferent  had  al- 
ready disappeared.  However,  Marples' 
( 1936 )   discovery  of  a  similar  mandibular 


artery  in  Sqiiatina,  and  his  statement  of 
the  existence  of  the  same  type  of  vessel 
in  Polyodon  and  certain  teleosts,  makes 
questionable  Allis'  interpretation  and  any 
evolutionary  theories  which  might  be  based 
upon  it. 

The  only  other  portion  of  the  arterial 
system  which  deviates  from  the  selachian 
pattern  is  the  efferent  branchial  series  of 
vessels.  In  contrast  to  the  distinct  loops 
created  in  selachians  (Hyman,  1942:  324) 
by  the  union  of  well-formed  pre-  and  post- 
trematic  arteries,  the  poor  development  of 
the  chimaerid  pretrematic  branches  creates 
discontinuities  in  the  posterior  three  collec- 
tor loops.  Allis  ( 1912)  believed  that  he  saw 
four  complete  loops,  but  Allen  ( 1905 )  and 
Parker  ( 1886 )  failed  to  find  any.  Without 
doubt,  the  pretrematic  vessels  seem  second- 
ary to  the  posttrematic  ones.  In  Chimaera 
coUiei,  the  ventral  ends  of  the  pretrematics 
diminish  in  size  and  the  distinct  ventral 
commissures  which  close  the  loops  were 
visible  only  in  well-injected  specimens. 
Again,  two  possible  explanations  present 
themselves:  either  the  chimaerid  situation 
represents  an  early  step  toward  loop-forma- 
tion or  it  is  the  result  of  a  modification 
associated  with  the  reduction  of  the  gill 
arch  apparatus  and  its  concealment  beneath 
the  operculum.  There  is  no  way  of  deciding 
which  theory  is  more  probable. 

In  conclusion,  then,  one  recognizes  that 
the  holocephalian  circulatory  system  resem- 
bles that  of  selachians  in  numerous  ways 
and  vet  differs  from  it  distinctlv  in  certain 
characteristics.  In  some  of  its  nonselachian 
features  the  chimaerid  system  parallels  the 
structure  of  bony  fish  while  in  others  it 
is  apparently  unique. 

Among  the  similarities  to  selachians  can 
be  listed  the  structure  of  the  heart,  the 
general  design  of  the  arterial  system,  the 
presence  in  the  venous  system  of  great 
sinuses,  and  the  arrangement  of  the  prin- 
cipal venous  sinuses.  Pursuing  further  sim- 
ilarities in  the  venous  system,  one  must 
mention  the  existence  of  a  pair  of  inferior 
jugular   \essels    which    are   more    like   the 


162 


Bulletin  Muscuin  of  Comparai'wc  Zoology.  Vol.  135,  No.  3 


inferior  jugulars  of  sharks  than  hke  those 
of  other  fishes,  of  similar  renal  portal  sys- 
tems, and  of  a  hepatic  portal  system  that 
is  certainly  closer  to  the  type  of  sNsteni 
found  in  selachians  than  to  that  found  in 
bony  fish.  Like  the  selachians,  the  holo- 
cephalians  have  a  system  of  cutaneous  veins 
and  lack  lymphatics.  As  a  last  point  of  sim- 
ilarity, even  the  valves  in  the  chief  vessels 
seem  to  be  located  at  the  same  places. 

Characters  which  distinguish  the  holo- 
cephalian  circulatory  system  from  that  of 
sharks  are  found  in  both  the  arterial  and 
venous  pathways.  It  is  perhaps  significant 
that  features  of  the  arterial  system  which 
are  unique  to  the  Holocephali  are  all  found 
in  the  head,  a  region  which  is  as  a  whole 
very  highly  specialized.  Although  the  in- 
complete collector  loops  and  the  absence 
of  the  anterior  extension  of  the  internal 
carotid  arteries  can  be  viewed  as  relatively 
small  modifications  of  the  selachian  plan, 
the  significance  of  the  branching  of  the 
mandibular  artery  from  the  afferent  pseudo- 
branchial  is  not  so  easy  to  interpret. 

In  the  venous  system,  the  absence  of 
lateral  abdominal  veins  represents  a  great 
divergence  from  the  selachian  pattern  and 
a  resemblance  to  that  of  ray-finned  fishes. 
With  those  vessels  missing,  the  brachial 
veins  empty  into  the  common  cardinals 
directly,  as  is  the  case  in  teleosts,  and  the 
femoral  veins  open  into  the  posterior  car- 
dinal sinuses.  The  entrance  into  these  sinuses 
of  blood  from  the  liver  is  surely  a  peculiar- 
ity developed  in  the  holocephalian  line,  as 
such  a  route  is  unheard  of  in  any  other 
group  of  fishes.  Whether  the  absence  of 
a  hyoid  sinus  is  also  to  be  interpreted  as  a 
secondary  development  or  as  a  primary 
arrangement  is  not  clear.  The  resolution  of 
this  question  may  depend  upon  the  solu- 
tion of  the  problem  of  the  status  of  the 
hyoid  arch  with  which  the  hyoid  sinus  is 
associated. 

In  surveying  the  circulatory  system  of  the 
Holocephali,  one  receives  the  impression 
that  the  resemblance  to  the  bony  fish, 
which  exists  chieflv  in  the  absence  of  the 


lateral  abdominal  veins,  is  fortuitous  rather 
than  representative  of  ancestral  connec- 
tions. Since  the  resemblance  to  the  circula- 
tory system  of  sharks  is  more  general,  the 
question  seems  to  be  whether  the  holo- 
cephalian system  is  a  derivative  of  the 
selachian  one  or  whether  it  has  developed 
in  its  own  path  from  a  system  which  char- 
acterized a  placodenn  group  ancestral  to 
both  holocephalians  and  selachians.  An 
opinion  as  to  which  of  these  alternatives 
is  the  more  probable  could  be  more  strongly 
supported  if  the  structural  arrangements  of 
other  systems  were  brought  into  evidence. 
Should  they  be  found  to  contain  characters 
too  primitive  to  have  been  derived  from 
the  more  specialized  homologues  of  sharks, 
resolution  in  favor  of  descent  from  separate 
placoderm  ancestors  would  be  indicated.  If 
the  other  systems  seem  entirely  shark-like 
or  differ  from  the  shark  pattern  in  minor 
ways  only,  a  direct  descent  from  some  shark 
group  cannot  be  ruled  out.  To  extend  the 
comparison  between  holocephalians  and 
selachians  beyond  the  confines  of  the  cir- 
culatory system,  then,  a  review  of  the 
nervous,  skeletal,  muscular,  urogenital,  and 
digestive  systems  is  undertaken  in  the  fol- 
lowing pages.  The  study  of  these  systems 
also  provides  a  check  upon  the  assumption 
made  here  that,  although  similarities  be- 
tween structures  of  holocephalians  and 
actinopterygians  do  appear,  there  is  little 
likelihood  of  an  ancestral  affiliation  be- 
tween the  two  groups. 

DISCUSSION  OF  PHYLOGENETIC  CLUES 
FROM  OTHER  ORGAN  SYSTEMS 

The   Nervous  System 

In  any  study  in  which  the  relationships 
of  the  Holocephali  are  reviewed,  the  ner- 
vous system  receives  primary  attention.  Its 
gross  anatomy  has  been  studied  in  detail 
(Braus,  1898;  Cole,  1896;  Carman,  1904; 
Fiirbringer,  1897;  Nicol,  1950;  Wilder, 
1877),  and  microscopic  work  has  been 
attempted  in  some  areas  (Kappers,  1911, 
1912;     Backstrom,     1924;     Johnston.     1910; 


Morphology  and  Relationships  of   Holocephali  •  Stahl        163 


Nicol,  1950).  From  the  evidence  presented 
in  these  papers,  it  seems  that  the  holo- 
cephahan  ner\ous  system  retains  some  char- 
acteristics that  must  surely  he  primitive  hut 
exhihits  pecuHarities  which  are  generally 
interpreted  as  secondary  phenomena. 

Although  the  form  of  the  spinal  cord  is 
quite  regularly  shark-like,  the  hrain  is  of  an 
extraordinary  nature.  In  its  posterior  por- 
tion it  resemhles  closely  the  selachian  organ, 
but  the  telencephalon  is  unique  in  the  ani- 
mal kingdom.  In  Chimacm  and  CoUorliyn- 
chus\  this  anterior  region  of  the  brain 
extends  forward  as  a  long  stalk  beneath  the 
interorbital  septum.  It  widens  finally  into 
a  subdivided  telencephalic  lobe  just  behind 
the  olfactory  bulbs  associated  with  the  nos- 
trils. The  uniqueness  of  the  arrangement 
argues  for  its  secondary  development.  Kap- 
pers  and  Carpentier  (1911)  have  considered 
the  elongation  of  the  telencephalon  and 
feel  that  it  has  taken  place  as  a  corollary 
to  the  enlargement  of  the  eyes.  Since  the 
eyes  have  encroached  upon  the  medial 
region  normally  occupied  by  the  cerebral 
hemispheres,  the  forepart  of  the  brain  has 
been  displaced.  Rather  than  being  com- 
pressed posteriorly,  in  the  ancestors  of  the 
Holocephali,  alone  among  all  the  animals 
that  have  developed  enormous  eyes,  the 
telencephalon  became  displaced  anteriorly. 
The  result  of  the  forsvard  gro\\'th  of  the 
telencephalon  has  been  the  creation  of  long 
brain  stalks  through  which  regular  connec- 
tions with  the  diencephalon  are  maintained. 
RJiinochimaera,  in  which  the  eyes  are 
smaller,  the  brainstalks  not  quite  so  elon- 
gated, and  the  olfactory  tracts  more  sela- 
chian-like, may  represent  an  earlier  stage 
in  the  evolution  of  this  curious  arrangement. 

Another  characteristic  of  the  telencepha- 
lon which  lends  itself  to  comment  of  a 
phylogenetic  nature  is  the  development  of 
the  pallium.  Holmgren,  who  has  studied 
forebrain  moq^hology  in  lower  vertebrates 
(1922),  has  investigated  the  pahial  region 
in  holocephalians,  selachians,  ganoids,  tele- 
osts,  and  lungfish.  He  points  out  that  the 
selachian  pallium  is  inverted  to  a  greater 


/.  supopK. n.       opt. I.    cbl. 

3T.r\.  I  I  / 

med  obi. 


Fig.  1.  Chimaera  coliiei.  Brain,  lateral  view,  cbl.,  Cere- 
bellum; eth.c,  ethmoid  canal;  hyp.,  hypophysis;  med.obl., 
medulla  oblongata;  oll.b.,  olfactory  bulb;  oph.prof.n., 
ophthalmicus  profundus  nerve;  opt. I.,  optic  lobe;  sup.oph.n., 
superficial  ophthalmic  nerve;  tel.,  telencephalon;  tel.st., 
telencephalic    stalk;    //,    optic    nerve.     (After    Gorman.) 


degree  than  is  the  case  in  the  other  fishes. 
By  inversion  he  means  a  rolling  medially 
of  both  left  and  right  edges  of  the  embry- 
onic neural  i^lates,  resulting  in  their  contact 
dorsally  if  the  two  masses  of  tissue  reach 
the  midline  as  they  do  in  selachians.  E vagi- 
nation  of  the  more  lateral  portions  of  the 
developing  forebrain  wall  gives  rise  to 
paired  cerebral  hemispheres.  If  nerve  cells 
mass  dorsally  over  the  ependymal  layer,  the 
dorsal  brain  wall  thickens  and  the  dividing 
furrow  between  the  hemispheres  may  be 
more  or  less  obliterated.  The  developmental 
mode  which  occurs  in  the  Holocephali,  how- 
ever, consists  of  a  lateral  rather  than  a  dor- 
sal concentration  of  nerve  cells.  The  brain 
roof  is  then  left  relatively  thin.  In  extreme 
manifestations  of  this  latter  tendency,  the 
lateral  brain  walls  grow  exceedingly  thick 
and  may  actually  evert,  leaving  the  roof  to 
be  covered  by  an  expanded  tela.  This  is 
the  case  in  actinopterygians.  A  less  extreme 
and  perhaps  more  primitive  version  of  the 
same  condition  is  found  in  the  lungfish 
Cemtodus  (Holmgren  and  Horst,  1925). 
The  dipnoan  has  a  broad  thin  roof  over  the 
pallium  but  the  moderately  thick  cerebral 
walls  are  not  everted. 

In  holocephalians  the  laterally  thickened 
pallium  is  inverted  but  never  becomes  con- 
tinuous over  the  dorsal  midline.  There  is 
always  a  small  strip  of  ependyma  bridging 


164         Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


A 


qpc  he 


D 


Fig.  2.  Transverse  section  of  the  forebrain  in  A,  Acanthias,  B,  Chimaera,  C,  Protopferus,  D,  Lepidosfeus,  to  show  position 
of  the  palhum.  epnd.,  Ependyma;  g.p.c,  general  pallial  cortex;  /i.e.,  hippocampal  lobe;  1. 1. p.,  lateral  limit  of  pallium; 
n.olf.l.,    nucleus   olfactorius   lateralis;   pal.,    pollium;    p.c,    pyriform    cortex;    sub. p.,    subpallium.     (After    Holmgren.) 


the  dorsal  gap.  The  palHum  is  evaginated 
to  form  two  separate  olfactory  lobes  which 
carry  separate  ventricles  forward  of  the 
foramen  of  Munro.  The  pallium  is  rather 
small,  being  confined  to  the  anterior,  en- 
larged portion  of  the  telencephalon.  The 
brainstalks  which  connect  the  anterior  en- 
largement with  the  remainder  of  the  brain 
are  composed  entirely  of  subpallial  tissue. 

Kappers,  who  reviewed  holocephalian 
brain  structure  in  his  compendium  on  the 
nervous  system  of  vertebrates  ( 1936 ) , 
grouped  the  Holocephali  with  the  lower 
actinopterygian  fishes  as  intermediate  be- 
tween selachians  and  teleosts.  He  regarded 
the  inversion  of  the  pallium  as  carried  over 
from  the  former  and  the  eversion  of  the 
brainstalk  walls  as  presaging  the  great  pal- 
lial eversion  of  the  latter.  Holmgren  dis- 
agrees with  the  conclusion  of  Kappers, 
however.  He  feels  that  eversion  of  the  sub- 
pallial tissue  of  which  the  brainstalks  con- 
sist cannot  be  regarded  as  an  early  stage  of 
the  pallial  eversion  seen  in  bony  fish.  In 
making  his  interpretation  of  the  phylo- 
genetic  position  of  the  Holocephali,  Holm- 
gren considers  only  the  true  pallium  whose 
limits  he  has  determined  by  histological 
study.  He  reasons  that  the  holocephalian 
pallium  resembles  most  nearly,  in  its  degree 
of  inversion  and  evagination,  what  must 
have  been  the  type  ancestral  to  that  of  ex- 
tant cartilaginous  and  bony  fishes.  The 
development  of  greater  inversion  with  re- 
sulting fusion  across  the  dorsal  midline 
would  lead  to  the  selachian  condition, 
whereas   the   development   of  thicker   and 


more  widely  separated  walls  would  lead  to 
the  lungfish-lower  actinopterygian-teleost 
sequence. 

Observations  upon  the  microscopic  struc- 
ture of  the  holocephalian  brain  have  been 
more  fragmentary,  and  no  clear-cut  indica- 
tions of  phylogenetic  position  arise  from 
them.  Kappers  (1912)  has  mapped  the  ar- 
rangement of  the  motor  nuclei  in  Chimaera 
monsfrosa  and  he  and  several  other  workers 
have  determined  the  course  of  some  of  the 
brain  tracts  in  the  chimaeroids  ( Kappers, 
1911;  Biickstrom,  1924;  Johnston,  1910). 
One  example  of  the  quandary  to  which  these 
studies  have  led  should  suffice.  The  sela- 
chians, with  which  investigators  have  sought 
to  compare  the  holocephalians,  are  charac- 
terized by  three  telencephalic  tract  decus- 
sations— one  dorsal  and  two  ventral.  Since 
the  left  and  right  pallial  masses  of  holo- 
cephalians do  not  fuse  in  the  dorsal  mid- 
line, the  dorsal  decussation  is  absent.  It  is 
not  known  whether  the  fibers  which  cross 
dorsally  in  selachians  are  channeled  through 
the  ventral  commissures  in  holocephalians 
or  whether  these  fibers  are  wholly  or  par- 
tially absent.  In  speaking  of  the  ventral 
decussation,  Biickstrom  goes  on  to  say,  "It 
is,  however,  possible  that  a  number  of  fibre 
connections  in  this  decussatio  existing  in 
Chimaera  are  lacking  in  selachians  or  vice 
versa"  (Biickstrom,  1924:  232). 

The  arrangement  of  the  cranial  nerves 
has  also  been  examined  by  a  worker  with 
the  phylogenetic  question  in  mind.  Cole, 
who  has  dissected  these  nerves  in  Chimaera 
in  detail  (1896;  Cole  and  Dakin,  1906),  was 


Morphology  and  Relationships  of  Holocephali  •  Stohl        165 


especially  interested  by  the  emergence  from  were  farthest  removed  from  the  beginning 

the  brain  and  the  distribution  of  nerves  V,  of  the  evolutionary  line.  He  had  detemiined 

VII,  and  X.    He  points  out  that  there  is  no  that  the  number  of  occipital  nerves  coming 

trigemino-facial    complex    in    Chimaera    as  through  foramina  at  the  back  of  the  skull 

there  is  in  sharks.   Nerve  V  emerges  by  two  ranged  from  five  in  notidanid  sharks  to  one 

roots  anterior  to  \'II  and  underneath  the  or  none  in  rays.    At  first,  after  finding  five 

buccal  branch  of  the  latter.    It  sends  sen-  such  nerves  in  Chimaera  and  four  in  Cal- 

sory  and  motor  fibers  to  the  usual  destina-  lorhynchus,  Fiirbringer  was  ready  to  place 

tions  without  e\'er  mingling  with  portions  these  fish  on  a  level  with  the  notidanids. 

of  VII.    Contact  between  these  two  cranial  On  closer  observation,  however,  he  noticed 

nerves   is   limited  to  a  variable  degree  of  that  only  the  first  two  resembled  the  occipi- 

binding  together  of  their  superficial  oph-  tal  nerves  of  the  selachians  in  appearance 

thalmic  branches.  Nerve  VII  can  be  divided  and  in   their  course  to  the  hypobranchial 

into   a   small   motor   portion   and   a   larger  muscles.    The  remainder  looked  very  much 

lateral    line    component.    Cole    recognizes  like  the  succeeding  spinal  nerves  and,  like 

that  the  isolation  of  the  lateral  line  fibers  the  latter,  sent  fibers  to  the  brachial  plexus, 

from  the  rest  of  the  cranial  nerve  is  also  Fiirbringer   distinguished   these    nerves    as 

characteristic   of  other   fishes   and   of   am-  spino-occipital   nerves,    explaining   that   in 

phibians  which  ha\e  a  lateral  line  system  the  Holocephali  two  or  three  of  the  anterior 

and   so   has    no   special    significance.    The  vertebrae  have  been  incorporated  into  the 

tenth  cranial  nerve  of  Chimaera  is  distinc-  skull  bringing  their  segmental  nerves  with 

tive,  though,  in  having  its  four  parts  in  addi-  them.   Thus  these  spino-occipital  nerves  are 

tion  to  the  laterahs  component  ( three  bran-  not  homologous  to  the  posterior  occipital 

chial  branches  and  one  visceral)  completely  nerves  of  simple  selachians  but  are  proof 

separate:    each  arises  separately  from  the  that  the  holocephalians  belong  to  a  "hohere, 

brain  and  each  has  its  own  ganglion.    Cole  mehr  specialisirte  Abtheilung"  ( Fiirbringer, 

dwelt  upon  the  evidence  of  the  primitive  1897:   446). 

position  of  Chimaera,  which  the  separation  The  true  spinal  nerves  of  holocephalians 
of  the  posterior  cranial  nerves  suggests,  and  bear  a  greater  resemblance  to  those  of 
concluded:  "The  discrete  nature  of  the  fifth,  selachians  than  to  those  of  bony  fishes  in 
seventh,  and  lateral  line  nerves  makes  Chi-  that  the  dorsal  and  ventral  roots  retain  the 
maera  a  very  unique  fish  as  regards  its  large  degree  of  independence  which  is 
cranial  nerves,  and  it  is  to  be  presumed  that  characteristic  of  the  former  group.  In  Chi- 
such  a  simple  condition  is  more  primitive  maera,  one  can  see  in  each  segment  of  the 
than  the  more  complex  fusions  and  inter-  trunk  two  roots  emerge,  give  off  dorsal 
minglings  that  obtain  in  other  fishes.  This  rami,  and  then,  as  ventral  rami,  gradually 
separation  may,  ho\\'e\'er,  be  purely  second-  come  together.  The  segmental  nerve  formed 
ary,  just  as  the  form  of  the  brain  of  Chi-  by  their  union  soon  divides,  and  the  nerve 
maera  undoubtedly  is,  but  on  the  other  continues  its  lateral  course  as  a  double- 
hand  the  vagus  is  also  in  a  very  simple  and  stranded  structure.  If  the  two  strands  rep- 
unfused  condition  in  Chimaera,  and  the  resent  the  reseparation  of  dorsal  and  ven- 
same  may  be  said  of  its  cranial  nerves  gen-  tral  root  fibers,  the  holocephalians  would 
erally"  (Cole  and  Dakin,  1906:  599).  then  show  a  very  limited  association  of  dor- 
While  Cole  was  upholding  the  primitive  sal  and  ventral  root  elements— an  arrange- 
position  of  Chimaera  suggested  by  the  ar-  ment  seemingly  closer  to  the  primitive  state 
rangement  of  its  cranial  nerves,  Fiirbringer  of  complete  separation  than  that  shown 
(1897)  was  concluding  from  his  compara-  even  by  selachians.  At  the  level  of  the  pel- 
tive  study  of  the  occipital  nerves  of  sela-  vie  fin,  Davidoffs  dissections  (1879)  show 
chians   and  holocephalians   that   the   latter  a  separation   of  the  strands   of  the  spinal 


166         Bulletin  Miisetim  of  Compnraiwe  Zoology,  Vol.  135.  No.  3 


nerves  followed  distally  by  a  recombination  ganglia  in  the  tail ) ;  but  these  ganglia  are 

involving  the  posterior  strand  in  one  seg-  haphazardly    connected   by    a    network    of 

ment  and  the  anterior  strand  from  the  seg-  nerve    fibers    and    communicate    with    the 

ment  behind.    In   Davidoff's  opinion,   this  spinal   nerves  by   white   rami   only.    Since 

arrangement  as  well  as  further  connections  there    are   minor   differences   between   the 

lietween  the  first  two  nerves  which  supply  systems  of  selachians  and  Chimaera,  Nicol 

the  fin  suggest  the  beginnings  of  a  more  is  of  the  opinion  that  the  chimaeroids  split 

complex   peKic  plexus   than   is   present  in  from  the  selachian  line  and  have  evolved 

other  fishes.    The  holocephalians'  lack  of  a  in  the  final  stages  on  a  separate  path, 

collector  nerve  in  the  pelvic  region,  accord-  A  study  of  the  sense  organs  produces  little 

ing  to  Van   der  Horst    {in    Bolk,   Vol.   II,  evidence   which   can   be   brought   to   bear 

1934),  also  sets  them  apart  from  selachians,  upon  the  phylogenetic  question.   The  olfac- 

dipnoans,  and  lower  actinopterygians.    The  tory,  optic,   and  otic  structures  in  holoce- 

pU'xus  at  the  level  of  the  pectoral  fin,  how-  phalians  are  similar  to  those  of  sharks,  and 

e\er,    resembles    that    of    selachians    and  there  are  no  fossil  remains  of  the  first  two 

actinopterygians    in    being    of    a    cervico-  types  of  sense  receptors  from  which  their 

brachial   nature.    The   holocephalians   and  hereditary  history  could  be  learned.    The 

these  fishes  are  distinguished  in  this  feature  design  of  the  inner  ear  in  early  vertebrates 

from  the  dipnoans  which  have,  like  tetra-  has   been   revealed    through   cranial    casts, 

pods,  two  separate  plexuses  in  this  region,  however,     making     comparisons     possible. 

Assessing  the  various  characteristics  of  the  Stensio's  ( 1963 )  cast  of  the  cavities  in  the 

spinal  nerves   and   the   plexuses   in   which  cranium  of  the  arthrodire  Kujdanowiaspis 

they  are  involved,  one  might  conclude  that  shows  a  general  arrangement  of  the  laby- 

the  holocephalians  may  show  the  retention  rinth   which   still   characterizes   both   holo- 

of  a  relatively  primitive  arrangement  which  cephalian   and  selachian  fishes.    Even  the 

has  been  modified  to  form  a  unique  pattern  endolymphatic  duct  appears,  rising  to  open 

in  the  pelvic  region.  upon  the  dorsal  surface  of  the  head.    The 

The  last  remaining  part  of  the  nervous  only  point  of  difference  between  chimaerids 

system  to  be  discussed,  the  autonomic  divi-  and  sharks,  of  which  Stensio  speaks,  con- 

sion,    has    been    described    thoroughly    by  cerns  the  structure  of  the  utriculus.    That 

Nicol  (1950).    His  study,  it  must  be  men-  of  selachians  is  divided,  while  the  utriculus 

tioned,  was  based  only  upon  Chimaera  col-  of  holocephalians  is  not.   Stensio  thinks  that 

liei.    Without  attempting  to  repeat  Nicol's  the  undivided  state  is  more  primitive  and 

description,  one  may  say  that  he  found  a  that  the  divided  utriculus  has  appeared  in 

very  close  similarity  between  the  autonomic  certain  orders  of  arthrodires  and  in  elasmo- 

systems    of   selachians    and    holocephalians  branchs  through  parallel  development.    If 

and  substantial  contrasts  between  their  type  Stensio's  speculation  is  correct,  the  ances- 

of  system  and  that  of  bony  fishes.    For  ex-  tors    of    holocephalians    and    early    sharks 

ample,   he   notes   that   teleosts   have   well-  would    have    been    separate    but    related 

ordered   sympathetic   chains   connected   to  stocks. 

the  spinal  nerves  by  Ijoth  gray  and  white  Both     Stensio     (1947)     and     Holmgren 

rami.    Although  the  dipnoans  show  a  less  (1942a)  have  included  a  study  of  the  lateral 

well-developed  pair  of  ganglionated  chains,  line  system  of  the  Holocephali  in  their  sur- 

the  presence  of  a  delicate  chain-structure  veys  of  lateral  line  systems  in   fishes  and 

differentiates    even   these   fishes    from   the  amphibians.     Although   these   authors    dis- 

selachians  and  from  Chimaera.    In  the  car-  agree  as  to  whether  a  general  pattern  of 

tilaginous  forms  there  is  a  more  or  less  seg-  head  canals  can  be  defined,  they  state  in 

mental  arrangement  of  sympathetic  ganglia  concert  that  no  explanation  of  the  evolution 

throughout  the  trunk   (and  an  absence  of  of  the  holocephalian  pattern  is  possible  at 


Morphology  and  Relationships  of  Holocephali  •  Stahl        167 


tliis  time.  Holmgren  finds  it  improliable 
tliat  the  holocephalian  arrangement  could 
be  derived  from  that  of  selachians  or  vice 
versa.  He  suspects  that  the  holocephalian 
system  has  been  reduced  from  a  more 
elaborate  pattern  although  the  absence  of 
embryological  studies  prevents  his  specu- 
lating upon  what  the  ancestral  state  might 
have  been.  He  is  forced  to  abandon  the 
problem  with  the  statement  that  the  holo- 
cephalian head  canals  "could  not  be  identi- 
fied with  lines  in  any  other  vertebrate" 
(Holmgren,  1942a:  21).  Stensio  would  have 
liked  to  have  made  a  comparison  between 
the  head  canals  of  holocephalians  and  those 
of  ptyctodonts,  but  unfortunately  the  latter 
have  not  been  preserved.  The  anatomy  of 
the  sensory  canals  of  holocephalians  reveals 
no  more  to  the  investigator  than  their  ar- 
rangement. As  Garman  ( 1888,  1904 )  and 
Reese  (1910)  have  shown,  the  sensory  cells 
lie  in  open  grooves  in  Chimaera,  in  slit 
tubules  in  Rhinodiimaera,  and  in  closed 
tubules  in  CaUorhynchiis.  It  is  not  possible 
to  detennine  which  of  these  arrangements 
is  primitive  or  if  any  one  of  them  is. 

Although  no  single  characteristic  of  the 
nervous  system  serves  as  a  key  to  the  evolu- 
tionary history  of  the  Holocephali,  it  is  pos- 
sible to  make  a  reasonable  speculation  based 
upon  the  group  of  anatomical  features  dis- 
cussed above.  The  survey  of  the  holoce- 
phalian nervous  system  has  shown  that  no 
portion  of  it  resembles  that  of  any  bony 
fish.  Although  the  anterior  extension  of  the 
telencephalon,  the  ramifications  of  the 
spinal  nerves,  and  the  pattern  of  the  sen- 
sory canals  are  unique,  the  posterior  parts 
of  the  brain,  the  sensory  organs,  and  the 
autonomic  nervous  system  are  strikingly 
like  the  shark  structures.  In  drawing  con- 
clusions based  upon  the  nature  of  the  ner- 
vous system,  then,  one  must  lay  aside  the 
possibility  that  the  Holocephali  might  be 
allied  to  any  line  leading  to  bony  fishes 
(despite  Holmgren's  view  that  the  dipnoan 
pallium  might  be  derived  from  the  holo- 
cephalian type)  and  predicate,  instead, 
some   degree   of  relationship   to   the   early 


sharks  or  their  ancestors.  If  one  agrees  with 
the  assumption  that  the  partially  inverted 
pallium  is  more  primitive  than  the  strongly 
inverted  selachian  structure  and  also  \\'ith 
the  interpretation  of  the  cranial  nerve  ar- 
rangement as  more  primitive  than  that 
found  in  sharks,  it  follows  that  the  Holo- 
cephali could  not  have  evolved  from  early 
sharks  in  which  the  more  complex  selachian 
organization  was  already  established.  One 
is  left  with  the  hypothesis  that  the  Holo- 
cephali have  descended  from  an  ancestral 
group  separate  from  that  of  selachians  but 
allied  to  it.  This  hypothesis  allows,  firstly, 
for  the  retention  in  holocephalians  of  the 
structures  assumed  to  be  primitive  even 
though  these  elements  are  modified  in 
sharks.  Secondly,  it  provides  an  explana- 
tion for  the  presence  of  similar  nervous 
structures  in  both  types  of  cartilaginous 
fishes,  since  these  elements  may  have  been 
characteristic  of  the  larger  group  to  which 
both  ancestral  stocks  belonged.  Thirdly,  the 
hypothesis  suggests  that  the  holocephalians 
developed  along  a  separate  line  long  enough 
to  permit  the  evolution  of  the  special  struc- 
tures which  are  unique  to  them.  The  alter- 
native theory,  that  holocephalians  are  de- 
scended from  an  early  shark  group,  \\'ith  its 
corollary  that  the  structures  of  the  nervous 
system  must  all  be  derived  by  modification 
of  the  selachian  plan,  seems  less  likely  than 
the  above  hypothesis  which  predicates  no 
such  close  relationship  between  the  two 
extant  groups. 

The  Skeletal  System 

Extant  holocephalians,  like  selachians, 
have  a  skeleton  constructed  entirely  of  car- 
tilage, their  only  hard  parts  being  isolated 
placoid  scales  and  the  large  dorsal  fin  spine. 
In  the  Jurassic  forms  Squaloraja  and  Myria- 
canthus,  ho\\'ever,  the  dermal  elements  are 
more  extensive.  The  fact  that  there  is  a 
greater  amoimt  of  hard  tissue  in  extinct 
holocephalians  than  there  is  in  modern  ones 
gives  added  support  to  the  idea  that  the 
cartilaginous  skeleton  characteristic  of  the 
Chondrichthyes  is  not  the  primitive  verte- 


168 


BuUetin  Museum  of  Comparative  Zoo/ogr/,  Vol.  135,  No.  3 


brate  framework  but  a  secondary  develop- 
ment. Although  this  hypothesis  does  not 
disallow  the  possibility  that  the  Holocephali 
split  from  the  selachian  line  after  the  dis- 
appearance of  bone,  it  invites  one  to  specu- 
late that  the  reduction  to  cartilage  may  have 
occurred  in  the  two  lines  separately. 

Comparative  studies  of  the  skeleton  pro- 
vide some  evidence  which  can  be  used  in 
trying  to  determine  where  the  holoce- 
phalian  and  selachian  fishes  diverged,  but 
one  feels  the  lack  of  sufficient  fossil  data  at 
every  turn.  Fossils  of  early  sharks  are  not 
abundant  and  among  those  which  have 
l)een  studied,  there  is  not  one  whose  char- 
acteristics suggest  that  it  might  have  served 
as  an  ancestor  for  the  holocephalian  line. 
Moy-Thomas  ( 1936 )  has  offered  the  coch- 
liodont,  Helodiis  simplex,  as  an  ancestral 
type,  but  the  bradyodonts  are  themselves 
distant  from  selachians.  Watson  (1938)  and 
0rvig  ( 1962 )  have  both  suggested  that  the 
fiolocephali  have  been  derived  from  pty- 
ctodonts,  and  thus  they  take  the  stand  that 
holocephalians  have  never  shared  the  sela- 
chians' evolutionary  pathway.  A  review  of 
the  holocephalian  skeleton  can  at  best,  then, 
only  attempt  to  define  the  degree  of  simi- 
larity between  it  and  that  of  selachians  and 
can  try  to  determine  whether  a  relationship 
to  ptyctodonts  or  cochliodonts  is  possible 
only  where  comparable  structures  ha\'e  been 
preserved. 

A  review  of  the  studies  of  the  skull  shows 
that  relatively  few  workers  have  tackled  the 
head  skeleton  in  its  entirety.  Only  Allis 
(1917,  1926),  DeBeer  and  Moy-Thomas 
(1935),  and  Holmgren  (1942b),  have 
looked  much  beyond  the  labial  cartilages. 
In  making  their  more  inclusive  studies,  they 
complain  of  the  lack  of  data  concerning  em- 
bryonic development:  literally  nothing 
exists  except  the  examination  of  a  few 
embryos  by  Dean  ( 1906)  and  Schauinsland 
( 1903).  It  is  a  pity  that  the  breeding  places 
of  these  laboratory-shy  fishes  are  not  well- 
known,  for  a  careful  review  of  a  series  of 
embryos  from  the  earliest  stages  would  go 


far  toward  settling  some  of  the  questions 
which  Allis,  DeBeer,  and  Holmgren  raise. 

The  first  of  these  questions  concerns  the 
developmental  interrelationships  of  the  eyes 
and  the  cranial  cavity.  As  has  been  men- 
tioned before,  Kappers  felt  that  the  depres- 
sion of  the  telencephalic  space  occurred  be- 
cause of  the  dorsomedial  expansion  of  the 
eyes.  With  this  conclusion  Holmgren  would 
agree.  Holmgren  surmises  from  this  point 
that  the  ancestors  of  the  Holocephali  must 
ha\'e  been  slightly  flattened  fonns  with 
rather  dorsally  placed  eyes.  Otherwise, 
Holmgren  reasons,  it  would  not  be  likely 
that  expansion  of  the  orbits  would  force 
the  brain  downward.  It  follows,  in  Holm- 
gren's thinking,  that  even  a  more  broad- 
headed  cochliodont  than  Helodus  would  be 
a  likely  ancestor  for  the  holocephalians. 

In  speaking  of  the  structure  of  the  cranial 
cavity,  both  Holmgren  and  Allis  take  issue 
with  the  opinion  of  DeBeer,  Moy-Thomas, 
and  Watson.  The  latter  workers  believe 
that  the  cranial  cavity  does  not  include  the 
passage  known  as  the  ethmoid  canal, 
through  which  the  superficial  ophthalmic 
nerves  run  forward  after  leaving  the  orbits. 
DeBeer  and  Moy-Thomas  ( 1935 )  state  that 
this  canal  is  roofed  over  by  a  dorsal  exten- 
sion of  the  orbito-nasal  lamina  beyond  the 
true  cranial  roof,  and  they  present  a  series 
of  drawings  of  hypothetical  evolutionary 
stages  from  the  uncovered  to  the  covered 
condition  of  this  supracranial  space.  In  his 
pul)lication  of  1936  in  which  he  presents 
the  case  for  the  descent  of  the  Holocephali 
from  the  cochliodont  Helodus  simplex,  Moy- 
Thomas  stresses  the  fact  that  Helodus  al- 
ready shows  a  dorsally-flared  orbito-nasal 
element. 

Allis  and  Holmgren  both  hold  that  since 
the  ethmoid  canal  is  continuous  with  the 
cranial  space,  it  is,  therefore,  a  part  of  it. 
Allis  ( 1926)  suggests  that  the  cranial  space 
anterior  to  the  orbits  was  cut  off  indirectly 
through  the  pressure  of  a  mysterious  em- 
bryonic "vesicle'  which  appears  between 
the  midbrain  and  the  forebrain.  As  the  fore- 
brain  is  pressed  downward,  the  trabeculae 


Morphology  and  Relationships  of  Holocephali  •  StaJiI        169 


are  squeezed  outward,  eventually  rising  up 
and  inward  to  cut  the  cranial  cavity  in  two. 
According  to  Allis'  theory,  the  trabeculae 
form  the  floor  of  the  ethmoid  canal  and  the 
roof  of  the  telencephalic  enclosure.  The 
floor  under  the  telencephalon  is  composed 
of  intertrabecular  tissue  with  perhaps  a  con- 
tribution from  the  fused  palatoquadrate. 

Holmgren's  interpretation  is  based  more 
on  anatomical  examination  and  less  upon 
flights  of  fancy.  In  his  study  of  the  heads 
of  fishes  (1942b),  Holmgren  presents  photo- 
graphs of  six  transverse  sections  through 
the  orbital  region  of  Chimaera  monstrosa 
in  which  he  points  out  a  fine  channel,  run- 
ning from  the  main  cranial  space  forward 
over  the  interorbital  septum  to  open  into 
the  ethmoid  canal.  He  remarks  that  in 
Rhinochimaera,  which  seems  to  be  the  most 
primitive  holocephalian,  this  channel  is 
much  wider,  making  even  clearer  the  con- 
tinuity of  the  two  spaces.  To  explain  the 
presence  of  a  floor  to  the  ethmoid  canal 
which  divides  it  from  the  telencephalic 
space  beneath,  he  suggests  that  this  carti- 
lage may  be  a  neomorph,  citing  its  very 
late  chondrification  as  shown  by  Schauins- 
land's  study  (1903)  of  a  CallorJiyncJius  em- 
bryo. He  goes  on  to  hypothesize  that,  as  in 
some  sharks,  the  superficial  ophthalmic 
nerves  of  the  ancestors  of  the  Holocephali 
may  ha\"e  run  in  left  and  right  preorbital 
canals  \\'hose  lateral  \\'alls  were  formed  by 
extensions  of  the  supraorbital  crest  carti- 
lages. Just  as  these  nerves  of  Pristiophonis 
come  inside  the  cranial  space  intennittently 
because  of  deficiencies  in  the  walls  medial 
to  the  preorbital  canals,  so  in  the  Holoceph- 
ali, through  complete  disappearance  of  the 
cranial  walls  in  this  area,  the  two  preorbital 
canals  may  have  merged  with  the  anterior 
brain  cavity.  It  is  by  such  a  change,  accord- 
ing to  Holmgren,  that  the  superficial  oph- 
thalmic nerves  may  have  come  to  run  \\'ithin 
what  he  considers  to  be  the  anterior  part  of 
the  cranial  cavity  in  the  holocephalians.  In 
disavowing  the  existence  of  an  ethmoid  ca- 
nal as  a  unique  holocephalian  character, 
Holmgren   removes   one   of   the   structures 


endl.d. 


Fig.  3.  Rhinochimaera  pacifica.  Neurocranium,  lateral  view. 
Cranial  cavity  with  brain  outlined.  anf.cer.v.,  Anterior 
cerebral  vein;  com.c,  communicating  channel  between 
cranio!  cavity  and  ethmoid  canal;  cr.cov.,  cranio!  cavity; 
endl.d.,  endolymphatic  duct;  eth.c,  ethmoid  conol;  int. orb. - 
sept.,  interorbital  septum;  oH.b.,  olfactory  bulb;  olf.tr., 
olfactory  tract;  orb.o.,  orbital  artery;  sup.oph.n.,  super- 
ficial ophthalmic  nerve;  tel.,  telencephalon;  //,  optic  nerve. 
(After  Holmgren.) 


upon  which  DeBeer  and  Moy-Thomas 
leaned  heavily  in  associating  the  Holoceph- 
ali with  the  cochliodonts. 

On  the  basis  of  what  has  been  said  about 
the  interrelationship  between  the  eyes  and 
the  cranial  cavity,  it  becomes  plausible  to 
conclude  that  the  chondrocranium  probably 
surrounded  a  brain  space  of  quite  ordinary 
dimensions  in  the  ancestors  of  the  Holo- 
cephali and  that  the  enlarging  eyes  press- 
ing an  interorbital  septum  between  them 
gradually  reduced  the  median  cavity  to  its 
present  divided  condition.  If  this  reason- 
ing is  correct,  the  Holocephali  must  have 
long  been  upon  a  separate  evolutionary 
pathway,  leading  from  a  form  like  Wiino- 
chimoera  to  one  like  Callorhynchus  and  fi- 
nally to  the  chimaerids.  This  sequence  of 
e\olution  is  supported  by  the  fact  that  the 
interorbital  septum  in  Chimaera  is  even 
more  extensive  than  that  of  CaUorhijnchus 
( Hubrecht,  1877 ) .  One  would  expect  the 
area  of  the  septum  to  be  largest  in  the 
group  which  shows  the  greatest  median 
expansion  of  the  eyeball. 


170        BtiUetin  Miiseinii  of  Comparotive  Zoologtj,  Vol.  135,  No.  3 


The  otic  region  of  the  holocephalian 
braincase  is  short,  the  ear  capsule  being 
pressed  close  against  the  back  of  the  orbit. 
Many  selachians  also  show  a  relatively 
short  otic  portion  of  the  skull.  If,  however, 
the  selachians  are  descended  from  primitive 
fomis  with  an  elongate  otic  and  occipital 
region  as  Romer  believes  (Romer,  1964)  it 
becomes  less  probalile  that  the  holoce- 
phalian fishes  di\  erged  from  early  selachian 
stock.  The  alternative  suggestion,  that  they 
diverged  from  the  shark  line  after  shorten- 
ing of  the  otic  region  had  occurred,  places 
the  origin  of  holocephalians  very  late,  per- 
haps in  the  Permian  or  even  in  the  Triassic 
period.  If  Dean  (1904)  is  correct  in  his 
identification  of  Menaspis — a  Pemiian  form 
apparently  not  in  the  selachian  line — as  an 
early  chimaerid,  it  would  be  better  to  seek 
a  separate  ancestral  group  for  the  Holo- 
cephali  among  the  Devonian  placoderms 
in  which  the  posterior  part  of  the  skull  was 
already  short.  0rvig  ( 1962 )  suggests  the 
pt\'ctodonts  as  such  a  group.  In  particular, 
he  describes  the  ptyctodont  CtcntircJki  as 
possessing  a  short  otic  region  set  behind 
large  orbits.  Since  the  ethmoid  region  of 
CteniircJJa  slopes  downward  anterior  to  the 
eyes,  the  general  form  of  the  skull  does  bear 
a  resemblance  to  that  of  the  holocephalians. 
The  holocephalians  are  unlike  the  sela- 
chians in  having  no  cartilage  wall  separat- 
ing the  otic  from  the  cranial  cavity.  Fossil 
remains  are  not  sufficiently  abundant  to 
indicate  whether  the  presence  of  a  parti- 
tion was  primary,  but  Stensio  (1963)  .states 
that  in  the  arthrodire  Kiijdanoiiias))i.s  the 
two  cavities  were  separated  b\'  a  thick  wall. 
If  the  condition  in  Kujdanoickispis  was  the 
general  one  in  arthrodires  as  it  is  in  modem 
selachians,  one  must  assume  that  the  con- 
fluence of  the  ear  and  brain  cavities  in  the 
Holocephali  is  a  secondary  development. 
Although  some  teleosts  show  a  confluence, 
it  seems  that  in  each  group  the  modifica- 
tion arose  separately. 

Another  characteristic  of  the  posterior 
end  of  the  braincase  which  is  very  probably 
secondarv  is  the  consolidation  \\'ith  the  oc- 


cipital region  of  two  or  three  vertebral  ele- 
ments. Rays  and  also  durophagous  fishes 
( with  disproportionate  development  of  the 
head  and  shoulder  region )  show  a  fusion 
of  skull  and  vertebral  elements  similar  to 
that  of  the  Holocephali.  Such  a  modifica- 
tion has  surely  obscured  the  ancestral  con- 
dition in  both  groups  of  fishes. 

In  comparing  the  braincase  \\  ith  that  of 
sharks,  Holmgren  mentions  particularly  the 
location  of  foramina.  The  entryway  into 
the  orbit  for  the  superficial  ophthalmic  nerve 
is  separate  from  that  of  the  other  nerves  in 
both  groups,  as  Moy-Thomas  also  claims  it 
to  be  in  the  cochliodont  HcJodtis.  The  fora- 
men for  the  entrance  of  the  internal  carotid 
artery,  however,  is  farther  forward  in  the 
Holocephali.  The  hypophysis  grows  ven- 
trally  into  a  depression  that  has  an  open 
passage  in  its  floor  in  the  Holocephali,  and 
the  notochord,  \\'hich  runs  toward  it  in  the 
base  of  the  cranium,  lies  in  a  groove  rather 
than  being  completely  embedded  in  the 
cartilage  as  in  sharks.  The  position  of  the 
notochord  shows,  in  fact,  not  only  a  dif- 
ference from  that  of  sharks,  but  also  a 
similarity  to  that  of  certain  arthrodires 
described  by  Stensio  ( 1963 ) .  In  his  recon- 
struction of  KujcJanowiaspis  he  shows  the 
notochord  lying  in  a  groove  on  the  dorsal 
surface  of  the  cranial  floor  as  it  does  in  the 
holocephalians.  This  similarity  to  the  arthro- 
dires (if  it  proves  to  be  general)  and  con- 
trast to  the  selachians  would  favor  the  idea 
that  the  Holocephali  have  direct  arthrodiran 
connections. 

The  holocephalians  also  differ  from 
sharks,  Recent  and  fossil,  in  the  possession 
of  a  palatoquadrate  cartilage  which  is  fused 
with  the  chondrocranium  rather  than  being 
suspended  in  amphistylic  or  hyostylic  fash- 
ion. Since  autostyly  is  known  in  extant  fish 
only  in  dipnoans,  the  palatoquadrate  in  the 
Holocephali  represents  a  remarkable  de- 
parture from  the  piscine  plan.  DeBeer, 
Moy-Thomas,  and  Holmgren  have  each  de- 
voted attention  to  the  holocephalian  palato- 
quadrate, and,  doubtless  due  to  the  scarcity 
of  the  embrvonic  material  available  to  them. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        171 


their  opinions  as  to  its  nature  are  divided. 
DeBeer  and  Moy-Thomas  ( 1935 )  see  it  as 
an  elongate  structure  extending  posteriorly 
to  the  mandibular  joint  b\'  an  otic  process 
which  fuses  to  the  cranium  against  the  ear 
capsule.  Holmgren  ( 1942b )  recognizes  the 
cartilaginous  lamina  between  jaw-joint  and 
ear  region  but  is  not  so  sure  that  it  is  truly 
a  part  of  the  palatoquadrate.  It  may  be, 
he  feels,  a  separate  cartilage  in  the  early 
embryo.  If  it  is,  then  the  Holocephali  are 
characterized  by  a  very  short  palatoquad- 
rate element,  and  only  fossil  fishes  having 
a  short  structure  should  be  sought  as  pos- 
sible ancestral  stock.  One  might  regard 
both  the  ptyctodonts  and  the  cochliodont 
Hclodus  as  having  a  short  palatoquadrate 
if  that  element  terminated  at  the  jaw-joint 
and  the  extension  to  the  otic  region  devel- 
oped separately,  as  Holmgren  believes  it 
does  in  holocephalians. 

Even  though  the  complete  fusion  of  the 
palatoquadrate  to  the  chondrocranium 
seems  so  distinctive  a  feature,  Holmgren 
minimizes  the  distance  that  this  fusion  puts 
between  holocephalians  and  sharks.  He 
points  out  that  in  shark  embryos  the  palato- 
quadrate is  connected  to  the  trabeculae  by 
a  membrane,  parts  of  which  chondrify.  One 
of  the  chondrifications  attaches  itself  to  the 
palatoquadrate  as  the  orbital  process;  an- 
other fuses  to  the  trabeculae  to  fonn  the 
subocular  shelf.  He  asks  whether,  if  the 
entire  membrane  should  chondrify,  firmly 
uniting  the  palatoquadrate  with  the  cra- 
nium, this  process  would  be  so  far  from 
what  occurs  in  selachians.  A  glance  at 
Schauinsland's  ( 1903 )  ilkistration  of  the 
developing  skull  of  the  65  mm  CaUorhijn- 
chus  embryo,  however,  suggests  that  Holm- 
gren's speculation  here  may  be  wide  of  the 
mark.  At  that  stage,  true  cartilage  already 
extends  from  the  palatoquadrate  area  near 
the  mandibular  joint  upwards  to  include 
the  lower  half  of  the  orbit.  The  region  in 
front  of  this  smooth  mass  of  cartilage,  as 
well  as  that  of  the  problematic  extension 
to  the  otic  capsule,  is  still  in  precartilaginous 
form.    It  seems  that  if  the  holostyly  of  the 


sup  oph  n 


rpm 


Fig.  4.  Callorhynchus  antarcticus.  Skull  of  60mm  embryo, 
lateral  view.  True  cartilage,  stippled;  precartilage,  white, 
mand..  Mandible;  pq.,  palatoquadrate;  rpm.,  medial  rostral 
process;  rpp.,  paired  rostral  processes;  sup. oph. n.,  super- 
ficial  opfithalmic    nerve.     (After  Schauinsland.) 


Holocephali  was  developed  through  further 
chondrification  of  a  shark-like  arrangement 
of  the  palatoquadrate  and  the  trabecula, 
some  indication  of  the  separate  nature  of 
these  elements  should  appear  in  this  early 
stage  before  the  palatoquadrate  is  devel- 
oped completely.  On  the  other  hand,  it  is 
possible  that  the  developmental  step  for 
which  we  are  searching  has  been  gradually 
abbreviated  to  the  point  of  disappearance. 
Thus,  there  are  two  alternatives:  either  the 
separate  palatoquadrate  never  existed  even 
as  an  embryonic  structure  in  fishes  at  the 
holocephahan  level,  or  its  development  was 
suppressed  later  as  the  line  evolved  to  the 
present  day.  Both  of  these  answers  imply 
an  evolutionary  path  long  separate  from 
that  which  led  to  modem  sharks. 

It  is  not  inconceivable  that  holocephalians 
might  ha\e  inherited  their  autostyly  from 
an  earlier  gnathostome  group.  That  condi- 
tion was  evolved,  according  to  Stensio 
( 1963 )  in  several  groups  of  arthrodires  and 
apparently  was  not  a  rare  occurrence.  0rvig 
admits,  however,  that  in  CtenurcUa  (the 
ptyctodont  that  he  regards  as  being  closely 
allied  to  the  holocephalian  line )  the  palato- 
quadrate was  not  fused  to  the  neurocra- 
nium.  Moy-Thomas,  in  advocating  a  coch- 
liodont ancestor  for  the  Holocephali,  points 
to  the  autostylic  suspension  of  the  palato- 


172        BuUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


quadrate  of  Helodus  as  an  important  simi- 
larity between  that  fonn  and  the  holoce- 
phaHans. 

A  second  pecuHarity  of  the  visceral  arch 
skeleton  in  holocephalians  concerns  the  dor- 
sal portions  of  the  hyoid  arch.  As  one  might 
expect  from  the  autostylic  suspension  of  the 
palatoquadrate,  no  part  of  the  second  arch 
is  enlarged  as  a  hyomandibular.    The  epi- 
hyal  and  pharyngohyal  resemble  their  serial 
homologues    in    the    successive    branchial 
arches.    The  question  arises  with  regard  to 
these    elements    of   the    hyoid    arch   as    to 
whether   their   state   is   truly    primitive    or 
whether  they  have  been   reduced   from  a 
specialized,  suspensory  condition  to  mimic 
the    simple    arrangement    of   the   posterior 
arches.   Holmgren  agrees  with  DeBeer  and 
Moy-Thomas  in  regarding  the  non-suspen- 
sory condition  of  the  hyoid  arch  as  primi- 
tive.   DeBeer  and   Moy-Thomas   have   ex- 
amined   the    holocephalian   hyoid    arch    in 
detail.    In  their  opinion  they  have  located 
all  its   parts,   including  the   pharyngohyal, 
and  judge  it  to  be  unmodified.   They  argue 
against  the  possibility  that  any  portion  of 
the  hyoid  could  be  fused  to  the  cranium 
and  so  lost  to  view.   To  make  this  supposi- 
tion, one  would  have  to  allow  the  migration 
of  the  cartilage  dorsal  to  the  lateral  head 
vein,  leaving  all  the  other  visceral  cartilages 
properly  ventral  to  it.    Judging   from   the 
unanimity  of  opinion  amongst  these  anato- 
mists, then,  it  would  appear  that  a  non-sus- 
pensory hyoid  is  one  of  the  primitive  char- 
acteristics   that    the    holocephalians    have 
carried   in   their   hereditary   baggage    from 
early    gnathostome    times.      As    an    early 
gnathostome  source  for  the  non-suspensory 
hyoid  arch,  Watson  would  have  offered  the 
placoderms    generally,    since    he    believed 
them  to  be  aphetohyoidean.    Stensio,  how- 
ever, is  of  the  opinion  that  the  early  placo- 
denns   possessed   a   suspensory   hyoid   and 
that  the  hyomandibular  was  reduced  to  a 
non-suspensory    bar    in    groups    in    which 
autostyli.sm    developed.     It   would   not    be 
feasible,  according  to  Stensio's  inteqoreta- 
tion,  to  seek  a  placoderm  ancestor  for  the 


Holocephali  if  their  hyoid  arrangement  is 
truly  a  primary  one. 

One  would  expect,  in  a  fish  with  an  ele- 
mentary hyoid,  to  find  a  full  gill  slit  an- 
terior  to   the   hyoid   arch   like  that  which 
Watson     predicated     for     aphetohyoidean 
placoderms.    Although  a  slit  does   appear 
in  the  embryo,  it  is  dorsally  placed  and  soon 
disappears.    The  space  between  the  hyoid 
and  the  mandible  is  later  crossed  by  three 
ligaments:    not  only  is  the  spiracle  absent, 
then,  but  the  area  has  been  completely  re- 
built.   It  is  probable  that  this  change  is  a 
modification   connected   with   the   forward 
displacement  of  the  visceral  skeleton  as  a 
whole  and,  one  might  add,  of  the  pectoral 
girdle  behind  it.    The  palatoquadrate  is  set 
far  forward  and  fused  to  the  cranium,  as 
we  have  seen;  the  hyoid  is  close  behind  the 
mandible  and  firmly  tied  to  it  by  the  above- 
mentioned    ligaments;    and   the    remaining 
five  arches  are  crowded  up  under  the  pos- 
terior end  of  the   cranium.    The  last  two 
pharyngobranchials   and   epibranchials   are 
squeezed  to  a  fusion  with  each  other,  cre- 
ating a  small  flat  disc  against  which  the 
scapula  abuts.    The  entire  gill  apparatus  is 
reduced  and  covered  by  an  operculum.  This 
arrangement  of  the  visceral  skeleton  con- 
trasts  sharply   with   that   of   extant   sharks 
which  have  five  arches,  or  in  the  notidanids 
and  ChlaimjdoscJachus  more  than  five,  in 
an  extensive  pharyngeal  region.  Fossil  forms 
with  a  short  pharyngeal  region  (and  bran- 
chial arches  crowded  forward  beneath  the 
posterior  end  of  the  braincase)   did  exist, 
and  might  be  a  more  logical  choice  as  a 
group  ancestral  to  the  Holocephali  than  the 
early  sharks.    The  ptyctodonts  have  been 
figured  by  Watson    (1938)    and  by  0rvig 
( 1962 )    as   having  only   a   small   branchial 
area,    and    Moy-Thomas    ( 1936 )    describes 
the  cochliodont  Helodus  as  having  the  pec- 
toral apparatus  set  close  behind  the  head. 
The  possession  of  a  single  median  rostral 
cartilage  also  distinguishes  the  Holocephali. 
In  Cltimacra  the  rostral  cartilage  is  short; 
in  Callurhynchus  it  is  longer  and  bent  ven- 
trally;  in  RhinucJiimaera  it  is  longest  and 


Morphology  and  Relationships  of  Holocephali  •  Stalil 


173 


Fig.   5.      He/odus  simplex.     Restoration  of  fish,   lateral  view.   (After  Moy-Thomas.) 


extends  directly  forward.  Carman  ( 1904 ) 
believes  that  the  longest  cartilages  are  the 
most  primitive.  This  supposition  seems 
reasonable  as  Rhinochimaeni,  with  the 
longest  rostrum,  also  shows  several  other 
characters  in  what  is  apparently  their 
earliest  form.  The  fossil  holocephalians 
Sqiialoraja  and  Myriacantluis  show  well 
developed  rostral  cartilages,  the  former 
exhibiting  some  calcification  of  the  element. 
There,  however,  the  trail  ends  mysteriously. 
Ischyodiis,  another  extinct  form,  is  figured 
by  Dean  ( 1895 )  as  having  a  short,  blunt 
head,  and  the  earlier  possible  ancestors,  the 
cochliodont  HcJodus  and  the  ptyctodont 
RJiampJiodopsis,  are  not  known  to  have 
possessed  rostral  structures.  CtemireUa, 
according  to  0rvig,  has  a  pair  of  rostral 
processes  but  not  a  medial  one.  It  may  be 
that  such  structures  were  not  preserved, 
but  in  any  case  the  rostral  cartilages  cannot 
now  be  used  as  Ariadne's  thread  to  reach 
the  light. 

Holocephalians,  like  sharks,  have  paired 
labial  cartilages.  Howe\'er,  in  their  number 
and  form  the  labial  cartilages  differ  from 
the  simple,  slim  bars — an  upper  and  a  lower 
one  on  each  side — which  meet  at  the  angle 
of  the  jaw  in  selachians.  At  the  mouth 
angle  in  holocephalians,  on  each  side,  there 
are  two  labial  cartilages  which  meet,  but 
the  small  superior  maxillary  element  and 
the  larger,  flattened  inferior  maxillary  are 
often  fused  in  the  adult.  Against  the  an- 
terior end  of  the  lower  jaw  there  may  be  a 


premandibular  labial  cartilage  (it  is  absent 
in  Chimaeni  coUici);  beside  the  upper  jaw 
there  are  always  a  large  prelabial  and  a 
smaller  premaxillary  element. 

The  labial  cartilages  of  the  Holocephali 
were  studied  for  two  reasons.  Comparative 
anatomists  examined  them  hopefully  as  pos- 
sible clues  to  the  history  of  the  descent  of 
modem  chimaeroids,  and  workers  inter- 
ested in  the  transition  from  agnathous  to 
gnathous  fish  sought  in  them  the  remains  of 
the  premandibular  visceral  arches.  Despite 
the  descriptions  given  bv  Allis  ( 1926),  Dean 
( 1906),  Carman  (1904),' Holmgren  ( 1942b), 
Hubrecht  (1877),  Luther  ( 1909),  and  others, 
the  significance  of  these  cartilages  has  not 
been  surely  decided.  Their  early  fossil  record 
is  dubious.  0rvig  finds  some  in  CtenureUa 
which  he  thinks  resemble  those  of  holo- 
cephalians rather  than  those  of  sharks. 
Holmgren  suggests  that  they  might  be  rep- 
resented in  three  small  elements  in  Rliam- 
pJwdopsis  which  Watson  ( 1938 )  had  iden- 
tified as  parts  of  the  hyoid  arch.  To  the 
suggestion  that  these  elements  are  modified 
premandibular  arches  there  are  at  least  two 
objections:  firstly,  they  are  lateral  to,  rather 
than  medial  to,  the  branchial  arteries;  and 
secondly,  they  show  no  close  resemblance  in 
number  or  design  to  visceral  arches.  Only 
their  position  against  the  upper  and  lower 
jaws  argues  for  the  assumption.  Taking  these 
objections  into  consideration,  Luther  (1909: 
32)  suggests  that  "Diese  Stiickchen  stellen 
aller  Wahrscheinlichkeit  nach  einer  ciino- 


174        Biillefin  Museum  oj  Comparaiive  Zoology,  Vol.  135,  No.  3 


0,i!iimi^'nm§ 


Fig.  6.  Reconstruction  of  the  head  and  shoulder  girdle  of  two  ptyctodonts;  lateral  view.  A,  Rhamphodopsis  trispinofus 
Watson;  B,  Ctenurella  glodbochens/s  0rvig.  o.d.,  Dorsal  arcualio;  a. v.,  ventral  arcualia;  has.,  basal;  bro.,  branchial  arch; 
c.hy.,  ceratohyal;  d.sp.,  dorsal  spine;  ep.hy.,  epihyal;  lab. cart  ,  labial  cartilage;  Mk.,  Meckel  s  cartilage;  m.tp.,  nnandib- 
ular  toothplate;  n.c,  notochord;  ph.hy.,  pharyngohyal;  pq.,  palatoquadrate;  pq./p.,  upper  toothplate;  rpm.,  medial  rostral 
process;   rpp.,   paired   rostral   processes;   sc.co.,  scapulocoracoid  ossification;   sp.,   spinale.     (A   after  Watson;    B  after  0rvig.) 


genetischen  Erwerb  dar,  der  speciellen 
mechanischen  Bediirfnissen  entsprang."  All 
that  can  really  be  said  with  certainty  is  that 
the  laliial  cartilages  are  quite  different  from 
those  of  sharks  in  their  number  and  fonn, 
and  in  having  muscles  inserted  upon  them, 
and  that  their  present  condition  implies  a 
long,  separate  evolution. 

The  remainder  of  the  axial  skeleton  is 
very  much  simpler  to  analyze  than  the  skull, 
but  no  more  directly  indicative  of  the  holo- 
cephalians'  ancestry.  The  vertebral  column 
presents  certain  distinctive  characteristics 
which  may  be  listed  in  a  straightforward 
manner.  Anteriorly,  it  is  consolidated  rad- 
ically— not  only  are  the  first  two  or  three 
vertebrae  fused  with  the  cranium,  but  the 
first  seven  elements  posterior  to  the  occip- 
ital articulation  are  broadly  fused  with  each 
other  to  support  the  strong  dorsal  spine  and 
an  accompanying  basal  fin-cartilage.  True 
centra  are  never  present;  in  CaUoilu/ncJiu.s 
the  notochord  is  unconstricted;  in  Wiino- 
chimaera  and  Chirnaero  cartilaginous  rings 
develop  within  the  notochordal  sheath. 
Rabinerson  (1925),  who  studied  the  com- 
parative anatomy  of  the  vertebrae  of  carti- 


laginous fishes,  was  of  the  opinion  that  the 
Holocephali  were  distinct  from  the  selachi- 
ans in  the  development  of  these  elements. 
He  recognized  that  the  supra-  and  hypo- 
chordal  arch  elements  of  holocephalians 
bore  a  greater  resemblance  to  those  of 
selachians  than  to  those  of  bony  fish,  but 
still  he  held  that  the  similarity  was  due 
to  convergence  rather  than  to  close  relation- 
ship. Although  the  holocephalians  have 
retained  the  primitively  unconstricted  noto- 
chord and  in  some  forms  surrounded  it  with 
a  variable  number  of  skeletal  rings,  they 
share  with  the  selachians  the  tendency  to 
develop  arches  and  intercalary  arches  above 
and  below  it.  If  Rabinerson  is  correct  in 
his  opinion  that  the  location  of  the  foramina 
for  the  spinal  nerves  differs  in  holocepha- 
lians and  in  sharks  and  that  the  arch  units 
in  the  two  types  of  cartilaginous  fishes  have 
been  laid  down  in  a  different  arrangement, 
it  would  seem  likely  that  the  holocephalians 
and  selachians  evolved  separately  from  a 
group  in  which  only  the  general  nature  of 
the  arch  elements  was  defined. 

Among  the  fossil  forms  which  have  been 
suggested  as  belonging  at  the  base  of  the 


Morphology  and  Relationships  of   Holocephali  •  Stahl        175 


holocephalian    line,    only    the    ptyctodonts  characterizes  the  bony  fishes,  too,  it  may 

seem    to    show    any    vertebral    structures  have  been  a  common  occurrence  in  early 

which  bear  a  special  resemblance  to  holo-  gnathostomes.    Watson   describes   a  pelvic 

cephalian    characters.     0rvig    ( 1962 )    has  girdle    of    this    type    in    the    ptyctodont 

stated  that  CtenureUa  possessed  a  fusion  of  Rhamphodopsis  and  Moy-Thomas  also  at- 

arch   elements    and   an    enlarged   basal   of  tributes  such  a  structure  to  Helodiis.    The 

the   dorsal   fin.    Although   Watson    (1938)  holocephalian  girdle  also  contrasts  with  that 

did  not  find  a  synarcual  in  R]iomp]wdo})sis,  of  selachians  in  developing  a  strong  dorsal 

he  does   figure   an   element  which  he   be-  process   and   foramina   for  the   passage   of 

lieved  to  be  the  enlarged  basal.  nerves  to  the  fin.    One  cannot  seize  upon 

In  reviewing  the  appendicular  skeleton  of  these  differences   as   demonstrating  a   sig- 

the  Holocephali,  one  is  struck  first  by  the  nificant    separation    of    the    holocephalian 

fact  that  its  general  structure  is  similar  to  line,   however,   as   Dean    ( 1909 )    describes 

that  of  selachians  and  quite  unlike  that  of  in    Cladoselache,    in    the    position    of    the 

bony  fish.    In  both  t)'pes  of    cartilaginous  pelvic  girdle,  a  structure  with  separate  left 

fishes  the  pectoral  girdle  takes  the  forni  of  a  and  right  portions. 

large  and  firm  U  which  embraces  the  body  The  pterygiophores  of  holocephalians 
from  the  ventral  side.  Articulated  with  this  and  selachians,  although  similar  in  their 
girdle  and  with  the  smaller  one  in  the  pelvic  general  extent,  do  differ  from  each  other, 
region  are  basipterygia  to  which  are  attached  The  basals  of  the  former  group  are  some- 
jointed  fin  radials  that  extend  halfway  out  what  more  compact,  there  being  two  rather 
upon  the  fin.  The  remainder  of  the  fin  than  three  in  the  pectoral  fin  and  one  rather 
is  supported  by  demial  rays.  On  closer  than  two  in  the  pelvic  fin.  The  radials  in 
inspection  of  the  holocephalian  skeleton,  both  forms  are  jointed,  although  those  of 
however,  distinctive  features  do  appear,  holocephalians  show  a  tendency  to  fuse  at 
The  pectoral  girdle  is  extraordinarily  mas-  their  proximal  ends.  Males  of  both  groups 
sive  and  contains  a  pair  of  channels  within  bear  pterygiophores  modified  as  claspers. 
it  for  the  passage  of  blood  vessels.  Its  If  one  accepts  Clodoselache,  with  its  broad- 
scapular  process  extends  extremely  far  dor-  based  fins,  long,  unjointed  radials,  and 
sally.  Whether  the  form  of  this  girdle  rep-  probable  lack  of  claspers,  as  typical  of  the 
resents  a  modified  selachian  type  or  a  ancestors  of  modem  selachians,  clearly  one 
different  development  is  not  possible  to  must  derive  the  holocephalians  from  se- 
decide.  Fossil  evidence  concerning  the  lachians  later  than  Cludoselache  in  which 
deep  elements  of  the  skeleton  in  the  shoul-  the  modern  type  of  fin  was  already  estab- 
der  area  is  meager.  Moy-Thomas  believed  lished  or  predicate  a  remarkable  convergent 
that  the  pectoral  girdle  of  the  cochliodont  evolution  in  the  two  groups.  Again  fossil 
Helodus  retained  separate  left  and  right  data  is  too  scanty  to  back  either  alternative 
haKes  and  if  so  \\'Ould  not  have  evinced  the  convincingly.  Both  Watson  and  0r\dg  claim 
consolidation  characteristic  of  the  holo-  that  the  ptyctodonts  they  have  examined 
cephalian  structure.  Neither  Watson  nor  probably  possessed  claspers,  and,  consider- 
0rvig  describes  the  internal  pectoral  girdle  ing  the  wide  variety  of  clasper-designs 
of  the  ptyctodonts.  Since  the  demial  annor  among  cartilaginous  fishes  shown  by  Leigh- 
of  the  shoulder  was  elaborate,  however,  Shaq^e  (1920  ff. ),  it  is  not  impossible  to 
one  may  speculate  that  inner,  non-demial,  imagine  their  having  evolved  from  more 
skeletal  elements  were  not  extensively  than  one  source.  There  is  no  evidence  of 
developed.  pterygiophores    in    ptyctodonts,     although 

The  pelvic  girdle  differs  from  that  of  Watson  speculates  that  the  pelvic  fins  in 
selachians  in  consisting  of  separate  left  and  Rhamphodopsis  were  probably  narrow- 
right    halves.     Although    this    arrangement  based  and  freely  movable. 


176        Bulletin  Museum  of  Coniparotwe  Zoology,  Vol.  135,  No.  3 


The  conclusion  to  which  this  review  of 
tlie  skeletal  s>'stem  leads  is  twofold.  First, 
although  the  skull,  xertebral  column,  and 
appendicular  structures  of  holocephalians 
have  distinctive  features,  there  exists  a  suf- 
ficient similarity  between  the  general  plan 
of  the  holocephalian  skeleton  and  that  of 
selachians  to  suggest  that  the  two  groups 
are  related  in  some  way.  Second,  it  seems 
obvious  that  the  holocephalians  have  very 
little  in  common  with  the  bony  fishes. 
There  are  occasional  similarities — the  ab- 
sence of  a  partition  between  the  otic  and 
cranial  cavities  and  the  existence  of  sep- 
arate halves  of  the  pelvic  girdle — but  no 
really  firm  basis  exists  for  postulating  a 
relationship  between  the  two  lines. 

The  nature  of  the  relationship  between 
holocephalians  and  selachians  demands 
analysis.  Certainly,  the  cartilaginous  nature 
of  the  skeleton  in  both  is  a  factor  to  be 
considered,  but  the  possibility  of  its  having 
been  evolved  separately  removes  the  obli- 
gation to  derive  the  holocephalians  from 
an  already  established  selachian  line.  It  is 
not  necessary  to  adhere  to  the  improbable 
theory  that  the  holocephalian  braincase, 
with  its  downward-sloping  ethmoid  and 
short  otic  regions,  was  derived  from  the 
early  selachian  chondrocraniiun.  If  the  non- 
suspensory  hyoid  is  truly  primitive,  a  non- 
selachian  origin  for  it  must  be  sought.  If 
it  is  a  secondary  development,  the  feasi- 
bility of  its  dedifferentiation  from  the  ex- 
panded selachian  hyomandibular  is  still 
questionable.  The  palatoquadrate  is  also 
different  in  its  proportions  from  the  se- 
Uichian  structure  if  the  point  of  articulation 
with  the  mandible  marks  its  posterior  limit. 
Its  fusion  to  the  braincase  seems  to  have 
been  an  early  event  rather  than  a  recent 
modification  if  its  already  cryptic  embry- 
onic development  has  any  significance. 
Finally,  labial  cartilages  are  structures  in 
the  head  which  it  is  difficult  to  visualize 
as  having  been  derived  from  their  counter- 
parts in  selachians.  Since  the  labial  carti- 
lages are  regarded  as  vestigial  in  the  latter 
group,  it  is  not  Hkely  that  they  would  have 


redeveloped  to  become  the  elaborate  ap- 
paratus of  the  holocephalians.  The  median 
rostral  cartilage  is  harder  to  assess.  The 
structure  is  unique  and  may  be  a  neomoq^h. 
The  postcranial  skeleton  of  the  Holo- 
cephali  shows  two  features  which  are  dis- 
tinct from  their  selachian  counteqoarts  and 
difficult  to  imagine  as  having  been  derived 
from  them.  The  circumchordal  elements  in 
chimaerids  may  be  independent  develop- 
ments rather  than  merely  reduced  versions 
of  selachian  centra.  The  absence  of  any 
type  of  centra  or  ring-like  structures  around 
the  notochord  in  CaUurJiijnclnis  is  possibly 
a  primitive  character.  The  same  may  be 
said  of  the  separate  halves  of  the  pelvic 
girdle  found  in  all  holocephalians. 

The  Muscular  System 

The  muscles  of  the  Holocephali  have 
been  described  by  several  investigators 
interested  in  evolutionary  relationships 
among  fishes.  Maurer  (1912)  made  a  sur- 
vey of  trunk  musculature,  whereas  Edge- 
worth  (1935),  Kesteven  (1933),  Shann 
(1919),  and  Vetter  (1878)  confined  their 
attention  to  the  muscles  of  the  head  and 
shoulder  regions.  Vetter  provided  the  most 
exhaustive  description  of  these  muscles  and 
assigned  names  to  them.  His  paper  is  ac- 
companied by  a  handsome  set  of  drawings 
which  are  helpful  in  interpreting  the  text. 

In  surveying,  first,  the  trunk  musculature, 
one  is  forced  to  recognize  the  similarity  of 
its  structure  in  all  fishes.  The  overriding 
demands  of  locomotion  as  perfomied  by  all 
but  a  relatively  small  number  of  specialized 
fonns  have  been  met  by  the  visibly  seg- 
mented, more  or  less  zigzagged  myotomes 
which  run  from  the  back  of  the  skull  and 
pectoral  girdle  to  the  caudal  fin.  Holo- 
cephalians share  this  general  arrangement 
of  the  trunk  muscles  with  other  fishes  but 
show  one  specialization  which  is  apparent 
as  soon  as  the  skin  is  removed:  the  anterior 
portion  of  the  ventral  hypaxial  musculature 
has  become  a  non-segmented  sheet  which 
rises  to  the  level  of  the  lateral  line,  covering 
the    more    dorsal    hypaxial    bundles.     This 


Morphology  and  Relationships  of  Holocephali  •  Stahl        177 


>  OS- 


Fig.  7.  Trunk  musculature:  anterior  part,  lateral  view.  A,  C/i/omydose/ochus  anguinens;  B,  Chimaera  monstrosa.  a,b,c,d. 
Divisions  of  hypaxial  musculature;  /.,  lateral  line;  o.in^,  inferior  oblique;  o.s.,  superior  oblique;  R.p.,  rectus  profundus;  line 
xy,    dorsal    limit   of    inferior   oblique.      (After    Maurer.] 


sheet  inserts,  as  one  would  expect,  upon  the 
pectoral  girdle.  Maurer  (1912),  who  di- 
vides the  hypaxial  muscles  into  superior 
oblique,  median  oblique,  and  inferior 
oblique  groups,  regards  the  holocephalian 
sheet  as  being  a  modification  of  the  in- 
ferior oblique  portion.  For  Maurer,  the 
state  of  the  inferior  oblique  in  the  Holo- 
cephali represents  a  more  highly  evolved 
condition  than  exists  in  any  other  carti- 
laginous fish.  In  the  arrangement  which 
Maurer  believes  is  primitive — that  seen  in 
Chlamydoselaclie  and  Heptanchus — there 
is  a  discontinuity  between  the  inferior 
oblique  and  the  median  oblique  ( line  x-y 
in  his  figures)  which  is  set  quite  far  ven- 
trally,  leaving  much  of  the  median  oblique 
visible.  In  the  course  of  evolution,  the  level 
of  the  discontinuity  rises.  The  inferior 
oblique  overlaps  the  median  oblique  and 
the  latter  is  gradually  reduced.  Maurer 
relates  this  change  to  the  growing  dom- 
inance of  the  pectoral  apparatus  to  which 
the  inferior  oblique  is  attached,  and  states 
that  the  Holocephali  represent  the  extreme 
expression  of  this  tendency.  (He  considers 
sharks  but  not  batoids.)  In  Maurer's  opin- 
ion, the  Holocephali  are  also  advanced  in 
lacking  a  ventral  rectus  muscle  of  the  sort 
that  Chlamydoseloche  shows.    That  shark 


has  the  two  most  ventral  muscle  bundles 
(c  and  d  in  Maurer's  figures)  rolled  medi- 
ally to  fonn  a  band  bordering  the  midline. 
In  the  sharks,  which  Maurer  regards  as 
more  highly  developed,  and  in  holocepha- 
lians  this  band  does  not  appear.  Through- 
out his  paper,  Maurer  emphasizes  the  pro- 
gression from  primitive  selachians  to  Holo- 
cephali. It  is  clear  that  he  regards  this  pro- 
gression as  having  taken  place  separately 
from  the  evolution  of  the  bony  fishes. 

Shann  (1924)  noted  that  fibers  of  the 
trunk  musculature  of  fishes  are  diverted  to 
hold  the  pectoral  girdle  in  place.  Although 
Shann  doubts  that  it  is  possible  to  draw 
homologies  between  the  various  shoulder 
muscles  with  absolute  accuracy  in  every 
case,  he  does  see  a  basic  likeness  between 
the  muscles  of  holocephalians  and  elasmo- 
branchs.  Shann  points  out,  however,  that 
the  shoulder  muscles  of  the  Holocephali 
show  a  far  greater  differentiation.  In 
sharks,  the  scapular  process  is  held  firm  by 
the  antagonistic  action  of  the  hypaxial  mus- 
cles and  the  cucullaris.  The  former  insert 
upon  the  posterior  border  of  the  scapular 
cartilage  and  the  latter  upon  its  anterior 
edge.  In  holocephalians,  however,  both  of 
these  groups  of  muscles  are  subdivided  into 
external  and  internal  portions.    The  origin, 


178        BttUefin  Mus-eiiw  of  Comparative  Zoology,  Vol.  135,  No.  3 


rd.p.         rlvp.e.      gdci. 


Fig.  8.  Chimaera  colliei.  Muscles  of  the  left  pectoral  region,  lateral  view,  add.s.,  Adductor  superficialis;  d. const. m.,  dorsal 
constrictor  muscle;  l-v.m.,  latero-ventral  muscle;  p.d.p.,  protractor  dorsalis  pectoralis;  r.d.p.,  retractor  dorsalis  pectorolis; 
r.l.v.p.e.,  retractor  latero-ventralis  pectorolis  externus;  r.p.i.,  retractor  pectoralis  superior;  scop.,  scapula;  tr.exf.m.,  trape- 
zius externus  muscle. 


insertion,  and  fiber  direction  of  each  differ 
slightly,  clearly  a  more  specialized  arrange- 
ment. Since  the  scapular  process  of  holo- 
cephalians  rises  above  the  level  of  the 
horizontal  septum,  there  are  also  epaxial 
fibers  which  insert  upon  it.  In  sharks  the 
epaxial  muscles  are  not  involved  in  the 
shoulder  musculature. 

In  contrast  to  the  more  highly  differen- 
tiated state  of  the  holocephalian  shoulder 
groups,  the  muscles  which  are  associated 
with  the  coracoid  region  may  be  simpler 
than  those  of  sharks.  The  bases  of  the 
coracobranchials  are  not  fused  into  common 
coracoarcuals  as  they  are  in  elasmobranchs. 
The  coracohyoid  muscles  actually  originate 
on  the  coracoid  cartilage  rather  than  on  the 
fascia  over  the  muscles  anterior  to  it.  These 
aspects  of  the  hypobranchial  musculature 
outweigh,  in  Shanns  mind,  the  seemingly 
special,  massive  development  of  the  cora- 


comandibularis,  and  he  emphasizes  his  im- 
pression that  the  Holocephali  are  in  these 
structural  arrangements  more  primitive  than 
the  sharks  and  rays. 

From  the  musculature  of  the  paired  fins 
few  inferences  may  be  drawn  concerning 
the  relationships  of  the  Holocephali.  Again, 
in  principle,  the  fin  muscles  of  all  fishes  are 
much  alike.  To  raise,  depress,  and  twist  the 
fins  all  that  has  proved  necessary  are  a 
dorsal  and  a  ventral  muscle  mass,  some 
fibers  of  which  are  drawn  into  the  fin 
over  an  oblique  course.  The  holocephalians 
present  l)ut  one  modification  of  the  general 
scheme.  The  proximal  portion  of  the  dorsal 
muscle  mass  associated  with  the  pectoral 
fin  is  differentiated  into  discrete  bands 
rather  than  existing  as  a  simple  sheet  of 
parallel  fibers.  The  most  superficial  band 
originates  on  fascia  at  the  level  of  the  lat- 
eral line  and  inserts  upon  the  anterior  edge 


Morphology  and  Relationships  of  Holocephali  •  StaliJ        179 


of  the  fin  through  a  small  tendoi).  From  the  comparable   to   the   quadrato-mandibularis 

girdle   another  band   of   fibers   extends   to  of  sharks  and,  anterior  to  it,  a  second  part 

the  front  edge  of  the  fin  and  another  to  the  which  is   regarded   as   homologous   to  the 

posterior   edge.    Between   the   latter   band  selachian   preorbitalis.     In   contrast   to  the 

and  the  former  two,  which  insert  anteriorly,  relative    size    of   these    muscles    in    sharks, 

the    deeper    fibers    which    cover    the    fin-  however,  the  posterior  part  of  the  adductor 

radials   lie   exposed.     The   distal    fibers   of  in  holocephalians  is  smaller  than  the  exten- 

the  dorsal  muscle  mass  are  unmodified  and  sive  preorbitalis.  The  preorbitalis  has  spread 

resemble  those  of  sharks.    A  dissection  of  upward  over  the  broad   wall   of  cartilage 

the    remaining    fin    muscles    in    either   the  created  in  front  of  the  eye,  by  the  fusion 

pectoral  or  pelvic   region   shows   that   the  of  the  palatoquadrate  cartilage  to  the  neuro- 

superficial  fibers  originate  upon  fascia  or  cranium,  and  the  development  of  the  high 

upon  parts  of  the  girdle  and  insert  upon  cartilage  wall  in  the  ethmoid  region.    The 

connective  tissue  over  the  fin  basals  and  levator  and  constrictor  elements  associated 

radials  in  the  usual  way.   The  deeper  fibers  with  the  selachian  mandibular  arch  are  not 

originate  and  insert  upon  the  fin  itself  as  present    in    holocephalians.     The    muscles 

they  do  in  sharks.  which  insert  upon  the  holocephalian  labial 

The  muscles  associated  with  the  anterior  cartilages,    however,    appear    in    no    other 

dorsal   fin    of  holocephalians   bear   special  group  of  fishes. 

mention.    They  consist  of  a  proximal  and  The  muscles  of  the  hyoid  and  successive 

a  distal  group  of  fibers  on  each  side.    The  arches     contrast     sharply    with     those     of 

proximal    muscle    mass    originates    on    the  sharks.    The    levator   fibers    in    holocepha- 

plate    formed    by    the    anterior    vertebral  Hans  are  grouped  in  external  and  internal 

fusion,  inserts  upon  the  base  of  the  dorsal  divisions,  as  was  mentioned  above,  rather 

fin  spine,  and  acts  to  elevate  the  spine.  The  than  existing  as  a  unified  cucullaris.    The 

distal    fibers    arise    from    the    broad    basal  individual  constrictor  muscles  of  the  pos- 

cartilage  of  the  fin  and  insert  at  the  base  tenor  arches,   identifiable  in   sharks,   have 

of  the  dermal  fin  rays,  allowing  the  web  disappeared.     Only    the    hyoid    constrictor 

of  the  fin  to  be  drawn  laterally.   This  com-  remains,  and  this  element  is  expanded  to 

bination    of   proximal    and    distal   muscles,  provide  the  musculature  of  the  operculum, 

which  is  not  found  in  any  other  cartilagi-  In  the  possession  of  a  hyoid  constrictor  of 

nous  fishes,  may  have  been  present  among  this    kind    and    in    the    reduction    of    the 

the  ptyctodonts  if  0rvig's  interpretation  of  musculature  associated  with  the  branchial 

the  skeletal  elements  of  Ctenurella  is  cor-  arches    covered    by    the    operculum,    holo- 

rect.    In  Ctenurella,  he  finds   a  synarcual  cephalians  bear  a  resemblance  to  the  bony 

element  beneath  the  dorsal  fin  and  a  basal  fishes.     Kesteven     (1942-1943),    who    ac- 

piece  which  could  have  served  as  sites  of  cepted    this    resemblance    as    evidence    of 

origin  for  the   proximal   and   distal   fibers,  evolutionary  relationship,  was  led  into  the 

respectively.  construction    of    an    evolutionary    scheme 

Much  more  has  been  written  about  the  which  is  untenable  in  the  face  of  recent 

musculature   of  the   head  and   pharyngeal  paleontological  findings.    It  might  be  more 

region   than   about  that  of  the   trunk  and  correct    to    suppose    that    the    similarities 

fins.    From  Vetters   ( 1878 )   description  of  which    do    exist    between    holocephalians 

the  branchial  muscles  of  the  Holocephali,  and  bony  fish  have   come   about  through 

one  sees  that  the  mandibular  arch  group  convergence. 

resembles  the  selachian   type,   lacking  the  One  could  assume,  then,  that  the  holo- 

complex  subdivision  shown  by  that  group  cephalian   branchial  musculature,  with   its 

in  bony  fishes.   The  adductor  mass  in  holo-  distinctive  specializations,  developed  in  cor- 

cephalians   consists   of  a  portion  which  is  relation  with  the  crowding  forward  and  the 


180        Btdletin  Museum  of  Comparative  Zoolof^y,  Vol.  135,  No.  3 


Fig.  9.  Chimaera  monsfroso.  Muscles  of  the  head,  lateral 
view.  C.max.,  Maxillary  cartilage;  C.plb.,  prelabial  car- 
tilage; C.pmnd.,  premandibular  cartilage;  l.a.o.a.,  levator 
anguli  oris  anterior;  l.a.o.p.,  levator  anguli  oris  posterior; 
ic.p.,  levator  of  prelabial  cartilage;  M.a.m.,  adductor 
mandibulae;  M.I. a.,  labialis  anterior  muscle;  M.l.i.,  lobiaiis 
inferior  muscle;  M.I. p.,  labialis  posterior  muscle;  M.pr., 
preorbitalis  muscle;  n.cap.,  nasal  capsule.  (Adapted  from 
Luther.) 


fusions  which  took  place  within  the  visceral 
and  cranial  skeleton  during  the  independent 
evolution  of  the  Holocephali.  As  the  gill 
arches  became  compressed  under  the  oc- 
cipital region  and  the  extrabranchial  carti- 
lages spread  to  form  an  opercular  cover, 
the  branchial  constrictor  muscles  gave  way 
in  favor  of  an  expanded  hyoid  constrictor 
sheet.  The  branchial  levators,  adductors, 
and  interbranchials  all  became  reduced  in 
accordance  \\'ith  the  reduction  and  com- 
pression of  the  cartilages  of  the  arches. 
Since  the  mandible  is  short  in  holocepha- 
lians  and  forms  only  a  shallow  curve,  the 
ventral  portion  of  the  hyoid  constrictor 
(which  reaches  the  midline  in  sharks  as 
the  interhyoideus )  apparently  shifted  the 
origin  of  its  most  anterior  fibers  forward 
to  the  connective  tissue  on  the  posterior 
ventral  edge  of  the  mandible.  There  being 
no  division  between  the  palatoquadrate  and 
the    ethmoid    region    of   the    cranium,    the 


muscles  innervated  by  the  trigeminal  nerve 
spread  over  the  entire  anterior  region  of 
the  head.  The  divisions  of  this  muscle 
which  insert  upon  the  labial  cartilages 
would  seem  to  be  late  developments.  If  the 
branchial  muscles  of  the  Holocephali 
evolved  as  suggested  here,  it  would  be 
logical  to  seek  an  ancestral  stock  in  which 
the  shortening  of  the  head  region  had 
already  begun.  The  ptyctodonts  show  such 
a  condition  and  may  thus  be  a  better  choice 
as  ancestral  material  for  the  holocephalians 
than  the  longer-headed  cochliodont  Helo- 
cJus  or  any  early  selachian. 

In    sum,    then,    one    recognizes    in    the 
muscular  system  of  the  Holocephali  a  num- 
ber of  similarities  to  the  system  of  sharks, 
many    characteristics    which    are    certainly 
specializations  peculiar  to  the  group,  and 
certain  features  which  are  comparable  to 
those    of    bony    fishes.     Among    the    holo- 
cephalian   muscles,   which   show   some   re- 
semblance  to    selachian    counterparts,    are 
the  trunk  and  fin  muscles,  the  hypobran- 
chial  muscles,  and  the  adductor  muscles  of 
the  mandibular  arch  group.    Within  each 
of  these  groups  of  muscles,  however,  some 
unique    arrangement    appears:     the   sheet- 
like, nonsegmented  inferior  obhque  among 
the    axial    muscles;    the    special    nature   of 
the  proximal  pectoral  extensors  among  the 
fin  muscles;  the  great  expansion  of  the  pre- 
orbitalis in  the  mandibular  arch  group.   Be- 
sides these  peculiarities,  the  complexity  of 
the  shoulder  musculature,  the  anterior  dor- 
sal   fin    muscles,    and    the    muscles    which 
insert  upon   the  labial   cartilages   must  be 
regarded  as  singular  and  non-selachian  in 
nature.    The  sole  resemblance  of  the  holo- 
cephalians to  the  bony  fish  lies  in  the  pres- 
ence of  an  expanded  hyoid  constrictor  and 
reduced  musculature  of  the  posterior  bran- 
chial arches.    In  assessing  this  similarity  as 
evidence  of  convergence  rather  than  rela- 
tionship, one  may  well  be  on  solid  ground. 
Estimating  the  significance  of  the  similar- 
ities between  holocephalians  and  selachians 
is  more  difficult,  however.    Since  the  mus- 
culature  of   the   holocephalians   shows   no 


Morphology  and  Relationships  of  Holocephali  •  Stahl        181 


characteristics  which  are  clearly  more  prim-  release  eggs  into  the  body  cavity,  but  the 

itive  than  those  of  any  shark — unless  the  ostium  of  the  oviduct  may  be  located  more 

absence  of  the  common   coracoarcuals  be  posteriorly  than   it  is   in  the   cartilaginous 

so  considered — the  possibility  of  its  evolu-  fishes  and  the  oviduct  itself  never  shows 

tion  from  a  generalized  selachian  pattern  the  specialized  areas  characteristic  of  the 

cannot  be  ruled  out.    On  the  other  hand,  oviducts    in    Chondrichthyes.     In    species 

the  axial  and  branchial  musculature  shows  which  are  descended  from  the  earlier  parts 

many    specializations     which    are    closely  of  the  bony  fish  line  ( Pohjpterus,  Acipenser, 

allied  to  the  design  of  the  skeleton.    If  one  Amia,  Lepisostcus),  the  ovary  is  unenclosed 

considers    the    evolution    of    the    muscular  but    is   either   more    elongated    or   located 

system  in  correlation  with  that  of  the  skel-  more    posteriorly.     The    oviduct    in    these 

eton,  it  seems  more  logical  to  suppose  that  forms  differs  in  design  from  that  in  carti- 

it  developed,  as  the  skeleton  seems  to  have  laginous    fishes.     Admittedly,   the   position 

done,  from  a  more  ancient  root  than  the  of   the    gonads    and    ducts    in    the    female 

early   selachian   fishes.    And   if   one   leans  lungfish  corresponds   more  nearly   to  that 

toward  the  idea  of  descent  from  a  ptycto-  of  the  Holocephali.    The  lungfish  ovary  is 

dont  rather  than  from  a  selachian  group,  it  much   longer,   however,    and   the   oviducts 

may   be   perhaps    because    it   is    easier   to  are  unspecialized  and  have  separate  ostia. 

imagine    building    holocephalian    muscula-  The  specialized  regions  of  the  holocepha- 

ture   upon   a  ptyctodont  frame,   especially  lian  oviduct  resemble  closely  the  selachian 

in  the  head  region,  than  it  is  to  derive  it  type.    Prasad,  who  made  a  series  of  histo- 

from  shark-like  origins.  logical   studies    of   such   specialized   areas, 

said,  ".  .  .  the  nidamental  glands  of  Hydro- 

The  Urogenital  System  lagtis  coUiei  exhibit  a  structure  very  similar 

Little  research  has  been  done  on  the  uro-  ^^    that    of    a    typical    oviparous    elasmo- 

genital  system  of  the  Holocephali.    Studies  t)ranch  .  .  .    (Prasad,  1948:  57).   One  could 

of    its    development    are    lacking    and    the  ^^y,    m    view   of   the    similar   reproductive 

histologv  of  its  component  organs  has  re-  habits  of  oviparous  elasmobranchs  and  hol- 

ceived    onlv    cursorv    attention    ( Burlend,  ocephahans,  that  their  similarly  specialized 

1910;    Leydig,    1851).     Its    gross    anatomv,  oviducts  were  a  parallel  development,  but 

which  is  known,  is  almost  exactlv  like  that  ^^^^^^  ^-^  "»  evidence  to  disprove  the  idea 

of  sharks  and  quite  different  from  that  of  ^1^'^^  t^^^^  fi"^^^^''*  '"'^y  l^^^e  inherited  both 

bonv  fishes  habits  and  the  structures  from  an  ear- 

A  glance 'at  the  reproductive  organs  of  li^'-even   a   very   much   earlier-common 

the    female    holocephalian    reveals    an    ar-  ^  ^^  ' 

rangement   which   is   exactly    like   that    of  ,  ^'\    searching    for    differences    between 

many    selachians.     Both    ovaries,    equally  ^^arks  and  holocephalians,  one  might  seize 

well  developed,  are  set  far  forward  in  the  ^1^^"  ^^^^  ^f  *  *^^'^*  f^''^\  female  sharks  have 

body  cavitv.    The  oxiducts   run   lateral  to  ^  ^^^^^^  ^^^^^^^  *^^^^  holocephalian  coun- 

the  ovaries' to  open  xxith  a  common  ostium  ^^'^f"'^'  ^^  "f  •    However,  the  importance 

in  the  extreme  anterior  end  of  the  coelomic  ^  ^^''^  P"")^  diminishes  when  one  sees  that 

space.     The   shark-like   nature   of   this   ar-  the  young  female  holocephalian  has  at  least 

rangement   is    emphasized   if   one   reviews  ^  ^^^P  urogenital  sinus  which  disappears 

the   female   genital   svstem   of  other  types  ^^  the  uteri  enlarge  and  press  outward  in 

of  fishes:    in  almost  all  teleosts  the  oviduct  the  maturing  animal.  The  one  unique  struc- 

is   continuous  with  the  ovary   so  that  the  ture  possessed  by  the  female  holocephalian 

eggs,  which  are  produced  in  large  numbers,  is  the  so-called  seminal  vesicle.   Hyrtl,  who 

are  at  no  time  free  in  the  coelom.    In  a  reported  in  1850  on  the  indented  blind  sac 

few  forms  like  the  trout,   the   ovar\'  does  which    opens    just    posterior    to    the    anus. 


182 


Bulletin  Museum  of  Comparative  Zoologij,  Vol.  135,  No. 


thought  that  it  functioned  as  a  "Samen- 
tasche,"  but  Rurlend  (1910)  showed  that  it 
was  glandular.  Redeke  ( 1898 )  sa\\'  in  this 
sac  a  possible  homologue  of  the  digitiform 
gland  of  sharks:  if  the  rectum  of  the  holo- 
cephalian  were  pulled  inward  from  the  sur- 
fact>,  drawing  the  "seminal  receptacle"  in 
\\  ith  it,  the  latter  structure  would  be  in  the 
same  relation  to  the  hindgut  as  the  gland 
of  the  shark.  It  is  probable  that,  whatever 
its  mode  of  formation,  the  blind  sac,  which 
is  not  found  in  any  other  xertebrate,  repre- 
sents a  minor  specialization  which  has 
occurred  in  the  later  evolution  of  the  Holo- 
cephali. 

The  reproductive  system  of  the  male  hol- 
ocephalian  is  as  shark-like  as  that  of  the  fe- 
male. In  both  types  of  fishes  the  testis  is 
connected  by  vasa  efferentia  to  a  highly 
coiled  epididymis  through  which  sperm  are 
conducted  to  the  more  posterior  and  wider 
poi-tion  of  the  vas  deferens.  The  vasa  ef- 
ferentia of  the  shark  represent  transformed 
anterior  kidney  tubules  which  lead  into  the 
embryonic  Wolffian  duct,  and  it  is  pre- 
sumed that  the  efferent  ductules  of  the 
Holocephali  are  homologous  structures.  The 
anterior  part  of  the  kidney  in  immature 
sharks  and  chimaerids  has  glomeruli  in  it, 
but  these  disappear  during  growth  toward 
sexual  matiuity.  The  anterior  part  of  the 
kidney  transforms  itself  from  an  excretory 
to  a  secretory  organ  and  is  then  known  as 
Leydig's  gland.  In  holocephahans,  as  in 
sharks,  its  secretion,  which  passes  through 
short  ducts  to  the  epididymis  and  vas  def- 
erens, serves  as  a  fluid  matrix  for  the  sus- 
pension of  the  sperm.  The  posterior  portion 
of  the  kidney  in  both  kinds  of  fishes  re- 
mains excretory,  sending  urine  through  one 
or  more  ureters  which  empty  into  a  urogeni- 
tal sinus.  In  commenting  upon  the  arrange- 
ment of  pathways  in  the  male  system.  Van 
Oordt  says,  "hinsichtlich,  der  Abfiihrung 
der  Spermien  stimmen  die  Holocephalen 
mit  den  Selachiern  iiberein"  (Van  Oordt, 
in  Rolk,  1938,  Vol.  V:  750).  In  resembling 
the  selachian  system  so  closely,  the  male 
reproductive  system  of  the  holocephalians 


is  markedly  different  from  that  of  the  bony 
fishes.  In  the  latter  group  one  finds  either 
a  duct  for  sperm  which  is  separate  from  the 
original  archinephric  duct  or  the  tendency 
to  develop  such  an  arrangement.  Even  in 
Acipenser,  where  the  expression  of  this 
tendency  is  minimal,  the  urogenital  system 
is  distinguished  from  the  selachian  and  hol- 
ocephalian  types  by  lacking  a  secretory 
portion  derived  from  the  anterior  end  of 
the  kidney.  No  bony  fish  develops  an  ac- 
cessory organ  comparable  to  Leydig's  gland. 

Given  the  great  degree  of  similarity  be- 
tween male  selachians  and  holocephalians, 
investigators  have  tried  to  define  the  rela- 
tively small  points  of  difference  which  do 
exist.  It  has  been  observed,  for  instance, 
that  the  number  of  vasa  efferentia  varies. 
In  contrast  to  one  in  ScylUum,  Chimacra 
mon.stro.sa  has  five  or  six.  Borcea  (1906: 
349),  who  made  an  extensive  study  of  the 
urogenital  system  of  elasmobranchs,  con- 
siders that  "le  nombre  des  vaisseaux  ef- 
ferents  est  plus  eleve  et  le  canal  longitudi- 
nal de  I'epididyme  est  plus  long  chez  les 
types  les  plus  primitifs."  In  making  this 
statement,  Borcea  had  in  mind  the  fact  that 
the  batoids  are  characterized  by  a  few  or 
only  one  vas  efferens. 

Another  minor  difference  concerns  the 
posterior  region  of  the  vas  deferens  which 
is  enlarged  to  form  an  ampulla  (Van  den 
Brock's  term)  or  a  sperm  vesicle  (Rurlend's 
tenn).  In  both  sharks  and  chimaerids,  the 
inner  wall  of  this  structure  is  thrown  into 
folds  which  divide  the  lumen  of  the  duct. 
In  sharks  like  Scyllium,  however,  the  par- 
titions are  as  simple  as  septa  in  a  mushroom 
cap,  whereas  the  inner  walls  in  a  large  sec- 
tion of  the  chimaerid  ampulla  run  into  one 
another  in  a  more  complex  fashion,  cutting 
up  the  space  \\'ithin  the  passage  into  inter- 
connecting compartments.  One  feels,  upon 
studying  these  septa,  that  their  different 
design  is  less  important  than  the  fact  of 
their  presence  in  both  holocephalians  and 
selachians.  The  appearance  of  these  struc- 
tures is  a  remarkable  point  of  similarity  in 
two  forms  whose  lines  ( in  consideration  of 


Morphology  and  Relationships  of  Holocephali  •  StaJil        183 


other  organ  systems)  seem  to  have  sepa- 
rated far  back  in  time. 

This  same  idea  may  be  emphasized  in  the 
matter  of  the  claspers  of  the  male.  Before 
descril)ing  the  differences  \\'hich  exist  be- 
tween these  structures  in  sharks  and  holo- 
cephaHans,  one  must  dwell  a  moment  upon 
the  fact  that  claspers,  which  are  not  a  com- 
mon vertebrate  characteristic,  do  appear  in 
a  generally  similar  form  in  both  of  these 
groups  of  fishes.  It  would  seem,  at  first, 
that  the  possession  of  such  claspers  is  signal 
proof  of  the  close  relationship  of  sharks  and 
holocephalians.  The  major  obstacle  to  the 
acceptance  of  this  idea  lies  in  the  fact  that 
CladoseJaclic,  a  fonn  apparently  anteced- 
ent to  modern  sharks,  shows  no  claspers. 
If  it  really  had  none,  then  the  holocepha- 
lians must  have  developed  their  claspers 
independently.  That  they  did  so  is  not  an 
impossible  assumption.  It  appears  that 
claspers  may  not  be  as  peculiarly  elasmo- 
branchian  a  character  as  one  would  assume 
from  a  study  of  extant  fishes.  If  Watson 
(1938)  and  0rvig  (1962)  are  correct  in 
postulating  the  presence  of  claspers  in 
Rhamphodopsis  and  Cteniirelki,  respec- 
tively, it  may  be  that  these  structures  were 
possessed  by  a  number  of  placoderm  groups. 
If  that  was  the  case,  holocephalians  and 
selachians  might  bear  claspers  inherited 
from  separate  ancestral  stocks.  In  support 
of  this  hypothesis  one  might  cite  0rvig's 
finding  of  a  pair  of  dermal  spines  anterior 
to  the  pelvic  girdle  of  CteiuireUa.  He  be- 
lieves that  these  spines  may  have  been  as- 
sociated with  anterior  claspers,  adjuncts  to 
the  reproductive  system  found  in  holoce- 
phalians but  not  in  elasmobranchs. 

The  elaborate  array  of  claspers  charac- 
teristic of  holocephalians  sets  these  fishes 
apart  from  other  cartilaginous  forms.  No 
other  type  of  fish  has  either  the  aforemen- 
tioned anterior  claspers  in  front  of  the  pel- 
vic fins  or  the  strange  median  frontal  clasper 
or  tenaculum  set  upon  the  dorsal  surface  of 
the  head.  In  all  extant  holocephalians  the 
anterior  claspers  are  represented  as  small, 
gripping  structures  which  are  concealed  in 


a  pouch  when  not  in  use.  Leigh-Sharpe 
( 1922 )  believes  that  the  prepubic  processes 
found  in  Squaloraja  supported  anterior 
claspers  in  that  Jurassic  form.  There  are 
no  reports  of  these  structures  in  earlier  fos- 
sils, however,  except  for  0rvig's  mention  of 
the  spines  in  CtemireUa.  Since  0rvig  found 
no  trace  of  a  tenaculum  in  CtenurcUa,  the 
earliest  form  of  that  structure  is  known 
from  Squaloraja  and  its  contemporary, 
Myriacanthiis.  The  tenaculum  in  those 
fishes  was  a  long  pointed  protuberance.  In 
living  holocephalians,  the  tenaculum  is 
smaller  and  rounded  at  its  distal  end. 

In  a  lengthy  series  of  papers  Leigh- 
Shaqoe  ( 1920  ff. )  presents  a  review  of  elas- 
mobranch  and  holocephalian  claspers.  He 
describes  the  claspers  of  Chimaera  and 
CaUorhijnchus  as  having  two  branches  and 
suggests  that  these  branches  represent  the 
ultimate  and  penultimate  pelvic  fin  radials. 
He  believes  that  claspers  of  this  type  are 
primitive.  However,  Rhinochimaem,  which 
is  thought  to  be  the  most  primitive  holo- 
cephalian in  terms  of  its  other  systems,  has 
an  unl)ranched  clasper  more  nearly  like  that 
of  sharks.  Leigh-Sharpe  ( 1922 )  includes  a 
drawing  of  a  clearly  preserved  clasper  of 
the  fossil  Squaloraja  which  shows  a  single 
but  unusually  broad  structure  terminated 
by  a  group  of  small,  dermal  hooks.  Since 
the  clasper  of  Squaloraja  is  unique  in  form, 
and  since  Squaloraja  lived  in  Jurassic  times 
when  the  holocephalian  line  was  already 
established,  one  cannot  be  sure  that  the 
claspers  of  this  fish  give  evidence  of  the 
original  nature  of  the  holocephalian  struc- 
tures. 

In  his  classification  of  the  cartilaginous 
fishes  according  to  the  type  of  clasper  they 
show,  Leigh-Sharpe  sets  the  Holocephali 
amongst  the  primitive  forms  for  still  another 
reason.  They  have  not  developed  the  ab- 
dominal structures — a  pair  of  muscular 
cavities  called  siphons — which  play  a  role 
in  sperm  passage  during  the  copulation  of 
most  elasmobranchs.  Holocephalians  do 
ha\e  a  different  sort  of  cavity,  though,  lo- 
cated in  the  proximal  portion  of  the  clasper. 


184        BtiUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


B 


ani    cl 


Fig.  10.  Claspers  of  various  holocepholian  forms.  A,  Squalorajo;  B,  Chimaera  monstrosa;  C,  Rhinochimoero  atlantica.  ant. 
c).,  Anterior  clasper;  bos.,  basal;  cl.,  clasper;  pel.gn.,  pelvic  girdle;  ppb.,  prepubic  processes;  r.,  fin-rays;  v.c,  vertebral 
column.     (After   Leigfi-Sharpe.) 


Leigh-Sharpe  interprets  this  cavity  as  ho- 
mologous to  that  of  Chlamijdosdachc  and 
so  brackets  these  fishes  together.  Surely  a 
common  category  for  these  forms  stands  on 
shaky  ground.  The  Holocephali  should 
probably  be  set  apart  even  here  if  the  pres- 
ence of  their  curious  frontal  and  anterior 
claspers  is  taken  into  consideration. 

Although  the  kidneys  have  not  been 
thoroughly  examined  histologically,  their 
gross  anatomy  and  their  relationship  to  the 
genital  organs  have  been  well  described 
(Burlend,  1910;  Leydig,  1851).  There  is  no 
doubt  that  these  organs,  too,  are  like  those 
of  elasmobranchs  and  quite  different  from 
those  of  other  fishes.  Unlike  the  kidneys 
of  the  cartilaginous  forms,  those  of  bony 
fishes  never  become  closely  involved  with 
reproductive  structures  in  the  male  and,  in 
both  sexes,  are  generally  unifonTi  in  tubule- 
structure  throughout  their  length,  under- 
going neither  transformation  nor  degenera- 
tion at  the  anterior  end  as  the  animal 
reaches  maturity.  It  is  not  necessary  to  lean 
entirely  upon  structural  resemblances  to 
predicate  a  possible  relationship  between 
the  Holocephali  and  the  Selachii  either. 
The  excretory  systems  of  both  groups  bear 
the  same  distinctive  functional  earmark: 
the    kidneys    resorb    urea    selectively    and 


maintain  that  substance  in  the  bloodstream 
in  unusually  high  concentration. 

In  adult  holocephalians,  as  in  sharks, 
urine  is  produced  in  the  posterior  regions 
of  the  kidney  and  drained  by  specially  de- 
veloped ureters.  This  arrangement  contrasts 
with  that  of  bony  fish  in  which  urine  is 
produced  throughout  the  entire  kidney  and 
is  removed  through  the  opisthonephric  duct. 
In  cartilaginous  fish  of  the  male  sex  the 
anterior  kidney  and  the  Wolffian  duct  be- 
come part  of  the  reproductive  system  as 
was  mentioned  before.  In  females,  despite 
there  being  no  secondary  use  for  the  an- 
terior region  of  the  kidney,  that  portion 
degenerates  and  the  Wolffian  duct  stretches 
forward  and  ends  blindly.  In  the  animals 
of  both  sexes  the  kidney  gives  some  hint  of 
its  originally  segmented  nature.  Especially 
in  the  anterior  region  traces  of  segmental 
divisions  remain.  The  segmental  blocks  are 
particularly  noticeable  in  the  male,  because 
ducts  leave  the  gland  of  Leydig  at  segmen- 
tal intervals. 

Borcea  ( 1906 ) ,  in  the  study  to  which 
reference  has  already  been  made,  is  plainly 
of  the  opinion  that  the  elasmobranchs  rep- 
resent the  primitive  vertebrates  from  which 
all  the  others  have  descended.  Although 
most  students  of  evolution  no  longer  agree 
with  that  premise,  they  still  admit  the  pos- 


Morphology  and  Relationships  of  Holocephali  •  Stahl        185 


sibility  that  certain  characteristics  of  car-  relationship  between  the  testis  and  the  kid- 

tilaginous  fish  may  have  been  carried  over  ney  similar  in  principle  to  that  which  ap- 

from  their  primitive  ancestors  at  the  placo-  pears  in  the  cartilaginous  fishes.    This  idea 

derni  level.    With  this  idea  in  mind  and  in  is  supported  further  by  the  emphasis,  again 

consideration  of  the  similarity  of  the  uro-  in  all  vertebrates  except  bony  fishes,  upon 

genital  systems  in  holocephalians  and  elas-  the  posterior  portion  of  the  kidney  as  the 

mobranchs,   one  may  find   interesting  the  part  chiefly  responsible  for  excretory  func- 

following    comment    of    Borcea:    "C'est    le  tion. 

groupe     des     Elasmobranches,     qui     nous  To  summarize  the  foregoing  points,  one 

montre  la  succession  de  ces  trois  stades  (of  may  state  that  the  urogenital  system  of  the 

the  evolution  of  the  vertebrate  kidney)  avec  holocephalians  resembles  the  selachian  sys- 

la    plus    grande    nettete.     Dune    part,    ils  tem  closely.    In  the  position  of  the  gonads, 

presentent  I'etat  nephridioide  .  .  .  plus  nette-  the   specialization   of  the   accessory   ducts, 

ment  que  n'importe  quel  autre  groupe  de  the  nature  of  the  kidney,  the  development 

Vertebres.     D'autre    part,    ils    sont    parmi  of  accessory  ureters,  and  the  possession  of 

ceux-ci,  les  animaux  les  plus  primitifs  chez  claspers  on  the  pelvic  fins  of  male  animals, 

lesquels  les  glandes  genitales  entrent  en  re-  the  two  groups  of  cartilaginous  fishes  are 

lation  avec  le  rein  et  son  uretere  primaire  remarkably  alike.    The  type  of  urogenital 

et  alors  la  serie  des  changements  se  montre  system  they  share  is  distinct  in  all  of  these 

d'une  fa9on  tres  manifeste.    Chez  les  Elas-  features  from  that  of  bony  fishes.   The  kid- 

mobranches  la  division  de  I'uretere  primaire  neys  of  holocephalians  and  selachians  are 

est  tout  a  fait  nette.   Chez  les  plus  primitifs  set  apart  from  those  of  all  other  vertebrates 

d'entre  eux  ce  n'est  qu'a  I'etat  adulte   (en  by  their  ability  to  resorb  urea  selectively 

relation   avec  la  maturite  sexuelle),   qu'on  and  return  it  to  the  circulating  blood.   The 

constate  la  modification  du  rein  superieur"  major   point   of   difference   between    holo- 

( Borcea,  1906:  251).  cephalians  and  selachians  lies  in  the  pos- 

Disregarding   Borcea's   use   of   the   term  session  by  the  former  of  claspers  anterior 

"etat  nephridioide "  which  summons  up  an  to  the  pelvic  fins  and  of  a  median  tenacu- 

argument  quite  apart  from  the  subject  of  lum. 

this  paper,  one  can  still  see  in  his  statement  Although  the  remarkable  similarity  of  the 
reasons  to  support  the  thesis  that  the  elas-  urogenital  system  of  holocephalians  to  that 
mobranch  urogenital  system  is  primitive  of  selachians  could  be  cited  as  evidence  of 
rather  than  secondarily  simplified.  If  the  the  evolution  of  the  Holocephali  from  the 
system  is  primitive,  then  there  is  an  alter-  selachian  line,  there  appears  to  be  an  alter- 
native to  the  theory  that  the  holocephalians,  native  to  that  hypothesis.  Since  it  seems 
whose  urogenital  organs  seem  shark-like,  possible  that  the  urogenital  system  of  car- 
must  therefore  have  diverged  from  the  tilaginous  fishes  is  truly  primitive  and  if  so 
elasmobranchs  relatively  late.  It  is  possible  may  have  existed  in  a  number  of  early 
to  speculate  that,  as  evidence  drawn  from  gnathostome  groups,  holocephalians  and 
other  structures  suggests,  the  holocephalian  selachians  could  have  evolved  from  two 
and  elasmobranch  lines  did  split  far  back  separate  ancestral  stocks.  Both  types  of 
among  their  placoderm  forebears,  and  that  cartilaginous  fishes  could  have  retained  the 
both  groups  of  fish  ha\'e  carried  to  modern  urogenital  system  in  its  ancient  fonn.  This 
times  the  type  of  urogenital  system  which  theory  presumes  that  the  holocephalian  and 
those  early  vertebrates  possessed.  That  a  selachian  claspers  were  not  derived  from 
system  of  this  type  may  have  become  wide-  the  same  source.  The  possibility  that  pty- 
spread  in  primitive  gnathostomes  generally  ctodonts  possessed  claspers  allows  one  to 
is  suggested  by  the  development  in  all  ver-  believe  that  there  may  have  been  more  than 
tebrates  except  the  bony  fishes  of  an  inter-  one  source  of  those  structures  at  the  placo- 


186 


Bulletin  Museum  of  Comparative  Zoolop^xj,  Vol.  135,  No.  3 


derm  level.  The  presence  of  anterior  and 
frontal  claspers  in  living  and  fossil  holo- 
cephalians  but  not  in  selachians  increases 
the  probability  of  the  existence  of  a  sepa- 
rate placoderm  ancestor  for  the  holocepha- 
lians. 

The   Digestive  System 

The  search  of  the  digestive  system  for 
e\idence  of  hereditary  relationships  turns 
up  a  profitable  thread  or  two  and  also  re- 
veals several  alleys  which  end  blindly.  As 
might  be  imagined,  an  investigation  of  the 
structure  of  the  teeth  gives  rise  to  specula- 
tions based  on  firmer  ground  than  does  an 
examination  of  the  digestive  tract  itself  or 
its  associated  glands. 

Holocephalians  have  three  pairs  of  tooth- 
plates.  The  smallest,  called  vomerine  plates, 
are  located  in  the  anterior  portion  of  the 
upper  jaw  immediately  in  front  of  the  larger 
palatine  pair.  The  mandibular  plates  of  the 
lower  jaw  are  the  largest,  being  equal  in 
length  to  the  other  two  combined.  A  com- 
parison of  the  sections  of  these  teeth  fig- 
ured by  Brettnacher  ( 1939 )  with  those  of 
cochliodonts  shown  by  Nielsen  ( 1932 )  sug- 
gests that  the  two  tooth-types  are  not 
similar,  as  Moy-Thomas  ( 1936 )  had  main- 
tained. A  difference  if  it  does  exist,  is  im- 
portant, because  the  structure  of  the  tooth- 
plates  was  one  of  the  main  supports  of  the 
theory  that  the  Holocephali  arc  descended 
from  bradyodonts.  If  the  teeth  of  the  two 
groups  are  truly  unlike,  and  if  the  presence 
of  holostylic  jaw  suspension  in  both  groups 
is  not  as  important  a  factor  as  Moy-Thomas 
thought  it  was,  then  the  case  for  close  rela- 
tionship becomes  very  much  weaker. 

The  discrepancies  in  tooth-type  become 
apparent  when  descriptions  of  the  internal 
structure  of  the  teeth  of  each  are  set  side 
by  side.  Eigil  Nielsen  (1952:  34)  gives 
the  now  classic  description  of  the  bradyo- 
dont  type:  "This  Bradyodont  structural 
type  is  especially  characterised  by  possess- 
ing a  system  of  numerous,  more  or  less 
parallel  vascular  canals  ascending  through 
the  greater  part  of  the  crown,  but  ending 


blindh'  just  below  the  tritoral  surface.  The 
ascending  canals  are  lined  with  layers  of 
dentine,  and  the  dentine  around  each  canal 
is  separated  from  that  around  the  other 
canals  by  a  hard  tissue,  described  as  enamel 
by  me  in  1932." 

The  chimaerid  toothplate  has  been  ex- 
amined, described,  and  figured  by  Barg- 
mann  (1933)  and  Brettnacher  (1939).  Their 
accounts  of  the  histology  of  the  toothplates 
agree,  although  the  terminology  that  they 
use  in  their  descriptions  is  not  exactly  the 
same.  The  outer  surface  of  the  crown  of 
each  plate  as  well  as  its  embedded  portion 
consists  of  a  type  of  dentine  which  Brett- 
nacher calls  "Hiillendentin"  and  Bargmann 
calls  "Manteldentin."  In  areas  where  epi- 
thelium comes  in  contact  with  the  tooth- 
plate,  there  is  a  superficial  layer  of  very 
hard  material  which,  for  Brettnacher,  is 
true  enamel,  and  for  Bargmann  merely  a 
specially  transformed  part  of  the  "Mantel- 
dentin." In  the  interior  of  the  tooth,  accord- 
ing to  both  men,  there  is  a  meslncork  of 
dentin  trabeculae  rather  than  parallel  den- 
tinal tubules.  Brettnacher  gives  these  tra- 
beculae the  special  name  of  "Balkendentin" 
(because  they  form  supporting  beams),  al- 
though he  docs  state  that  they  are  formed 
by  an  extension  of  the  odontoblast  layer 
which  creates  the  "Hiillendentin."  Barg- 
mann uses  the  temi  "Manteldentin"  to 
embrace  the  trabeculae  as  well  as  the  pe- 
ripheral layer.  The  spaces  in  the  trabecular 
region  are  pulp  channels  which  Bargmann 
says  are  slowly  obliterated  in  the  pressure- 
receiving  parts  of  the  plate  by  deposition 
of  circumpulpar  dentin. 

Jacobshagen,  who  relies  upon  Brett- 
nacher's  work,  has  included  the  chimaerid 
toothplate  in  his  review  of  the  structure  of 
selachian  teeth  (1941).  As  he  presents  his 
figures  and  comparative  descriptions,  one 
sees  that  there  could  be  logic  in  his  reason- 
ing that  the  internal  arrangement  of  the 
holocephalian  plate  is  a  primitive  variant  of 
the  dentinal  pattern  still  in  existence  in 
extant  elasmobranchs.  Both  holocephalians 
and  selachians  show  the  outer  "Hiillenden- 


Morphology  and  Relationships  of  Holocephali  •  Sfahl        187 


tin "    covering    an    inner    trabecular    mesh-  ( 1951 )   would  classify  as  "tubular  dentin" 

work.     The    categories    that    Jacobshagen  and  in  Nielsen's  figures  look  singularly  dif- 

establishes   depend  upon   the   thickness   of  ferent  from  anything  produced  by  the  Holo- 

the  outer  dentin  layer  and  the  amount  and  cephali.   The  "Balkendentin"  which  fills  the 

distribution  of  the  inner  "Balkendentin."  chimaerid   toothplate  seems  more   akin   to 

Jacobshagen  does  make  a  separate  cate-  0rvig's  osteodentine  in  its  arrangement  and 

gory  for  the  toothplates  of  the  Holocephali,  its  apparent  mode  of  development, 

not  only  because  of  their  plate-like  struc-  If  it  is  not  correct  to  associate  holocepha- 

ture,  but  also  because  they  contain  a  unique  lian  and  cochliodont  teeth  with  each  other, 

material  which  both  Brettnacher  and  Barg-  one  is  free  to  seek  other  relationships.    It 

mann   describe.    Brettnacher  calls   it   "pri-  seems  not  unreasonable  to  connect  the  chi- 

mary  dentin"  and  Bargmann  uses  the  old  maerid  structure  with  that  of  ptyctodonts. 

tenn  "Kosmin"  to  refer  to  it.  This  substance  Ptyctodont  plates  have  been  studied  histo- 

is  found  within  the  pulp  channels  in  several  logically  most  recently  by  Gross  ( 1957 )  and 

regions  within  each  plate.    Sometimes  the  0rvig  (1957).    Gross  found  very  little  dif- 

Kosmin  appears  in  pearl-like  masses  strung  ference  between  the  teeth  of  Rhynchodus 

in  rows;   in  some  teeth  the  "pearls"  seem  and  Ptijctodus,  and  his  general  description 

coalesced  to  form  an  elongated  bar.   All  the  reveals  a  surface  layer  of  dentin  supported 

investigators   who  have   discussed   Kosmin  from  within  by  dentinal  trabeculae  which 

regard  it  as  an  ancient  vestige.    Schauins-  formed  a  network.    Against  these  internal 

land  thought  it  represented  the  remains  of  trabeculae  in  tritoral  areas,  what  Gross  calls 

fused  cylindrical  teeth.    Bargmann  discards  a  secondary  dentin  was  laid  down.  It  would 

this  idea,  however,  for  the  teeth  of  younger  have  been  interesting  if  Gross  had  referred 

specimens   show   Kosmin   in   its   undivided  to  the  work  of  Brettnacher  and  Bargmann. 

bar-like  form.    The  rather  periodic,  pearl-  Without  such  a   reference  one   cannot  be 

like  division,  he  feels,  is  a  later  manifesta-  sure  whether  Gross  considered  the  dentin 

tion.    Bargmann   has  his  own   theory:     he  material  which  he  mentions  equivalent  or 

compares  the  structure   of  Kosmin  to  the  similar  to  that  of  the   Holocephali.     It   is 

structure  of  the  surface  knobs  on  Cepha-  impossible  from  Gross'  description,  for  in- 

hspis  plates,  and  speculates  that  in  the  e\'0-  stance,  to  tell  whether  he  saw  something 

lution  of  the  Holocephali  this  early  type  of  like  Kosmin.    It  appears  that  he  did  not. 

hard  tissue  may  have  sunk  inward.  0rvig's  description  of  PoIeomyJus  is  more 

Brettnacher  and  Jacobshagen  point  out  puzzling.    He    states    that    the    PaJeomyJus 

that   dentin  in  general  may  have  evolved  toothplate  is  much  like  those  of  Ptyctodus 

from  a  relatively  soft  substance,  penetrated  and  Rhynchodus,  and  in  the  number  of  its 

by  widely  spaced,  branching  tubules  to  a  tritoral  columns  even  more  like  the  Mesozoic 

much    harder    material    with    close-ranked  and  Cenozoic  Holocephali.  But  he  describes 

parallel   tubules.    With   this   idea  in   mind  these  tritoral  columns  as  being  separated  by 

they  both  consider  that  the  dentin-tissue  in  acellular  bone,  while  in  holocephalians  they 

the  Holocephali  is  of  the  primitive  type,  the  are  separated  by  an  interstitial  substance 

toothplate   deriving   its   strength   from    the  "not  unlike  enamel."    He  refers  to  the  chi- 

arrangement    of    the    dentinal    trabeculae  maeroid  columns  as  being  of  a  peculiar  tu- 

rather  than  from  the  hardness  of  the  dentin  bular  dentin  ".s?//  ficncris."  Since  describing 

itself.  the  Pcdcomylus  toothplate  in  1957,  however, 

As  these  workers  describe  and  discuss  the  0rvig  has  revised  his  terminology.  For  hard 

structure    of   holocephalian    toothplates,    it  tissues  which  grow  inward  toward  the  basal 

seems  less  and  less  likely  that  these  plates  region  of  the  toothplate,  including  tritoral 

have   much   in    common   with   cochliodont  columns,    he    has    introduced    the    name 

teeth.    The    latter    consist    of    what    0rvig  "pleromic  hard  tissue."    Although  he  does 


188        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


not  mention  Paleoinyhis  specifically,  in  a 
forthcoming  book  he  indicates  similarities 
between  the  pleromic  hard  tissue  of  ptycto- 
dont  arthrodires  and  holocephalians.  He 
emphasizes  the  difference  in  arrangement 
of  the  pleromic  tissues  of  ptyctodonts  and 
holocephalians,  on  the  one  hand,  and  of 
]:)radyodonts,  on  the  other,  by  classifying 
the  pleromic  material  of  the  former  as 
columnar  and  of  the  latter  as  coronal. 

Although  it  is  usual  to  analyze  the  histo- 
logical structure  of  teeth  in  an  effort  to 
derive  evidence  of  phylogenetic  signifi- 
cance, it  might  be  well  to  keep  in  mind  the 
possibilit)'  that  convergent  evolution  could 
have  brought  about  similar  structure  where 
no  relationship  exists.  Radinsky  (1961), 
who  has  found  similar  patterns  in  the  den- 
tin of  bradyodonts,  batoids,  selachians,  and 
dipnoans,  is  of  the  opinion  that  the  internal 
structme  of  teeth  may  be  adaptive  and  that 
classification  should  therefore  not  be  based 
entirely  upon  it.  Despite  this  consideration, 
however,  the  results  of  a  comparison  of 
cochliodont,  holocephalian,  and  ptyctodont 
teeth  seems  useful.  The  difference  between 
holocephalian  teeth  and  those  of  cochlio- 
donts,  although  the  latter  fishes  were  ap- 
parently durophagous,  should  be  kept  in 
mind.  The  resemblance  between  the  struc- 
ture of  ptyctodont  and  holocephalian  teeth 
may  be  significant  in  combination  with 
other  evidence. 

One  should  not  leave  a  discussion  of  chi- 
maerid  toothplates  without  mentioning  the 
problem  of  their  origin.  Their  plate-like 
structure  is  unusual  and  has  dictated  com- 
parisons between  the  Holocephali  and  other 
vertebrates  like  Dipnoi  that  also  possess 
plate-like  formations  in  the  mouth.  These 
comparisons  founder,  however,  upon  one 
point.  The  folates  of  lungfish,  the  teeth  of 
most  cochliodonts,  and  the  pavement  denti- 
tion of  rays,  all  can  be  shown  to  be  com- 
pounded of  units  which  arise  first  as 
separate  entities.  In  holocephalians  no 
amalgamation  of  individual  denticles  is  de- 
monstrable. Even  in  the  early  embryos 
which  Schauinsland  studied  there  were  no 


indications  of  a  fusion  of  teeth  or  tooth 
buds.  It  is  possible  that  the  Holocephali 
descended  from  forms  whose  teeth  lost 
their  discrete  nature  and  that,  as  the  group 
evolved,  ontogenetic  evidence  of  fusion  was 
suppressed.  Since  it  has  not  been  demon- 
strated that  all  fossilized  toothplates  evolved 
through  a  compounding  of  individual  units, 
however,  it  may  be  that  holocephalian 
toothplates  were  derived  from  pre-existing 
integral  structures.  As  antecedents  of  holo- 
cephalian toothplates,  ptyctodont  plates 
might  be  preferable  to  large  cochliodont 
teeth  produced  through  fusion. 

In  turning  from  the  toothplates  to  the 
digestive  tract,  one  reaches  a  series  of  struc- 
tures whose  evolutionary  history  is  even 
harder  to  define.  All  the  Holocephali  show, 
beyond  the  mouth  and  pharynx,  an  esopha- 
gus which  leads  to  the  intestine  directly, 
without  the  intervention  of  a  differentiated 
stomach  expansion.  The  obvious  question — 
is  the  lack  of  a  stomach  a  primitive  or  a 
degenerate  character? — has  found  no  sure 
answer.  Since  the  stomachless  condition  is 
found  in  a  number  of  unrelated  fishes,  one 
could  argue  that  it  represents  a  common 
type  of  degeneration  which  has  occurred 
independently  in  several  lines.  On  the 
other  hand,  the  absence  of  a  stomach  in 
cyclostomes  may  be  a  remnant  of  the  ear- 
liest vertebrate  plan.  At  least  one  worker, 
Fahrenholz  (1915),  assumes  that  this  is  true 
in  the  case  of  the  Holocephali.  Since  one 
answer  seems  as  logical  as  the  other,  neither 
can  be  relied  upon  to  carry  much  weight 
in  the  solution  of  the  evolutionary  problem. 

The  same  may  be  said  about  the  holo- 
cephalian spiral  intestine.  All  the  chimaerid 
fishes  show  an  intraintestinal  fold  which 
takes  one  slow  turn  throughout  the  greatest 
part  of  the  intestinal  tube  and  then  makes 
two  and  a  half  tighter  turns  at  the  posterior 
end.  The  edge  of  the  fold  is  free  in  the 
loosely  coiled  forepart  and  caught  up  in  the 
center  of  the  corkscrew  tunis  at  the  end. 
This  arrangement  seems  to  be  a  combina- 
tion of  the  "gerollte"  type  which  Jacobshagen 
(1915)  described  as  existing  in  a  few  sharks 


Morphology  and  Relationships  of  Holocephali  •  Stahl        189 


and  the  "gedrehte"  type  which  he  dechiied 
to  be  much  more  common  amongst  the 
selachians.  The  pecuhar  nature  of  the  spiral 
valve  can  be  interpreted  in  either  of  two 
ways.  Firstly,  as  Fee  (1925)  and  Dean 
( 1906 )  see  it,  the  viscera  of  the  chimaerid 
fishes,  believed  by  them  to  be  modified  re- 
latively late  from  sharks,  have  been  crowded 
into  a  shortened  body  cavity.  The  stomach 
dilation  fails  to  develop  and  "the  intestinal 
valve,  instead  of  undergoing  the  further 
spiral  development  of  sharks,  makes  but  a 
few  tunis  (about  four)  .  .  ."  (Fee,  1925: 
179).  The  view  of  the  valve  arrangement 
as  secondary,  as  set  forth  here,  might  be 
supported  by  Jacobshagen's  contention  that 
reduction  in  the  intestinal  fold  always  takes 
place  from  the  anterior  end.  In  fishes 
which  bear  a  degenerate  spiral  valve  or  a 
vestigial  one,  the  parts  of  it  that  remain  are 
in  the  posterior  region  of  the  intestine. 
Secondly,  the  holocephalian  valve  might 
be  held  as  primitive,  especially  in  its  histo- 
logical structure.  Evidence  for  this  conten- 
tion has  been  presented  by  Jacobshagen 
(1934),  who  has  made  a  detailed  compara- 
tive study  of  the  spiral  intestine  in  sela- 
chian, dipnoan,  ganoid,  and  jawless  fish. 
He  points  out  that  the  valvular  infolding 
in  sharks  includes  only  the  mucosa  and  the 
muscularis  mucosae.  Since  the  ammocoetes 
larva  shows  inclusion  of  circular  muscle  as 
well,  Jacobshagen  suggests  that  the  primi- 
tive fold  was  an  indentation  of  the  whole 
intestinal  wall  which  lay  within  the  envel- 
oping serosa.  Significantly,  the  holocepha- 
lians  are  the  only  fish  that  show  portions 
of  the  main  circular  muscle  of  the  intestine 
still  included  in  the  adult  valvular  fold.  Of 
course,  Jacobshagen's  idea  may  be  incor- 
rect, and  the  inclusion  of  the  muscle  may 
not  be  a  primitive  condition  in  either  ani- 
mal. 

As  one  advances  to  a  consideration  of  the 
glands  associated  with  the  digestive  tract, 
one  finds  less  and  less  information  avail- 
able. Scammon,  who  has  studied  the  sela- 
chian liver,  reports  in  his  account  of  it  that 
"the  histology   of  the   adult   elasmobranch 


liver  was  first  briefly  described  by  Leydig 
from  observations  on  Chimaera"  (Scammon, 
1915:  245).  Since  Scammon  does  not  even 
think  to  distinguish  the  holocephalian  from 
the  selachian  organ,  it  is  apparent  that  their 
characteristics  must  be  very  much  alike. 
Scammon  holds  that  the  elasmobranch  liver 
differs  from  that  of  other  vertebrates  by  its 
unique  type  of  lobulation,  its  accumulation 
of  fat  within  the  hepatic  cells,  and  its  com- 
paratively slight  development  of  the  bile 
duct  system.  It  is  impossible  to  decide 
whether  these  characteristics  are  peculiar 
to  the  shark  line  or  whether  they  arose  deep 
within  the  placoderm  stock. 

The  holocephalian  pancreas  has  appar- 
ently not  been  studied.  Siwe,  writing  in 
1926,  does  not  mention  the  chimaerid  struc- 
ture in  his  paper  on  the  comparative  anat- 
omy of  that  gland.  The  only  other  glandular 
organ  associated  with  the  digestive  tract  of 
the  Holocephali  that  has  received  attention 
is  an  intraparietal  mass  of  tubules  located 
at  the  posterior  end  of  the  spiral  valve. 
Citterio  ( 19.32 )  discusses  this  gland,  first 
described  by  Leydig,  suggesting  that  it 
might  be  homologous  to  the  digitiform 
gland  of  selachians  and  more  primitive  in 
its  intraparietal  location. 

Another  structure  which  may  have  a  sela- 
chian homology  is  the  mass  of  lymphomye- 
loid  tissue  dorsal  to  the  skin  of  the  palate. 
Extant  sharks  and  rays  have  a  pair  of  struc- 
tures, similar  in  their  histology,  built  into 
the  sides  of  the  esophagus  (Fahrenholz, 
1915 ) .  The  tissue  itself  seems  of  a  like  con- 
struction in  the  Holocephali  and  the  sela- 
chians: both  show  several  different  types 
of  myeloid  cells  set  in  a  fibrous  stroma 
which  is  highly  vascular.  Kolmer  ( 1923 ) 
who  examined  the  tissue  in  Chwmero  mon- 
strosa  regarded  it  as  hemopoietic.  Its  dis- 
tribution in  the  Holocephali  is  singular. 
There  is  none  in  the  esophageal  wall,  but 
it  exists  in  a  large  mass  not  only  over  the 
palate  but  also  within  each  orbit  and  in  the 
ethmoid  canal.  The  tissue  masses  are  con- 
nected by  strands  which  run  through  foram- 
ina from  one  area  to  another.   There  seems 


190 


Bitlk'tin  Museum  of  Coiuparativc  Zoolofiy,  Vol.  135,  No.  3 


to  be  a  relatively  small  mass  of  it,  isolated 
from  the  rest,  within  a  pair  of  ventral  chan- 
nels in  the  pectoral  girdle.  Kolmer,  im- 
pressed by  the  fact  that  much  of  this  tissue 
was  surrounded  by  cartilage,  refers  to  it  as 
"knockenmarkahnliche  Gewebe."  However, 
all  of  it  seems  to  be  external  to  the  peri- 
chondrium. The  presence  of  this  tissue 
raises  more  questions  than  it  answers.  No 
one  has  dared  to  guess  whether  it  is,  in  its 
present  extent  in  the  Holocephali,  a  spe- 
cialization lately  developed  or  another 
primitive  vestige. 

Conclusions  from  the  nature  of  the  diges- 
tive tract  are  difficult  to  draw.    The  Holo- 
cephali are  extraordinary  in  the  structure 
of  their  teeth,  the  lack  of  a  stomach,  the 
design  of  the  intestinal  valve,  and  the  pres- 
ence in  association  with  the  gut  of  unique 
masses    of    glandular    and    lymphomyeloid 
material.    Examination  of  these  character- 
istics, however,  does  not  produce  extensive 
evidence   of   value   in    solving   the    phylo- 
genetic    problem.      Some    clues    may    be 
gleaned,  nevertheless.    The  greater  resem- 
blance between   the   internal   structure   of 
holocephalian    and    ptyctodont   teeth    than 
between  those  of  holocephalians  and  coch- 
liodonts  suggests,  if  such  similarities  are  at 
all  significant,  that  there  is  more  likelihood 
of  a  relationship  between  the  Holocephali 
and  the  fonner  than  the  latter  group.    The 
contrast  between  the  integral  structure  of 
holocephalian  toothplates  and  the  tendency 
towarcl  fusion  of  teeth  which  Moy-Thomas 
( 1936 )  describes  as  being  exhibited  by  the 
cochliodont    Helodus    makes    it    seem    im- 
probable that  this  type  of  cochliodont  was 
ancestral  to  the  Holocephali. 

A  hint  of  similarity  to  selachians  lies  in 
the  likeness  of  the  liver  in  the  two  groups 
of  cartilaginous  fishes.  The  affinities  of  the 
remaining  soft  parts  of  the  digestive  system 
of  holocephalians  defy  analysis.  It  is  impos- 
sible to  determine  whether  the  lack  of  a 
stomach  and  the  minimal  development  of 
the  spiral  valve  are  primitive  or  secondary 
conditions.  The  evolution  of  the  glandular 
mass  at  the  posterior  end  of  the  intestine 


and  of  the  lymphomyeloid  matter  in  the 
pharyngeal  region  is  equally  obscure.  One 
must  admit,  then,  that  little  can  be  derived 
from  an  analysis  of  the  digestive  organs  to 
reinforce  either  the  theory  of  a  selachian  or 
a  non-selachian  origin  of  the  Holocephali. 

CONCLUSION 

The  study  of  the  venous  system  of  Chi- 
maera  coUiei  was  undertaken  in  an  attempt 
to  clarify  the  evolutionary  history  of  the 
Holocephali.  The  fishes  of  this  group  have 
been  long  regarded  as  an  offshoot  from 
the  shark  line  and  as  such  have  been  placed 
with  selachians,  bradyodonts,  and  batoids, 
in  the  class  Chondrichthyes.  The  non- 
replacement  of  their  toothplates  resulted  in 
their  association  with  the  bradyodonts,  and 
through  the  work  of  Moy-Thomas  (1936) 
the  theory  was  established  that  they  might 
have  descended  from  a  cochliodont  of  that 
group.  Of  late,  however,  0rvig  ( 1962 )  has 
argued  that  the  Holocephali  are  more  prob- 
ably derived  from  a  ptyctodont  ancestor 
and  so  only  distantly  related  to  sharks. 

In  an  effort  to  re-evaluate  the  position  of 
the  Holocephali,  the  anatomy  of  the  venous 
system  was  examined  for  similarities  and 
differences  between  it  and  that  of  other 
fishes.  Undoubted  resemblances  to  the 
selachian  system  were  found  in  the  pres- 
ence and  arrangement  of  sinuses  and  in  the 
existence  of  a  subcutaneous  network  of 
veins.  The  hepatic  portal  system,  while  not 
exactly  like  that  of  sharks,  resembled  the 
selachian  system  more  nearly  than  that  of 
bony  fishes.  The  two  main  points  of  differ- 
ence from  selachians  lay  in  the  absence  of 
lateral  abdominal  veins  and  the  opening  of 
the  hepatic  veins  into  the  posterior  cardinal 
sinuses.  Further  examination  of  the  cir- 
culatory system  brought  forth  no  similar- 
ities to  the  bony  fishes  but  a  heart  of  the 
selachian  type,  and  a  unique  arrangement 
of  arteries  in  the  head  region.  It  was 
obvious  from  the  study  of  the  circulatory 
system  that  holocephalian  structure  agreed 
with  that  of  bony  fishes  only  in  the  lack 
of  lateral  abdominal  veins,  and  that  it  bore 


Morphology  and  Relationships  of  Holocephali  •  Stahl 


191 


a  much  greater  resemblance  to  the  selachian 
type.  The  peculiarities  of  holocephalian 
vessel  arrangement  gave  no  clue  as  to  their 
derivation.  It  was  impossible  to  detemiine 
whether  they  represented  modifications 
from  the  selachian  plan  or  whether  they 
had  been  inherited  from  a  non-selachian 
source. 

A  review  of  the  holocephalian  nervous, 
skeletal,  muscular,  urogenital,  and  digestive 
systems  was  made  in  the  search  for  char- 
acteristics whose  derivation  could  be  more 
clearly  interpreted.  Since  each  system  dis- 
played distinct  differences  from  the  com- 
parable system  of  bony  fish,  and  the  sim- 
ilarities to  selachian  structure  were  often 
marked,  the  degree  and  the  implications 
of  the  resemblance  to  selachians  became 
the  focal  problem. 

A  strong  similarity  between  holocepha- 
lian and  selachian  structure  allows  the 
possibility  of  the  origin  of  the  former  from 
the  latter  group  but  does  not  necessitate  it. 
The  possession  of  similar  structures  might 
also  have  occurred  through  their  inheri- 
tance from  a  common  ancestor  at  a  lower 
level  of  the  vertebrate  line.  In  the  case  of 
a  single  structure,  its  presence  may  be  the 
result  of  parallel  evolution.  The  existence 
of  characters  which  seem  unlikely  to  be 
derived  from  selachian  structures  or  of 
those  which  seem  more  primitive  than  their 
homologues  in  sharks  might  be  less  equiv- 
ocal. If  it  can  be  shown  that  a  structure 
is  basically  unlike  its  selachian  counterpart 
or  that  it  is  not  a  secondary  simplification 
of  a  form  which  exists  in  a  more  specialized 
state  in  sharks,  one  could  conclude  that  the 
Holocephali  should  logically  be  traced  back 
to  placoderm  stock  by  an  independent  line 
rather  than  to  an  early  shark  group. 

The  review  of  the  nervous  system  re- 
vealed likenesses  to  selachians  in  the  ar- 
rangement of  the  autonomic  fibers  and 
the  anatomy  of  the  sense  organs  and  pos- 
terior regions  of  the  brain.  Although  the 
unusual  form  of  the  telencephalon  could 
have  originated  as  a  modification  from  the 
selachian  plan,  it  does  not  appear  likely  that 


the  structure  of  the  pallium  itself  or  the 
simple  arrangement  of  the  cranial  nerves 
could  have  been  so  derived.  It  appears 
doubtful  too,  that  the  pattern  of  the  sen- 
sory canals  came  from  a  selachian  source. 

The  fact  that  the  skeleton  of  both  holo- 
cephalians  and  sharks  is  completely  carti- 
laginous was  once  thought  to  be  indicative 
of  close  relationship,  but  it  has  become 
apparent  that  that  conclusion  is  not  the 
only  possible  one.  Since  it  seems,  now,  that 
a  transition  from  bone  to  cartilage  occurred 
in  several  vertebrate  lines,  one  must  allow 
that  the  cartilaginous  skeletons  of  sharks 
and  holocephalians  may  have  developed 
independently.  If  one  can  look  beyond  the 
similarity  of  the  skeletal  material,  holo- 
cephalians can  be  seen  to  have  several 
skeletal  characters  that  would  be  difficult 
to  derive  from  early  sharks.  Their  form  of 
autostyly  is  distinctive.  Although  it  ap- 
pears that  autostyly  has  developed  several 
times  among  vertebrates,  it  is  hard  to  be- 
lieve that  the  arrangement  in  the  Holo- 
cephali could  be  a  modification  of  selachian 
structure.  If  it  were,  one  would  expect  to 
find  a  longer  palatoquadrate  element  rather 
than  a  short  one  with  a  process  extending 
postero-dorsally  in  finger-like  fashion  to 
reach  the  otic  region.  Also,  the  hyoid  would 
be  expected  to  show  some  sign  of  its  former 
involvement  in  the  jaw  suspension.  In  holo- 
cephalians it  does  not,  being  to  all  appear- 
ances exactly  like  the  succeeding  arches 
even  in  its  dorsal  part.  In  addition  to  the 
difference  of  the  palatoquadrate  and  hyoid 
elements  from  the  shark  type,  the  presence 
of  elaborate  labial  and  rostral  cartilages  and 
the  general  proportions  of  the  skull,  with 
its  short  otic  and  steeply  sloped  ethmoid 
areas,  distinguish  holocephalians  from  early 
sharks. 

A  study  of  the  muscular  system  produces 
less  that  is  clearly  significant.  The  similar- 
ity of  the  musculature  of  fishes  generally 
and  the  difficulty  of  ascertaining  homol- 
ogies are  obstacles  to  meaningful  analysis. 
Peculiarities  in  holocephalian  axial,  appen- 
dicular, and  branchial  muscles  are  appar- 


192        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


ent,  but  there  is  nothing  to  indicate  whether 
they  were  or  were  not  derived  from  the 
selachian  plan.  There  seems  to  be  no  sure 
ground  for  denying  that  they  could  ha\'e 
been. 

The  urogenital  system  of  holocephalians 
resembles  that  of  sharks  very  closely  in  the 
nature  of  kidneys,  the  gonads,  the  accessory 
ducts,  and  the  interrelationship  between 
those  structures.  Because  that  interrelation- 
ship is  characteristic  of  most  extant  verte- 
brates (bony  fish  are  the  cardinal  excep- 
tion ) ,  it  is  possible  to  interpret  the  arrange- 
ment as  one  which  was  widespread  among 
early  gnathostomes  and  so  obviate  the  ne- 
cessity of  deriving  the  holocephalian  system 
from  a  specifically  selachian  source.  If  one 
is  free  to  seek  its  forerunner  in  a  wide 
variety  of  early  vertebrate  groups,  one 
might  consider  the  ptyctodonts  as  having 
had  a  system  which  could  have  been  ances- 
tral to  the  holocephalian  type.  Although  no 
evidence  of  soft  organs  remains,  it  seems 
that  ptyctodonts  may  have  had,  associated 
with  the  reproductive  system,  accessory 
claspers  similar  to  those  of  holocephalians. 
No  trace  of  those  structures  appears  in  any 
other  fossil  group. 

The  digestive  system  of  the  Holocephali 
is  unlike  that  of  selachians  in  its  lack  of 
a  stomach  and  poor  development  of  the 
spiral  valve.  Among  the  soft  organs,  the 
liver  is  the  only  structure  which  bears  a 
striking  resemblance  to  its  selachian  coun- 
terpart. While  the  evolutionary  history  of 
the  soft  parts  of  the  digestive  system  is  not 
clear,  the  holocephalian  toothplates,  which 
show  no  evidence  of  having  developed 
through  a  fusion  of  separate  teeth,  seem 
not  to  ])e  derived  from  any  known  shark 
structures. 

The  general  conclusion  to  be  drawn  from 
this  study  is  that,  although  similarities  be- 
tween holocephalians  and  selachians  are 
numerous,  holocephalians  possess  certain 
characteristics  which  suggest  that  these 
fishes  evolved  from  other  than  a  selachian 
stock.  The  existence  in  sharks  and  holo- 
cephalians of  like  structures  does  not  con- 


tradict this  hypothesis,  since  such  structures 
may  have  been  carried  over  from  a  common 
ancestor  or  developed  convergently.  Even 
the  derivation  of  the  Holocephali  from  the 
bradyodont  sharks  can  be  questioned.  Al- 
though the  cochliodont  HeJodus  shows, 
according  to  Moy-Thomas,  a  number  of 
similarities  to  holocephalians,  the  teeth  of 
that  fish  show  a  tendency  toward  fusion  of 
which  there  is  no  hint  in  the  Holocephali. 
Hclodus  was  apparently  autostylic,  as  are 
the  holocephalians,  but  autostyly  has  arisen 
repeatedly  in  vertebrate  groups  and  cannot 
be  considered  as  weighty  evidence  in  favor 
of  the  holocephalian-cochliodont  relation- 
ship. There  is  as  good,  or  better,  evidence 
in  favor  of  a  relationship  between  holo- 
cephalians and  ptyctodonts.  Although  the 
ptyctodont  palatoquadrate  was  not  fused 
to  the  cranium,  the  toothplates  appear  to 
have  been  integral  structures,  and  the  body 
form,  with  the  large,  short  head,  was  sim- 
ilar to  that  of  holocephalians.  If  one  will 
concede  that  the  dennal  skeleton  of  the 
ptyctodonts  could  have  disappeared  as  the 
evolution  of  the  group  continued,  then  the 
presence  of  labial  cartilages,  rostral  proc- 
esses, anterior  and  pelvic  claspers,  a  synar- 
cual,  and  a  dorsal  fin  supported  by  radials 
posterior  to  the  dorsal  spine,  stand  forth 
as  a  substantial  and  therefore  possibly  sig- 
nificant number  of  characteristics  suggest- 
ing linkage  between  the  ptyctodont  and 
holocephalian  lines. 

In  sum,  one  may  assume  from  available 
evidence  that  holocephalians  are  not  de- 
rived from  selachians  or  bradyodonts  but 
have  evolved  along  an  independent  line. 
However,  anatomical  similarities  between 
extant  holocephalians  and  selachians  which 
set  both  groups  apart  from  the  bony  fishes 
suggest  that  these  cartilaginous  forms 
shared  a  common  ancestor.  Tliis  ancestral 
stock  must  have  existed  at  the  placoderm 
level  or  even  earlier  among  unknown  ante- 
cedent forms.  Although  the  specific  group 
of  placoderms  from  which  sharks  originated 
is  imknown,  the  ptyctodonts  may  represent 
the  root  of  the  holocephalian  line. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        193 


ACKNOWLEDGMENTS 

I  want  to  express  my  gratitude  to  Dr. 
Alfred  S.  Romer  of  Harvard  University  for 
the  guidance  and  encouragement  that  he 
has  given  me  throughout  my  research  and 
the  preparation  of  this  paper.  I  am  in- 
debted, also,  to  Dr.  Richard  Snyder  of  the 
University  of  Washington  and  to  Dr.  Nor- 
man J.  Wilimovsky  of  the  University  of 
British  Columbia  through  whose  efforts  I 
obtained  the  specimens  of  CJiimacra  colJiei 
that  I  used.  Instruction  in  the  technique  of 
latex-injection  was  given  me  by  Dr.  Richard 
Thorington  who  was,  at  the  time,  a  grad- 
uate student  at  the  Harvard  Biological 
Laboratories.  Mrs.  Myvanwy  Dick  allowed 
me  to  preserve  and  store  my  fishes  in  the 
Fish  Department  of  the  Museum  of  Com- 
parative Zoology  and  was  kind  enough  to 
lend  me  a  specimen  of  CaUorhijnchus.  The 
list  of  people  who  answered  my  letters  of 
inquiry  is  very  long.  I  owe  thanks  especially 
to  Dr.  D.  L.  Gamble  of  Ward's  and  to  Mr. 
Thomas  E.  Powell,  Jr.  of  Carolina  Bio- 
logical Supply  Company  who  sent  advice 
about  injecting  frozen-and-thawed  material. 
Lastly,  I  should  like  to  thank  Dr.  David  G. 
Stahl  \\'ho  encouraged  me  to  undertake  this 
work  and  whose  patience  and  good  \\\\\ 
enabled  me  to  finish  it. 

REFERENCES  CITED 

Allen,  \V.  F.      1905.     Tlie  blood  vascular  system 

of  the  Loricati,  the  mail-cheeked  fishes.    Pioc. 

Washington   Acad.   Sci.,   7:    27-157. 
Allis,  E.   p.     1912.     The  branchial,  pseudobran- 

chial,  and  carotid  arteries  in  Chimaera  coUiei. 

Anat.  Anz.,  42:   10-18. 
.      1916.     The   so-called  mandibular  arter\' 

and  the  persisting  remnant  of  the  mandibular 

aortic  arch  in  the  adult  selachian.    J.  Morph., 

27:  99-118. 
.      1917.     The    prechordal    portion    of    the 

chondrocranium    of    Chimaera    coUiei.     Proc. 

Zool.  Soc.  London,   1917:   105-14.3. 

1926.      On    the    homologies    of    the   pre- 


Bargmann,  W.  19.33.  Die  Zahnplatten  von 
Chimaera  tuowitrosa.  Z.  Zellforsch.,  19(3): 
537-561. 

BoRCEA,  J.  1906.  Recherches  sur  le  systeme 
urogenital  des  elasmobranches.  Arch.  Zool. 
(Paris),  Ser.  4,  4:   199-484. 

Braus,  H.  1898.  Uber  die  Innervation  der 
paarigen  Extremitaten  bei  Selachiem,  Holo- 
cephalen,  und  Dipnoen.  Ein  Beitrag  zur 
Gliedmassenfrage.  lena.  Z.  Med.,  31:  239- 
468. 

Brettnacher,  H.  1939.  Aufbau  und  Struktur 
der  Holocephalenziihne.  Z.  Mikr.  Anat. 
Forsch.,  46:   584-616. 

Burlend,  T.  H.  1910.  The  urogenital  organs  of 
Chimaera  monstrosa.  Proc.  Zool.  Soc.  Lon- 
don, 1910:  510-5.34. 

Citterio,  V.  19.32.  Richerche  sul  canale  di- 
gerente  di  Chimaera  momtrosa.  Atti  Soc.  Ital. 
Milano,  71:  9.3-118. 

Cole,  F.  J.  1896.  The  cranial  nerves  of  CJii- 
maera  moruirosa.  Trans.  Rov.  Soc.  Edinburgh, 
38:  49-56. 

Cole,  F.  J.  and  W.  J.  Dakin.  1906.  Further 
observations  on  the  cranial  nerves  of  CJii- 
maera.    Anat.  Anz.,  28:   59.5-599. 

Daniel,  J.  R.  1934.  The  elasmobranch  fishes. 
Berkeley,  334  pp. 

Davidoff,  M.  v.  1879.  Beitrage  zur  vergleich- 
enden  Anatomie  der  hinteren  Gliedmasse  der 
Fische.  I.  Haie,  Chimaera  und  Ganoidei 
chondrostei.     Morph.   Jahrb.,   5:    450-520. 

Dean,  B.  1895.  Fishes  living  and  fossil.  Mac- 
millan,  New  York,  300  pp. 

.      1904.     In  the  matter  of  the  Permian  fish 

Mena.spi.s.    Amer.  Geol.,  34:   49-53. 

.  1906.  Chimaeroid  fishes  and  their  de- 
velopment. Carnegie  Inst.  Washington,  Pub. 
No.  32,   172  pp. 

1909.     Studies   on   fossil  fishes    (sharks, 


chordal    portions    of   the    skull    of    the    Holo- 
cephali.   J.  Anat.,   60:   33.5-340. 
Backstrom,  K.      1924.     Contributions  to  the  fore- 
brain   morphology   in   selachians.     Acta   Zool., 
5:    12.3-240. 


chimaeroids,  arthrodires).  Mem.  Amer.  Mus. 
Nat.  Hi.st.,  9:  207-287. 

DeBeer,  G.  R.  and  J.  A.  Moy-Thomas.  1935. 
On  the  skull  of  the  Holocephali.  Phil.  Trans. 
Roy.  Soc.  London,  B  224:  287-312. 

Edgeworth,  F.  H.  1935.  The  cranial  muscles 
of  vertebrates.    Cambridge,  300  pp. 

Fahrenholz,  C.  1915.  Uber  die  Verbreitung 
\on  Zahnbildungen  und  Sinnesorganen  in 
Vorderdarm  der  Selachier  und  ihre  phyloge- 
netische  Beurteilung.  Jena.  Z.  Naturwiss., 
53:   .389-444. 

Fee,  a.  R.  192.5.  The  histology  of  the  colon 
and  its  contained  spiral  valve  of  the  Pacific 
Coast  dog-fish  (Squahis  sucklii)  with  an  in- 
vestigation of  the  phylogeny  of  the  intestinal 
valve.  Proc.  Trans.  Roy.  Soc.  Canada,  Sect. 
V,  19:  169-193. 

FiJRBRLNGER,  M.  1897.  Uber  die  spino-occipi- 
talen  Nerven  der  Selachier  und  Holocephalen 


194         Bulletin  Museum  of  Comparative  Zoologij,  Vo/.  135,  No.  3 


unci  ihre  vergleichende  Morphologic.  In: 
Festschr.  70  Gelnirtstage  C.  Gcgenbaur,  Leip- 
zig, 3:   349-788. 

Carman,  S.  1888.  On  the  lateral  canal  system 
of  the  Selachia  and  Holocephalia.  Bull.  Mus. 
Comp.  Zool.,   17:  57-119. 

.      1904.     The  chimaeroids,  especially  Wii- 

nochimacra  and  its  allies.  Bull.  Mus.  Comp. 
Zool.,  41:  243-272. 

Cross,  W.  1957.  Mundziihne  und  Hautziihne 
der  Acanthodier  und  Arthrodiren.  Paleon- 
tographica,    109(A):    1-40. 

Holmgren,  N.  1922.  Points  of  view  concerning 
forebrain  morphology  in  lower  vertebrates. 
J.   Comp.   Neur.,   34:    391-440. 

.  1942a.  Ceneral  morphology  of  the  lat- 
eral sensory  line  system  of  the  head  in  fish. 
K.  Svenska  Veteiiskaps-Acad.  Handl.,  Ser.  3, 
20(1):    1-46. 

1942b.      Studies   on   the   head   of  fishes, 


an  embryological,  moiphological,  and  phylo- 
genetical  study.  3:  The  phylogeny  of  elasmo- 
branch  fishes.    Acta  Zool.,  23:   129-261. 

Holmgren,  N.  and  C.  J.  van  der  Horst.  1925. 
Contributions  to  the  morphology  of  the  lirain 
of  Ccratodus.    Acta  Zool.,   6:    59-165. 

Hurrec:ht,  a.  A.  W.  1877.  Beitriige  zur  Kennt- 
niss  des  Kopfskeletts  der  Holocephalen.  Nie- 
derl.  Archiv.  f.  Zool.,  3:  255-276. 

Hyman,  L.  H.  1942.  Comparative  vertebrate 
anatomy.    2nd  ed.,  Chicago,  544  pp. 

Hyrtl,  J.  1853.  Ueber  weibliche  Oviducte  bei 
miinnlichen  Chimaeren,  imd  cine  miinuliche 
Vesicula  seminalis  bei  Weibchen.  Sitzungsber. 
K.  Akad.  Wiss.  (Vienna),  11(5):   1078-1087. 

JACOH.SHAGEN,  E.  1915.  Untcrsuchungeu  iiber 
das  Darmsystem  der  Fische  und  Dipnoer. 
Jena.  Z.   Natur%viss.,  53:  44.5-556. 

.      1934.     Das     Problem    des    Spiraldarms. 

Morphol.  Jahrb.,  73:  392-445. 

.      1941.      Crundlinien  einer  vergleichenden 

Anatomic  des  Zahnbeines  und  der  Ziihne 
niederer  Wirbeltiere.  Z.  Mikros.  Anat.  Forsch., 
49:  225-272. 

Johnston,  J.  B.  1910.  A  note  on  tlic  forebrain 
of  Chinmera.    Anat.  Anz.,  36:  233-242. 

Kappers,  C.  U.  a.  1912.  The  arrangement  of 
the  motor  nuclei  in  Chimacru  moristrosa  com- 
pared with  other  fishes.  Proc.  K.  Akad. 
Wetensch.,   14:   1176-1180. 

Kappers,  C.  U.  A.  and  F.  W.  Carpentier.  1911. 
Das  Gehirn  von  Chimaera  monstrosa.  Folia 
Neurobiologica,  5:   127-160. 

Kappers,  C.  U.  A.,  G.  C.  Huber,  and  E.  C. 
Crosby.  1936.  The  comparative  anatomy 
of  the  nervous  system  of  vertebrates.  3  Vols., 
New  York,   1845  pp. 

Kesteven,  H.  L.  1933.  The  anatomy  of  the 
head  of  Callorhynchtis  (iittdicticus.  J.  Anat., 
67:  443-474. 


.      1942-1943.     The  evolution  of  the  skull 

and  the  cephalic  muscles.  Part  I.  The  fishes. 
Mem.  Aus.  Mus.,  8:  1-132. 

Kolmer,  W.  1923.  Uber  das  Vorkommen  eines 
knockenmarkahnlichcn  Gewebes  bei  einem 
Selachier  ( Knoipelmark  bei  Chimaera  luon- 
struosa).    Anat.  Anz.,  56:   529-534. 

Lankester,  E.  R.  1878.  On  the  hearts  of  Cera- 
todtis;  Protopterus,  and  Chimaera.  Trans. 
Zool.   Soc.  London,   10:   493-505. 

Lek;h-Sharpe,  W.  H.  1920-1926.  The  com- 
parative morphology  of  the  secondary  sexual 
characters  of  elasmobranch  fishes.  J.  Morph., 
34:  245-265;  35:  359-380;  36:  221-243; 
39:    558-577;  42:    307-308. 

Leydig,  F.  1851.  Zur  Anatomic  und  Histologic 
der  Cliimaera  mon.stri)sa.  Arch.  Anat.  Physiol., 
1851:  241-271. 

Luther,  A.  1909.  Beitriige  zur  Kenntnis  von 
Musculatur  imd  Skelett  des  Kopfes  des  Haies 
Stefiostoma  tifiriniim  Cm.  und  der  Holo- 
cephalen.   Acta  Soc.  Sci.  Fennicae,  37:  1-60. 

Marples,  B.  J.  1936.  The  blood  vascular  system 
of  the  elasmobranch  fish  Sciuatina  squatimi 
(Linne).  Trans.  Rov.  Soc.  Edinburgh,  58: 
817-840. 

Maurer,  Fr.  1912.  Die  ventrale  Rumpfmusku- 
latur  der  Fische.  lena.  Z.  Naturwiss.,  49: 
1-118. 

Moy-Thomas,  J.  A.  1936.  On  the  structure  and 
affinities  of  the  Carboniferous  cochliodont 
Helodus  simph'x.    Ceol.  ^h^g.,  73:  488-503. 

NicoL,  J.  A.  C.  1950.  The  autonomic  ner\'ous 
system  of  the  chimaeroid  fish,  Hijdrola^us 
coUiei.    Quart.   J.   Micros.   Sci.,  91:    379-400. 

Nielsen,  E.  1932.  Permo-carboniferous  fishes 
from  East  Greenland.  Medd.  om  Gr0nland, 
92(3):  1-63. 

.      1952.     On    new    or    little-known    Edes- 

tidae  from  the  Permian  and  Triassic  of  East 
Greenland.  Medd.  om  Gr0nland,  144(5): 
1-55. 

O'Donoghue,  C.  H.  1914.  Notes  on  the  circula- 
tory system  of  elasmobranchs,  I.  The  venous 
system  of  the  dogfish  (Sei/Uium  cauietda). 
Proc.  Zool.  Soc.  London,  19i4:  435-455. 

0RVIG,  T.  1951.  Histologic  studies  of  placo- 
derms  and  fossil  elasmobranchs,  I.  The  endo- 
skeleton  with  remarks  on  the  hard  tissues  of 
lower  xertebrates  in  general.  K.  Svenska 
Vetenskap.-Akad.,  Ark.  Zool.,  Ser.  2,  2(2): 
.321-454. 

.      1957.      Notes    on    some    Paleozoic   lower 

vertebrates  from  Spitzbergen  and  North  Amer- 
ica.   Norsk  Geol.  Tids.skr.,  37(3-4):  285-353. 

.      1962.      Y  a-t-il  une  relation  directc  entre 

les  arthrodircs  ptyctodontides  et  les  holo- 
cephales?  Problemes  Actuels  de  la  Paleon- 
tologie,  Centre  Nat.  Recher.  Sci.,  No.  104: 
49-61. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        195 


Pander,  C.  H.  1858.  Uber  dir  Ctenodipterinen 
des  devonischen  Systems.  St.  Petersburg, 
65  pp. 

Parkeh,  T.  J.  1886.  On  tlie  blood  vessels  of 
Mustelus  (into  ret  if  us.  Phil.  Trans.  Roy.  Soc. 
London,  Ser.  B,   177:  685-731. 

Patterson,  C.  1965.  The  phylogeny  of  the 
chimaeroids.  Phil.  Trans.  Roy.  Soc.  London, 
Ser.  B,  249:  101-219. 

Prasad,  R.  R.  1948.  Observation  on  the  nida- 
mental  glands  of  Hydrohigus  colliei.  Raja 
rhiiui  and  Plati/ihinoidis  tri.scriatti.s.  Copeia, 
No.   1:  54-57.  ' 

Rabinerson,  a.  1925.  Beitriige  znr  vergleichen- 
den  Anatomie  der  Wirbelsaule  der  Knorpel- 
fische.    Anat.  Anz.,   59:   433-454. 

Radinsky,  L.  1961.  Tooth  histology  as  a  taxo- 
nomie  criterion  for  cartilaginous  fishes.  J. 
Morphol.,    109:    73-92. 

Redeke,  H.  C.  1898.  Onderzoekingen  betref- 
fende  het  Urogenitaalsysteem  der  Selachiers 
en  Holocephalen.    Amsterdam,  85  pp. 

Reese,  A.  M.  1910.  The  lateral  Hue  system  of 
Chimaera  colliei.    ].  Exp.  Zool.,   9:   349-370. 

RoMER,  A.  S.  1964.  The  braincase  of  the  Paleo- 
zoic elasmobranch  Tamiohatis.  Bull.  Mus. 
Comp.   Zool.,   131(4):    89-105. 

Scammon,  R.  E.  1915.  The  histogenesis  of 
selachian  liver.    Amer.  J.  Anat.,  17:  245-315. 

ScHAUiNSLANu,  H.  H.  1903.  Beitriige  zur  Ent- 
wicklungsgeschichte  und  Anatomie  der  Wir- 
beltiere,  I.,  Sphenodon,  Callorhijnchus,  Cha- 
maeleo.    Zoologica,  39:   1-98. 

Shann,  E.  W.  1919.  The  comparative  myology 
of  the  shoulder  girdle  and  pectoral  fin  of 
fishes.  Trans.  Roy.  Soc.  Edinburgh,  52:  531- 
570;  continued  in  Proc.  Zool.  Soc.  London, 
1924:   195-215. 


SiWE,  S.  A.  1926.  Pancreasstudien.  Morphol. 
Jahrb.,  57:  84-307. 

Stensio,  E.  1947.  The  sensory  lines  and  dermal 
bones  of  the  cheek  in  fishes  and  amphibians. 
K.  Svenska  Vetenskaps.-Akad.  Handl.,  Ser.  3, 
24(3):    1-195. 

.      1963.     Anatomical  studies  on  the  arthro- 

diran  head.  Part  I.  K.  Svenska  Vetenskaps.- 
Akad.  Handl.,  9(2):  1-419. 

Van  den  Broek,  A.  J.  P.  1938.  Gonaden  und 
Ausfiihrungsgiinge.  In:  Bolk,  L.  et  al.,  Hand- 
buch  der  vergleichenden  Anatomie  der  Wir- 
beltiere.  Vol.  6:    1-154. 

Van  Gelderen,  Chr.  1938.  Venensystem,  mit 
einem  Anhang  iiber  den  Dotter-  und  Plazen- 
tarkreislauf.  7/i:  Bolk,  L.  et  al.,  Handbuch 
der  vergleichenden  Anatomie  der  Wirbeltiere, 
Vol.  6:   685-744. 

Van  der  Horst,  C.  I.  1934.  Spinalnerven.  In: 
Bolk,  L.  et  al.,  Handbuch  der  vergleichenden 
Anatomie   der   Wirbeltiere,   Vol.   2:    505-540. 

Van  Oordt,  G.  J.  1938.  Besondere  Morphologie 
der  Harnorgane  der  Wirbeltiere.  In:  Bolk, 
L.  et  al.,  Handbuch  der  vergleichenden  Anat- 
omie der  Wirbeltiere,  Vol.  5:   727-836. 

Vetter,  B.  1878.  Untersuchungen  zur  verg- 
leichenden Anatomie  der  Kiemen-  und  Kiefer- 
muskeln  der  Fische.  II  Teil.  Jena.  Z.  Natur- 
wiss.,   12:   431-550. 

Watson,  D.  M.  S.  1938.  A  ptyctodont  from  the 
middle  Old  Red  Sandstone  of  Scotland.  Trans. 
Roy.  Soc.   Edinburgh,  59(2):   .397-410. 

Wilder,  B.  1877.  On  the  brain  of  Chinuiera 
monstrosa.  Proc.  Acad.  Nat.  Sci.  Philadelphia, 
1877:  219-250. 

(Received  27  Septend)ci\  1965.) 


196        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


-*- 

«^ 

o 

-C 

X 

**-' 

-Q 

a 

a 

-6 

<U 

o 

.^ 

E 

> 

c 

c 

Qi 

>. 

> 

'53 

> 

> 

o 

o 

c 

-6 
E 

3 

o 

c 



"*~ 

o 

o 

c 

^ 

3 

12 

OJ 

o 

U 

o 

> 

3 

-Q 

-D 

a 

3 

o 

>' 

c 

3 

o 

(U 

c 
o 

o 

u 

C) 

'-*— 

c 

-D 

3 

.- 

< 

r9 

c 

.- 

o 

*~ 

> 

0 

d 

D 

u 

D 

o 

o 

-g 

O 

S 

c 

-D 

>' 

a 

o 

^ 

^ 

C 

3 

>i 

U 

P 

3 
J3 

o 

TD 

^ 

X 

u 

lU 

i 

$ 

c 
o 

c 

_D 

-Q 

t/i 

3 

a> 

a 

> 

u 

"5 

o 

trt 

(J 

Q- 

D 
0) 

o 

o 

(U 

o. 

O 

u 

^ 

— ' 

d 

^ 

"5 

to 

^ 

>- 

.5 

o 

TD 

P 

— ; 

-□ 

c 

d 

3 

o 

o 

jn 

U 

i 

o 

-c 

> 

■*- 

o 

a 



1o 



OJ 

u 

o 

o 

o 

o 

T3 

"x 

a 

0) 

>^ 

D 

(U 

3 

U 

O 

"5 

-C 

6 

u 

"o 

3 
J2 

0 

D 

0) 

c 

- 

6 

i 

> 

> 

3 

o 

D 

3 

■^ 

3 

o 

U 

>■ 

o 

u 

0) 

.^ 

lU 

^ 

^ 

Q. 

^ 

c 

_^ 

X 

O 

u 

o 

o 

c 

~D 

o 

3 

3 

3 

o 

"D 

-Q 

d 

-Q 

<U 

>^ 

3 
C 

o 

"5 

j: 

o 

m 

.^ 

-c 

1 — 

c 

Q. 

3 

C 

OJ 

> 

(U 

_ 

"o 

"5 

O 

o 

B 

X 

o 

o 

c 

ID 

Q_ 

o 

"D 

~o 

> 

Morphology  and  Relationships  of  Holocephali  •  StaJiI        197 


a.v.  int.v. 


L-L,V. 


v.  in+.v. 


Plate  2.  The  hepatic  portal  system  of  Chimaera  colliei.  Diagrammatic  view.  X  0.75.  a.d.int.t.,  Anterior  dorsal  intestinal 
tributary;  oux.  spl.v.,  auxiliary  splenic  vein;  a. v. Inf. v.,  anterior  ventral  intestinal  vein;  h.p.v.,  hepatic  portal  vein;  i-i.v.,  intra- 
intestinal  vein;  /-p. v.,  lieno-pancreatic  vein;  mes.v.,  mesenteric  vein;  p.d.tnt.v.,  posterior  dorsal  intestinal  vein;  p.v./nt.v., 
posterior  ventral    intestinal   vein. 


198        Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  3 


Plate  3.  The  systemic  and  renal  portal  veins  of  Chimaera  colliei.  Diagrammatic  view.  X  0.5.  o.br.v.,  Anterior  brachial 
vein;  ant. card.,  anterior  cardinal  sinus;  onf.cer.o.,  anterior  tributary  of  the  anterior  cerebral  vein;  ant.cer.p.,  posterior 
tributary  of  the  anterior  cerebral  vein;  ant.sbct.v.,  anterior  subcutaneous  vein;  br.i.,  brachial  sinus;  br.s.mid-v.exf.,  mid- 
ventral  extension  of  brachial  sinus;  coud.v.,  caudal  vein;  efh.v.,  ethmoidal  vein;  iem.v.,  femoral  vein;  h.v.,  hepatic  vein; 
inl.jug.v.,  inferior  |ugular  vein;  il.v.,  iliac  vein;  lot. cut. v.,  lateral  cutaneous  vein;  m-f.v.,  maxillo-facial  vein;  o-n.v.,  orbito- 
nasal vein;  orb.s.,  orbital  sinus;  ov.s.,  oviducol  sinus;  par. v.,  parietal  vein;  p.br.v.,  posterior  brachial  vein;  post. card., 
posterior  cardinal  sinus;  post. card. mid-v.exf.,  mid-ventral  extension  of  posterior  cardinal  sinus;  post.cer.v.,  posterior  cere- 
bral vein;  postorb.v.,  postorbital  vein;  preorb.v.,  preorbital  vein;  prescap.f.,  prescapular  tributary;  rect.trib.,  rectal  tributary; 
rn.v.,  renal  vein;  r.p.v.,  renal  portal  vein;  sbsc.s.,  subscapular  sinus;  sbsc.tnb.,  subscapular  tributary;  s.v.,  sinus  venosus; 
v-a.par.v.,   ventro-anterior   parietal   vein;    v-p.par.v.,   ventro-posterior  parietal   vein. 


Morphology  and  Relationships  of  Holocephali  •  Stahl 


199 


on^.cer.  a. 
etW.v. 
o-n.  V. 
preorb.v. 
orvt-cer  p. 
orb.  s. 


rr\ 


-f. 


posiorbv. 
cxni.cord. 

post.cer.v. 

iiaf  JLLci.v. 

sbsc.s. 

to  airuxm 

s.v. 

br.  s. 

br.  s.mui-v.  ejd. 

p  ost .  card .  m  ui- V.  ext . 

post.  cord. 

V-a.par.v. 

oar.  V. 

from  V\.W. 
ov.  S. 


V-p.  parv. 
or.  V, 


rect.-trib. 
"fem.v: 

d.v. 


caud.v. 


200        BiiUetin  Miisciiu}  of  Comparative  Zoology,  Vol.  135,  No.  3 


Plate  4.  A,  The  subcutaneous  veins  of  the  closper  and  pelvic  fin.  Ventral  view.  X  1-  B,  The  heart  and  vessels  of  the 
hypobranchial  region.  Ventral  view.  Coracomandibuloris  muscle  and  right  half  of  pectoral  girdle  removed.  X  1.  ob.p., 
Abdominal  pore;  aff.brn.a.,  afferent  branchial  artery;  onf. cl.,  anterior  closper;  br.o.,  brachial  artery;  br.n.,  brachial 
nerve;  br.s.,  brachial  sinus;  c.o.,  conus  arteriosus;  c-brn.m.,  coracobranchialis  muscle;  c-h.m.,  coracohyoideus  muscle;  cl.v., 
closper  vein;  c-m.m.,  coracomandibuloris  muscle;  com. card.,  common  cardinal  vein;  cor.c,  coracoid  cartilage;  hy.c,  hyoid 
cartilage;  hyp.m.,  hypaxial  muscle;  hypobrn.n.,  hypobranchial  nerve;  ml. jug. v.,  inferior  jugular  vein;  mand.c,  mandibular 
cartilage;  m.w.g.c,  medial  wall  of  gill  chamber;  pect.l.,  pectoral  fin;  post. card.,  posterior  cardinal  sinus;  sbct.v.pe/.f.,  sub- 
cutaneous veins  of  pelvic  fin;  s.v.,  sinus  venosus;  frib./nf.|ug.,  inferior  jugular  tributary;  v.o.,  ventral  aorta;  x,  fine  vein 
accompanying    ventral    aorta. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        201 


P 


eWlcVln 


sbc^.y.pe\,f, 


clasper 


A 


marud.c. 
C-rr\.m. 
aff  brn.Q, 

C-brn.  m. 


Irlb.im.juflT 


C-h.m. 


v.a. 

X 


Wypobrn.a. 
TOscLol 


ventricle 
atnum 
cor.c.  (cui) 
hyp. 

br.  s 


202         Bulletin  Muscuiu  of  Comparative  Zoologij,  Vol.  135,  No.  3 


Plate  5.  A,  Origin  of  right  inferior  jugular  vein,  showing  drainage  of  tissues  immediately  posterior  to  mandible.  Ventral 
view.  Coracomandibuloris  muscle  cut  and  deflected  toward  midline.  XI-  B,  The  brachial  veins.  Postero-dorsal  view  of 
right  pectoral  fin,  proximal  region.  XI-  C,  The  systemic  veins  entering  the  sinus  venosus.  Diagrammatic  view.  X  0.5. 
D,  The  anterior  cerebral  vein  and  its  tributaries.  Lateral  view.  Cartilage  removed  to  show  ethmoid  and  cranial  cavities. 
X  1-  a.br.v..  Anterior  brachial  vein,-  ant. card.,  anterior  cardinal  sinus;  ant.cer.a.,  anterior  tributary  of  the  anterior  cere- 
bral vein;  ant.cer.p.,  posterior  tributary  of  the  anterior  cerebral  vein;  onf.cer.v.,  anterior  cerebral  vein;  ont.v.const.m., 
anterior  ventral  constrictor  muscle;  a-v.,  antero-ventral;  far.o.,  brachial  artery;  br.s.,  brachial  sinus;  cor/.,  cartilage;  cfa/., 
cerebellum;  cer.o.,  cerebral  artery;  c-m.m.,  coracomandibuloris  muscle;  com. card.,  common  cardinal  vein;  ent.orb.s.,  en- 
trance to  orbital  sinus;  ep.,  epiphysis;  eth.v.,  ethmoidal  vein;  hyp.m.,  hypaxial  muscle;  inf. jug. v.,  inferior  jugular  vein; 
Inf.hy.m.,  interhyoideus  muscle;  inf. orb. sept.,  interorbital  septum;  lev.m.,  levator  muscle;  lig.,  ligament;  lym.,  lymphomyeloid 
tissue;  mond.c,  mandibular  cartilage;  n.cop.,  nasal  capsule;  nos.,  nostril;  o-n.v.,  orbito-nasal  vein;  opt. a.,  optic  artery; 
opt. I.,  optic  lobe;  p.br.v.,  posterior  brachial  vein;  peel. I.,  pectoral  fin;  pect.gir.,  pectoral  girdle;  poit.card.,  posterior  car- 
dinal sinus;  psb.o.,  pseudobronchial  artery;  sfasc.s.,  subscapular  sinus;  scop.,  scapula;  sp.n.,  spinal  nerve;  sup.oph.n.,  super- 
ficial ophthalmic  nerve;  s.v.,  sinus  venosus;  tel.,  telencephalon;  thy.gld.,  thyroid  gland;  tr.inl.m.,  trapezius  internus  muscle; 
V. const. m.,  ventral    constrictor   muscle;    //,  optic   nerve;    III,   oculomotor  nerve;    IV,   trochlear  nerve;   X,   vagus   nerve. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        203 


A 


man.ci.c. 
insertion  c-m.m 
Itq.onl.v.const.m. 
•fromf^ro?TnoatW 
orx^.v.  const,  m 


V.  const. m. 


^'1//!/"'^ int-Wym. 

WW//////;^ C-m.m. 


C-m.m. 
tky.qld. 


.?. 


n 


V. 


B 


bra. 
UQamen.-t 
5ca^.  (cjui.) 
tocom.cara. 

to  a-v.  »ul£ 
of  pecl.T. 
channel  through 
pect.qlr- 
Lev.  m. 
p.  br  V. 
pect.f. 


C 


sbsc.s. 

oni.cxxrd. 
tr.  lnk.ir\. 
or\t.coj-dl. 
scop- (cut) 
irtP.  juq.v. 
com.  card. 

S.v. 


D 


Ini. orb.  sept.  SUp.opb.a 


204         Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  3 


Plate  6.  A,  Systemic  veins  and  related  structures  in  the  postero-dorsai  region  of  the  head.  Lateral  view.  X  1-  B, 
Veins  draining  dorsal  region  of  trunk.  Lateral  view.  Epaxial  muscles  cut  and  partially  removed.  Scapular  cartilage  cut 
and  deflected  ventrally.  X  1-  o.ep.v.,  Anterior  epaxial  vein,-  anf.cord.,  anterior  cardinal  sinus;  a.v.s.c,  anterior  vertical 
semicircular  canal;  brn.n.,  branchial  nerve;  chcr.,  chondrocranium;  com. cord.,  common  cardinal  vein;  d. const. m.,  dorsal 
constrictor  muscle;  d.f.s.,  dorsal  fin  sinus;  d.sp.,  dorsal  spine;  endl.d.,  endolymphatic  duct;  ep.m.,  epaxial  muscle;  lot. cut. v., 
lateral  cutaneous  vein;  lym.,  lymphomyeloid  tissue;  m.d.v.,  median  dorsal  vein;  m-l.v.,  maxillo-faciol  vein;  of. cop.,  otic 
capsule;  p.cbr.s.,  posterior  cerebral  sinus;  post. card.,  posterior  cardinal  sinus;  posf.cer.v.,  posterior  cerebral  vein;  pos/orb.v., 
postorbital  vein;  post.scap.fnb.,  postscapular  tributary;  sbsc.tnfa.,  subscapular  tributary;  s-b.v.,  spino-basal  vein;  scap.,  scap- 
ula; sp.n.,  spinal  nerve;  tr.int.m.,  trapezius  internus  muscle;  V,  trigeminal  nerve;  VII,  facial  nerve;  VII,hyo.,  hyomandlbular 
branch    of   facial    nerve;    IX,    glossopharyngeal    nerve;    X,    vagus    nerve. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        205 


cbr.s. 


cKcr. 


scap. 
s-b.v, 


ont.card 
tr.  inlm 

brn.rv 


d.'. 


to  post. 

scop.trib. 

s-b.v. 

a.cp.v. 
post.  cand. 
porl.cer.  V. 
ot.  cap. 

X 

ant.carj. 
to  com. canal. 

d. const. m. 

scop.  Uid) 
sbsc.  -brib. 


206        Bulletin  Museum  of  Coniixiidtive  Zoology,  Vol.  135,  No.  3 


Plate  7.  A,  The  maxillo-facial  vein  and  its  tributaries.  Lateral  view.  Lower  portion  of  adductor  mandibulae  muscle  re- 
moved. XI-  B,  Deep  veins  associated  with  lymphomyeloid  tissue  dorsal  to  mouth  cavity.  Lateral  view.  Palatoquodrate 
cartilage  cut  and  partially  removed.  XI-  C,  The  femoral  vein.  Lateral  view.  Right  side.  X  0.75.  cart.,  Cartilage; 
d.lob.v.,  deep  labial  vein;  etf.rn.v.,  efferent  renal  vein;  ex.ov.op.,  external  oviducal  opening;  fern. a.,  femoral  artery; 
lem.v.,  femoral  vein;  hy.c,  hyoid  cartilage;  lab. cart.,  labial  cartilage;  /ofa.s.,  labial  sinus;  lym.,  lymphomyeloid  tissue; 
mond.arf.,  mandibular  articulation;  m-f.v.,  maxillo-facial  vein;  m.tp.,  mandibular  foothplate;  n.cop.,  nasal  capsule;  o-n.v., 
orbito-nasal  vein;  orb.s.,  orbital  sinus;  ov.s.,  oviducal  sinus;  pel.gtr.,  pelvic  girdle;  post. card.,  posterior  cardinal  sinus; 
postorb.v.,  postorbital  vein;  pq.,  palatoquodrate;  preorb.m,,  preorbitalis  muscle;  preorb.v.,  preorbitol  vein;  psb.o.,  pseu- 
dobranchial  artery;  rect.trib.,  rectal  tributary;  r.p.v.,  renal  portal  vein;  som.,  samentosche;  v. const. m.,  ventral  constrictor 
muscle;  v.tp.,  vomerine  toothplate;  y,  hyoid  tributary;  z,  possible  venous  pathways;  V,  trigeminal  nerve;  VII,  facial  nerve; 
V//,hyo.,   hyomondibular  branch  of  facial   nerve;  VII, pal.,   palatine  branch  of  facial   nerve. 


Morphology  and  Relationships  of  Holocephali  •   Stahl 


207 


r\. 


A 


Ell.pcd. 

po-slorb.v^. 


r  p.v. 

^em.Q. 

ex.cv.op. 

liaamertV. 

recttrvh, 

pd.  qIt. 
(cut) 


208        Bulletin  Musctim  of  Comparative  Zoology,  Vol.  135,  No.  3 


Plate  8.  A,  The  renal  portal  vein.  Lateral  view.  Left  side.  XI.  B,  The  hepatic  portal  system.  Dorsal  view.  X  1- 
o.d.int.f.,  anterior  dorsal  intestinal  tributary;  aif.rn.v.,  afferent  renal  vein;  a. v. int. v.,  anterior  ventral  intestinal  vein;  b.d., 
bile  duct;  esoph.,  esophagus;  iem.a.,  femoral  artery;  fern. v.,  femoral  vein;  g.b.,  gall  bladder;  b.p.v.,  hepatic  portal  vein; 
hyp.m.,  hypaxiol  muscle;  i-i.a.,  intra-intestina!  artery;  i-i.v.,  intra-intestinal  vein;  il.v.,  iliac  vein;  k.,  kidney;  mes.,  mesen- 
tery; mei.v.,  mesenteric  vein;  ov.s.,  oviducal  sinus;  pan.,  pancreas;  pan.d.,  pancreatic  duct;  par.v.,  parietal  vein;  p. d. int. v., 
posterior  dorsal  intestinal  vein;  pel.gir.,  pelvic  girdle;  p.mes.o.,  posterior  mesenteric  artery;  post. cord.,  posterior  cardinal 
sinus;    p. V. int. v.,    posterior   ventral    intestinal    vein;    r.p.v.,    renol   portal  vein;  spl.,  spleen. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        209 


.V.  int.v. 


210        Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  3 


Plafe  9.  A,  Hepatic  veins.  Lateral  view.  Right  side.  XI-  B,  Hepatic  veins.  Lateral  view.  Left  side.  X  1-  br.s. 
Brachial  smus;  epid.,  epididymis;  ien.mem.,  fenestrated  membrane;  g.b-,  gall  bladder;  h.p.v.,  hepatic  portal  vein;  h.v. 
hepatic  vein;  L.gl.,  Leydig's  gland;  mes.,  mesentery;  pect.gir.,  pectoral  girdle;  post. card.,  posterior  cardinal  sinus;  sem.ves. 
seminal  vesicle;  test. a.,  testicular  artery;  t.s.,  transverse  septum;   v.d.,  vas  deferens. 


Morphology  and  Relationships  of  Holocephali  •  Stahl        211 


.card,     sem.ves. 


L.qLN   ^v.d. 


eft  epid. 


212        Bulletin  Miiseiivi  of  Comparative  Zoology,  Vol.  135,  No.  3 


Plate  10.  A,  The  hepatic  portal  system:  veins  draining  the  intestine.  Xl-  B,  The  hepatic  portal  system:  veins  draining 
the  pancreas  and  the  spleen.  X  1-  a.d.int.t.,  Anterior  dorsal  intestinal  tributary;  aux.spl.v.,  auxiliary  splenic  vein;  a. v. int. v., 
anterior  ventral  intestinal  vein;  b.d.,  bile  duct;  coei.a.,  coeliac  artery;  esoph.,  esophagus;  g.b.,  gall  bladder;  h.o.,  hepatic 
artery;  h.p.v.,  hepatic  portal  vein;  i-i.a.,  intra-intestinal  artery;  t-i.v.,  intra-intestinal  vein;  l-p.v.,  lieno-pancreatic  vein; 
mei.v.,  mesenteric  vein;  pan.,  pancreas;  pon.d.,  pancreatic  duct;  pan. v.,  pancreatic  vein;  p. d. int. v.,  posterior  dorsal  intes- 
tinal vein;  p.mes.a.,  posterior  mesenteric  artery;  p-s.a.,  pancreatico-splenic  artery;  p. v. in/. v.,  posterior  ventral  intestinal 
vein;   spL,  spleen;   spl.v.,   splenic  vein. 


Morphology  and  Relationships  of  Holocephali  •  Staid        213 


esopK 


L-i,v. 
jan.d 


a.dL.irck.t 


p  V.  trtt.v. 


).ci,vrv^.v. 


rectum . 


B 


cx>el.a 


p.v.  ini.v. 
p,  d.int.v. 


'^mi.'W'i^iVu'?. 


SulUiin 


OF     TH 


seum   o 


■tXiJ^'UV--^'  h!K 


A    Review    of   the    Mesochrysinae    and 
Nothochrysinae    (Neuroptera:    Chrysopidae) 


PHILLIP  A.  ADAMS 


HARVARD   UNIVERSITY 

CAMBRIDGE,  MASSACHUSETTS,   U.S.A. 


VOLUME    135,   NUMBER  4 
FEBRUARY  24,   1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

MEMoms  1864-1938 

JoHNSONiA,  Department  of  Mollusks,  1941- 

OccASiONAL  Papers  on  Mollusks,  1945- 


Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects.  $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).   $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.  $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.   ( Complete  sets 
only, ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©  The  President  and  Fellows  of  Harvard  College  1967. 


A  REVIEW  OF  THE  MESOCHRYSINAE  AND 
NOTHOCHRYSINAE  (NEUROPTERA:  CHRYSOPIDAE) 


PHILLIP    A.    ADAMS^ 


ABSTRACT 

In  this  preliminary  subfamilial  classifica- 
tion of  the  Chrysopidae,  Mesochrysopidae 
is  reduced  to  subfamilial  rank.  Keys  to  sub- 
families and  to  genera  of  Nothochrysinae 
are  presented.  The  Nothochrysinae,  as 
newly  constituted,  is  characterized  by  re- 
tention of  jugum  and  frenulum,  lack  of  alar 
tympanal  organ,  archaic  pseudomedia  (ex- 
cept in  Nothochn/sa  and  Dyspetochrysa) 
and  little  sclerotized  prosternum.  It  includes 
PaIeochnjsa,Arc]}aeochrys(i,Dyspetochrysa, 
Trihoch  ryso,  Dictyoch  rysa,  Triploch  rysa, 
HypocJirysa,  Kimochrysa,  Pamochrysa,  Pi- 
niachrysa,  and  Nothochryso.  The  living 
species  of  Nothochrysinae  are  cataloged, 
and  the  North  American  species  described 
and  illustrated. 

New  fossil  taxa  are:  Archaeochrysa,  new 
genus  for  Paleochrysa  creed i  Carpenter, 
fracta  (Cockerell)  and  parancrvis  n.  sp. 
(Florissant,  Colo.);  Dyspetochrysa,  n.  gen. 
for  Trihochrysa  vetuscuht  Scudder.  New 
Recent  species  are:  Pimaclirysa  albicostoles-, 
Baja  Calif.,  and  Arizona;  fiiseo,  intermedia 
and  nigra,  southern  Calif. 

INTRODUCTION 

This  study  comprises  the  first  section  of 
a  taxonomic  revision  of  the  North  American 
Chrysopidae.  The  concept  of  the  subfamih' 
Nothochrysinae  as  herein  employed  is  en- 

^  California  State  College,  Fullerton. 

Bull.  Mus.  Comp. 


tirely  new,  necessitating  a  preliminary  re- 
classification of  the  family.- 

There  appears  no  justification  for  retain- 
ing in  a  separate  family  the  Mesozoic  forms, 
which  are  placed  in  the  family  Meso- 
chrysopidae. These  already  have  achieved 
the  alignment  of  the  inner  gradate  veins 
which  precedes  the  pseudomedia  of  the 
later  forms.  In  addition,  they  show  the  fun- 
damental chrysopid  characteristics:  many 
straight,  unforked  branches  of  Rs,  which 
diverge  from  R  at  a  wide  angle;  two  reg- 
ular gradate  series;  MPl  and  MP2  few- 
branched,  intersecting  the  wing  margin 
near  the  base,  so  that  most  of  the  discal 
area  is  occupied  by  the  R-Rs-MA  system. 
These  basic  features  of  proportion,  while 
difficult  to  define,  readily  separate  the 
Chrysopidae  from  all  other  families. 

The  Apochrysinae,  a  distinctive  and  ho- 
mogeneous group,  has  recently  been  revised 
by  Kimmins  ( 1952b ) .  The  Chrysopinae 
constitutes  a  receptacle  for  the  remaining 
vast  and  \arious  assemblage  of  forms, 
doubtless  requiring  subdivision  when  bet- 
ter understood. 

All  described  genera  of  Mesochrysinae 
and  Nothochrysinae  are  included  in  the 
keys,  and  all  species  have  been  cataloged. 
Descriptions  and  illustrations  have  been 
included    for    all    living    North    American 

-  Although  this  paper  was  submitted  tor  publica- 
tion prior  to  the  appearance  of  Tjeder's  paper, 
1966,  it  has  been  possible  to  include  his  new  taxa. 
He  delimits  the  Nothochrysinae  (=Dictyochr>sinae) 
similarly. 


ZooL,  135(4):    215-238,  February,  196' 


215 


216        Bulk'iin  Miiscuni  of  Comparative  Zoology,  Vol.  135,  No.  4 


species,  and  for  such  fossil  or  Old  World 
species  as  are  of  particular  interest.  I  have 
not  examined  material  of  Dictyochrysa  or 
TripJoclin/sa,  which  were  reviewed  b\' 
Kimmins   (1952a). 

It  is  interesting  that  the  southwestern 
United  States  should  have  such  a  rich  rep- 
resentation (two  genera,  six  species)  of 
this  archaic  group,  which  has  changed  but 
little  since  the  Miocene  (Adams,  1957). 
This  concentration  of  these  relics  is  rivaled 
by  South  Africa,  with  two  genera  and  four 
species,  and  Australia  and  Tasmania,  also 
with  two  genera  and  four  species.  Such  a 
distribution  contributes  to  the  mounting 
evidence  that  the  southwestern  United 
States  has  served  as  an  evolutionary  re- 
fugium  for  the  Neuroptera. 

ACKNOWLEDGMENTS 

Grateful  acknowledgment  is  made  of  the 
help  and  encouragement  of  F.  M.  Carpenter 
and  P.  J.  Darlington,  Jr.,  of  the  Museum 
of  Comparative  Zoology  (MCZ),  Harvard. 
Ellis  MacLeod,  of  the  Biological  Labora- 
tories, Harvard,  has  made  many  valuable 
criticisms,  and  has  given  generously  of  his 
time  during  the  preparation  of  the  manu- 
script. D.  E.  Kimmins  has  kindly  examined 
type  material  and  has  made  useful  sugges- 
tions. Material  has  also  been  made  avail- 
able through  the  courtesy  of  C.  D.  MacNeil, 
California  Academy  of  Sciences  (CAS),  J. 
D.  Powell,  University  of  California,  Berke- 
ley, California  Insect  Survey  (CIS),  J.  N. 
Belkin,  University  of  California,  Los  Angeles 
(UCLA),  P.  R.  Timberlake,  University  of 
California,  Riverside  (UCR),  L.  Stange, 
University  of  California,  Davis  (UCD),  J. 
Lattin,  Oregon  State  College  (OSC),  E.  I. 
Sleeper,  Long  Beach  State  College,  J.  E.  H. 
Martin,  Entomology  Research  Branch,  Can- 
ada Department  of  Agriculture,  Ottawa 
(CNC),  Hugo  Rodeck,  University  of  Colo- 
rado Museum,  and  Floyd  Werner,  Univer- 
sity of  Arizona. 

Financial  assistance  has  been  provided 
through  grants-in-aid  from  Sigma  Xi-RESA, 
the   American   Academv   of   Arts    and    Sci- 


ences,   and    the    University    of    California, 
Santa  Barbara. 

DISCUSSION  OF  CHARACTERS 

The  wing-coupling  apparatus  consists  of 
a  large  jugal  lobe  on  the  fore  wing,  and 
a  frenulum,  bearing  several  long  setae,  on 
the  hind  wing;  this  is  essentially  the  same 
apparatus  as  in  Hemerobiidae,  etc.  The 
Chrysopinae  may  have  a  weak  frenulum, 
but  the  jugal  vein  is  thin,  and  there  is  no 
jugal  lobe.  Loss  of  the  wing-coupling  ap- 
paratus appears  associated  with  narrowing 
and  strengthening  of  the  wing  base,  and 
is  probably  of  great  adaptive  significance. 
A  similiar  modification  has  occurred  in 
the  evolution  of  the  Myrmeleontidae  and 
Ascalaphidae  from  an  osmyloid  ancestor 
( Adams,  1958 ) ,  and  in  the  Mantispidae. 

In  the  Nothochrysinae,  there  is  no  obvi- 
ous tympanal  organ  (Friedrich,  1953;  Er- 
hardt,  1916)  in  the  base  of  R  in  the  fore 
wing,  and  the  stem  of  M  is  easily  visible 
extending  in  a  straight  line  adjacent  to  R. 
In  the  Chrysopinae,  the  tympanal  organ 
fonns  a  conspicuous  bulge  in  R,  at  the 
point  where  Cu  diverges;  the  base  of  M 
usually  is  coalesced  with  R,  but  if  visible, 
makes  a  detour  posteriorly  around  the  tym- 
panal organ.  Probably  this  is  an  auditory 
organ  (Adams,  1962). 

In  No])i]inus  ( Apochrysinae),  the  tym- 
panal organ  involves  a  large  area  between 
R  and  M,  but  is  longer  and  does  not  form 
a  bulge  on  the  underside  of  R,  as  in 
Chrysopinae. 

Another  character,  probably  of  great 
adaptive  significance,  is  the  pseudomedia, 
which  differs  fundamentally  in  the  more 
primitive  chrysopids,  and  in  the  Chryso- 
pinae— Apochrysinae.  In  Pimachnjsa  (Figs. 
1,  2),  Ili/pochry.sa  (Fig.  5),  and  in  most 
of  the  fossil  genera,  Psm  is  composed  of 
crossveins  alternating  with  the  branches 
of  Rs  +  MA,  and  is  merely  a  basad  exten- 
sion of  the  inner  gradate  series.  The  course 
of  the  longitudinal  veins  and  composition 
of  the  jirimitive  pseudomedia  is  particularly 
clear   in    I h/pochrysd.    In   the   Chrysopinae 


Mesochrysinae  and  Nothochrysinae  •  Adaim        217 


(Fig.  45),  and  Apochrysinae  (Fig.  44),  and 
in  Nothoch)-yso  (Fig.  3),  Psm,  at  least 
basally,  is  composed  oi  overlapping  zig- 
zagged branches  of  Rs  and  M,  with  no 
crossveins  between  them.  ( Some  specimens 
of  N.  fulviceps  show  no  overlap. )  In  a  few 
genera  of  Chrysopinae,  such  as  Ytimachrysa 
and  Chrysopielki,  there  is  a  transition  from 
the  primitive  arrangement  apically,  with 
no  overlap,  to  the  more  advanced  arrange- 
ment basally,  with  overlapping  veins. 

Primitively,  the  longitudinal  veins  forked 
at  the  posterior  wing  margin,  except  for 
the  anals  in  both  wings,  and  CuP  in  the 
hind  wing.  This  condition  may  be  seen  in 
the  hind  wing  of  Archacochnjsa  (Figs.  40, 
41).  There  is  a  tendency  for  the  point  of 
furcation  to  move  proximally  until  it  reaches 
the  outer  gradate  crossveins;  when  this 
occurs,  the  pseudomedial  cells  appear  each 
to  give  rise  to  two  marginal  veinlets.  This 
process  begins  at  the  wing  base,  and  pro- 
ceeds apically;  the  sequence  is  best  seen 
in  the  fore  wing  of  Archaeochryso  para- 
nervis  (Fig.  40).  Frequently  the  longi- 
tudinal veins  fail  to  fork  at  all. 

The  pseudocubitus  has  evolved  like  the 
pseudomedia,  by  alignment  of  longitudinal 
veins  and  gradate  crossveins.  In  the  Notho- 
chrysinae it  usualh-  is  more  strongly  de- 
veloped than  Psm,  especially  in  the  fore 
wing.  In  the  Chrysopinae,  where  the  longi- 
tudinal veins  may  overlap  at  Psc,  it  is  im- 
possible to  detemiine,  in  most  cases,  which 
marginal  veinlets  have  arisen  from  which 
longitudinal  veins.  For  this  reason  the  over- 
lap at  Psc  shown  in  Figures  44  and  45  is 
conjectural,  although  consistent  with  the 
tracheation  of  Psc  in  Chrysopa  sipmta 
Walker,  as  demonstrated  by  Tillyard 
(1916). 

Evolution  of  the  basal  Banksian  cell  of 
the  hind  wing  has  been  described  by  Car- 
penter ( 1935 ) .  The  archaic  condition  is 
exemplified  by  Archacochrysa,  where  MPl 
is  connected  to  Rs  +  MA  by  the  sectorial 
crossvein,  which  intersects  MPl  in  a  Y- 
formation.  More  advanced  genera  exhibit 
slight  basad  migration  of  MF,  and  fusion 


of  MPl  with  Rs  +  MA.  This  evolutionary 
sequence  can  be  seen  by  comparing  Figures 
41,  5,  4,  and  2. 

In  Chrysopinae  and  Apochrysinae,  the 
wing  flexes  along  a  line  immediately  an- 
terior to  Psm  (dotted  line,  Figs.  44,  45). 
To  facilitate  this  flexion  in  the  fore  wing, 
the  first  sectorial  crossvein  (first  crossvein 
distal  to  the  base  of  Rs  +  MA )  and  the 
branches  of  Rs  +  MA  are  interrupted  or 
articulated  at  the  point  of  intersection  with 
the  pseudomedia.  In  the  Nothochrysinae, 
the  sectorial  crossvein  is  always  interrupted, 
but  the  branches  of  Rs  +  MA  never  are  so 
(except  in  Notliochiysa).  Flexion  along  the 
pseudomedia  is  accomplished,  in  Hypo- 
chrysa,  by  folds  traversing  the  crossveins 
of  Psm  (Fig.  5).  In  Pimachrysa,  there 
appears  to  be  no  distinct  line  of  flexion,  ex- 
cept perhaps  in  P.  nigra  (Fig.  4).  In  this 
species  the  veins  show  no  obvious  weaken- 
ing or  articulation,  but  most  specimens 
have  a  slight  wrinkle  in  the  membrane 
parallel  and  anterior  to  Psm.  A  similar  line 
of  flexion  occurs  between  MA  and  MP  in 
most  other  families  of  Neuroptera.  In 
Chrysopidae,  this  line  has  been  lost  early  in 
evolution;  the  pseudomedial  fold  is  its  func- 
tional, but  not  morphological,  equivalent. 
As  an  interesting  example  of  parallel  evo- 
lution, the  shape  of  the  wing,  and  course  of 
the  pseudomedial  fold  in  Chrysopinae  is 
strikingly  like  that  of  the  smaller  Nymphi- 
dae  ( e.g.  Osmylops,  Nesydrion ) .  In  the 
Nodita-Leucochrysa  complex  (Chrysopi- 
nae), Psm  curves  up  to  join  the  outer  gra- 
date series;  this  evolutionary  line  has  cul- 
minated in  the  Apochrysinae,  of  which  the 
nymphid,  MyiodactyJiis,  is  a  counterpart. 

Another  striking  feature  of  the  Notho- 
chrysinae is  the  near  uniformity  in  thickness 
of  most  of  the  veins  at  the  wing  base.  In 
Chrysopinae  and  Apochrysinae,  R  is  ex- 
panded basally,  and  Cu  is  always  inflated 
near  the  intersection  with  the  first  medial 
crossvein. 

The  Apochrysinae  appear  to  be  special- 
ized derivatives  of  the  Nodita-Leucochrysa 
complex  ( Chrysopinae ) .  The  closest  resem- 


218        Bulletin  Museum  of  Comparaiwe  Zoology,  Vol.  135,  No.  4 


blance  is  to  forms  such  as  Gonzaga,  from 
which  the  less  speciaHzed  Apochrysinae, 
such  as  Synthochn/sa  (Fig.  44),  differ 
only  in  the  loss  of  the  basal  subcostal  cross- 
vein,  less  distance  between  Psm  and  Psc, 
condition  of  MP2,  and  slight  changes  in  the 
proportions  of  the  wing. 

In  Gonzaga  (Fig.  45),  MP2  bends 
close  to  CuA,  then  up  toward  MP,  thence 
down  to  CuA  again,  which  it  joins  to  form 
part  of  Psc.  If  MP2  were  to  shift  posteriorly 
so  that  it  joined  CuA  directly,  eliminating 
the  zig-zag,  a  configuration  like  that  in  the 
Ajx)chrysinae  would  result.  It  may  be  noted 
that  the  second  apparent  medial  crossvein 
of  the  Apochrysinae  ( Fig.  44 )  lies  in  ex- 
actly the  same  relationship  to  the  first 
sectorial  (sxv)  and  cubital  crossveins  as 
does  the  medial  fork  in  Gonzaga.  This  is 
significant,  since,  if  MP2  had  instead  moved 
anteriorly  to  coalesce  with  MPl,  the  ap- 
parent second  medial  crossvein  would  be 
expected  to  intersect  CuA  at  the  middle  of 
the  second  cubital  cell. 

The  degree  of  overlap  of  the  branches  of 
Rs  -I-  MA  in  Apochrysinae  is  the  same  as 
in  Chr\'sopinae,  and  may  be  verified  by 
matching  branches  of  Rs  +  MA  with  the 
veinlets  between  Psm  and  Psc,  working 
basally  from  the  distal  end  of  Psm.  Such  a 
count  would  not,  of  course,  bear  on  the 
question  of  whether  MP2  had  disappeared 
by  moving  forward  and  coalescing  with 
MPl,  or  whether  it  may  have  moved  pos- 
teriorly and  fused  with  CuA.  But  it  pre- 
cludes the  possibility  that  the  first  median 
crossvein  is  MP2,  as  suggested  by  Kimmins 
( 1952b),  since  in  that  case  one  would  have 
to  account  for  an  extra  crossvein  between 
MP  and  CuA. 

Investigation  of  male  genitalic  structures 
in  a  wide  variety  of  neuropterous  types  has 
led  me  to  believe  that  in  the  Planipennia, 
the  ninth  gonocoxites  primitively  articulate 
on  the  arms  of  the  gonarcus.  In  the  Chry- 
sopidae,  the  "parameres"  of  most  authors 
appear  to  be  the  gonocoxites  and  are  re- 
ferred to  as  such  below. 


Key  to  the  Siibfainilies  of  Chrtjsopidae 

1.  Media  posterior  1  runs  in  an  even  curve  to 

margin,  not  annulate  at  intersections  with 
l)asal  inner  and  outer  gradate  crossveins. 
These  crossveins  intersect  MP  at  nearly 
a  right  angle,  so  that  MP  does  not  appear 
interrupted  by  either  Psm  or  Psc  ( Meso- 
zoic)    Mesochrysopinae. 

-  Media  posterior   1   usually  definitely  angu- 

late  at  intersections  with  liasal  outer  and 
inner  gradates,  so  that  it  appears  inter- 
rupted by  a  definite  pseudocubitus,  and 
(usually)    a  pseudoniedia   2. 

2.  Jugal    lobe    of    fore    wing    large;    frenulum 

present  on  hind  wing;  tympanal  organ  of 
fore  wing  absent  Nothochrysinae. 

-  fugal   lobe   of  fore   wing  and   frenulum   of 

liind  wing  reduced  or  absent;  tympanal 
organ  of  fore  wing  present  ( reduced  in 
Apochrysinae)  3. 

3.  In  fore  wing,  basal  subcostal  crossvein  and 

intramedian  cell  present;  space  between 
Psm  and  Psc  relatively  wide  (Fig.  45) 
Chrysopinae. 

-  In  fore  wing,  basal  subcostal  crossvein  and 

intramedian  cell  absent;  space  between 
Psm  and  Psc  relatively  narrow  ( Fig.  44 ) 
__    Apochrysinae. 

Subfamily  MESOCHRYSOPINAE  Handlirsch 

Mesochrysopidae  Handlirsch,  1908,  Die  fossilen 
Insekten:  612,  pi.  48,  fig.  14.  Type:  Meso- 
chnjsupa  Handlirsch  1908. 

Genus  MESOCHRYSOPA  Handlirsch 

Mc.wehnjsopa  Hamllirsch,  1908,  Die  fossilen  In- 
sekten: 612.  Type  (by  monotypy):  Hagen- 
ioteimes  zitteli  Meunier,   1898. 

Mesochrysopa  ziffeli  Meunier 

Ihigeniotc'iines  zitteli  Meunier,  1898,  Arch.  Mus. 
Teyl.  (2)  6:  34,  pi.  2,  fig.  2  (reference  from 
Handlirsch ) . 

Mesochrysopa  zitteli:  Handlirsch,  1908,  luc.  cit. 
Tillyard,  1916,  Proc.  Linn.  Soc.  N.S.  Wales 
41:  245-248,  text-fig.  8.  Martynova,  1949, 
Trudy  Palcontol.  Inst.,  Akad.  Nauk  SSSR 
20:   169.  Adams,  1957,  Psyche  63:   72. 

Discussion.  From  the  Jurassic  limestone 
of  Bavaria.  A  myrmeleontid  appearance  de- 
rives from  the  long  slender  wings,  with  Sc 
and  R  apparently  fused  apically,  and  with 
many  branches  of  Rs.  But  the  absence  of 
pectinate  branching  of  MP  and  CuA  pre- 
cludes    relationship     with     the     osmyloid- 


Mesochrysinae  and  Nothochrysinae  •  Adams 


219 


myrmeleontoid  families.  Although  the  inner 
and  outer  gradate  series  are  somewhat  ir- 
regular, the  basal  inner  gradates  are  strongly 
inelined,  forming  a  well-defined  pseudo- 
media,  which  does  not  include  MP.  The 
multiplication  and  irregularity  of  veins  is 
likely  a  function  of  the  insect's  large  size 
(fore  wing  36  mm  long);  the  arrangement 
is  essentially  like  that  of  the  following 
genus. 

Genus  MESYPOCHRYSA  Martynov 

Mesijpochnjsa  Martynov,  1927,  Izvestia  Akad.  Xauk 
SSSR  21:  764.  Type  species  ( liy  monot\p\  ) : 
Mesijpochnjsa  hitipentiis  Mart\no\  .  Mart>no\a, 
1949,  Trudy  Paleontol.  Inst.  Akad.  Xauk 
SSSR  20:    169. 

Mesypochrysa   laiipennis   Martynov 

Mesijpochnjsa  latipcnnis  Martyno\',  1927,  Izvestia 
Akad.  Nauk  SSSR  21:  765,  figs.  10-12  (hind 
wing  and  body  structure).  Martynova,  1949, 
Trudy  Paleontol.  Inst.  Akad.  Nauk  SSSR  20: 
169.  Type  from  the  Jurassic  of  Kara-tau  ( Tur- 
kestan), Paleontol.  Inst.  Acad.  Sci.  USSR,  not 
examined. 

Discussion.  Martynov's  figure  (redrawn 
as  Fig.  37)  shows  the  body  and  hind  \\'ing 
in  some  detail.  The  course  of  MP  is  prob- 
ably drawn  too  straight;  one  would  expect 
slight  zig-zagging  at  the  intersection  of  the 
basal  gradate  crosswing.  Probably  two  cross- 
veins  were  present  between  the  branches 
of  MP,  as  in  nearly  all  chrysopids. 

This  fossil  is  of  particular  interest  in 
that  the  hind  \\'ing  is  exactly  what  one 
would  postulate  in  the  ancestor  of  Archae- 
ochnjsa.  The  basal  Banksian  cell  is  similar, 
but  the  inner  crossvein  is  longer  and  more 
strongly  inclined  in  Mesijpochnjsa.  The 
basal  branches  of  Rs  +  MA  are  only  slightly 
zig-zagged,  not  strongly  bent  as  in  Archac- 
ochnjsa.  It  differs  from  more  advanced 
chrysopids  in  the  short,  transverse  pro- 
notum. 

Subfamily  NOTHOCHRYSINAE  Novas 

Nothochrysinos    i\a\as,    1910,    Broteria    9:    .38-59. 

Type:   NotJiochnjsa  McLachhm. 
Nothochrysini  Navas,  191.3,  Ann.  Soc.  Sci.  Bruxel- 

les  .37:   303.  Type:   NotJuxlinisa  McLachlan. 


Dictyochrysinae  Esben-Petersen,  1918,  Ark.  Zool. 
11  (26):  26.  Tjeder,  1966,  p.  246.  Tvpe: 
Dicttjochnjsa  Esben-Petersen.  NEW  SYNON- 
YNH    (subjective). 

Description.  Wing  veins  of  nearly  uni- 
form diameter  near  base,  tapering  e\enly 
apically;  MP  extends  adjacent  to  R,  but  is 
not  fused  with  it;  no  tympanal  organ  ap- 
parent on  \entral  side  of  R  in  fore  wing; 
jugal  lobe  of  fore  wing  conspicuously  pro- 
duced; Psm  (except  in  Nothoclinjsa  and 
DyspetocJinjsa)  composed  of  branches  of 
Rs  alternating  with  crossveins.  Prosternal 
area  largely  membranous.  Color  rarely 
green. 

Discussion.  Navas  (1910)  based  his 
tribal  division  of  the  Chrysopidae  on  rela- 
tively trivial  venational  characters.  His 
tribe  Nothochrysinos  included  all  the  forms 
which  had  the  intramedian  cell  (the  cell 
immediately  distal  to  the  fork  of  MP) 
either  (juadrangular,  or  absent.  This  re- 
sulted in  the  inclusion  of  NotJiochnjsa  ( most 
species  of  which  belong  in  Italochnjsa), 
LeucocJinjsa,  and  Nesochiysa.  With  the 
exception  of  three  species  of  Nothochnjsa, 
all  of  these  are  members  of  the  Chrysopi- 
nae.  Hijpochnjsa  was  placed  in  the  "Chry- 
sopinos." 

There  is  some  question  as  to  the  validity 
of  Navas'  1910  names.  The  taxa  were  clearly 
designated  as  tribes  and  names  properly 
constructed  but  for  the  colloquial  plural 
endings.  These  names  bore  correct  termina- 
tions in  the  1913  paper,  and  have  priority 
from  that  date  at  latest.  While  Navas'  usage 
of  the  name  makes  it  nomenclatorially  avail- 
able, the  concept  of  the  subfamily  here  pro- 
posed is  entirely  new. 

The  Tertiarv'  genera  differ  from  living 
forms  in  position  of  the  basal  subcostal 
cross\'ein  onh.  Furthermore,  Kimochnjsa 
africana  (Recent)  has  the  crossvein  as  in 
the  fossils.  Two  Tertiary  chrysopids  appear 
to  ha\e  been  described  from  Eurasia,  both 
belonging  to  Recent  genera  {Nothochnjsa, 
Chnjsopa).  Martynova  (1949)  reports  hav- 
ing collected  a  series  of  wing  imprints  from 
the  Miocene  of  Stavropol  (North  Cau- 
casus), but  these  ha\e  not  been  described. 


220        Bulletin  Mitseitni  of  Coniparative  Zoology,  Vol.  13.5,  No.  4 


Nutliochiy.sa  is  included  here  because  oi 
its  many  archaic  features,  not  found  in  any 
member  of  the  Chrysopinae.  However,  the 
development  of  the  pseudomedia  is  more 
advanced  than  in  an\'  other  nothochrysine 
(except  Dy.spetuclinj.m),  and  does  not  dif- 
fer from  that  of  the  Chrysopinae.  Scleroti- 
zation  of  the  prosternum  is  intermediate 
between  that  of  Pimachrysa  and  Ilypo- 
clirysa,  and  Chrysopinae.  These  are  both 
important  evolutionary  ad\ances  over  the 
rest  of  the  Nothochrysinae,  and  tribal  sep- 
aration might  be  advisable,  but  for  the 
small  number  of  genera  involved. 

Biolo'^y.  Biological  information  on  this 
subfamily  is  meagre.  Nothochrysa  (Killing- 
ton,  1937;  Toschl  1966)  and  Uypochrysa 
( Principi,  1958)  lay  stalked  eggs.  The  larvae 
of  Hypochrysa  (Brauer,  1867)  and  Notho- 
chrysa (Killington,  1937;  Toschi,  1966) 
have  been  described.  Both  N.  capitata  and 
N.  fiihiceps  may  occasionally  carry  trash, 
but  have  exceptionally  small  thoracic  tu- 
bercles. There  is  some  indication  of  pref- 
erence by  adults  for  ancient  host-plants: 
Nothochrysa  fulviceps  associates  with  oak, 
and  N.  capitata  with  pine  (Killington),  N. 
calif ornica  with  both  conifers  ( W.  Wade, 
pers.  comm.)  and  oak.  Pimachrysa  inter- 
media feeds  on  willow  pollen;  other  species 
of  Pimachrysa  have  been  taken  in  bait 
traps  (fusca,  albicostales)  or  on  flowers 
{nifi,ra,  on  Ccanothus)  and  may  be  pollen 
or  nectar  feeders. 

In  Nothochrysa  and  llyj)Ochrysa  nohilis, 
a  pale  mass  of  material  is  deposited  on  the 
dorsal  surface  of  the  female  abdomen.  This 
has  been  erroneously  interpreted  as  a 
spermatophore  (Killington,  Principi).  How- 
ever, Toschi,  1966,  has  observed  that  in 
female  specimens  of  N.  californica  bearing 
such  a  dorsal  mass,  the  spermatophore  is 
to  be  found  internally  in  the  bursa.  In 
Pimachrysa  fusca,  and  presumably  the  other 
species,  a  sac-like  spermatophore  ( Fig.  21, 
sp)  may  olten  be  found  protruding  from 
the  gonocoxites  ( "gonapophyses  laterales") 
which  marginally  bear  hooked  hairs,  prob- 
ably for  its  retention  (Fig.  20). 


Key  to  the  Genera  of  Nothochrysinae 

1.  Basal    subcostal    crossvein    of    fore    win^ 

arises  distally  to  origin  of  Rs  +  MA,  ap- 
proximately opposite  the  medial  fork 
(MF,  Fig.  42);  mostly  fossil  genera  -— 
„  2. 

-  Basal    subcostal    crossvein    of    fore    wing 

arises  basally  to  origin  of  Rs  +  Ma,  or 
subcostal  crossveins  numerous;  living 
genera  6. 

2.  Innermost  branch  of  Rs  +  MA  coalesced 

with  MP  1+2  Dyspetochrysa  gen.  n. 

-  Innermost  liranch  of  Rs  -|-  MA  connected 

to  MP  1  +  2  by  a  crossvein  3. 

3.  In    fore    wing    Rs  -|-  MA    arises    basally, 

nearer  the  first  medial  crossvein  than 
to  MF;  in  hind  wing  MP  and  Rs  joined 
by  a  crossvein  Archaeochrysa  gen.  n. 

-  In  fore  wing,  Rs  -|-  MA  arises  more  api- 

cally,  nearer  MF  than  to  first  medial 
crossvein;  in  hind  wing  MP  coalesces 
with  Rs  for  a  short  distance  4. 

4.  Inner  gradate  series  forms  a  smooth  curve 

continuous   with   Psm   _.     5. 

-  Apical     inner     gradate     series     closer     to 

Rs  +  MA  than  is  Psm,  thus  forming 
a  broken  curve,  with  basalmost  inner 
gradate  of  apical  series  arising  prox- 
imally  to  most  distal  inner  gradate  of 
basal    series    Trihoclmjsa. 

5.  Subcosta  and  costa  fused  apically;  living 

species     11. 

-  Subcosta   and   costa   not   coalesced;   fossil 

species  Paleochrysa. 

6.  More  than  two  series  of  gradates  in  both 

wings    7. 

-  Only  two  series  of  gradates  in  both  wings 

8. 

7.  Three   regular   series   of  gradates   

-   Triplochrysa. 

-  Six  or  seven  irregular  series  of  gradates; 

discal  area  divided  into  a  meshwork 
of  polygonal   cells   Dictyochrysa. 

8.  Proximal     branch     of     Rs  +  MA     usualh 

coalesced  for  a  short  distance  with  MP 
1;  in  living  species,  pseudomedial  fold 
follows  entire  length  of  Psm  in  both 
wings   Nothochrysa. 

-  Proximal   branch   of   Rs  +  MA   never   co- 

alesced with  MP  1,  so  that  Psm  consists 
of  alternating  longitudinal  veins  and 
crossveins;  pseudomedial  fold  intersects 
first  sectorial  crossvein  above  intra- 
medial  cell,  and  may  cross  some  basal 
gradates  of  the  Psm,  but  never  extends 
the  full  length  of  Psm  9. 

9.  Hooked  hairs  on  ninth  gonocoxites  ("gon- 

apophyses laterales")  of  female;  basal 
crossveins  of  Psm  not  interrupted  by 
pseudomedial  fold  Pimachrysa. 


Mesochrysinae  and  Nothochrysinae  •  Adains        221 


—  No   hooked   hairs   on   ninth   gonocoxite   ot 

female;  basal  crossveins  of  Psni  inter- 
rupted l)y  pseudoniedial  fold  10. 

10.  In  fore  winji,  2A  and  3A  fused  apically; 

male  epiproct  fused  with  ninth  termite, 
with  ventral  processes  Hijpochnjsa. 

—  In  fore  winji,  2A  and  3A  connected  by  a 

cross\ein;  male  ectoproct  demarked 
from  ninth  tergite,  with  no  ventral  pro- 
cesses        11. 

11.  Intramedian  cell  rhomboid,  almost  as  high 

as  long;  fore  pterostigma  shorter  tlian 
half  wing  width   Pamochnjsa. 

—  Intramedian   cell  about  twice  as   long  as 

high;  fore  pterostigma  as  long  as  two- 
thirds  wing  width  Kimochnjsa. 

Genus  DICTYOCHRYSA   Esben-Petersen 

Dictyochnjsa  Esben-Petersen,  1917,  Proc.  Linn. 
Soc.  New  South  Wales  42:  214-215  (type, 
by  original  designation,  D.  fulva  Esben-Peter- 
sen); 1918,  Arkiv  for  Zool.  11(26):  1-37. 
Kimmins,  1952,  Ann.  Mag.  Nat.  Hi.st.  (12)5: 
70-72   ( key  to  species ) . 

Dictyochrysa  fulva   Esben-Petersen 

D.  fulva  Esben-Petersen,  1917,  Proc.  Linn. 
Soc.  N.S.  Wales  42:  214-215,  pi.  13,  fig.  10. 
Holotype,  from  Queensland,  Australia,  Frog- 
gatt  Collection,  CSIRO,  Canberra  (not  seen). 
Kimmins,  1952,  Ann.  Mag.  Nat.  Hist.  (12)5: 
71-72. 

Dictyochrysa   peferseni  Kimmins 

D.  peterseni  Kinmiins,  1952,  Ann.  Mag.  Nat. 
Hist.  (12)5:  70-71,  fig.  2  (wings,  head,  pro- 
notum).  Holotype,  from  Mt.  Kosciusko,  New 
South  Wales,  6000  ft.,  12  December  1931, 
R.  J.  Tillyard,  coll.  In  Brit.  Mus.  (Nat.  Hist.), 
not  seen. 
Additional  distribution.  Mt.  Maria,  Tasmania? 

Dicfyochrysa   latifasciata    Kimmins 

D.  latifasciata  Kimmins,  1952,  Ann.  Mag.  Nat. 
Hist.  (12)5:  71-72,  fig.  3  (male  genitalia). 
Holotype  male,  from  Mt.  Wellington,  S.  Tas- 
mania, 1300-2300  ft.,  12-21  March  1913. 
Abdomen  in  balsam.  Brit.  Mus.  (Nat.  Hist.), 
not  seen. 


Triplochrysa   pallida    Kimmins 

T.  pallida  Kimmins,  1952,  loc.  cit.,  fig.  1 
(wings).  Holotype  female,  from  Bunya  Mt., 
Queensland,  Australia,  27  January  1951,  in 
Brit.   Mus.    (Nat.    Hist.),   not  seen. 

Genus  HYPOCHRYSA  Hagen 

Hijpochnjsa  Hagen,  1866,  Stettiner  Entomol.  Zeit. 
27:  377.  Type  species  (by  monotypy): 
Chrysopa   nohilis  Schneider. 

Hypochrysa    nobilis   (Schneider) 

Figures  5,  32,  33 

Chrysopa  elc^ans  Burmeister,  1839,  Handb.  d. 
Entomol.:  981.  Type  locality  "Harze,"  Saxe- 
sen  collector,  probably  in  the  Zoological  Mu- 
seum, Univ.  of  Halle  (not  seen).  This  name 
has  priority,  but  is  a  nomen  oblitiim  under 
Rule  231).  Not  to  be  confused  with  Hcm- 
erohius  elegans  Ciuerin  1838,  which  was  trans- 
ferred to  Chrysopa  by  Schneider  in  1851,  and 
probably  is  a  Gonzafia;  this  name  is  pre- 
occupied by  Henicrobiiis  elegans  Stephens 
1836. 

Chrysopa  nohilis  Schneider,  1851,  Symb.  ad 
Monogr.  Gen.  Chry.sopae:  142-144,  pi.  51. 
(  Name  replaces  elegans  Burmeistei . ) 

Hypochrysa  nobilis:  Hagen,  1866,  Stettiner 
Entomol.  Zeit.  27:  377.  Brauer,  1867,  Verb. 
Zool.-Bot.  Ces.  Wien  17:  27-29,  pi.  9,  fig.  1 
(larva).  Navas,  1913,  Insecta  28:  129-1.30, 
fig.  1;  1915,  Arx.  In.st.  Cien.,  Barcelona  3(2): 
88.  Principi,  1956,  Atti  Acad.  Sci.  Inst.  Bolo- 
gna, Rend.  (XI)3:  1-3,  pi.  1  (Ethology, 
photographs  of  adult,  egg,  habitat);  1961, 
Mem.  Mus.  Civ.  Stor.  Nat.  Verona  9:  109. 

Description.  Male  ninth  tergite  fused 
with  ectoprocts,  not  articulated  with  ninth 
sternite.  Eighth  and  ninth  sternites  not 
fused,  but  moveable.  Ectoprocts  (Fig.  32, 
ect. )  each  posteroventrally  bearing  heavily 
sclerotized  toothed  process.  Gonarcus  flat- 
tened, heavily  sclerotized,  black;  mediuncus 
bilobed;  gonocoxites  ( "parameres" )  weakly 
sclerotized,  dorsolateral  to  mediuncus  lobes. 

Di.itri])ution.  Widespread  in  southern 
Europe. 

Remark.  Hypochrysa  arg.entina  Navas, 
1911,  is  a  Chrysopa  (Tjeder,  1966). 


Genus  TRIPLOCHRYSA  Kimmins 

Triplochrysa  Kimmins,  1952,  Ann.  Mag.  Nat.  Hist. 
(12)5:  69-70  (type,  by  original  designation, 
7'.  pallida  Kimmins). 


Genus  KIMOCHRYSA  Tjeder 

Kiniochrysa  Tjeder,  1966,  S.  African  Anim.  Life 
12:  254.  Type  .species  (by  original  designa- 
tion):  Kimochrysa  inipar  Tjeder. 


0  •')•') 


Bulletin  Miiscmii  of  Comparative  Zoology,  Vol.  135,  No.  4 


Kimochrysa   impar  Tjeder 

Kii)n>chnj.sii  impar  Tjeder,  ibid.:  256—259,  figs. 
820-834.  Holotype  male,  Kleinmond,  Cape 
Province,  S.  Africa,  in  the  South  African 
Museum,  Cape  Town  (not  seen). 

Di.scussio)}.  This  is  the  only  species  oi 
the  Nothochrysinae  with  numerous  sub- 
costal \  ('inlets  in  the  fore  wing.  In  the  male, 
the  ninth  abdominal  tergite  is  not  hinged  to 
the  eighth  and  ninth  sternites,  and  these 
sternites  are  completely  fused. 

Kimochrysa   africana    (Kimmins) 

Hypochrtjsa   africdiui    Kimmins,    1937,   Ann.    Mag. 

Nat."  Hist.    (10)    19:    307-308,  fig.    Holotype 

female,    from    Worcester,    Cape    Province,    S. 

Africa,  in  Brit.  Mus.   (Nat.  Hist.)    (not  seen). 
Kimocliriisci    africana:     Tjeder,     1966,    S.    African 

Anim.  Life   12:  259-261,  figs.  835-838. 

Kimochrysa    raphidioides   Tjeder 

Figures  10,  29-31 

Kimochrysa  rapJiidioide.s  Tjeder,  1966,  ibid.:  261- 
262,  figs.  839-844.  Holotype  female,  from 
"Cap  Drege"  ( abl^rexiation  for  "Cape  of 
Cood  Hope,  Drege,  collector"),  in  the  MCZ 
( examined ) . 

Description.  Head  marked  with  dark 
as  in  Figure  10.  Pronotum  transverse,  a 
thin  transverse  black  line  at  outer  edge  of 
furrow,  a  short  longitudinal  line  at  posterior 
corner.  Subcostal  crossvein  distal  to  origin 
of  Rs  +  MA.  Male  ninth  tergite  hinged 
to  ninth  sternite  (Fig.  31);  fusion  of  eighth 
and  ninth  sternites  demarked  by  internal 
sclerotized  ridge.  Gonocoxites  ( Figs.  29,  30, 
gcx )  paddle-shaped. 

Discussion.  In  addition  to  the  type,  a 
male  with  no  data  is  also  in  the  MCZ.  This 
species  is  doubtfully  distinct  from  africana, 
from  which  it  differs  only  by  the  position 
of  the  subcostal  crossvein  (proximal  to  the 
origin  of  Rs  +  MA  in  africana),  and  by 
greater  sclerotization  of  the  spermatheca 
(which  may  be  due  to  different  degrees  of 
maturity  in  the  specimens).  Tjeder  dif- 
ferentiates these  two  species  principally  on 
the  position  of  the  second  medial  crossvein. 
Hilt  in  africana,  it  varies  in  position  from 
distal   to   jMoximal   to   the   fork    (Kimmins, 


pers.  comm.).  In  the  male  of  raphidioides, 
the  crossvein  intersects  MP  at  the  fork.  In 
the  specimen  of  K.  impar  figured  by  Tjeder, 
the  second  medial  crossvein  is  proximal  to 
the  fork  on  one  wing,  and  distal  on  the 
other. 

The  coloration  of  the  two  species  is  sim- 
ilar, although  the  frontal  markings  below 
the  antennae  are  absent  on  the  type  of 
africana,  and  faint  on  the  second  specimen; 
the  interantennal  marks  are  faint  on  both 
specimens  of  africana  ( Kimmins,  pers. 
comm. ) . 

Genus  PAMOCHRYSA  Tjeder 

Pamochrijsa  Tjeder,  1966,  S.  African  Anim.  Life 
12:  248.  Type  species  (by  original  designa- 
tion):  Pamochry.sa  stcUata  Tjeder. 

Pamochrysa   sfellata   Tjeder 

Pamochry.sa  stcllata  Tjeder,  ibid.:  250-253,  figs. 
797-815.  Holotype  male,  from  Cathedral 
Peak,  Forestry  Reserve,  Indumeni  River, 
Drakensberg,  Natal,  S.  Africa,  March  1959, 
in  the  Natal  Museum,  Pietermaritzburg,  not 
seen. 

Genus  NOTHOCHRYSA  McLachlan 

Nothochrysa  McLachlan,  1868,  Monograph  of  the 
British  Neuroptera-Planipennia.  Trans.  En- 
tomol.  Soc.  London  (1868):  195.  Type 
species,  Chrysopa  fidviceps  Stephens,  desig- 
nated by  Banks,  1903,  Trans.  Amer.  Entomol. 
Soc.  29:'  142.  Tjeder,  1941,  Entomol.  Tidskrift 
(1941):  30-31.  Principi,  1946,  Bol.  Inst. 
Entomol.   Univ.   Bologna   15:    86. 

Nathanica  Navas,  1913,  Trans.  2nd  Intern.  Congr. 
Entomol.,  Oxford,  1912,  2:  181  (type  [by 
subsequent  designation  of  Tjeder,  1941,  ibid.]: 
Hemcrobius  capitatii.s  Fabricius). 

As  McLachlan  failed  to  designate  a  type 
for  this  genus.  Banks'  listing  of  N.  fulviceps 
as  type  constitutes  a  valid  type  designation. 
Principi  (1946)  restricted  Nothochry.'ia  to 
include  fulviceps  and  capitata  only,  erect- 
ing a  new  genus,  IfalocJirysa,  for  N.  italica 
Rossi.  It  appears  that  the  great  majority  of 
Old  World  species  formerly  included  in 
Notfjochry.sa  belong  instead  to  Italochnjsa. 

Description.  Head  wide,  vertex  low, 
eyes   small,   labrum   emarginate.   Antennae 


Mesochrysinae  and  Nothochhysixae  •  Adams        223 


about  as  long  as  fore  wing;  scape  about  as 
long  as  wide,  flagellar  segments  longer  than 
wide.  Prosternum  small  (Fig.  35).  Tarsal 
claw  either  with  basal  enlargement  {fulvi- 
ceps)  or  simple  {copitata  and  californica). 
Fore  wing  (Figs.  3,  43):  basal  subcostal 
crossvein  opposite  midpoint  between  first 
medial  crossvein  and  medial  fork.  Rs  origi- 
nates basad  of  medial  fork;  two  rows  of 
gradates.  Psm  formed  by  overlapping  lon- 
gitudinal veins,  may  extend  beyond  basal 
inner  gradates;  pseudomedial  fold  present; 
MP2  parallels  MPl,  connected  by  a  cross- 
vein;  second  medial  crossvein  intersects 
MP2  midway  between  Mp  and  CuA.  Psc 
parallels  Psm;  runs  into  outer  gradate  series. 
lA  forked  apically;  2A  and  3A  simple.  Jugal 
lobe  prominent.  Hind  wing:  frenulum  well 
developed,  MP  coalescent  with  Rs  for  a 
moderate  distance;  two  rows  of  gradates. 
Psm  \ve\\  developed,  but  longitudinal  veins 
do  not  overlap  ( except  basally  in  fiilviceps ) ; 
instead  each  intersects  the  next  where  it 
bends  shaiply  apicad,  so  that  the  veins  be- 
tween Psm  and  Psc  are  exactly  opposite  the 
basal  branches  of  Rs  ( a  very  conspicuous 
and  characteristic  feature ) .  Pseudomedial 
fold  present.  Psc  runs  into  outer  gradate 
series. 

Male:  (Fig.  17)  ninth  abdominal  tergite 
heavily  sclerotized  anteroventrally,  with  ir- 
regular margin;  fused  with  ectoproct.  Eighth 
sternite  distinct  or  only  weakly  fused  with 
ninth  sternite.  Cuticular  gland  openings 
present.  Gonarcus  (Figs.  18,  19)  bears  wide 
triangular  mediuncus;  gonocoxites  small. 
Female:  No  hooked  hairs  on  ninth  gono- 
coxites; eighth  sternum  bilobed,  lobes  fit- 
ting into  cuplike  ninth  vaKulae  ("divertic- 
ulos  linguiformes"  of  Principi )  so  as  to  close 
the  genital  cavity  anteriorly,  as  in  Chryso- 
pinae. 

Key  to  the  Recent  Species  of  Nothochrysa 

1.  Proiiotum  membranous  medially;  Psm  in 
lore  wing  extends  beyond  inner  gradate 
series;   head   orange   and   lilack;    i\orth 

American  californica. 

Pronotum    sclerotized    medially;    Psm    in 
fore  wing  runs  into  inner  gradate  series; 


head  concolorous,  orange  or  orange- 
brown;  European  -_ 2. 

2.  Thorax  with  median  pale  band,  tarsal 
claws  with  basal  expansion;  several  mar- 
ginal veinlets  from  Psc  forked;  large 
species   ( expanse  37-48  mm )   ....  fulviceps. 

—  Thorax  orange-brown,  no  median  band; 
tarsal  claws  simple;  marginal  veinlets 
from  Psc  rarely  forked;  smaller  (expanse 
27-36  mm)   capitata. 

Nofhochrysa   fulviceps    (Stephens) 

Cbrysopa  fulviceps  Stephens,  1836,  lUustr.  Brit. 
Entomol.  Mand.,  6:  101.  Types:  4  syntypes 
in  the  Brit.  Mus.   (Nat.  Hist.),  not  seen. 

Notlwchiysa   ftilvice})s:    McLachlan,    1868,   Trans. 

Entomol.  .Soc.  London   (1868):   207. 
Nathanica    fulviceps:     Navas,     1913,    Trans.    2nd 

Intern.    Congr.    Entomol.,    Oxford,     1912,    2: 

181.    Killington,  1937,  Monogr.  Brit.  Xeuropt. 

2:   236-242,  fig.   114   (male  genit.),  fig.   115 

(female  genit.),  pi.  27,  fig.  2  (wing),  pi.  30, 

fig.  3   (larva). 
Nathanica  fulviceps  var.  flavida  Navas,  1919,  Bol. 

Soc.   Entomol.   Espana  2:    55. 
Nothochn/sa    fulviceps:     Tjeder,     1941,    Entomol. 

Tidsicrift    (1941):    30-31. 

Di,stril)tdion.     Europe. 

Nofhochrysa   capitafa    (Fabricius) 

Hemerobius   capitatus    F'abricius,    1793,    Entomol. 

Syst.  2:   82.    Type  locality  "Germania.   Dom. 

de   Paykull."   Zool.   Mus.,   Univ.   of   Kiel,   not 

seen. 
Chrysopa    capitata:    Curtis,    1834,    Brit.    Entomol. 

"  pi.  520. 
Nothochrysa    capitata:    McLachlan,    1868,    Trans. 

Entomol.   Soc.   London    (1868):   207. 
Nathanica     capitata:     Navas,     1913,     Trans.     2nd 

Intern.    Congr.    Entomol.,    Oxford,    1912,    2: 

181.    Killington,  1937,  Monogr.  Brit.  Neuropt. 

2:  242-246,  fig.  115c  (female  genit.),  pi.  27, 

fig.  3    (wings);  pi.  30,  fig.  4    (larva). 
Nothochrysa     capitata:     Tjeder,     1941,     Entomol. 

Tidskrift   (1941):  .30. 

Distrihtition.     Europe. 

Nofhochrysa   californica    Banks 

Figures  3,  8,  17,  18,  19,  22,  .35 

Nothochrysa  californica  Banks,  1892,  Trans.  Amer. 
Entomol.  Soc.  19:  373.  Male  holotype  from 
Los  Angeles,  Calif.,  MCZ  No.  11406,  examined. 
1903,  Trans.  Amer.  Entomol.  Soc.  29:  142,  pi. 
2,  fig.  3  (wing  base);  1904,  Cat.  Neuropt.  In- 


224        BtiUcfiii  Museum  of  Comparative  Zoology,  Vol.  135,  No.  4 


sects  I'.S.,  Philadelphia:  26.  MacGillivra\ , 
1894,  Canad.  Entoniol.  (1894):  171.  Smith, 
1932,  Ann.  Entoniol.  Soc.  Anier.  35:  582,  pi. 
1,  fig.  5  (body  and  wings,  color).  Bickley 
and  MacLeod,  1956,  Proc.  Entoniol.  Soc. 
Washington    58:    182-183. 

Description.  Head  (Fig.  8):  antennal 
socket.s  large,  margin.s  blaek.  A  black  line 
extending  down  .suture  from  antennal 
.socket  to  anterior  tentorial  pit;  black  line 
connecting  pit.s  bent  posteriorly  in  middle; 
dark  mark  on  clypeu.s  connected  to  this 
line  laterally;  black  genal  .spot,  near  man- 
dibular articulation.  Vertex  elevation  slight; 
median  and  lateral  stripes  connected  to 
circumantennal  marks.  Antennae  black; 
scape  wider  than  long.  Palpi  dark,  pale  at 
joints. 

Pronotum  divided  by  median  longitudinal 
pale  membranous  area;  lateral  sclerites  dark 
umbraceous,  lateral  margins  pale.  Alinotum 
dark,  mesoprescutum  with  posterior  pale 
band  interrupted  by  black  median  longitu- 
dinal suture;  mesoscutum  pale  anteriorly. 
Propleuron  and  sternum  dark.  Meso-  and 
metapleura  dark,  pale  posteriorly.  Coxae 
dark,  trochanters  dark  basally;  femora  with 
dorsal  and  ventral  longitudinal  dark  stripes, 
confluent  on  hind  femur.  Tibiae  pale  with 
dark  basal  and  apical  marks;  tarsi  fuscous, 
dark  apically.   Claws   lack   basal   tooth. 

Fore  wing  ( Fig.  3 ) :  veins  black,  except 
C,  R,  3A  and  posterior  marginal  vein  pale 
basally;  sensory  area  posterior  to  apex  of 
3A  and  posterior  marginal  vein  pale  basally; 
sensory  area  posterior  to  apex  of  3A,  and 
jugal  lobe,  fuscous;  stigma  fuscous,  spaces 
between  apical  subcostal  crossveins  ivory. 
MPl  desclerotized  at  intersection  with  basal 
sectorial  crossvein;  Psm  continues  beyond 
basal  inner  gradates.  Hind  wing:  venation 
dark,  R  white  to  stigma;  Rs  and  MPl  white 
basally;  CuP  white;  stigma  as  in  fore  wing. 

Abdomen:  tergites  black,  narrowly  pale 
apically;  sternites  may  be  pale  both  basally 
and  apically.  Female  ninth  tergite  with 
pale  spot  posterolaterally;  tenth  tergite  pale, 
callus  cerci  and  posterior  margin  black. 
Eighth  sternum  (Fig.  22)  bilobed.  Dorsal 
postcopulatory  mass  white  or  creamy  yel- 


low. Male  ninth  sternum  pale  vcntrally. 
Callus  cerci  fused  to  ninth  tergum;  ninth 
sternum  partially  coalescent  with  eighth 
(Fig.  17).  Cuticular  gland  openings  lack- 
ing on  heavily  sclerotized  anteroventral  area 
of  ninth  tergum,  and  tenth  tergum.  Medi- 
uncus  (Figs.  18,  19)  broad,  hoodlikc;  gono- 
coxites  small,  platelike. 

Measurements  ( mm ) :  Fore  wing  length 
12.5-15.  (13.5);  width  4.5-5.5  (5.1);  an- 
tenna 10-12  (11.2). 

Distribution:  CALIFORNIA:  Alameda 
Co.:  E.  C.  Van  Dyke  (USNM);  Berkeley, 
28-IV-12,  9,  J.  C.  Bridewell  (USNM);  Oak- 
land, 17-IV-15,  E.  P.  Van  Duzee  (MCZ); 
Hills  back  of  Oakland,  30-IV-1911,  6,  9- 
V-09,  6,  E.  C.  Van  Dyke  (CAS).  Contra 
Costa  Co.:  Near  Orinda,  21-IV-50,  W.  Wade 
(CNHM).  Marin  Co.:  Mill  Valley,  28-11- 
26,  9,  6,7-111-26,  2  9,  E.  P.  Van  Duzee 
(CAS),  25-III-52,  2  6,  19,  H.  B.  Leach 
(CAS),  8-V-56,  E.  S.  Ross  (CAS);  Muir 
Woods,  23-IV-1911,  9,  E.  C.  Van  Dyke 
(CAS);  San  Geronimo,  20-IV,  O.  Sacken 
(  MCZ ) .  Mendocino  Co. :  Caspar  Lumber 
Camp,  12  mi.  E.  Fort  Bragg,  20-VI-38,  9 , 
Van  Dyke  ( CAS ) ;  Van  Damme  State  Park, 
18-V-47,  i,  H.  Welsh  (CAS);  Yorkville, 
l-V-24,  9,  E.  P.  Van  Duzee  (CAS).  Santa 
Clara  Co.:  Stevens  Cr.  16-III-41,  9,  E.  S. 
Ro.ss  (CAS);  Palo  Alto,  27-V-92  (MCZ); 
Hills  back  of  Palo  Alto,  29-IV-1928,  9, 
(CAS);  Stanford  Univ.  5-IV-04,  9,  (MCZ). 
San  Mateo  Co.:  King's  Mt.,  5-VII-46,  c^, 
E.  S.  Ro.ss  (CAS). 

OREGON:  Benton  Co.:  Corvallis,  19- V- 
45,  9,  Marge  Johnson  (OSC);  McDonald 
For.,  5  mi.  N.  Corvallis,  26-V-56,  9,  J.  R. 
Mori  (OSC);  Sulfur  Spgs.,  6  mi.  N.  Cor- 
vallis, ll-V-56,  9,  J.  F.  O'Brien  (OSC); 
Rock  Cr.,  12  mi.  W.  Corvallis,  4-V-56,  9, 
N.  E.  Johnson  (OSC);  1  mi.  N.  Dawson, 
2-V-62,  4  9 ,  1  c5 ,  W.  Barnett,  ( OSC ) ;  2  mi. 
S.E.  Summit,  12-IV-57,  c? ,  J.  Lattin  (OSC). 

WASHINGTON:  Kittitas  Co.:  Easton,  A. 
Koebele  (USNM). 

BRITISH  COLUMBIA:  Bowser,  5-IV-55, 
20-VI-55,  1  <^,  1  9,  J.  W.  Brown  (CNC); 
So.   Pender   Is.,   30-\'-50,   on   Douglas   Fir, 


Mesochrysinae  and  NoTHOCHRYSiNAE  •  Aclcims        225 


1  5,  (CNC);  Vancouver,  31-V-31,  on  snow, 
4700  ft.,  Seymour  Mt.,  H.  B.  Leech  (MCZ); 
Victoria,  4-V-lS,  W.  B.  Anderson  (Smith, 
1932). 

Diagnosis.  This  species  differs  markedly 
from  N.  fulviccps  and  capitata  in  head 
coloration,  in  Psm  extending  beyond  the 
basal  inner  gradates,  and  in  the  peculiar 
separation  of  the  lateral  pronotal  sclerites 
by  a  median  membranous  area.  It  is  re- 
tained in  Notliochnjsa  because  it  shares 
with  the  other  species  the  well  developed 
Psm  and  Psc. 

Discussion.  The  paucity  of  material  from 
Oregon,  Washington,  and  western  Canada 
probably  indicates  less  collecting  activity, 
rather  than  scarcity.  In  the  San  Francisco 
Bay  area  it  sometimes  is  locally  fairly  nu- 
merous. W.  Wade  (pers.  comm.)  collected 
eggs  on  conifers  in  the  Orinda,  California, 
area.  D.  Breedlove  (pers.  comm.)  has  noted 
adults  abundantly  on  oak  near  Oakland, 
California.  Throughout  its  range,  it  appears 
to  be  restricted  to  moist  forest  areas,  not 
far  from  the  coast.  The  type  is  probably 
mislabelled;  in  spite  of  intensive  collecting 
around  Los  Angeles  since  1892,  no  addi- 
tional specimens  have  been  taken. 

Nothochrysa   praeclara  Statz 

Figure  43 

Nothochrysa  praeclara  Statz,  1936,  Bechiana  93: 
215-216,  fig.  3  (photograph  of  fore  wing), 
fig.  4  (drawing  of  fore  wing).  Holotype, 
from  Rott  am  Siebengebirge,  West  Germany, 
mid-Tertiary,  in  the  Los  Angeles  County 
Museum   ( examined ) . 

This  species,  known  from  a  single  fore 
wing,  differs  from  capitata  only  in  having 
fewer  overlapping  veins  in  Psm.  In  all  the 
specimens  of  capitata  examined,  at  least 
four  branches  of  Rs  +  MA  coalesce  with  the 
next  proximal  vein  at  the  Psm,  whereas 
only  one  basal  branch  is  so  coalesced  in 
praeclara.  This  may  not  be  significant;  in 
fulviccps  the  venation  is  very  irregular,  and 
some  specimens  show  no  overlap  at  all. 
Praeclara  differs  from  fulviccps  in  smaller 
size  (fore  wing  length  16.33  mm,  16 
branches  of  Rs  -f  MA ) . 


Genus  PIMACHRYSA  Adams 

Pimachrysa  Adams,  1957,  Psyche  63:  67-70.  Type 
(  by  original  designation ) :    P.  Virata  Adams. 

Description.  Vertex  moderately  elevated; 
anterior  tentorial  pits  large;  labrum  barely 
emarginate.  Antennal  length  variable:  scape 
short,  little  swollen;  apical  flagellar  seg- 
ments more  than  twice  as  long  as  wide. 
Tarsal  claws  simple,  without  basal  enlarge- 
ment. Prosternum  unsclerotized  anteriorly 
to  sternal  pits  (Fig.  34).  Male  with  ninth 
tergite  clearly  separated  from  ectoproct; 
ninth  sternite  clearly  demarked  from  eighth, 
posterior  margin  acute  (Figs.  11,  14). 
Mediuncus  slender,  apex  angulate  (Figs. 
12,  13,  15,  16).  Cuticular  gland  openings 
present.  Female  with  enlarged  ninth  gono- 
coxites  bearing  marginally  a  band  of  spoon- 
shaped  setae  (Figs.  20,  21).  Eighth  sternite 
(subgenital  plate)  and  ninth  valvulae  lie 
on  membrane  of  genital  opening,  so  that 
it  is  permanently  closed  anteriorly.  Sper- 
matophore  ovoid,  protruding  from  gonocox- 
ites. 

Fore  wing:  costal  area  narrow;  basal  sub- 
costal crossvein  slightly  basal  to  origin  of 
Rs;  Rs  arising  basally  to  cubital  fork.  MP2 
not  fusing  with  MPl,  but  turning  to  wing 
margin  at  apex  of  intramedian  cell.  Second 
mediocubital  crossvein  near  MF,  usually 
distal  to  it.  Two  gradate  series,  extending 
smoothly  into  Psm  and  Psc  (except  in  P. 
nigra);  no  overlap  of  adjacent  longitudinal 
veins  on  Psm  or  Psc.  Pseudomedial  fold 
undeveloped  except  for  weak  spot  at  inter- 
section of  first  sectorial  crossvein  and  MPl. 
Anals  neither  branched  nor  anastomosed. 
Jugal  lobe  prominent,  with  jugal  vein. 

Hind  wing:  frenulum  present.  Costal  area 
narrow.  Rs  and  MP  merely  touching,  or 
fused  for  a  short  distance;  basal  Banksian 
cell  large.  Psm  and  Psc  similar  to  those  of 
fore  wing.  Anals  neither  forked  nor  anas- 
tomosed. 

Discussion.  This  genus  is  most  similar 
to  H\ipochnjsa,  but  differs  in  having  en- 
larged ninth  gonocoxites,  bearing  modified 
hairs,  probably  serving  to  hold  the  sper- 
matophore    in    place.    In    Hijpochnjsa,    a 


226         BiiUctiii  Miisciini  of  Compdiative  Zoology,  Vol.  135,  \o.  4 


postcopulatory  mass  is  placed  on  the  dorsal 
surface  of  the  female  abdomen,  the  second 
mediocubital  crossvein  is  more  basal,  and 
some  anal  veins  are  anastomosed;  tl\e  eyes 
are  smaller  in  relation  to  the  remainder 
of  the  head,  than  in  PimacJirysa. 

Key  to  the  Species  of  Pimachrysa 

1.  A  slight  break  lietween  inner  gradates  and 

Psni;  only  1  or  2  Ijranches  of  Rs  forked 
marginally;  color  yellow-orange  and 
black  P.   ninra  sp.   n. 

-  Inner  gradates  merge  smoothly  with  Fsm: 

nearly  all  branches  of  Rs  forked  mar- 
ginally    2. 

2.  Head  unmarked  except  for  interantennal 

mark;  body  mostly  pale  yellow;  stigma 
white  P.  ^rata  Adams 

-  Head  with  dark  marks  on  face  and  xertex; 

subcostal  veinlets  of  stigma  conspicu- 
ously dark-bordered;  body  gray-prui- 
nose  -        3. 

3.  Discal   veins    of  hind   wing   mostly   black 

P.   inicrmcdia  sp.  n. 

-  Veins  of  hind  wing  (except  Sc)  all  pale  4. 

4.  Costal  \einlets  of  fore  wing  pale;  in  hind 

wing  MPl  touches  Rs  +  MA  at  one 
point  only  P.  alhicostales  sp.  n. 

-  Costal  veinlets  of  fore  wing  black;  in  liind 

wing  MPl  coalesced  with  Rs  +  MA  as 
far  as  length  of  first  radial  crossvein 
P.  fusca  sp.  n. 

P'tmachrysa   grata   Adams 

Figures  9,  23 

Pinuiclny.sd  grata  Adams,  1957,  Psyche  63:  67-70 
(1956),  figs.  1-5  (head  and  pronotum,  geni- 
talia, wings ) .  Holotype  female,  from  Madera 
Canyon,  Santa  Rita  Mts.,  Arizona,  26  August 
1949,  at  light.  P.  Adams,  col.  MCZ  No.  29624, 
examined. 

Description.  Head  (Fig.  9):  straw  yel- 
low, marked  with  black;  flagelhim  black. 
Pronotum  ivory,  with  three  fuscous  stripes. 
Meso-  and  metanota  yellow,  scuta  infus- 
cate  posterolaterally.  Meso-  and  meta- 
pleurae  bright  yellow,  with  fuscous  marks. 
Legs  ivory,  femora  with  wide  apical  fus- 
cous band. 

Wings  (Adams,  1957,  Fig.  1)  more  acute 
th;in  in  fiisca;  inner  gradate  series  sinuous, 
so  that  in  middle  of  series,  the  distance  be- 
tween inner  and  outer  gradates  equals  half 
or  less  the  distance  from  inner  2;radates  to 


Rs.  Second  medial  crossvein  of  fore  wing 
basal  to  MF.  Most  marginal  veinlets  forked. 

Fore  wing:  costa,  subcosta  and  R  white, 
stigma  white;  other  veins  fuscous  except  at 
wing  base. 

Abdomen  (  9  )  short,  slender,  terminal 
segments  greatly  enlarged.  Gonocoxites 
larger  than  in  other  species,  with  more 
hooked  setae.  Subgenital  plate  (Fig.  23) 
ligulate,  with  transverse  ridge  on  ventral 
(anterior)  surface.  (The  illustration,  from 
Adams,  1957,  shows  a  ventral  view,  with 
the  distal  margin  uppermost.  In  the  other 
species  a  posterior  view  is  shown,  with  the 
distal  margin  below. ) 

Measurements  (mm):  Antennae  16;  fore 
wing  length  14. 

Diagnosis.  Immediately  distinguishable 
from  fnsca  by  its  pale  body  and  venation. 

The  following  three  closely  related  species 
are  grey  with  prominent  stigmas,  and  look 
much   alike. 

Pimachrysa   albicostales   sp.    n. 

Figures  27,  28 

HoJotijpe.  Male  from  Mexico;  Raja  Calif., 
22  mi.  N.  of  Punta  Prieta,  9-XII-58,  H.  R. 
Leech,  col.,  in  the  California  Academy  of 
Sciences.  The  specific  name  refers  to  the 
white  costal  veinlets. 

Description.  Eight  or  nine  black  spots 
on  apical  subcostal  crossxeins  of  fore  wing, 
six  or  seven  in  hind  wing.  Intramedian  cell 
of  fore  wing  about  four  times  as  long  as 
high,  branches  of  Rs  +  MA  more  sharply 
inclined  to  Psm  than  in  fusca,  and  cells  be- 
t^\'een  Rs  and  Psm  longer.  MPl  of  hind 
wing  merely  touches  Rs  +  MA,  not  co- 
alesced over  a  long  distance,  as  in  fusca. 
Wing  setae  black;  on  dorsum  of  fore  wing 
equal  to  length  of  longest  costal  veinlet. 

Abdomen  pale,  tergites  2-8  \\'ith  triangu- 
lar dark  marks  (apices  anterior);  9  with 
small  mediodorsal  dark  rectangle,  ectoprocts 
fuscous.  Sternites  light  fuscous,  paler  medi- 
ally; ninth  dark-lineate  basally,  dark  api- 
cally.  Apical  beak-like  extension  of  ninth 
sternite  longer  than  in  fnsca.  Cuticular 
glands   of  basal   segments   sparser  than   in 


Mesochrysinae  and  Nothochrvsinae  •  Adams        227 


fusca.  Gonarcus  shorter,  mediuncus  (Figs. 
27,  28)  longer,  spoon-shaped  distally,  with 
apieal   tooth. 

Measurements  (mm).  Fore  wing  10.0- 
11.3  (10.6). 

Diagnosis.  Head  and  thorax  marked  sim- 
ilarly to  fusca,  but  paler.  Wing  veins  colored 
as  in  fusca,  but  costal  veinlets  of  fore  wing 
pale.  Membrane  not  brown  in  center  of 
cells  as  in  fusca. 

Record  (Paratype).  ARIZONA,  Tucson, 
Tumamoc  Hill,  week  ending  5-XI1-61,  <5 , 
in  ethylene  glycol  pollen  trap,  Sandra  Ray 
Johnson,  col.  (MCZ,  in  alcohol). 

Pimachrysa   fusca   sp.    n. 

Figures  1,  7,  11-13,21,24 

Holotype.  Male  from  L.  Covington  Flat, 
Joshua  Tree  Nat.  Mon.,  Riverside  Co., 
Calif.,  19-III-61,  E.  L.  Sleeper,  col.  In  the 
California  Academy  of  Sciences.  The  name 
refers  to  the  dark  appearance  of  this  species. 

Description.  Head  pale,  marked  with 
black  ( Fig.  7 ) ;  scape  pale  with  medial  and 
lateral  black  stripes;  pedicel  and  flagellum 
fuscous.  Pronotum  about  as  long  as  broad, 
lateral  margins  strongly  deflexed;  no  trans- 
verse groove;  pale  ivory  with  fuscous  me- 
dian and  marginal  stripes;  a  short  narrow 
longitudinal  black  stripe  each  side  between 
the  fuscous  bands,  connected  to  the  median 
stripe  posteriorly  by  a  narrow  transverse 
black  band;  black  spot  in  posterolateral 
corner.  Meso-  and  metanota  with  median 
and  lateral  fuscous  stripes;  mesoscutum 
pale  near  wing  base,  with  black  spots  ( areas 
devoid  of  microtrichia )  on  inner  and  outer 
margins  of  lateral  stripe.  Pleurae  fuscous. 
Membrane  anterior  to  prosternum  pale. 
Legs  pale,  femora  with  dark  preapical  band; 
tibiae  with  dark  bands  at  base,  about  two- 
fifths  of  distance  to  apex,  and  at  apex.  Tarsi 
fuscous. 

Wing  venation  as  in  Figure  1.  Fore  wing: 
veins  black  except  for  C,  apex  of  Sc,  R, 
and  bases  of  Cu  and  anals,  white.  Stigma 
pale,  with  4-6  apical  subcostal  crossveins 
conspicuously  black-bordered.  Marginal 
area    posterior    to    3A    black;    membrane 


hyaline,  usually  brown-tinted  in  central 
portion  of  all  cells.  Intramedian  cell  about 
3  times  as  long  as  high.  Setae  on  upper  sur- 
face black,  long,  stiff,  sparse.  Hind  wing: 
veins,  except  Sc,  all  pale;  posterior  mar- 
ginal black  at  intersections  of  marginal 
veinlets.  MPl  coalesced  with  MA  +  Rs  for 
a  short  distance.  Macrotrichia  about  one- 
third  length  of  those  of  fore  wing. 

Abdomen  in  male  dark  fuscous.  Ninth 
sternite  with  a  beak-like  posterior  projec- 
tion, fused  with  eighth  sternite  medially 
(Fig.  11).  Pores  of  cuticular  glands  sparse 
on  apical  segments,  on  ninth  sternite  con- 
fined to  anterolateral  area.  Mediuncus  (  Figs. 
12,  13)  widened  between  attachment  of 
membrane  and  apical  tooth.  Female  abdo- 
men mostly  fuscous;  ninth  tergite  pale  lat- 
erally. Ninth  gonocoxites  ( Fig.  21 )  not  so 
large  as  in  P.  grata,  and  with  fewer  hooked 
setae.  Eighth  sternite  ( Fig.  24 )  with  blunt, 
anterodorsally  directed  horn;  a  basal  frag- 
ment present,  shaped  as  a  narrow  trans- 
verse band.  Ninth  valvulae  appear  as  slen- 
der rods.  Spermatophore  (  Fig.  21,  sp )  ovoid, 
with  cuplike  apical  indentation,  and  tubular 
process  extending  into  genital  opening. 

Measurements  (mm).  Fore  wing  length 
8.8-11.1  (10.3);  maxillary  palpus  (apical 
three  segments )  0.67-0.83  ( 0.73 ) ;  scape 
0.24-0.32  (0.28). 

Diagnosis.  In  shape  of  head  and  wings, 
and  in  venation,  this  species  is  closer  to  P. 
grata  than  to  P.  nigra.  It  may  be  recognized 
by  the  gray-pruinose  body,  smoky  wings, 
prominent  dark  pterostigma,  dark  costal 
veinlets  of  fore  wings,  and  pale  veins  in 
the  hind  wing. 

Records  ( Paratypes ) .  CALIFORNIA : 
Riverside  Co.,  Joshua  Tree  National  Monu- 
ment, L.  Covington  Flat  #1,  19-III-61  6; 
#2,  5-III-61  9;  #4,  25-XI-60  6,  5-XI-60; 
U.  Covington  Flat  #1,  5-XI-60  i;  Smith- 
water  Wash  #1,  25-XI-60  9,  4-I1I-61  9; 
all  collected  by  E.  L.  Sleeper  in  molasses 
bait  traps  (Long  Beach  State  College,  MCZ, 
Adams  Collection).  Los  Angeles  Co.:  Aliso 
Can.  chapparal,  26-X-58  9,  R.  X.  Schick 
(UCLA). 


228        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  4 


Pimachrysa    infermedia   sp.    n. 

Figures  2,  20,  25 

Holotypc.  Female  from  Snow  Creek, 
1500',  White  Water,  Rixerside  Co.,  Calif., 
8-III-1955,  W.  R.  M.  Mason,  col.  Canada 
Department  of  Agriculture,  Ottawa.  The 
name  means  intermediate. 

Description.  Head  marked  with  black 
similarly  to  fmca,  but  dark  border  of  an- 
tennal  socket  not  connected  to  lateral  ver- 
tex stripe,  median  \ertex  mark  larger,  and 
genae  black-bordered  posteriorly.  Maxillaiy 
palpi  and  scape  both  shorter  than  in  fiisca. 
Pronotum  longer  than  in  fusca,  pale,  with 
lateral  and  median  red-fuscous  stripes, 
broader  posteriorly;  between  these  each 
side  a  brown  stripe,  extending  to  furro\\' 
(which  is  barely  perceptible).  Meso-  and 
metanota  dark;  yellow  spots  laterally  on 
prescuta  and  scuta.  Propleura  and  cervical 
sclerites  pale  fuscous,  prosternum  dark- 
fuscous;  membrane  anterior  to  sternum  pale, 
with  median  and  lateral  fuscous  stripes, 
overlaid  with  a  rufous  transverse  band. 
Meso-  and  metapleurae  fuscous.  Legs  pale, 
femora  with  a  broad  preapical  dark  band; 
tibiae  with  thin  dark  band  at  about  one- 
third  distance  from  base  to  tip;  fore  and 
mid  tibiae  slightly  infuscated,  hind  with 
two  narrow  dark  stripes  on  ventral  surface. 

Wing  venation  as  in  Figure  2;  few  mar- 
ginal veinlets  forked.  R  (basal  two-thirds) 
and  3A  in  both  wings,  and  bases  of  lA,  2A, 
and  Cu  in  hind  wing,  pale;  other  veins  all 
dark.  Stigma  white,  with  dark  clouds  on 
apical  subcostal  crossveins;  membrane  prox- 
imal to  end  of  3A  in  fore  wing  fuscous. 

Abdomen  fuscous,  segments  narrowly 
pale  at  anterior  and  posterior  margins;  mus- 
cle scars  on  tergites  black,  surrounded  by 
pale  areas.  Female  external  genitalia  sim- 
ilar to  those  of  fmca;  subgenital  plate  bears 
knob  on  basal  margin,  basal  fragment  much 
wider,  and  less  sclerotized  than  in  fusca 
(Fig.  25). 

Measurements  ( mm  ) .  Fore  wing  9.2; 
apical  three  segments,  maxillary  palpus, 
0.49;  scape  0.21. 

Diagnosis  and  discussion.    In  the  features 


by  which  this  species  differs  from  typical 
fusca — dark  venation,  smaller  number  of 
gradate  cells,  fewer  forked  marginal  vein- 
lets,  short  palpi,  color  and  shape  of  pro- 
notum— it  grades  toward  nigra.  Further- 
more, it  was  taken  geographically  close  to 
a  nigra  locality.  It  may  be  a  unique  hybrid 
individual,  or  a  representative  of  an  intro- 
gressed  population. 

Biology.  The  crop  of  the  type  is  filled 
with  Salix  pollen  (identified  by  D.  Walk- 
ington,  California  State  College,  Fullerton). 

Pimachrysa   nigra   sp.    n. 

Figures  6,  14-16,  26,  34 

Holotypc.  Male  from  Gavilan,  Riverside 
Co.,  Calif.  19-III-36,  on  Ceanothus  crassi- 
folius,  Timberlake,  col.  (California  Acad- 
emy of  Sciences ) .  The  name  is  descriptive 
of  the  dark  coloration. 

Description.  Head  long,  eyes  small;  color 
shades  from  ivory  anteriorly,  to  orange  on 
vertex;  dark-fuscous  markings  as  in  Figure 
6;  mark  between  antennae  extending  an- 
teriorly to  frontal  suture,  and  broadly  over 
antennal  sockets  and  vertex.  Maxillary  palpi 
short,  dark,  pale  at  joints.  Occipital  fora- 
men broadly  dark-margined.  Scape  pale, 
broad  medial  and  lateral  dark  stripes;  ped- 
icel and  flagellum  dark.  Pronotum  long, 
brown-fuscous  with  t\\'o  narrow  submedian 
pale  stripes.  Meso-  and  metanota  dark,  two 
disconnected  yellow  stripes.  Propleurae, 
sternum,  and  cervical  sclerites  dark;  rufous 
median  area  on  membrane  anterior  to  pro- 
sternum.  Meso-  and  metapleurae  dark.  Legs 
dark;  femora  pale  basally  and  apically; 
tibiae  (especially  metathoracic)  paler  ex- 
teriorly, with  apical  dark  band. 

Fore  wing  veins  dark;  bases  of  C,  R, 
3A,  and  marginal  vein  in  jugal  area,  pale. 
Stigma  black;  conspicuous  ivory  spots  on 
membrane,  between  apical  subcostal  cross- 
veins.  Hind  wing  similar,  but  R  pale  for 
two-thirds  length. 

Venation  ( Fig.  4 ) :  A  break  in  alignment 
of  inner  gradates  and  Psm  in  both  wings; 
weak  pseudomedian  fold  in  both  wings 
(especially  hind),  but  no  definite  articula- 


Mesochrysinae  and  Nothochrysinae  •  Adams        229 


tions,  as  are  found  in  NotJiochrysa;  usually 
only  1  or  2  branches  of  Rs  forked  margin- 
ally ( 3  and  4  in  hind  wings  of  one  female ) . 

Abdomen  dark,  segments  narrowh'  pale- 
banded  posteriorly.  Male  8th  and  9th 
sternites,  and  9th  tergite-ectoproct  more 
strongly  fused  than  in  fusca  (Fig.  14).  Many 
cuticular  gland  openings,  except  on  ecto- 
proct.  Margin  of  mediuncus  between  apical 
tooth  and  attachment  of  membrane,  nearly 
straight  ( Figs.  15,  16 ) .  Female  terminalia 
similar  to  those  of  fusca;  8th  sternite  (Fig. 
26)  preceded  by  broad  basal  fragment, 
ventrally  with  blunt  horn.  Ninth  valvulae 
reduced  to  dotlike  sclerites. 

Measurements  (mm).  Fore  wing  7.5-9.0 
(8.0)  long;  apical  three  segments,  maxillary 
palpi,  0.48-0.53  (0.51);  antennae  7-8  (7.5). 

Diagnosis.  This  small  species  may  im- 
mediately be  distinguished  from  all  other 
species  of  Pimachrysa  by  its  brown-black 
and  orange  markings.  It  strikingly  resem- 
bles Nothochnjsa  calif  omica  Banks,  for 
which  it  has  been  mistaken  by  several 
workers,  including  the  writer.  In  shape  of 
head,  and  condition  of  pseudomedia,  it  is 
closer  to  Nothoclirysa  than  is  any  other 
species  of  Pimachrysa.  However,  these 
genera  are  structurally  so  different  that  the 
colorational  resemblance  is  undoubtedly  co- 
incidental. 

Records  (Paratype).  CALIFORNIA: 
Gavilan,  l-IV-38,  2,  on  Ceanothus  crassi- 
foUus,  Timberlake,  col.  (MCZ).  Pinnacles, 
San  Benito  Co.,  23-111-40,  R.  L.  Usinger, 
col.  (CIS).  Lebec,  Kern  Co.,  l-IV-39,  9, 
E.  S.  Ross  (CAS). 

DYSPETOCHRYSA  gen.  n. 
Figure  42 

Type  species.  Tribochrysa  vetuscida 
Scudder.  The  feminine  name  means  "diffi- 
cult-chrysopid,"  referring  to  the  intricacy 
resulting  from  the  superimposition  of  fore 
and  hind  wings  in  the  fossil. 

Description.  Basal  subcostal  crossvein 
distal  to  origin  of  Rs,  opposite  MF;  Rs  + 
MA  originates  far  basal  of  MF.  First  sec- 
torial  crossvein   intersects   MPl   verv  near 


MF.  Intramedian  cell  long.  Proximal  branch 
of  Rs  -I-  MA  coalesces  with  MPl,  and  sec- 
ond branch  coalesces  with  proximal  branch, 
so  that  basal  portion  of  Psm  contains  no 
crossveins.  Psc  strongly  developed,  two 
marginal  veinlets  from  third  medial  cell, 
basal  one  forked;  at  least  four  additional 
non-forked  marginal  veinlets  from  Psc.  Hind 
wing  with  zig-zagged  Psc. 

Discussion.  The  fore  wing  base  is  similar 
to  that  of  ArcJiaeochrysa,  except  for  the 
overlap  of  branches  of  Rs  +  MA  forming 
Psm.  It  is  most  interesting  that  the  wing 
should  demonstrate  such  a  generalized 
structure  of  Rs  +  MA  and  the  intramedian 
cell,  while  having  a  pseudomedia  which,  at 
least  basally,  resembles  that  of  Nothochrysa. 
In  this  regard,  Dyspetochrysa  is  more  ad- 
vanced than  all  other  Nothochrysinae,  save 
Nothochrysa  itself. 

Dyspeiochrysa   vefuscula   Scudder 

Figure  42 

Tribochrysa  vetuscida  Seudder,  1890,  U.  S.  Geol. 
Geogr.  Survey  Terr.,  Kept.,  13:  170,  pi.  14, 
fig.  9.  Holotype:  Florissant,  Colorado,  Mio- 
cene; MCZ  No.  245,  tip  of  abdomen  and  distal 
half  of  fore  wing  missing:  hind  wing  venation 
mostly  not  visible,  examined. 

Paleochrysa  vetusciila:  Cockerell,  1908,  Canad. 
Entomol.  40:  90. 

Paleochrysa  stricta:  Carpenter,  1935,  J.  Paleontol. 
9:  263  (not  Scudder). 

Description.  Head  large,  interocular 
width  1.0  mm,  eye  0.5  mm.  Scape  short, 
flagellar  segments  0.2  X  0.1  mm.  Pronotum 
wide,  anterior  margin  straight,  corners 
slightly  angulate;  dark,  a  four-lobed  pale 
central  area;  width  1.44  mm;  length,  ex- 
treme 1.0  mm,  midline  0.7  mm.  Costal  area 
narro\\',  tallest  cell  0.63  X  0.55  mm.  Intra- 
median cell  0.35  X  1.5  mm,  pseudomedial 
area  as  wide  as  pseudocubital  area. 

ARCHAEOCHRYSA   gen.   n. 

Figures  40,  41,  46,  47 

Type  species.  Paleochrysa  creedei  Car- 
penter. The  feminine  name  means  "old- 
chrysopid"  referring  to  the  archaic  \enation. 


230         Bulletin  Miificuw  of  Contpanitive  Zoology,  Vol.  135,  No.  4 


Description.  Fore  wing:  Basal  subcostal 
crossvein  distal  to  origin  of  Rs  +  MA  op- 
posite MF.  Rs  +  MA  arises  nearer  to  basal 
medial  crossvein  than  to  MF.  Sectorial 
crossvein  intersects  MPl  slightly  distally  to 
MF.  Proximal  branch  of  Rs  +  MA  not 
coalesced  with  MPl,  but  connected  by  a 
gradate  crossvein;  Psm  strongly  zig-zagged. 
Psc  well  developed. 

Hind  wing:  Base  of  Rs  +  MA  moderately 
long;  not  coalesced  with  MPl.  Psm  zig- 
zagged, with  no  overlap  of  veins.  Psc  poorly 
developed,  strongly  zig-zagged,  so  that  the 
gradate  series  apparently  extends  to  MP2; 
marginal  veinlets  from  Psc  all  forked. 

Pronotum  broad,  squarish;  cervical  scle- 
rites  and  prosternum  as  in  Pimachrysa.  Ter- 
minal abdominal  segments  of  male  broad, 
short;  ectoprocts  forcipate  (Fig.  46). 

Discussion.  This  is  the  most  archaic  of 
the  Miocene  genera.  It  is  unique  in  its  lack 
of  fusion  of  MPl  and  Rs  +  MA  in  the  hind 
wing.  The  unspecialized  Psc  of  the  hind 
wing  is  found  elsewhere  only  in  Faleo- 
chnjsa  icickhami  Cockerell. 

Archaeochrysa   creedei    (Carpenter) 

Figure.s  41,  46,  47 

Paleochnisa  creedei  Carpenter,   1935,  J.  Paleontol. 

9:   265,  lis.   3;    1938,   Psyche  45:    108.   Holo- 

type:   Creede,  Colorado,  Miocene,  A.  Caplan, 

coll.;  MCZ  No.  4316,  examined. 
Paleochrijsa  .striata:    Carpenter,    1938,   Psyche   45: 

108,  fiji.   1. 

Description.  Costal  area  narrow,  0.25  mm 
wide,  cells  about  as  high  as  long.  Ten 
branches  of  Rs  +  MA  in  fore  wing,  and  7-8 
in  hind  wing.  Gradate  cells  about  twice 
as  long  as  wide.  Intramedian  cell  narrow, 
about  0.375  X  1.5  mm.  Flagellar  segments 
0.1.3  X  0.21  mm.  Pronotum  (Fig.  47)  squar- 
ish, anterior  margin  not  clear;  0.94  mm 
wide,  0.91  mm  (0.625  mm  on  midline)  long. 
Ectoprocts  (Fig.  46)  lateral,  forcipate; 
gonarcus  small. 

Discussion.  The  lateral  margins  of  the 
conspicuous  cervical  sclerites  apparently 
correspond  to  the  pronotal  margins  in  Car- 
penter's (1935)  figure. 


The  prosternum  apparently  was  un- 
sclerotized.  The  basal  Banksian  cell  is  nor- 
mal for  ArcJuieochry.sa;  on  the  type,  the 
proximal  branch  of  Rs  +  MA  is  very  in- 
distinct, and  was  not  shown  by  Carpenter, 
but  is  present  on  all  the  other  specimens 
of  this  species. 

Records.  Creede,  Colorado,  Miocene; 
MCZ  No.  4462-4471. 

Archaeochrysa   paranervis   sp.    n. 
Figure  40 

Paleochrysa     vetiiseula     Cockerell,     1908,     Canad. 

Entomol.   40:    90.    (Erroneous   identification.) 
Paleoehnjsa  .siricta:   Carpenter,  1935,  J.  Paleontol. 

9:   263,  fi.u.    1.    (Erroneous  identification.) 

Holotype.  Univ.  Colo.  Mus.  No.  4419, 
Florissant,  Colorado,  Miocene,  expedition 
1907,  examined.  The  name  means  "beside- 
vein,"  referring  to  the  juxtaposition  of  the 
radial  crossveins  and  the  branches  of  the 
radial  sector. 

Description.  Venation  as  in  Figure  40; 
fore  wing  broad;  costal  area  broad;  Sc  ap- 
proaches C  near  stigmal  base;  12  branches 
of  Rs  +  MA,  several  branches  opposite 
radial  crossveins;  longest  gradate  cell  2.5 
times  as  long  as  broad.  Fore  wing  length: 
14.2  mm. 

Diagnosis.  Differs  from  creedei  in  the 
length  and  arrangement  of  branches  of  Rs, 
and  in  width  of  the  costal  space.  This  speci- 
men was  the  basis  for  Carpenter's  ( 1935 ) 
figure  and  redescription  of  FaJeochrysa 
strict  a. 

Archaeochrysa   fracfa    (Cockerell) 

Paleochrysa  fracta  Cockerell,  1914,  J.  Geol.  22: 
716,  fig.  2.  Holotype:  Florissant,  Colorado, 
Miocene,  H.  F.  Wickham,  col.;  MCZ  No.  4501 
(formerly  No.  .3349),  isolated  fore  wing, 
examined.  Carpenter,  1935,  J.  Paleontol.  9: 
264,  fig.  7,  p.  268. 

Description.  Costal  area  moderately 
wide,  the  cells  0.95  mm  X  0.5  mm;  veinlets 
inclined  about  10°,  17  branches  of  Rs  +  MA, 
Rs  hardly  zig-zagged.  Longest  gradate  cells 
3.1  times  as  long  as  wide;  intramedian  cell 
2.0  X  0.5  mm,  acute  proximally,  distal  cross- 
vein  oblique.    lA  forked. 


Mesochrysinae  and  Nothochrysinae  •  Adams        231 


Diagnosis.  This  species  differs  from 
crecdei  in  the  shape  of  the  subcostal  area, 
more  branches  of  Rs  +  MA,  longer  gradate 
cells,  and  forked  lA. 

Carpenter's  figure  of  this  specimen  was 
inadvertently  transposed  with  that  of  Tri- 
hochnjsa  firmata,  and  appeared  on  page 
268,  as  his  Figure  7. 

Genus  PALEOCHRYSA  Scudder 

Figures  38,  39 

Paleochnjsa  Scudder,  1890,  U.  S.  Geol.  Geogr. 
Survey  Terr.,  Rept.,  13:  166.  Type  species 
(by  nionot\pv') :  Paleochnjsa  stricta  Scudder. 
Cockerell,  1908,  Canad.  Eutomol.,  40:  90. 
Carpenter,   1935,  J.  Paleontol.  9:   262. 

Lithochnjsa  Carpenter,  1935,  J.  Paleontol.  9:  265. 
Type  species  ( by  original  designation )  Paleo- 
chnjsa  icickhami  Cockerell.   xew   synonymy. 

Description.  Basal  subcostal  crossvein 
distal  to  origin  of  Rs  +  MA;  opposite  MF. 
Rs  +  MA  arises  near  MF;  proximal  branch 
of  Rs  +  MA  not  coalesced  with  MPl,  hence 
no  overlapping  veins  in  Psm.  Gradate  series 
run  into  Psm  and  Psc  in  an  even  curxe  in 
both  wings.  Psc  of  hind  wing  either  reg- 
ularly zig-zagged  with  forked  marginal 
veinlets  basally  (P.  wickhatni),  or  with 
furcations  more  proximal,  so  zig-zagging  is 
irregular,  as  in  Pimachnjsa,  Hijpochnjsu, 
etc.  (P.  stricta). 

Discussion.  The  type  of  P.  stricta  has 
all  four  wings  superimposed,  making  inter- 
pretation excessively  difficult.  But  the  short 
distal  stem  of  Rs  +  MA,  in  the  fore  wing 
and  the  coalescence  of  Rs  +  MA  with  MPl, 
forming  a  quadrangular  Banksian  cell  in 
the  hind  wing,  are  clear;  thus  this  species 
has  the  essential  characteristics  of  Litho- 
chnjsa Carpenter.  Paleochrysa  icickhami, 
the  type  of  Lithochrijsa,  does  not  differ 
from  stricta  in  any  important  respect  save 
the  more  archaic  structures  of  Psc  in  the 
hind  wing.  The  species  of  Pimachnjsa  show 
such  variation  in  the  site  of  furcation  of  the 
marginal  veinlets  (e.g.,  cf.  Figs.  1  and  2) 
that  I  do  not  consider  this  difference  in 
structure  of  Psc  of  sufficient  importance  to 
warrant  maintaining  a  generic  distinction. 


Paleochrysa   sfricta   Scudder 
Figures  38,  39 

Paleochnjsa  stricta  Scudder,  1890,  U.  S.  Ceol. 
C;eogr.  Survey  Terr.,  Kept.,  13:  166,  pi.  14, 
figs.  13,  14.  Holotype:  Florissant,  Colorado, 
Miocene,  S.  H.  Scudder;  MCZ  No.  242  a-b, 
e.xamined. 

(not)  Trihochnjsa  vettiscttla:  Cockerell,  1908, 
Canad.  Entomol.  40:  90.  Carpenter,  1935,  J. 
Paleontol.,  9:  263. 

Description.  Costal  area  moderately  wide, 
cells  0.69  mm  long,  0.875  mm  tall.  Twelve 
branches  of  Rs  +  MA.  Gradate  cells  0.5  X 
1.0  mm;  intramedian  cell  0.35  X  1.25  mm; 
8  unbranched  veinlets  from  Psc  in  fore 
wing.  In  hind  wing,  5  unbranched  veinlets 
from  Psc  to  wing  margin;  CuA  3-branched, 
the  distal  branch  with  a  marginal  fork. 
Fore  wing  length,  16.0  mm,  width  5.0  mm. 

Paleochrysa   wickhami   (Cockerell) 

Paleochnjsa  wickhami  Cockerell,  1914,  J.  Geol. 
22:  717,  fig.  3.  Holot>pe:  Florissant,  Colo- 
rado, Miocene,  H.  F.  Wickhani,  col.;  MCZ 
No.  4499,  e.xamined. 

Lithochnjsa  wickhami:  Carpenter,  1935,  ].  Paleon- 
tol., 9:  265,  fig.  4. 

Description.  Costal  area  moderately 
broad,  cells  about  0.46  x  0.5  mm.  Winlets 
inclined  about  15  degrees.  Nine  branches 
of  Rs  +  MA  in  fore  wing,  and  hind  wing. 
Intramedian  cell  0.313  X  0.88  mm.  Gradate 
cells  5.25  X  0.75  mm.  Costal  space  nar- 
rows abruptly  before  stigma,  as  in  Llypo- 
chnjsa  nobilis-  C  and  Sc  may  be  coalesced 
for  a  short  distance.  Radius  runs  in  a 
smooth  curve  at  wing  apex,  not  angulate 
near  stigma  as  in  Hijpochnjsa. 

Eye  fairly  large  (0.44  mm  diam.).  Pro- 
notum  rounded  anteriorly,  sides  straight, 
parallel. 

Paleochrysa   concinnula   Cockerell 

Paleochnjsa  concinnula  Cockerell,  1909,  Canad. 
Entomol.  41:  218,  fig.  5.  Holotype:  Floris- 
sant, Colorado,  Miocene;  Colo.  Uni\ .  Mus., 
not  examined. 

Lithochnjsa  concinntila:  Carpenter,  1935,  J.  Pa- 
leontol. 9:  266,  fig.  5. 

Paleochrysa  ferruginea  Cockerell,  ibid.-.  218,  fig.  6. 
Holotype:  Florissant,  Colorado,  Miocene;  Colo. 


232         Bulletin  Museum  of  Ciunpaiative  Zoology.  Vol.  135,  No.  4 


I'liiv.    Mils.,    not    ixaiiiiiR'cl.     (^aipt'iitt  r.    Inc. 
(it.:  266-267. 

Genus   TRIBOCHRYSA   Scudder 

TrihtnluiiMi  ScucUIlt,  LS85,  in  Zittrl-Banois,  Traite 
PaU'ont.  1 :  777.  Type  species  ( by  monotypy ) : 
Tiil>(>chrt)sa  iudciinalis  Scudder.  Scudder, 
1890,  U.  S.  Ceol.  Ccogr.  Surv.  Terr.,  Kept. 
13:    168. 

Description.  Eyes  large  (0.5  mm);  .scape 
large,  flagellar  segments  0.15  x  0.25  mm. 
I'roiiotuin  short,  broad,  margin  arcuate. 
.Subcostal  crossNciu  obscured  in  all  speci- 
lucns.  Hs  -f  M.\  arises  slightly  basally  to 
Ml'\  \o  oNcrlap  of  veins  in  P.sm.  Inner 
gradates  in  two  series,  so  that  the  fourth 
gradate  (numbered  from  base)  is  much 
closer  to  Rs  than  is  the  second  or  third.  Psc 
well  de\eloped  in  fore  wing.  In  hind  wing, 
Hs  +  MA  coalesces  with  MPl,  and  Psc  is 
less  well  developed  than  in  fore  wing. 

Discu.ssion.  The  irregularity  in  the  inner 
gradate  series  is  a  necessary  precondition 
for  separation  from  Psm,  as  has  taken  place 
in  man\-  Chrysopinac.  It  is,  therefore,  not 
surprising  that  .several  other  taxa  exhibit  a 
similar,  but  less  emphasized  break  in  the 
inner  gradate  series  [Fimachrijsa  ni<^ra  (Fig. 
4),  lliipoclinisa  (Fig.  5),  Notlioclinjsa  cali- 
f arnica  (  Fig.  3)]. 

Tribochrysa   inaequalis   Scudder 

Trihoclin/.sa  ina((j\iali\  Scudder,  1S85,  ;'/i  Zittel- 
Barrois,  Traite  Paleoiit.  1 ;  777,  fijf.  982.  Holo- 
t\pe:  Floris.sant,  Colorado,  Mioceue;  MCZ 
.\o.  24.3,  examined.  Scudder,  1890,  U.  S. 
Ceol.  Ceoyr.  Surv.  Terr.,  Rept.  13:  170.  Coek- 
erell,  1908,  Canad.  Entouiol.  40:  90.  Carpen- 
ter.   1935,  J.  i^deoutol.  9:   267,  fig.  6. 

Tribochrysa   firmafa   Scudder 

Tribochrysa  finiidta  Si^udder,  1890,  U.  S.  Ceol. 
Ceogr.  Siu\.  Terr.,  Kept.  13:  172,  pi.  14,  figs. 
6,  7,  10.  11.  Cotypes:  Florissant,  Colorado, 
Miocene-  MCZ  No.  241,  Xo.  4127,  examined. 
Carpenter.    195.5.   j.    Paleontol.   9:    267,   fig.   7. 

LITERATURE  CITED 

AuA.Ms,  P.  A.  19.57.  .\  new  genus  and  new 
species  of  Chr\sopidae  from  the  western 
United  States,  with  remarks  on  the  wing 
xcnatinn   of  tlie   famil\'.    i'svehe  6.'i:    67-74. 


— .  1958.  Studies  in  the  Neuroptera,  with 
special  reference  to  wing  structure  and  evolu- 
tion in  the  Osmyloidea.  Ph.D.  thesis,  Har- 
vard  University   (unpubl.). 

1962.     A  stridulatory  structure  in  Chry- 


sopidae.  Pan-Pac.  Entomol.  38:  178-180. 
Bk.\uer,  a.      1867.     Larve  von  Hypochnjsa  nohilis 

Heyd.    Verhandl.    Zool.-Bot.    Ces.    Wien    17: 

27-29,  pi.  9,  fig.  1. 
CARPENTER,  F.  M.      1935.     Tertiary  insects  of  the 

family  Chry.sopidae.  J.  Paleontol.  9:  259-271. 
Ehh.'\hi)t,     E.      1916.     Zur     Kenntnis     der     Inner- 

vierung  und  der  Sinnesorgane  der  Fliigel  von 

Insekten.  Zool.   Jahrb.    ( Anat. )    39:    295-3.34, 

pis.  17-18. 
I'lUEDRicH,  H.      1953.    Neuroptera.  Bronns  Klas.sen 

u.  Ordnungen  des  Tierreichs,  5  Band,  3  Al)t. 

Xll  Buch,  Teil  a.  Leipzig.  148  pp. 
KiLLiNGTON,   F.   J.     1937.     A   monograph    of   the 

British  Neuroptera.  Ray  Society,  London.  269 

pp.,  15  pis. 
Ki.Mxnxs,    D.    E.      1952a.     Some    new    Australian 

Chrysopidae.    Ann.   Mag.  Nat.   Hist.    (12)   5: 

68-81. 
.      1952b.     A  revision  of  the  genera  of  the 

Ap()chr\sinae  ( fam.  Chrvsopidae ).  Ann.  Mag. 

Nat.  Hi.st.   (12)  5:  929-944. 
M.AHrvxovA,        O.      1949.     Mesozoic        lacewings 

( Neuroptera )    and   their  bearing  on   concepts 

of   phylogeny    and    systematics    of   the    order. 

(In    Russian.)    Trudy   Paleontol.    Inst.,   Akad. 

Nauk   SSSR   20:    150-170. 
Nav.As,  L.      1910.     CriscSpidos  nuevos.  Broteria  9: 

38-59. 
.      1913.      Les  Chrysopides  (Ins.  Nevr. )  du 

Musee   de    Londres.   Ann.    Soc.    Sci.    Bruxelles 

37:  292-330;  38:  73-114. 
Phixcipi,  M.  M.      1946.      Contributi  alio  studio  dei 

Neurotteri    Italiani.    IV.    Ni)tlu>cluy.su    italica 

Rossi.  Boll.  1st.   Entomol.  Univ.   Bologna    15: 

cS,5-102. 
.      1958.      Neurotteri     dei     Monti     Sibillini. 

Mem.  Mus.  Ci\ .  Storia  Nat.  Verona  6:   175- 

189. 
Scudder,     S.     H.      1890.     The     fossil     insects     of 

North  America,  with  notes  on  some  European 

.species.    II.  The  Tertiary  in.sects.  Rept.  U.  S. 

Ceol.  Ceogr.  Surv.  Terr.' 13:  1-734,  pis.  1-28. 
Tn.EYAHi),     R.     J.      1916.     Studies     in     Australian 

Neuroptera.  No.  3.  The  wing-venation  of  the 

Chrysopidae.    Proc.    Linn.    Soc.    New    South 

Wales  41   (2):  221-248,  pis.  X,  X  bis,  XI. 
TjEUER,  Bc).      1941.      Some  remarks  on  the  generic 

names    of    the    British    Neuroptera.    Entomol. 

Tidskrift   (1941):  24-31. 
.    1966.    Neuroptera-Planipennia.   The  lace- 
wings   of    Southern    Africa.    5.    Chrysopidae. 

South   African   Animal   Life    12:    228-534. 
T()sc:iTi,    C.    A.      1966.     The    taxonomy,    life    his- 
tories, and  mating  beha\'ior  of  the  green  lace- 


Mesochrysinae  and  Nothochrysinae  •  Adams        233 


wings   of  Strawberry   Canyon.    Hilgardia   36: 
391-435. 

(Received  21    October  1965) 

ABBREVIATIONS   USED  IN   FIGURES 

b — basal  Banksian  cell;  CUA — cubitus  anterior; 
CUP — cubitus  posterior;  CV — cervical  sclerite;  CX 
— coxa;  ect — ectoproct;  fr — frenulum;  g.c. — gradate 
cell;  gcx —  9 ,  nintb  gonocoxite  ( gonopophyses 
laterales ) ,  i  ,  gonocoxites  (  =  "paranieres,"  "ento- 
processus");    gs — gonarcus;    hyp.    i. — hypandriuni 


internum;  i.g. — inner  gradate  crossvein;  im — intra- 
median  cell;  J— jugal  vein;  jl— jugal  lobe;  MA— 
the  proximal  branch  of  the  fused  media  anterior 
and  Rs,  presumably  media  anterior;  memli — mem- 
brane; MF— medial  fork,  fork  of  MP;  MP— media 
posterior;  mu — mediuncus;  o.g. — outer  gradate 
crossvein,  PL — propleuron;  pni — paramere;  psc — 
pseudocu!)itus;  psm— pseudomedia;  R— radius;  RS' 
— proximal  branch  of  radial  sector;  sp — spermato- 
phore;  ST — sternum;  SXV — first  sectorial  cross- 
vein;  lA,  2A,  3A— anal  veins;  Im— first  medial 
crossvein. 


234        Bulletin  Mu.si'uni  of  Conijxiiative  Zoology,  Vol.  135,  No.  4 


Wings:  Fig.  1.  Pimachryio  lusca  sp.  n.  Fig.  2.  P.  inlermedia  sp.  n.  Fig.  3.  Nothochrysa  californica.  Fig.  4.  P.  nigra  sp.  n. 
Fig.  5.  Hypochryso  nob/7is.  Heads:  Fig.  6.  P.  nigra.  Fig.  7.  P.  fusca.  Fig.  8.  N.  ca//7ornica.  Fig.  9.  P.  gra(o.  Fig.  10. 
K.  raphidioides. 


Mesochrysinae  and  Nothochrysinae  •  Adatm        235 


Figs.  11-13.  Pimachrysa  fusco.  Fig.  11.  Mole  abdomen,  lateral  view.  Fig.  12.  Gonorcus,  lateral  view.  Fig.  13.  Same, 
posterior  view.  Figs.  14-16.  P.  nigro.  Fig.  14.  Male  abdomen,  lateral  view.  Fig.  15.  Gonorcus,  posterior  view.  Fig.  16. 
Some,  loterol  view.  Figs.  17-19.  Nofhochryso  calilornica.  Fig.  17.  Male  abdomen,  lateral  view.  Fig.  18.  Gonorcus,  lateral 
view.  Fig.  19.  Some,  posterior  view,  and  hypandrium  Internum.  Fig.  20.  P.  intermedia,  setae  on  margin  of  right  ninth 
gonocoxite,  lateral  view.  Pig.  21.  P.  fusca,  female  abdomen,  with  spermotophore.  Figs.  22-26.  Female  eighth  sternum  or 
subgenitoi  plate.  Fig.  22.  N.  californica.  Fig.  23.  P.  grata.  Fig.  24.  P.  fusco,  anterior  view  below,  posteroventral  above. 
Fig.   25.   P.    intermedia,  onteroventrol   view.     Fig.    26.    P.   ni'gro,    onteroventral   view. 


236        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  4 


Figs.  27,  28.  P/machrysa  olbicosfa/es,  gonarcus,  lateral  and  ventral  views.  Figs.  29-31.  Kimochrysa  ahicana.  Fig.  29. 
Gonarcus,  ventral  view;  Fig.  30.  Same,  lateral  view.  Fig.  31.  Male  abdomen,  lateral  view.  Figs.  32,  33.  H.  nob///s.  Fig. 
32.  Male  abdomen,  lateral  view.  Fig.  33.  Gonarcus,  dorsal  view.  Figs.  34-36.  Prothorax,  ventral  view,  membranous  areas 
stippled.  Fig.  34.  P.  nigra.  Fig.  35.  N.  calilornica,  left  coxa  removed  to  show  sternal  apophysis.  Fig.  36.  Chrysopiella 
sp.    (Chrysopinae). 


Mesochrysinae  and  NoTHOCHRYsiNAE  •  Acluuis        237 


— ■'::q=^=^-4=^  /  V  /  1 1  / 1 T^ 


Fig.  37.  Mesypochryso,  hind  wing  (after  Martynov).  Figs.  38,  39.  Paleochrysa  ilricta,  fore  and  hind  wing  bases.  Fig.  40. 
Archaeochryio  paranervis  type,  wings.  Fig.  41.  A.  creedei,  wings  (from  type,  and  from  MCZ  specimens  No.  4464  and  No. 
4462). 


238        BiiUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  4 


Fig.  42.  Dyspefochfysa  veluscula,  type,  fore  wing  base;  arrows  indicate  position  of  basal  subcostal  crossvein,  and  point  of 
fusion  of  MPl  and  MA.  Fig.  43.  Nothochrysa  praechra,  fore  wing  base.  Fig.  44.  Synfhochryso  (Apochrysinoej,  fore  wing 
base,  showing  overlap  of  veins  at  Psnn,  probable  overlap  at  Psc  and  probable  course  of  MP2  (modified  from  Kimmins,  1952b). 
Fig.  45.  Gonzogo  |Chrysopinae),  fore  wing  base,  showing  overlap  of  veins  at  Psm,  and  course  of  MP2.  Fig.  46.  Archaeo- 
chryso  creedei,  tip  of  male  abdomen,   showing  forcipate  gonarcus  (from  MCZ  specimen  No.  4466).     Fig.  47.  A.  creedei  type. 


bulletin  OF  THE 

Museum   of 

Comparative 

Zoology 


-  ■-■  •-'-'•' 


Marine    Nematodes    of   the    East    Coast   of 
North    America.    I.    Florida 


WOLFGANG  WIESER  and  BRUCE  HOPPER 


HARVARD   UNIVERSITY  VOLUME    135,   NUMBER  5 

CAMBRIDGE,   MASSACHUSETTS,   U.S.A.  APRIL  24,    1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

Memoris  1864-1938 

JoHNSONLv,  Department  of  Mollusks,  1941- 

OccASioNAL  Papers  on  Mollusks,  1945- 

Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.   Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).   $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.    $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.    ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©   The    President   and    Fellows   of   Horvard    College    1967. 


MARINE  NEMATODES  OF  THE  EAST  COAST  OF  NORTH 
AMERICA.    I.    FLORIDA 

WOLFGANG  WIESER^  AND   BRUCE  HOPPER- 


CONTENTS 

Abstract   240 

Introduction  240 

Systematic  Section 242 

Family   Leptosomatidae   242 

Genus  Anticoina  242 

Oxystomatidae  246 

Halalaimits  246 

Tripyloididae  247 

Halanonchus    __ 248 

CijtoIaiDiium    249 

Bathylaimus  250 

Phanodermatidae  251 

Phanodcrmopsis  251 

Enoplidae  251 

Enoploides  .._ 251 

Mesacanthoides  254 

Oncholaimidae 255 

Oncholaimus    255 

Oncholainihim  255 

Metoncholaimus    256 

Prooncholdimus  258 

Viscosia    258 

Enchelidiidae    260 

Eiinjstomina    260 

Illiiim   261 

Pohjgastrophora 262 

Cyatholaimidae   263 

Pomponcma  264 

Longicyatholaim  us    265 

Xijzzors    266 

Paracanthonchus  267 

Paracyatliolaimus 268 

Halichoanolaimus    269 

Neotonchus    270 

Desmodoridae    270 

Spirinia    271 

S.    (Perspiria)    272 

Chromaspirina  272 

Metachronuidoid   273 


1  Zoologisches  Institut,  Universitat  Wien,  Wien, 
Austria. 

-  Nematology  Section,  Entomology  Research  In- 
stitute Research  Branch,  Canada  Department  of 
Agriculture,  Ottawa. 


Paradesmudora    276 

Dcsmodoia    277 

D.  ( Pseudochromadora )  277 

Monoposthia    277 

Monoposthioides   278 

Microlaimidae    279 

Pcirainicmhiimus    279 

Chromadoridae    279 

Hypodontulaimus    279 

Rhips    281 

Chromadora  281 

Tinimia    281 

SpilophoreUa   282 

ProchromadorcUa    282 

Chiomadorella  282 

Euchrotuadora     284 

Atrochromadora  287 

Comesomatidae   287 

Mesonchiuin    287 

Sahatieria  288 

A.xonolaim idae  289 

Axonolainitis    289 

OdontopJiora   290 

Purodontophora     290 

Leptolaimidae   292 

Ahiimella    292 

Camacolaimidae    __ 292 

Camacolaimits  292 

Diplopeltidae     292 

Parataivaia    292 

Didelta  293 

Linhomoeidae    __ 293 

Terschellingia    __ 293 

Monhysteridae     295 

PanimonJiystera  295 

Stcincria     295 

Thcristiis    _____ 296 

T.   (  Penzancia )   296 

T.  (  Daptonema )  298 

T.    (  Trichoiheristus )    _ _____ 298 

T.     (Cyliudrotheristii.s)    300 

Monhystcra    __ 302 

Scaptrella  ___. 303 

Xenolainius    303 

Acknowledgments  ___ 303 

References     Cited     304 

Table  1,  List  of  collecting  sites  in  Florida  __._  242 
Talkie  2,  List  of  species  reported  from  Florida  _    244 


Bull.  Mus.  Comp.  ZooL.  135(5):   239-344,  April,  1967        239 


240        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


ABSTRACT 

A  beginning  is  made  on  a  series  of 
papers  in  which  the  nematode  fanna  in- 
habiting the  east  coast  of  North  America 
from  Florida  to  Newfonndland  is  to  be  de- 
scribed. The  present  paper,  deahng  with 
90  species,  extends  the  number  of  species 
known  from  Florida  to  118.  One  new  genus, 
Paratarvaia  within  the  Diplopeltidae,  and 
48  new  species,  are  described,  viz.:  Hala- 
laimus  meyersi,  Bathyhimus  aithropappus, 
Enoploidcs  ])isulciis,  E.  ij^ryplius,  Mcsa- 
canthoidcs  fiJ)uJatiis,  M.  psittacus,  Mcton- 
cholaimus  intermedins,  M.  simplex,  M. 
scis.m.s,  Prooneholaimus  hosiatus,  Viscosia 
onchohimelloides,  lUium  lihidinosum,  Poly- 
il(i.stro})liora  eddx,  Pomponema  tesselotum, 
Loniiieyatholaimus  annae,  Xyzzois  inii,Iisi, 
Paracanthonehus  platypus,  Paracyatholai- 
muspesavis,  Spirinia  hamata,  Chromaspirina 
inaurita,  Metochwmadora  pulvinota,  M. 
meridiana,  Paiadesmodora  toreutes,  Mono- 
posfJiioides  mayri,  Paramiewhiimus  luna- 
tus,  Uypodontolaimus  intcrruptus,  Chwma- 
dorella  tiilix,  C.  vanmeterae,  Euchromadoni 
pectinata,  E.  meadi,  Atrochwmadom  dentic- 
idata,  Sabatieria  paradoxa,  S.  poracupida, 
AxomtJaimus  Iwxapihis,  Odontophoni  varia- 
bilis, Paratarvaia  seta,  Tersehellingia  Jongi- 
spieuhita,  T.  monohystcra,  Paramonhystera 
canicida,  Steineria  ampullaeea,  Thcristus 
calx,  T.  osientator,  T.  floridanus,  T.  e rectus, 
T.  galeatus,  T.  fistidatus,  T.  tortus,  and  T. 
.xyaliformis.  The  Ualanonchinae  n.  sub- 
fam.,  with  UaJanonchus  Cobb,  1920,  Rhab- 
docoma  Cobb,  1920,  and  Cytolaimium 
Cobb,  1920,  is  created  within  the  Tripyloidi- 
dae.  The  Halanonchinae,  via  Trefusia,  is 
considered  to  link  the  Tripyloididae  with 
the  Oxystomatidae. 

INTRODUCTION 

The  nematodes  from  the  coasts  of  the 
American  continents  are  poorly  known,  a 
fact  that  is  regrettable  since  this  group 
represents  such  an  important  component 
of  the  fauna  of  every  marine  habitat. 

To  improve  this  situation  we  decided  to 


collect  material  towards  a  monograph  on 
the  marine  nematodes  inhabiting  the  east 
coast  of  North  America.  In  order  not  to 
overburden  this  monograph  with  descrip- 
tions of  new  species  and  with  taxonomic 
discussions,  we  shall  publish  the  results  of 
our  investigation  in  a  numbers  of  papers 
each  dealing  with  a  portion  of  coastline 
between  Florida  and  Newfoundland.  In 
each  paper,  descriptions  and  figures  will  be 
given  for  all  the  species  found  except  for 
those  in  which  agreement  between  our 
specimens  and  representations  in  the  litera- 
ture were  considered  very  good.  In  addi- 
tion, a  list  containing  all  species  reported 
from  the  particular  region  will  be  included. 

In  the  monograph,  the  information  con- 
tained in  the  special  reports  will  be  con- 
densed. Short  clescriptions  and  figures  will 
be  given  of  all  species,  together  with  keys 
to  most  of  the  genera  and  families  of  ma- 
rine nematodes. 

A  portion  of  the  material  for  this  under- 
taking was  collected  separately  by  us  on 
excin\sions  to  the  eastern  seaboard  during 
the  period  1958-1961,  or  it  was  sent  to  us 
by  various  colleagues  whose  cooperation 
we  shall  acknowledge  in  the  relevant  sec- 
tions of  this  series.  The  bulk  of  the  ma- 
terial, however,  was  collected  by  us  be- 
tween May  7  and  June  8,  1963,  on  a  trip 
in  which  we  covered  the  area  from  south- 
ern Florida  to  Maine. 

The  following  remarks  should  suffice  to 
clarify  our  approach  to  the  whole  project: 

Habitats. — At  each  location  we  tried  to 
cover  the  important  types  of  habitat  that 
could  be  reached  by  manual  sampling. 
Shallow  water  dredge  or  core  sampling 
was  carried  out  in  Miami,  Florida,  Beau- 
fort, North  Carolina,  Lewes,  Delaware,  and 
Woods  Hole,  Massachusetts.  In  addition 
we  received  shallow  or  deep  water  samples 
from  various  sources  which  we  shall 
acknowledge  later. 

Methods. — In  general,  nematodes  were 
extracted  from  the  substrate  utilizing  a 
combination  of  the  sieving  and  decanting 
processes.    Live  nematodes  were  relaxed  in 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        241 


an  oven  maintained  at  54 "C  prior  to  fixa- 
tion in  27c  formaldehyde.  As  the  number 
of  specimens  recovered  in  most  instances 
was  considerable,  only  a  representative 
fraction  of  the  total  was  picked  out  and 
processed  into  dehxdrated  glycerine.  The 
remainder  of  the  material  was  also  pre- 
served and  is  being  kept  for  reference  in 
the  Nematology  Section,  Entomology  Re- 
search Institute,  at  Ottawa. 

The  dehydration  process  used  was  an 
adaptation  of  Seinhorst's  methyl  alcohol 
method  (Seinhorst,  1959).  The  preserved 
specimens  were  placed  in  a  B.P.I,  watch- 
glass  containing  107f  glycerine  in  methyl 
alcohol.  The  watchglass  was  then  placed 
in  an  oven  maintained  at  54°C  to  hasten 
the  evaporation  of  the  methyl  alcohol,  a 
process  which  took  less  than  30  minutes. 
The  watchglass  containing  the  specimens 
was  then  transferred  to  a  desiccator  for 
24-48  hours  to  insure  complete  dehydration 
of  the  glycerine  prior  to  the  construction  of 
slides.  Nearly  all  marine  nematode  species 
do  well  when  subjected  to  this  dehydration 
process.  Sporadic  distortion  of  varying  in- 
tensitv  results  with  certain  members  of  the 
Desmodoridae  and  a  few  isolated  genera 
of  other  groups. 

Systcmotics. — When  we  write  the  planned 
monograph  we  hope  to  be  able  to  arrange 
the  species  and  genera  of  marine  nematodes 
into  a  more  satisfactorv  classification  than 
has  been  in  use  so  far.  For  the  special  re- 
ports, however,  we  shall  adhere  to  the  old 
Filipjev-Micoletzky  system  as  used  by 
Schuurmans-Stekhoven  (1935),  with  a  few 
improvements  suggested  by  Chitwood,  in 
Chitwood  and  Chitwood  (1950),  and  Chit- 
wood  ( 1951 ) .  We  shall  proceed  only  from 
the  family  downward,  ignoring  the  position 
and  status  of  higher  systematic  categories. 

Descriptions. — W^e  feel  that  too  much 
\^'eight  has  been  attached  in  the  past  to  the 
use  of  formulas  in  the  descriptions  of  nema- 
tode species.  Purely  relative  formulas  like 
those  suggested  by  de  Man  or  by  Cobb  are 
of  little  value.  Formulas  in  which  absolute 
dimensions   are  used  suggest   a   degree  of 


precision  in  the  construction  of  nematode 
species,  which  is  fairly  unrealistic.  We 
shall,  therefore,  in  our  descriptions  give  as 
many  absolute  measurements  as  possible 
of  body  and  organ  dimensions  but  refrain 
from  assembling  them  into  formulas.  In- 
stead, we  would  like  to  stress  the  need  for 
giving  as  precise  a  representation  as  pos- 
sible of  the  morphology  of  the  animal,  es- 
pecially of  the  stmctiu-e  of  the  male  genital 
apparatus. 

Various  authors  (i.e.,  Wieser,  1955,  Chit- 
wood, 1960,  Inglis,  1962)  have  drawn  at- 
tention to  the  necessity  of  the  presence  of 
a  male  specimen  of  a  species  as  a  prerequi- 
site to  a  valid  description  of  the  species. 
Inglis  (1962)  goes  as  far  as  to  state,  ".  .  . 
I  will  not,  in  general,  accept  identifica- 
tions based  solely  on  females  or  larvae  and 
I  will  insist  on  treating  almost  all  species 
based  originally  on  females  or  larvae  alone 
as  species  chi])iae."  We  are  in  complete 
sympathy  with  this  view  so  far  as  the  "lar- 
vae" are  concerned.  With  regard  to  the 
female,  however,  we  must  support  the  con- 
cept with  certain  limitations.  If  such  a  rule 
were  to  be  followed  with  no  exceptions, 
some  genera  would  receive  little  or  no  at- 
tention, e.g.,  Illiiim  in  this  paper.  Males 
are  scarce  or  unknown  in  certain  groups, 
particularly  the  Plectoidea,  and  the  ad- 
vancement of  our  knowledge  of  these 
groups  could  be  curtailed  if  Inglis'  sug- 
gestion were  adhered  to  without  reserva- 
tion. Therefore,  we  feel  that  the  action 
suggested  by  Inglis  might  best  be  limited 
to  those  genera  in  which  males  are  com- 
monly known  to  occur. 

In  additi(m  to  species  descriptions  and 
figures,  general  discussions  and  keys  will 
be  supplied  wherever  necessary. 

Florida  collection. — The  Florida  coast- 
line is  one  of  the  least  known  with  respect 
to  the  nematode  fauna  of  all  the  regions 
investigated.  Only  Cobb  (1920,  1922), 
Chitwood  (1951,'  1956),  and  Hopper 
(1961a,  1966)  have  reported  a  few  species, 
mainlv  from   Miami   and   Kev  \\est.    Our 


242         Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


Table   1.      Lisi    of  Collecting  Sites  in  Florida 


Sample 

No. 


Date 
1963 


Kt.'\-  Biscayne,  Bear  Cut  area;  about  low  water  level;  seaweeds  scraped  off  mangrove 

roots.    Canadian  National  Collection  of  Nematodes  No.  4066. 

Key  Biscayne,   Bear  Cut  area;   shallow  water   (about  20   cm  at  low   tide),   close  to 

submerged  patch  of  ThaJa.ss'm  and  Syrini^odium;  fine  sand  and  debris.    C.N.C.  of  N. 

No.  4067. 

Key  Biscayne,  Bear  Cut  area;  flat  around  high-water  level,  with  Uca;  fine  to  medium 

sand,  debris.    C.N.C.  of  N.  No.  4068. 

Off  Rickcnbacker  Causeway;  very  soft  mud,  stagnant  water  at  low    tide.    C.N.C.   of 

N.  No.  4069. 

Virginia    Key;    beach    on    grounds    of    Marine    Institute;    clean    sand,    mid-tide    level. 

C.N.C.  of  N.  No.  4070. 

Everglades    National   Park,    Florida    Bav   near    Flamingo;    clavish    mud    with    a    little 

sand;  below  low-tide  level.    C.N.C.  of  N.  No.  4071. 

Everglades  National  Park,   Florida   Bay  near  Flamingo;   upper  part  of  shore;   sandy 

mud.    C.N.C.  of  N.  No.  4072. 

Biscayne  Bay,  dredgings  in  Tliala.ssia  beds,  about  four  meters  deep;  sand,  shells  and 

mud;  several  subsamples.    C.N.C.  of  N.  No.  4073. 

Lauderdale-by-the-Sea;    verv    exposed    beach    around    mid-tide    level;    clean,    coarse 

sand,  shells.    C.N.C.  of  N.  No.  4074. 

Vero  Beach.     Sheltered,  muddy  sand,  with   Uca   and   mangroves   nearby.     C.N.C.    of 

N.   No.  4075. 


M-1 

Nhiy 

16 

M-2 

Id. 

M-3 

Id. 

M-4 

.May 

17 

\[-5 

Id. 

M-fi 

May 

18 

M-7 

Id. 

Nr-8 

May 

20 

L 

May 

21 

V 

Id. 

collection    increases    the    total    of    species 
known  to  118. 

Most  of  onr  collecting  was  carried  out  in 
Biscayne  Bay,  in  the  vicinity  of  the  Insti- 
tute of  Marine  Science  of  the  University  of 
Miami,  Virginia  Key,  Miami;  additional 
collecting  sites  were  in  the  Everglades, 
near  Lauderdale-by-the-Sea,  and  near  Vero 
Beach.  A  list  of  the  collecting  sites  is  given 
in  Table  1,  and  a  list  of  all  the  species 
found,  together  with  the  species  reported 
in  the  literature,  in  Table  2. 

SYSTEMATIC   SECTION' 
LEPTOSOMATIDAE 
ANTICOMA  Bastion,  1865 

Type    species.— Anf/coma    eberfhi    Bastion, 
1865:  141,  pi.  11,  figs.  143-145. 

In  this  genus  it  appears  possible  to  dis- 
tinguish two  groups  of  species  character- 
ized by  the  position  of  the  excretory  pore 


^  Abbreviations  used  in  this  paper  are  as  fol- 
lows: a.b.d.,  anal  body  diameter:  c.b.d.,  corre- 
sponding l)od\-  dianictcr:  L,  length;  \'u,  \'ul\a: 
w,  widtli. 


and  the  length  of  the  terminal  excretory 
duct.  The  correlation  between  these  two 
characters  is  fairly  well  established,  al- 
though, due  to  a  few  insufficient  descrip- 
tions, there  remains  some  doubt  as  to  its 
general  apjDlicability.  The  two  groups  (A 
and  B )  coincide,  respectively,  with  group 
A  and  groups  B  +  C  in  Wieser  (1953). 
Filipjev  ( 1927 )  had  previously  arranged 
the  species  of  this  genus  in  a  similar  man- 
ner. 

Characteristics  of  Species  of 
Anticoma  Group  A 

Excretory  pore  situated  between  poste- 
rior cervical  setae  and  nerve  ring;  terminal 
excretory  duct  short  ( never  longer  than  the 
width  of  the  excretory  ampulla ) . 

Key  to  Species  of  Anticoma  Croup  A 

1.  Gubernaculum  absent  -..  A.  daJiIi  Wieser,  19.53 

A.  major  .\hrwson,  1956 
Gubernaculum    present    2 

2.  Excretory     pore     a     short     distance     behind 

cervical     setae     —  3 

E.xcretory  pore  about  halfway  between  cervi- 
cal setae  and  nerve  ring 4 


Florida  Marine  Nematodes  •   Wicscr  and  Hopper        243 


Excretory  pore  a  short  distance   in  front  of 
nerve  ring  6 

3.  Tail  short  A.  camphcUi  Allgen,  1932 

Tail  long,  filiform  ._  A.  filicaiida  Mawson,  1956 

4.  Longer  cephalic  setae  24-25  ij.  =  about  one 

head   diameter  long  

— A.   eberthi  Bastian,  1865 

Cephalic  setae  less  than  10  ^  =  one-half  to 
two-thirds    of   head    diameter    long    5 

5.  Cuticularized   portion   of  spicula    ( excluding 

velum )   more  or  less  parallel   

A.  lata  Cobb,  1898 

Cuticularized     portion     of    spicula     strongly- 
dilated   near  proximal   end  

A.  cohimha  Wieser,  1953 

Syn.  A.  aiistrulls  Mawson,  1956 

6.  Supplement    opening    posterior    to    proximal 

end  of  spicula  ___  ,    A.  Utoris  Chitwood,  1936 
Supplement    opening    distinctly    anterior    to 
proximal   end   of  spicula   7 

7.  Amphids  one-third  to  one-fourth  of  cephalic 

diameter  wide  A.   trichina  Cobb,  1891 

Amphids  one-sixth  of  cephalic  diameter  wide  8 

8.  Tail   short    (4-5   a.b.d.)    

A.  insulaealbae  Filipjev,  1927 

Tail  long   (8-10  a.b.d.)    

A.   tenuicaudata  Filipjev,  1946 

Others.  A.  typica  Cobb,  1891  is  related 
to  the  above  three  species  but  too  incom- 
pletely known  for  its  position  to  be  deter- 
mined with  certitude. 

A.  kergiielensi^  Mawson,  195Sb  is  closely 
related  to  A.  lata,  perhaps  even  identical. 

Characteristics  of  Species  of  Anticoma 
Group  B 

Excretory  pore  situated  on  a  level  with 
or  in  front  of  cervical  setae;  terminal  excre- 
tory duct  long  ( at  least  as  long  as  the  width 
of  the  excretory  ampulla). 

The  species  within  this  group  will  not  be 
differentiated  in  this  work.  However,  for 
the  reader's  convenience,  a  list  of  the  recog- 
nizable species,  including  the  most  recent 
synonymy,  is  given: 

A.  acuminata  (Eberth,  1863)  (Syn.: 
Odontobius  acuminatus  Eberth,  1863; 
Stenolaimus  lepturus  Marion,  1870;  Anti- 
coma  JimaJis  Bastian,  1865  pt.;  A.  tijrrhenica 
de  Man,  1878;  A.  caheti  de  Rouville,  1903; 
A.  pontica  Filipjev,  1918;  A.  zosterae 
Schulz,  1932;  A.  similis  Cobb,  1898  [see 
Gerlach,    1962];    and    A.    profunda    Mico- 


letzky,  1930  [see  Gerlach,  1962]);  A.  pel- 
hicida  Bastian,  1865  (Syn.  A.  hmilis  Bastian, 
1865  pt.);  A.  subsimilis  Cobb,  1914  (see 
Mawson,  1958);  A.  arctica  Steiner,  1916 
(Syn.  A.  procera  Micoletzky,  1930  [see 
Gerlach,  1962]);  A.  minor  Filipjev,  1927; 
A.  murmanica  Filipjev,  1927;  A.  extcnsa 
Wieser,  1953;  A.  stekhoveni  Wieser,  1953 
(Syn.  A.  acuminata  of  Schuurmans-Stek- 
hoven,  1950,  nee  Eberth,  1863);  A.  iciescri 
Mawson,  1958  (Syn.  A.  stekhoveni  Maw- 
son, 1956,  nee  Wieser,  1953). 

AnVicoma  lata  Cobb,  1898 
Plate  I,  fig.  1,  a-e 

Anticoma  lata  Cobb,   1898:   384,  385. 

Anticoma    ditlevseni    Micoletzkv,    1930:    255-258, 

fig.  2. 

L  =  2.04-2.06  mm;  w  =  58  /x;  esophagus 
390-412  ii;  tail  215-220  ^  (  9  ,  6.5  a.b.d., 
6,  5.5  a.b.d.).  Cephahc  setae  4-4.5  +  3- 
3.5  /x.  Buccal  cavity  conical.  Amphids  3-4 
p.  wide,  11  JUL  behind  anterior  end.  Cervical 
setae  33  fx  behind  anterior  end.  Excretory 
pore  115  /J.  from  anterior  end.  Spinneret 
delicate  and  pointed,  directed  dorsally 
(always?).  Spicula  61-70  ^u,  dorsal  and 
ventral  contours  nearly  parallel,  \'elum 
present.  Gubernaculum  21-24  fi.  Supple- 
ment 56-70  IX  in  front  of  anus. 

Representation  in  samples  studied. — M- 
1,  Key  Biscayne. 

Gcoiiraph  ical  distribution. — Australia 
(Cobb,  1898),  Sunda  Islands  (Micoletzky, 
1930),  Maldives  (Gerlach,  1962),  Red  Sea 
( Gerlach,  1958c ) . 

Remarks. — The  spicula  are  somewhat 
longer  than  reported  by  Micoletzky  and 
by  Gerlach  ( 61-70  /j.  as  against  40-47  p. ) , 
but  our  specimens  seem  to  agree  in  all  other 
points. 

Anticoma  frichura  Cobb,  1898 
Plate  I,  fig.  2,  a-d 

Anticoma  trichura  C()l)b,   1898:   385,  386. 

L  =  3.01-3.04  mm;  w  =  49-52  ^;  esopha- 
gus 390-445;  tail  in  9  ,  530-565  /x  ( 18  a.b. 
d.),  in  i,  460-480  p.  (14  a.b.d.).  Cephalic 
diameter  18-19  /x.   Cephalic  setae  15  +  8  /x. 


244        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


Table  2.     List  of  Free-lring  Nematodes  Reported  from  the  Coasts  of  Florida   (Numbers  of 

Specimens  found  are  given  in  parentheses)^ 


Leptosomatidae 

Auticoma  lata  Cobb,   189S 

A.  trichura  Cobb,  1898 
Oxystomatidae 

Ualalaimus  cf.  fleichcri  Mawson,   1958 
//.   mcyersi  n.  sp. 

H.  pachydermotus  (Cobb,  1920)  Syn.  Tijcnodora  ]>. 
Poiocoina  strkita   Cobb,    1920 
Litiniuin  acqitalc  Cobb,   1920 
Tripyloididae 

Hcihinonrlius  niacruni.s-  Cobb,   1920 
Ciitolaimium  exile  Colli),   1920 
Bathylaimus  australis  Cobb,   1893 

B.  arthropappus  n.  sp. 
Phanodennatidae 

Thanodermopsis  longisetae  Chitwood,   1936 
Enoplidae 

Enoplnide.s  l)i.sulciis  ii.  sp. 
E.  gryphus  n.  sp. 
Mesaeanthoides  fihulattis  n.   sp. 
M.  psittacus  n.  sp. 
Ironidae 

Tri.ssoiirhulus  oceami.s  Cobb,   1920 
Oncholaimidat' 

Anoplostoma  heterunmi    (Cobb,   1914) 

Syn.   Oncholaimelhi^s-  h. 
A.  vwiparum    (  Bastian,   1865 ) 

Syn.  Symplocostoma  v. 
Oiiehohiiniiis  dujdrdinii  de  Man,   1878 
Onchokiiiiiiuin  iippeadiculatiim  Cobb,   1930 
O.  domesticuni    (Chitwood  &  Chitwood,   1938) 
Metoiuholaimiis  inienuedius  n.  sp. 
M.  s-hnplex  n.  sp. 
M.  .S'cm-u.y  n.  sp. 
Pwourholaimus-  hastaius  n.  sp. 
Vlscosia  onchohiimelhndes  n.  sp. 
\".  papillata  Chitwood,   1951 
V''.   macramphida  Chitwood,    1951 
Enchclidiidac 

Eurysiomina  mintitisctdae  Chitwood,   1951 
lUium  exile  Cobb,   1920 
/.   lihidiitosiim  n.  sp. 
Pnlygasirophora  edax  n.  sp. 
Calyptronema  eohhi  W'iescr,    1953 

Syn.   Caldhdiuu-s  aeuminatus  Cobb,    1920 
Cyatholaimidac 

Ponipoiieiiia  tcs.selatum  n.  .sp. 
Longieyathohiiiinis  annae  n.  sp. 
Xyzzors  inglisi  n.  sp. 
Paracanthnnchus  ]il(itypu.s  ii.  sp. 
P.   trtincatus   (Cobb,    1914) 

Syn.  Cyatholavmus  t. 
Pararydtholdirnus  pesdri.s  n.   sp. 
II(di(li()diii)lainiu.s  cpidlhiordecittipapilLittis 

Chitwood,  1951 
H.  duodecimpapillatiis  Timm,   1952 
Neotonchiis  liitosus  Wiescr  and   Hopper,    1966 
Syiiouehiinn   ohtu.sum   Cobb,   1920 


M-l(6) 

M-2(5) 

M-2 

M-2(3) 

Cobb,   1920,  Key  West 
Cobb,   1920,   Biscayne  Bay 
Cobb,   1920,  Miami 

M-8(2),  Cobb,   1920,   Biscavne  Bay 
M-2(l),  Cobb,   1920,  Biscayne  Bay 

V(6) 
M-5(9) 

V 

M-2  (13) 
M-5(6) 
M-3(4) 
L(3) 

Cobb,  1920,  Miami 


Chitwood,   1951,  Ocala 
Hopper,   1961a,  Panama  City 

M-l(24) 

M-l(l),  M-2(l) 

M-l(2) 

M-5(23) 

M-5 

M-5 

M-8(1S) 

M-3(4),  M-8(16) 

M-7(5),  M-8(l),  V(6) 

M-2(S),  M-8(23) 

M-l(4),  M-7(l) 
M-8(3) 
M-8(  1 ) 
M-8(6),  V(3) 

Cobb,  1920,  Biscayne  Bay 

M-3(l) 

M-3(13),  M-8(l) 

V(5) 

M-8(5) 

Chitwood,   1951,   Silver  Springs 

M-6(3) 

M-2(3),  M-6(7),  M-8(l) 

V 

M-4(2) 

Cobb,   1920,  Miami 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        245 


Desmodoridae 

Spirinia  parasitifera   ( Bastian,   1865) 

S.  hamata  n.  sp. 

CJtronui.spirina  inaurita  n.  sp. 

Metachromadora  piilvinata  n.  sp. 

yf.  onijxoides  Cliitwood,   19.36 

M.   nicridUina  n.  sp. 

Paradesnwdora  toreutes  n.   sp. 

Desmodora  qiiadripapiUata   { Daday,   1899) 
Syn.  Fseiidocdiromadora  (j. 

Xennella  cephaluta  Cobb,   1920 

LcptoncmcUa  cincta  Cobb,   1920 

Monopu-sthia  mirahilis  Schidz,   1932 

Monoposthioides  maijri  n.  sp. 
Miciolaimidae 

ParamicroJaunus  lunatiis  n.  sp. 
Chromadoridae 

Hypodontolaimus  interruptus  n.  sp. 

H.  pandispicuhtiis  Hopper,   1961 

Rhips  ornata  Cobb,  1920 

Actinonemo  pachijdermata  Cobb,   1920 

Chromadora  nuiciolaimoides  Steiner,   1915 

Timmio  parva   (Timm,   1952) 
Syn.   Parachromadora  p. 

Spilophorclh  paradoxa   (de  Man,   1888) 

ProchromadorcUa  mcditerranca   ( Micoletzky,  1922) 

Chwmadorella  filifonnis   (Bastian,   1865) 

C.  trUix  n.  sp. 

C   vanmeterac  n.  sp. 

Euchromadora  gaulica  Inglis,   1962 

E.  pectinata  n.  sp. 

E.  mcadi  n.  sp. 

Atrocliioinadora  denticulata  n.  sp. 
Comesomatidae 

Mesonchiitm  peUiicidum    (Cobl\   1920) 
Syn.  Pepsonema  p. 

M.  poriferum  Cobb,  1920 

Sahaticria  paradoxa  n.  sp. 

S.   paracupida  n.   sp. 

Laimella  longicauda  Cobb,  1920 

Nanunlaimus  giittatu.s  Cobb,   1920 
Axonolaimidae 

Axonolaimus  hexapihis  n.  sp. 

Odontophom  variabilis  n.  sp. 

Parodontophora  hrevamphida    (Timm,   1952) 

Araeolaimus  punctatus   (Cobb,   1920) 
Syn.  Coinonema  p. 
Leptolaimidae 

Alainiclla  cincta  Cobl?,   1920 

Cyniira  iiniformis  Cobb,   1920 

Haliplectus  floridanus  Cobb  in  Cliitwood,   1956 

H.  hickncri  Chihvood,   1956 
Camacolaimidae 

Camacolainnis  prijtherchi  Chitwood,   1933 

Onchium  oceUatum  Cobb,   1920 

O.  metocellatum  Wieser,  1956 

Syn.  Onchidella  ocellata  Cobb,   1920 

Neiirelki  simplex  Cobb,  1920 


M-4(10),  V(.55) 
M-3(14) 

M-3(4) 

V(2) 

M-5(58) 

M-2(l),  V(103) 

M-3(7) 

M-2(l),  M-3(l) 
Cobb,   1920,  Key  West 
Cobb,   1920,  Miami 
M-2 
M-3(9) 

M-2(8) 

M-l(3),  M-3(2),  M-5(17) 

M-3(25),  M-7(2) 

M-5(29),  Cobb,   1920,  Miami,  Key  West 

Cobb,   1920,  Key  West 

M-l(.30),   M-4(9),  M-5(l),  M-7(l),   M-8(3; 


M-l( 

M-l( 

M- 

M-l( 
M-l( 
M-2( 

M-2( 
M-l( 
M-8( 
M-2( 
M-l( 


12) 

3), 

8(6 

13) 

19) 

31) 

14) 

.38) 

8) 

3), 

1), 


,  V(2) 
M-2(12) 


M-3(l),  M-6(5),  M-7(2), 


M-8(l) 
M-2(2) 


M-S(4) 

Cobb,   1920,  Key  West 

M-7(2),  V 

V 

Cobli,   1920,  Biscayne  Bay,  Key  West 

Cobb,   1920,  Biscayne  Bay 

M-5(3) 

V(ll) 

V(5) 

Cobb,   1920,  Biscayne  Bay,  Key  West 

M-2(l),  Cobb,   1920,  Biscayne  Bay 

Cobb,  1920,  Miami 

Chit\\ood,   1956,  Long  Key 

Clritwood,   1956,   Atwood  Gro\e,  Ellenton 

V(3) 

Cobb,   1920,   Key  West 

Cobb,   1920,  Biscayne  Bay 
Cobb,   1920,  Key  West 


246 


Bulletin  Miiscuni  of  Comparaiive  Zoology,  Vol.  135,  No.  5 


loncnui  cohhi   ( Steiner,   1916) 

Syn.  loiicDui  ocrllatiiiii  Colili,   1920 
Nemclla  uccllatu  Cobb,   1920 
Diplopeltidae 

Pdnitarvaia  seta  n.   gen.,  n.   sp. 

Didclta   maculatinn   Cob):),   1920 
Linhomoeidae 

Terschclliii^id  lon<iicaud(it(i  dc  NFan,   1907 

r.  lon^ispiculata  ii.  sp. 

T.  iijonohijstera  n.  sp. 

Aiifici/dtlnis  tcntiicandatiis  Colili.    1920 

lluUuciua  spinosuin   Cobb,   1920 

Pandinhomoeus  fuscacephalum   (Colil),   1920  i 
Syn.   Crysttdloneina  f. 

P.  .simdc   (Cobb,   1920) 
Syn.  Cry.stallonema  s. 

Linhomocihi  rxilis  Cobb.   1920 
Monhysteridae 

Paramonhy.stera  canicula  n.  sp. 

Stcincrifi  ampidlacca  n.  sp. 

Thoi.stus  tiictaflcvcn.sis  Gerlach,   1955 

T.  calx  n.  sp. 

T.  poliichactophilus  Hopper,   1966 

T.  ostcntator  n.  s\}. 

T.  floridaniis  n.  sp. 

T.  erectiis  n.  sp. 

T.  galcatus  n.  sp. 

T.  oxyuroides  ( Schuurmans-Stekhoven,   1931' 

T.  fistidatus  n.  sp. 

T.  tortus  n.  sp. 

T.  xyaliformi.s  n.  sp. 

Monhystcra   parva    (Bastian,    1(S65) 

SvaptrcUa  c'lucta   Cobl),    1917 

Xenohiiniii.s  .stiiatus  Col)b,   1920 
Desmoscolecidae 

GreefficUa  dasyura  Colib,   1922 


Cobb,   1920,  Biscayne  Bay 


M-2 

M-8(i: 


Cobb,   1920,  Key  West 


M-4(l),  M-6(13),  M-7(89),  M-S(l 

M-7(24),  M-8(l) 

V(12) 


V(9: 


Colib,   1920.   Miami 

Cobb,   1920,  Biscayne  Bay 

Colili,   1920,  Miami 

Cobb,  1920,  Miami 

Cobb,   1920.   Biscayne  Bay 

M-l(l),  M-3(6) 

V(l) 

M-4(ll) 

M-4(3),  V(33) 

Hopper,   1966,  Virginia  Key 

M-3(5S) 

M-2(5) 

M-2(3),  M-3(S3),  M-4(l), 

V(2) 

M-5(3) 

M-2(l),   M-4(19),   M-5(l), 

M-S(l 

M-3(2) 

M-2(5) 

M-2(3) 

M-l(7),  M-7(l) 

M-2(2) 

M-2(9),  Cobb,   1920,  Bisca) 

/ne   Bay 

Cobb,   1922.  Biscayne  Bay 


1  The  niiinber  of  specimens  indicated  in  this  table  represents  only  those  that  were  present  in  the  fraction  of  the  total 
sample  examined.  In  this  manner  the  relative  abundance  of  each  species  is  somewhat  suggested.  In  some  cases,  where 
a  species  was  represented  by  only  a  few  specimens,  an  effort  was  made  to  locate  additional  material  from  the  preserved 
remainder  of  the  sample.  These  additional  specimens  are  not  recorded  in  this  table.  Species  for  which  the  number  of 
specimens  is  not  indicated  were  not  iiresent  in  the  original  fraction  examined.  These  were  subsequently  recovered  from 
the   remainder   of  the   sample. 


Amphids  5  /x  wide  =  one-third  of  body 
diameter.  Buccal  cavity  conical.  Cervical 
setae  35  /x  behind  anterior  end.  Excretory 
pore  170  p.  behind  anterior  end,  i.e.,  85% 
of  distance  anterior  end  to  nerve  ring. 
Spicula  59-63  ^,  proximally  bent,  dorsal  and 
ventral  contours  nearly  parallel,  with  slight 
hump  near  proximal  end.  Velum  present, 
smooth  (not  striated  as  figured  by  Gerlach, 
1962).  Cubernaculum  about  20 /x.  Supple- 
ment 55-60  ji  in  front  of  anus. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne. 

Geoii^raphical  distribution. — Australia 
(Cobb,  1898),  Sunda  Islands  (Micoletzky, 


1930),  Maldives  (Gerlach,  1962),  Antarc- 
tic, Subantarctic  (Mawson,  1958b,  Allgen, 
1959). 

OXYSTOMATIDAE 
HALALAIMUS  de  Man,  1888 

Type  species.— Ho/a/a/mus  gracilis  de  Man, 
1888:    3,  4,  pi.  1,  fig.  1. 

A  key  to  the  species  of  this  genus,  ex- 
cluding the  subgenus  PacJu/odora,  was  given 
by  Mawson  ( 1958b ) . 

There  are  three  species  which  are  set 
apart  from  the  rest  by  the  occurrence  of  a 
distinct  circle  of  labial  setae,  i.e.,  H.  papil- 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        247 


lifer  Gerlach,  1956,  //.  fletcheri  Mawson, 
1958,  and  H.  filicoUis  Timm,  1961.  In  the 
first  species  the  tail  is  rounded  at  the  tip, 
in  the  latter  two  the  tail  is  filiform  and  its 
tip  is  bifid.  The  Miami  material  contained 
one  species  \\'hich  seems  to  correspond  in 
all  essential  features  with  H.  fletcheri,  al- 
though the  indistinctness  of  the  amphids 
and  the  somewhat  sketchy  figures  of  the 
type  leave  room  for  doubt. 

The  second  species  in  our  material  is 
closely  related  to  H.  supercirrhatiis  Gerlach, 
1955,  and  //.  longiseto^us  Hopper,  1963,  but 
is  distinguished  by  the  much  more  elon- 
gated, filiform  tail,  the  spicula  which  have 
a  ventral  hump,  and  the  gubernaculum  with 
its  lateral  guiding  pieces.  Examples  of  H. 
longisefosus  have  been  recovered  from  the 
Charleston,  South  Carolina,  samples,  and 
its  status  will  be  clarified  in  the  paper  deal- 
ing with  the  specimens  collected  from  that 
area. 

Halalaimus  (H.)  cf.  fletcheri  Mawson,  1958 

Plate  II,  fig.  3,  a-c 

Halalaimus  (H.)  cf.  fletclwri  Mawson,  19.58:    .332. 
fig.  13,  a,  b. 

L  =  2.6-2.37  mm;  w  =  26-35  ju.;  tail  in  <^ , 
310  ^  (18  a.b.d.),  in  9  275  ,x.  Head  at  level 
of  first  cephalic  setae  about  4  /x  wide  and 
6  ^  high.  Six  labial  setae,  2  /x  long.  Cephalic 
setae  6  /x  long,  arranged  in  two  circles  ( 6  + 
4)  about  4  fx  apart.  Amphids  beginning 
about  10  /,<,  from  anterior  end,  indistinct, 
particularly  the  posterior  end  which  appears 
to  run  into  lateral  alae.  Its  length  in  one 
specimen  is  probably  17  p..  Anterior  portion 
of  neck  exceedingly  drawn  out,  narrow  for 
about  two-thirds  its  length.  Tail  with  bifid 
tip,  each  prong  10  /j.  long.  Spicula  27  jx 
long,  with  velum.  Gubernaculum  strongly 
developed. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

Geograpliical  distribution . — Macquarie 
Isl,  Kerguelen  Isl.   (Mawson,  1958b). 

Halalaimus  (H.)  meyersi  new  species 
Plate  II,  fig.  4,  a-c 

L  =  2.26-2.49  mm;  w  =  17  /x;  Vu  =  47%; 


tail  in  i  ,  540  /x.  Head  about  4.5  p.  wide,  7 
/x  high.  Cephalic  setae  20  /x  long,  arranged 
in  two  circles  ( 6  -f  4 ) .  Cuticle  finely 
striated.  Amphids  38-40  /x  long,  beginning 
20-30  fx  from  anterior  end.  Excretory  pore 
50  p.  from  anterior  end,  ampulla  at  base  of 
amphids.  Terminal  excretory  duct  8-10  /x 
long.  Narrowed  portion  of  neck  about  one- 
fourth  its  total  length.  Spicula  21  /x  long, 
with  ventral  swelling  at  the  end  of  its 
proximal  third.  Gubernaculum  with  lateral 
guiding  pieces.  Tail  very  thin,  elongated, 
with  narrow  tip. 

Holotypespecimen.~Mc\\e;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa  Collection  Num- 
ber 4067,  Type  slide  No.  60.  Type  locality, 
M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne. 

Remarks. — This  species  is  named  in 
honor  of  Dr.  Samuel  P.  Meyers,  our  Miami 
host. 

TRIPYLOIDIDAE 
HALANONCHINAE  new  subfamily 

Type     genus.— Ho/anoncfius     Cobb,     1920: 
266. 

Diagnosis. — Tripyloididae,  with  three 
large,  well-separated  lips,  six  labial  papillae, 
and  10  cephalic  setae  in  two  circles,  the 
anterior  circle  consisting  of  six  jointed, 
mostly  elongated  setae;  large,  unarmed 
conical  or  cylindrical  buccal  cavity  without 
partitions;  spiral  or  tubular  amphids;  spicula 
short,  bent;  gubernaculum  plate-shaped, 
simple,  without  apophysis;  pre-  and  some- 
times postanal  supplements  present  in 
males. 

Discussion. — In  Cobb's  paper  of  1920 
one  finds  the  description  of  three  closely 
related  genera,  the  systematic  position  of 
which  has  been  doubtful  ever  since.  These 
genera  are:  1)  Cijtolaimium  [with  the 
species  C.  exile  Cobb,  1920,  and  C.  obtusi- 
caudatum  Chitwood,  1936,  the  latter,  ac- 
cording to  Gerlach  ( 1962 ) ,  being  a  synonym 
of  the  former],  2)   Rliabdocoma  [with  the 


248         Bulletin  Museum  of  Coiiii>ar(itivc  Zoology,  Vol.  135,  No.  5 


species  R.  americana  Cobb,  1920,  R.  artic- 
iilata  Gerlach,  1955,  R.  brevicauda  Schuur- 
mans-Stekho\'en,  1950,  R.  cyJindricaiida 
Schiuirnians-Stekhoxen,  1950,  R.  macnira 
Cobl),  1920],  and  3)  Hcdanonchiis  (syn. 
Latilai))iii.s  Allgen,  1933)  [with  the  species 
//.  macramphidus  Chitwood,  1936,  and  H. 
macrurus  Cobb,  1920]. 

All  tliree  genera  are  characterized  by  a 
smooth  cuticle,  an  arrangement  of  cephalic 
sense  organs  in  three  circles  of  6  +  6  +  4, 
the  second  circle  consisting  of  the  longest 
and  distinctly  jointed  setae,  three  large, 
deeply  cut  lips,  a  buccal  cavity  which  is 
either  conical  or  cylindrical,  amphids  which 
are  either  spiral  with  a  single  turn  and 
a  posterior  break  (Cytolaimiiim,  RJuiJydo- 
coma),  or  more  tubular  or  pocket-shaped 
with  a  circular  opening  (Halanonchus), 
simple  spicula,  plate-shaped,  simple  guber- 
nacula,  and  the  occurrence  in  males  of 
many  supplements,  not  only  preanally  but 
also  postanally  and  even  in  the  cervical 
region. 

The  number  (though  not  the  arrange- 
ment) of  the  cephalic  setae,  the  deeply  cut 
lips,  the  shape  of  the  amphids  (although 
the  situation  in  Halanonchii.s-  is  not  quite 
clear),  and  the  presence  of  a  large  buccal 
cavity  suggest  relationship  with  the  Tripy- 
loididae,  to  which  family  Halanonchus  and 
Rlial)d()coma  ha\c  been  referred  by  Filipjev 
( 1934 ) .  On  the  other  hand,  as  Schuurmans- 
Stekhoven  ( 1950 )  and  Gerlach  ( 1955,  1962) 
have  pointed  out,  both  Rliahdocoma  and 
Cyfohimium  have  a  number  of  featiues  in 
common  with  Trcftt.mi  de  Man,  1893,  which 
has  no  buccal  cavity,  pocket-shaped  or  spiral 
amphids  and  no  supplements,  and  which 
so  far  has  been  considered  an  oxystomatid 
of  somewhat  uncertain  position.  Finally. 
Chitwood  (1936,  1951)  placed  Cytohimium 
and  Rhahdocoma  with  the  Monhysteridae. 

We  consider  that  by  virtue  of  the  deeply 
cut  lips,  the  jointed  setae,  the  large  Iniccal 
cavity  (particularly  in  Halanonchus),  and 
the  spiral  amphids  (in  Cytokiimium  and 
RhaJydocoma ) ,  the  three  genera  mentioned 
belong  to  the  family  Tripyloididae.    How- 


exer,  because  of  the  simplicity  of  the  buccal 
cavity,  the  arrangement  of  the  cephalic 
setae  in  two  circles  (instead  of  one  circle), 
and  the  different  structure  of  the  male 
genital  armature,  a  distinct  subfamily  should 
be  created  for  them,  for  which  we  propose 
the  name  Halanonchinae  new^  subfamily, 
with  Halanonchus  Cobb  as  the  type  genus. 

This  new  subfamily,  via  Trcfusia,  links 
the  Tripyloididae  with  the  Oxystomatidae 
and  thus  allows  a  more  satisfactory  place- 
ment of  the  former  family  which  so  far  has 
occupied  a  rather  isolated  position  either 
within  the  order  Araeolaimoidea  (Schu- 
lu-mans-Stekhoven,  1935),  or  the  super- 
family  Chromadoroidea  (Chitwood,  1951). 

\\'ithin  the  genus  Trcfusia  and  the  sub- 
family Halanonchinae  it  seems  as  if  a  transi- 
tion of  the  amphidial  shape  from  spiral  to 
pocket-shaped  or  tubular  had  taken  place, 
thus  stressing  the  intennediary  position  of 
this  group  of  genera  between  the  Oxysto- 
matidae and  the  Tripyloididae.  This  is  dem- 
onstrated not  only  by  a  comparison  of 
Halanonchus  with  Cyfolaimium  and  Rhah- 
docoma, but  also  by  Trcfusia  varians  Ger- 
lach, 1955,  in  which  the  juxeniles  have  spiral- 
shaped,  and  the  adults  tubular-shaped  am- 
phids not  too  different  from  the  shape 
which  we  observed  in  H.  macrurus  (see 
below).  Further  proof  of  the  intennediary 
position  of  the  new  subfamily  might  be 
the  fact  that  Rhahdocoma  is  reported  to 
have  just  one  posterior  ovary  (as  is  the 
case  in  many  oxystomatids),  whereas  Cy- 
folaimium has  two  ovaries  (like  the  Trip- 
yloidinae ). 

HALANONCHUS  Cobb,  1920 

Type  spedes.— Halanonchus  macrurus  Cobb, 

1920:   266,  fig.  51. 

Halanonchus  Cobb,  1920:    266. 
Lotihimus  Allgen,  1933:    90. 

Halanonchus  macrurus  Cobb,  1920 
Plate  III,  fig.  5,  a-c 

Halanonchus  macrurus  Cobb,   1920:    266,  fig.  51. 

L  =  2.00  mm;  w  =  40  /x;  esophagus  =  265 
/A.    Head  diameter  (on  level  of  4  cephalic 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        249 


setae)  19  fi.  Three  large,  deeply  cut  lips, 
on  each  lip  2  thin  labial  setae  and,  shortly 
behind,  2  larger,  t\vo-jointed  cephalic  setae, 
measuring  3-3.5  /x.  Further  behind,  there 
are  the  four  setae  of  the  second  cephalic 
circle,  measuring  about  5  ^  and  being  non- 
jointed.  There  are  many  yellowish  granules 
which  are  scattered  throughout  the  epider- 
mis of  the  body,  although  a  certain  arrange- 
ment into  longitudinal  rows  can  be  dis- 
cerned. The  amphids  give  the  impression 
of  sawed-off  and  slightly  bent  pieces  of 
tubing  with  an  indistinct,  more  or  less 
circular  orifice;  they  are  8  /x  long  and 
situated  20  p.  behind  the  anterior  end.  Cobb 
shows  the  amphids  more  pocket-shaped 
with  a  distinctly  circular  orifice.  Each  lip 
seems  to  be  supported  by  a  large  oval 
structure  which  apparently  was  mistaken 
for  the  amphids  by  Allgen  ( 1933 )  in  his 
description  of  Lotikiiiniis  zosferac.  In  the 
cervical  region  one  can  distinguish  5-6 
ventral  bumps  which  might  be  papillae. 
The  same  organs  have  been  observed  in 
undcscribed  species  of  Rhahdocoma  and 
HaJammchus  by  Gerlach  (1962).  Buccal 
cavity  with  strong  walls,  20-22  jx  long. 
Esophagus  weakly  dilated  posteriorly,  no 
bulb.  A  small  triangular  cardia  is  present. 
Spicula  slender,  32  jx  long,  gubernaculum 
either  absent  or  consisting  of  a  thin  lamella, 
lying  parallel  to  the  spicula.  There  are 
about  12  preanal  supplements,  the  posterior 
six  being  more  distinct  than  the  more  an- 
terior ones.   Tail  610  jx  long,  a.b.d.  28  /x. 

Rcpresenfafion  in  samples  studied. — M-8, 
Biscayne  Bay. 

Geographical  distribution. — Biscayne  Bay 
(Cobb,  1920). 

CYTOLAIMIUM  Cobb,  1920 

Type     species.— Cy/o/o/'m/um     ex/7e     Cobb, 

1920:  251,  fig.  31. 
Cytolaimium  exile  Cobb,  1920 

Plate  III,  fig.  6,  a-d 

Cyfokiimium  exile  Cobb,  1920:    251,  fig.  31. 
Cytolaimium  obtiisicauclatiim  Chitwood,  1936:    13, 
"fig.  3,  J-L  (cf.  Gerlach,  1962). 

L  =  3.2  mm;  w  =  32  fx;  diameter  at  base 


of  esophagus  30  /x.  Esophagus  250  /x  long. 
Head  diameter  20-25  /x.  Head  with  six 
setose  labial  papillae,  six  segmented  ce- 
phalic setae  and  four  subcephalic  setae. 
Cephalic  setae  in  female  20  /x  long,  in  male 
somewhat  longer  but  distorted  so  as  to 
render  precise  measurement  impossible. 
Amphid  8  /x  wide  and  located  20-23  /x  from 
anterior  end.  Subcephalic  setae  10  /x  long 
in  male,  5-6  /x  in  female  and  located  35  /x 
from  anterior  end.  Body  with  several  short 
(3-4  ij.)  cervical  setae  beginning  90  /x  from 
anterior  end  and  ceasing  at  nerve  ring  and 
with  a  few,  very  thin,  somatic  setae  10-15 
IX  in  length.  Female  didelphic,  amphidelphic, 
ovaries  reflexed.  Male  diorchic,  testes  out- 
stretched. Spicules  37  /x  long,  their  chord 
28  /x.  Male  preanally  with  9  pairs  of  discoid 
supplements  and  2  pairs  of  fleshy,  setose 
papillae  and  postanally  with  3  pairs  of 
discoid  supplements  and  5  pairs  of  fleshy 
setose  papillae.  Tail  length  extremely  vari- 
able, 15  /x  long  for  one  female,  420  /x  for 
the  other,  and  740  /x  for  the  male. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne. 

GeograpJncal  distribution. — Biscayne  Bay 
(Cobb,  1920).  Beaufort  North  Carolina 
(Chitwood,  1936),  Maldives  (Gerlach.  1962). 
1962). 

Remarks. — Tlie  data  for  our  specimens 
( 1  6  ,  2  9)  lend  support  to  the  view  of 
Gerlach  (1962)  that  C.  obtusicaudatum 
Chitwood,  1936,  is  a  synonym  of  C.  exile 
Cobb,  1920.  The  tail  length  in  our  three 
specimens  ranged  from  15  ^  in  one  female 
to  740  fx  in  the  male.  In  the  specimens  with 
shorter  tails,  the  terminus  appears  abnonnal 
and  suggests  the  phenomenon  of  \\ound- 
healing  as  discussed  for  this  species  by 
Gerlach.  Even  in  the  case  of  the  longest- 
tailed  specimen  there  is  the  possibility  of 
a  missing  portion,  as  no  typical  spinneret 
can  be  recognized. 

The  setose  papillae  associated  with  the 
male  tail  appear  to  be  similar  to  the  papillae 
seen  on  the  discoid  supplements,  the  only 
difference  being  the  lack  of  the  disc. 


250        BuUctiii  Mus(  iini  of  Comparative  Zoology,  Vol.  135,  No.  5 


BATHYLAIMUS  Cobb,  1893 
Type  spedes.—Bafhyloimus  australis  Cobb, 
1893:    409,  410,  fig.  9,  l-IV. 

Our  material  contained  two  species  of 
the  genus  BafliyJaimus,  one  of  wliich  ap- 
pears to  be  the  type  species,  B.  amtialis 
Cobb,  1S93.  The  other  represents  an  un- 
described  species  closely  related  to  B. 
co}xicoms  Hopper,  1962,  from  which  it  can 
be  distinguished  b\  lia\'ing  the  shorter  of 
the  submedian  cephalic  setae  equal  in 
length  to  the  basal  segment  of  the  longer. 
In  B.  capacosus  the  shorter  of  the  pair  is 
approximately  half  as  long  as  the  basal 
segment  of  the  longer. 

Bathylaimus  ausfralis  Cobb,  1893 
Plate  IV,  fig.  7,  a-e 

Bathylaimus  australi.s   Cohh,    1893:     409-410,   fig. 

9,  I-I\'. 
Bathi/laiinus  assimilis  de   Man,   1922b:     119,    120, 

fiu.  2-2e.   NEW  SYNONYMY. 

h=  6  ,  1.5-2.2  mm,  9  ,  1.4  mm;  w  =  45- 
53  II-  diameter  at  base  of  esophagus  41-44  /x. 
Esophagus  270-,370  ^.  Vu  =  51%.  Tail  9S- 
135  /x  (3-3.5  a.b.d.).  Head  diameter  22-24  /x. 
Labial  setae  about  4  ix.  Cephalic  setae  20  + 
9  /x,  the  longer  set  comprised  of  four  seg- 
ments. Cervical  region  bearing  eight  rows 
of  somatic  setae;  those  anterior  to  nei-ve 
ring  being  5  /x  long.  Buccal  cavity  37-39  /x 
long,  bipartite  (29-31  +  8-9  ^);  without 
armature.  Amphid  20  /x  from  anterior  end, 
positioned  over  posterior  half  of  the  anterior 
portion  of  buccal  cavity;  internal  amphidial 
ponch  5  /;,  wide,  its  orifice  approximately 
3  /-.  Spicula  4(S-50  /x  long,  gubernaculum 
50  /x  long.  Female  tail  without  setae,  male 
tail  bearing  .setae  as  illustrated   (Fig.  7,  b, 

c)- 

Representation    in    .samples-   studied. — ^V, 

Vero  Beach. 

Gc()'j,i(i))hical  distribution. — Cosmopoli- 
tan. 

Remarks. — In  considering  the  synonymi- 
zation  of  i^.  assimilis  de  Man  with  B. 
au.^tralis  Cobb,  the  following  facts  are  per- 
tinent.    \\'i(scr.    1956,    separated    the    two 


species  on  the  basis  of  different  values  for 
the  lengths  of  both  the  buccal  cavity  and 
the  spicula.  According  to  the  information 
presented  by  that  author,  B.  au.stralis  has 
a  buccal  cavity  of  33  /x  and  spicula  of  simi- 
lar length,  while  in  B.  assimilis  the  figures 
were  50-55  fx  and  45  ^,  respectively.  Fur- 
ther, W'ieser  states  that:  "B.  australis  and 
B.  assimilis  are  very  closely  related  and  the 
differences  in  the  length  of  both  spicula 
and  buccal  cavity  are  the  only  ones  I  can 
find."  An  examination  of  the  original  de- 
scription of  B.  australis  shows  that  Cobb's 
animal  had  a  buccal  cavity  of  50  /x  and 
spicula  of  approximately  40  fx.  These  fig- 
ures closely  approach  those  presented  by 
de  Man  in  his  original  description  of  B. 
assimilis  (buccal  cavity  40-43  ^,  spicula  37 
fi).  On  this  basis  we  consider  B.  as.similis 
de  Man,  1922b,  along  with  its  synonyms,  to 
be  a  junior  synonym  of  B.  australis  Cobb, 
1893. 

B.  setosicaudatus  Timm,  1961,  while  ex- 
tremely close  to  B.  au.stralis..  can  be  sepa- 
rated by  the  fact  that  the  spicular  cephali- 
zation  is  reduced  while  that  of  B.  au.stralis 
is  prominent.  Also  the  spicula  are  uni- 
formly bent  in  B.  seto.sicaudatus,  while  in 
B.  aiLsiralis  most  of  the  cur\'ature  is  limited 
to  the  mid  region. 

Bathylaimus  arfhropappus  new  species 
Plate  IV,  fig.  8,  a-f 

L  =  3.1-3.3  mm;  w  =  6  ,  52-55  /x,  9  ,  62 
fx.  Diameter  at  base  of  esophagus,  $  ,  44- 
48  /x,  $  ,  53  IX.  Esophagus  i  ,  687-750  ix, 
9  ,  820 IX,  Vu  =  56% .  Tail  6  ,  118-130  ^  (3.4- 
3.7  a.b.d.),  9,  220  il  (5.4  a.b.d.).  Head 
diameter,  34-37  ^.  Labial  setae  about  20  ^. 
Cephalic  setae  47-52  +  1^19  /x.  Both  the 
labial  setae  and  longer  cephalic  setae  are 
segmented,  with  the  former  having  three 
segments  and  the  latter  four.  Cervical  re- 
gion bearing  eight  rows  of  somatic  setae 
which  are  more  or  less  arranged  in  circles. 
First  circle,  6  ix  long,  near  base  of  amphid. 
Second  circle,  17  /x  long,  half  the  distance 
from  the  anterior  end  to  the  nerve  ring. 
Third  circle,  8  /x  long,  just  anterior  to  nerve 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        251 


ring.  Remaining  two  circles,  8-9  ix  long, 
posterior  to  nerve  ring,  the  last  70  /x  poste- 
rior to  nerve  ring.  Buccal  ca\'ity  28-35  p- 
long,  without  armature.  Amphid  in  i  ,  12- 
13  |U,  in  9  ,  8  /x  wide,  located  posterior  to 
buccal  cavity,  37-38  fx  from  anterior  end. 
Spicula  50-57  /x  long,  proximally  cepha- 
lated.  Gubernaculum  50  jx  long.  Male  tail 
bearing  setae  as  illustrated  ( Fig.  8,  f ) ;  sub- 
terminal  setae  40-45  ix  long.  Four  gland 
cells  are  associated  with  the  spinneret  ap- 
paratus in  both  sexes. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4070,  Type  slide  No.  61. 
Type  locality,  M-5,  Virginia  Key. 

Representation  in  samples  studied. — M- 
5,  Virginia  Key. 

Geo^iraphica]  disiribufion. — The  species 
also  occurs  at  Gulf  Shores,  Alabama  (un- 
published observation). 

Retnarks. — The  region  of  the  esophagus 
directly  posterior  to  the  buccal  cavity  is 
constructed  in  such  a  manner  that,  if  the 
esophageal  musculature  were  to  exert  a 
pull  in  this  region,  a  cavity  might  arise 
which  could  be  mistaken  for  a  second  com- 
partment of  the  buccal  cavity.  This  subject 
was  also  brought  up  in  the  description  of 
B.  capacosus  Hopper,  1962. 

PHANODERMATIDAE 
PHANODERMOPSIS  Ditlevsen,  1926 

Type  species.— Phanodermopsis  groenland- 
ica  Ditlevsen,  1926:  13,  14,  pi.  7,  figs. 
1,  2;  pi.  8,  fig.  5. 

Our  material  contained  typical  represen- 
tatives of  P.  longisetoe  Chitwood,  1936. 

The  following  species  have  been  de- 
scribed since  a  key  to  the  species  of  this 
genus  was  given  b\-  ^^'ieser  (1953):  P. 
conicauda  Filipjev,  1946,  and  P.  in^irami 
Mawson,  1958,  belonging  to  W'ieser's  group 
A,  and  P.  ohtusicauda  Filipjev,  1946,  be- 
longing to  group  B.  P.  necta  Gerlach,  1957, 
does  not  appear  to  belong  to  this  genus, 
since  it  has  a  well  de\eloped  supplement 


and  spicula  as  in  Phanoderma.    P.  suecica 
Allgen,  1953,  is  a  species  inqiiirenda. 

Mawson  ( 1958a )  raised  the  question  of 
the  position  of  this  genus  since  Ditlevsen 
(1926)  did  not  designate  a  type  species. 
However,  Filipjev  ( 1927 )  in  the  appendix 
to  his  paper,  established  synonymy  of  his 
genus  Galeonema  with  Phanodermopsis 
and  designated  P.  fi,roenIandica  Ditlevsen 
as  the  type  species. 

Phanodermopsis  longisetae  Chitwood,  1936 
Plate  V,  fig.  9  a-c 

PJianodermopsis  longisetae   Chitwood,    1936:    209, 
210,  pi.  26,  fiss.  16-19. 

L  =  3.85  mm;  w  =  80  /.;  tail  250  /x.  Head 
diameter  15  /i,  capsule  weakly  developed. 
Labial  papillae  distinct,  conical.  Cephalic 
setae  15  +  10  jx  long.  Amphids  6  ix  wide  = 
40'yr  of  head  diameter,  10  ^  behind  anterior 
end.  Excretory  pore  55  /x  behind  anterior 
end.  Spicula  360  ix.  Gubernaculum  56  fx. 
Caudal  setae  arranged  in  characteristic 
pattern,  the  setae  being  of  two  types:  one 
fleshy  and  S-shaped,  the  other  slender  and 
straight  or  slightly  curved. 

Representation  in  samples  studied. — V, 
\'ero  Beach. 

GcograpJi ical  distribution.  —  Beau- 
fort, North  Carolina  (Chitwood,  1936). 

ENOPLIDAE 
ENOPi.O/DES  Saveljev,  1912 

Type  species.— Enop/o/c/es  fypicus  Saveljev, 
1912:    115. 

In  this  genus,  classification  is  possible 
only  on  the  basis  of  the  male  genital  arma- 
ture. Consequently,  we  have  to  insist  that 
all  species  known  from  juveniles  or  by  fe- 
males only  are  to  be  regarded  as  species 
inquirendae.  This  includes,  in  addition  to 
the  doubtful  species  mentioned  by  Wieser 
( 1953),  the  following:  E.  labiatus  BiitscWi, 
1874  [Synonymy  of  this  species  with  E. 
spiculohamatus  Schulz,  1932,  cannot  be 
proven  in  an\-  way  and  should  be  aban- 
doned, as  advocated  by  Brunetti,  1950.], 
E.    tridentatus    Sa\'elje\-,    1912,    E.    brevis 


252        Bulletin  Museum  of  Comparative  Zoology,  Vol  135,  No.  5 


Filipjev,  1918,  E.  hratistrorni,  E.  parakibio- 
tus,  E.  icdiictu.s,  and  E.  lon^icaudatus  all 
Wieser,  1953,  E.  oU<iotricha  Mawson,  1956 
(syn.  E.  oli'j^ochiictus  Mawson,  1956),  E. 
ptcro^^natlius  Mawson,  1956,  and  E.  ker- 
'^ucJcnsc  Mawson,  1958. 

The  remaining  species  can  be  separated 
into  two  gronps,  one  with  short  spicula,  the 
other  with  extremely  long  spicula.  The 
former  group  comprises  only  two  species, 
viz.,  E.  ciirhatus  Filipjev,  1918,  and  E. 
tyrrhenicus  Brunetti,  1949  (cf.  Gerlach, 
1952),  for  which  most  likely  a  new  genus 
or  subgenus  should  be  established. 

The  group  with  long  spicula  is  very  uni- 
fonn.  Since  the  gubernaculum  represents 
one  of  the  best  distinguishing  characters, 
the  shape  of  this  organ  in  all  the  species 
belonging  to  this  group  (except  E.  ti/picus, 
of  which  no  figures  were  given)  is  shown 
in  Text-figure  1.  A  number  of  species  can 
be  separated  immediately  by  the  shape  of 
the  gubernaculum,  viz.,  E.  hirsiitits  Filipjev, 
1918,  E.  hnmettii  Gerlach,  1952,  E.  vcctis 
Gerlach,  1957  (syn.  E.  brunettii  var.  vcctis), 
and  E.  harpax  \\'ieser,  1959.  In  the  remain- 
ing species  the  gubernaculum  is  more  or 
less  S-shaped,  although  differences  in  shape 
between  the  species  can  be  found.  The 
species  comprising  this  remaining  group 
can  be  separated  by  use  of  the  following 
key. 

Key  to   Species   of   Enoi'I.oides    Possessing 
s-shaped  gubernacula 

1 .  Cephalic  setae  of  equal  length   

E.   ccplialophorus   ( Ditlevsen,   1919) 

Cephalic  setae  of  unequal  length   2 

2.  Longest  cephalic  setae  about  1.2  head  diam- 

eters long,  shorter  setae  measuring  l:{  of 
longer  ones.  Gubernaculum  with  cliar- 
acteristic  ventral  knob  .,  £.  ^njphtts  n.  sp. 
Longest  cephalic  setae  measuring  not  more 
than  one  head  diameter,  shorter  ones 
about  'j  that  length.  Cmbernaculum 
rather    smoothly    S-shaped    —  3 

3.  Tip  of  gubernaculum  apparently  3-pronged; 

supplement  measiuing  about  half  the  cor- 
responding    body     diameter     (description 

and   figures   not   (luite   clear)  

_„ E.    spictiloluimatus    Schnlz,    1932 

Tip  of  gubernaculum  2-pronged;  supplement 
much  shorter  4 


4.  Distal    end    of    spicula    with    moljile    spine; 

mandibles   55-60   m   long    

E.  amphioxi  Filipjev,   1918 

Distal  end  of  spicula  without  mobile  spine; 
mandil^les    45    m    <'i"    less    5 

5.  Spicula    smooth,    distal    end    pointed,    then 

dilated;   mandibles   about   45   fj.    (extrapo- 
lated from    Southern's   figmes   and   text) 

E.  labrosiriatiis  (Southern,  1914) 

Spicula  vertically  striated,  in  their  distal 
half  with  a  diagonal  break  (Fig.  10,  c); 
mandibles  30-35  m  long  -  E.  I)isulcus  n.  sp. 

The  type  species,  E.  typicus  Saveljev, 
1912,  is  poorly  known  since  no  figures  were 
given.  It  seems  to  differ  from  all  other 
species  by  the  small  dimensions  of  its 
organs  ( cephalic  setae  only  8  ^a  long,  buccal 
ca\ity  only  9  /x  deep! ) . 

Enoploides  bisulcus  new  species 
Plate  V,  fig.  10,  a-d 

L  =  3.5-4.2  mm;  w  =  115-120  //;  diam- 
eter at  base  of  esophagus  95-105  ^a,.  Esoph- 
agus 750-800  fx;  nerve  ring  at  25%.  Vu  = 
52.5%.  Eggs  165x75  /x,  one  per  uterus. 
Tail  in  6  ,  170-200  /x  (4  a.b.d.),  in  9  ,  215- 
225  IX  (3.7-4.5  a.b.d.),  in  juvenile  4.8  a.b.d. 
long.  Head  diameter  50-56  /x.  Labial  setae 
25  pi.  Cephalic  setae  45-50 -f  25-28  /x. 
Mandibles  30-35  /x  long,  deeply  notched 
anteriorly,  with  strong  apophyses  and  mus- 
cles that  link  them  to  the  buccal  capsule. 
Subventral  teeth  small,  nearly  parallel  to 
mandibles  and  thus  difficult  to  obsei-ve  in 
lateral  view.  Their  presence  can  best  be 
ascertained  in  en  focc  view.  Dorsal  tooth 
seemingly  absent.  Peribuccal  portion  of 
esophagus  strongly  de\eloped,  muscular. 
Spicula  420-475  /x,  knobbed  proximally, 
vertically  striated  throughout,  with  a  diag- 
onal break  in  their  distal  end  that  runs  from 
dorsal  to  ventral,  tip  pointed.  Gubernacu- 
lum S-shaped,  its  proximal  end  open,  ill- 
defined;  distal  end  forked.  Supplement  20 
/x  long,  120-150  /x  preanal.  Tail  with  pair 
of  fleshy,  S-shaped  spines  and  a  number 
of  setae.  Setae  in  anal  area  and  paired 
terminal  setae,  17  y.  long. 

Holotype  specimen. — Male;  Canadian 
National    Collection   of   Nematodes,    Ento- 


Florida  Marine  Nematodes  •   Wicscr  and  Hopper        253 


^ 


a=^    &<:. 


cF=y.    (r^ 


Figure  1.  Gubernacula  of  several  species  of  Enop/oides.  a — E.  hirsutus:  b — E.  brunettii;  c — E.  vecfis;  d — E.  harpax; 
e — £.  amphioxi;  f — E.  labrostriatus;  g — E.  cepho/ophorus;  hi — E.  sp/cu/ohomofus;  i — E.  b/su/cus;  k — E.  gryphus.  All 
gubernacula   copied    from    original    descriptions   of   respective   species. 


mology  Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  62.  Type 
locality,  M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

Enoploides  gryphus  new  species 

Plate  II!,  fig.  lie;  plate  V,  fig.  1 1  a,  b 

L  =  3.0-3.7  mm;  w  =  8.5-100  /^;  diameter 
at  end  of  esophagus  70-85  fi.  Esophagus 
800-900  ix.  Tail  190-240  /x;  a.b.d.  45-60  ix. 
Head  diameter  45-57  fi.  Labial  setae  28-32 
/._.  Cephalic  setae  60-80  +  20-27  ix.  Man- 
dibles 26-28  fx.  Apart  from  the  dimensions, 
the  structure  of  the  buccal  armature  is  the 
same  as  in  E.  hisidciis.  Spicula  230-260  /x, 
6-7  /x  wide,  knobbed  proximally,  vertically 
striated  throughout,  without  break,  tip 
pointed.    Gubernaculum  hook-shaped,  with 


characteristic  ventral  knob.  Supplement 
80-110  fx  preanal.  Tail  with  pair  of  slender 
postanal  spines  (not  S-shaped  as  in  the 
foregoing  species)  and  several  setae.  (Note: 
All  the  somatic  and  caudal  setae  on  this 
species  are  shorter  and  more  slender  than 
on  E.  bisidciis.) 

HoJotype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomol- 
ogy Research  Institute,  Ottawa,  Collection 
Number  4070,  Type  slide  No.  63.  Type 
locality,  M-5,  Virginia  Key. 

Representation  in  .samples  .studied. — M- 
5,  X'irginia  Key. 

Remarks. — This  species  resembles  the 
foregoing  one  in  most  respects.  The  main 
distinguishing  features  are  the  longer 
cephalic  setae  and  the  shorter  spicular  ap- 
paratus. 


254        Bulletin  Museuni  of  Conipanitivc  Zoology,  Vol  135,  No.  5 


MESACANTHOIDES  Wieser,  1953 

Type      species.— Mesacanf/io/des      scufpfilis 

Wieser,  1953:    86,  fig.  46,  a-d. 

Through  the  addition  of  two  new  species 
this  genus  is  becoming  more  heterogeneous 
as  far  as  the  male  genital  appaiatus  is  con- 
cerned. Howexer,  its  most  characteristic 
feature,  the  shape  and  texture  of  the  man- 
dibles, suffices  to  separate  it  clearly  from 
related  genera.  Classification  is  largely 
based  on  the  genital  armature  ( see  \\' ieser, 
1959).  Consequently,  M.  wieseri  Mawson, 
1956,  has  to  be  considered  a  species  in- 
quirenda,  although  it  appears  to  be  closely 
related  to  M.  Jatignaihus. 

Key  to  Species  of  Mesacanthowes 

1.  (;ubernaculum    reduced,    no    supplement 

M.   latignathus    ( Ditlevsen,   1919) 
Gubernaculum  and  supplement  present  2 

2.  Supplement   small,   tubular   -  3 

Supplement   \erv   large,   "wrench-like"    

M.  sculptilis  Wieser,    1953 

3.  Spicula  more  than  two  anal  diameters  long, 

tail    filiform,    with    flagellum    

M.   fiJjuhifus   n.    sp. 

Spicula   about   one   anal   diameter   long,    tail 
plump  or  elongate,  never  filiform  4 

4.  Gubernaculum    more    or    less    plate-shaped; 

head  with  four  circles  of  cephalic  setae___ 

M.  caputmediisae   (Ditlevsen,  1919) 
Cubcrnaculuni     S-shaped;     head     witli     2-3 

circles   of   cephalic  setae   5 

.5.    Implantation   of  cephalic  setae   near  middle 
of  cephalic  capsule;  spicula  strongly  arcu- 
ate, tip  pointed  -^    M.  sinuosus  Wieser,  1959 
Implantation    of    cephalic    setae    at    posterior 
edge    of    cephalic    capsule;    spicula    nearly 

straight,    tip   elaborately    armed    

M.    p.sittacus   n.    sp. 

Mesacanthoides  fibuiatus  new  species 
Plate  VII,  fig.  12a-c 

L  =  4.3  mm;  w  =  75  /x.  Esophagus  =  675 
//.  Tail  =  350  /v..  Head  diameter  33  p. 
Labial  setae  10  /i.  Longest  cephalic  setae 
42  ij.,  setae  of  submedian  pairs  sticking  to- 
gether. One  circle  of  subcephalic  setae  in 
male.  Cephalic  capsule  deeply  lobcd.  Ce- 
phalic organ  present,  in  front  of  lateral 
cephalic  seta.  Mandibles  18  X  10  /x,  solid 
as   ill    Knoploides   but   with    an    additional 


transverse  bar  near  the  anterior  end;  more- 
over, the  sclerotization  is  not  uniform  but 
gives  a  mottled  impression.  The  tips  of  the 
claws  are  darker  than  the  rest  of  the  man- 
dibles. Teeth  well  developed,  about  half 
as  long  as  the  mandibles.  Stomodeal  ring 
forming  three  "brackets"  around  the  man- 
dibles which  serve  as  muscular  attach- 
ments. Spicula  125-135  p,  about  4  anal 
diameters,  cephalate  proximally,  with  a 
break  just  before  the  distal  sixth.  Guber- 
naculum slightly  S-shaped,  15  p  long.  Sup- 
plement small,  about  90  p  in  front  of  anus. 
Tail  at  first  conical,  then  abruptly  attenu- 
ated and  drawn  out  into  an  extremely  long 
and  whip-like  flagellum,  about  five  times 
the  length  of  the  spicula.  In  the  circumanal 
region  there  are  scattered  setae. 

Holotijpe  specimen. — Male;  Canadian 
Xationai  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4069,  Type  slide  No.  64. 
Type  locality,  M-3,  Key  Biscayne. 

P.epresenfation  in  samples  studied. — M- 
3,  Key  Biscayne. 

Mesacanthoides  psittacus  new  species 
Plate  VI,  fig.  13a-e 

L  =  2.32  mm;  w  =  50  p;  esophagus  =  540 
IX-  tail  =  222  p..  Head  diameter  35  p.  Lips 
plump,  labial  setae  stout,  15  p.  Cephalic 
setae:  lateral  62  p,  submedian  56  +  27  p, 
implanted  near  posterior  edge  of  cephalic 
capsule.  The  latter  with  straight  edge,  17 
p.  high.  Mandibles  19  X  11  p.,  powerful, 
plump,  of  typical  shape  and  texture,  each 
claw  with  an  oval  apophysis  on  its 
"shoulder."  Teeth  well  de\eloped,  about 
half  as  high  as  the  mandibles.  Spicula 
plump,  nearly  straight,  50  /x  long,  their  tips 
broacl,  each  with  a  three-pronged  process 
on  the  caudal  edge.  Gubernaculum  S- 
shaped,  about  26  p.  long,  powerful.  Supple- 
ment small,  90  p  in  front  of  anus.  Tail  5 
anal  body  diameters  long,  with  scattered 
setae. 

HoIntj/))c  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute.  Ottawa,  Collec- 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        255 


tion  Number  4074,  Type  slide  No.  65.  Type 
locality,  L,  Lauderdale-by-the-Sea. 

Representation  in  samples  studied. — L, 
Lauderdale-by-the-Sea. 

ONCHOLAIMIDAE 
ONCHOLAIMUS  Dujardin,  1845 

Type  species.— Onc/io/a/mus  aftenuatus  Du- 
jardin, 1845:   236. 
Keys  to  the  species  of  Oncholaimiis  can 
be  found  in  the  works  of  Kreis  ( 1934 )  and 
Wieser  (1953). 

Oncholaimus  dujardinii  de  Man,  1878 
Plate  VII,  fig.  14,  a,  b 

Oncholaimus  dujardinii  de   Man,   1878:    94,   pi.   7, 
fig.  4,  a-c. 

L  =  2.4-2.95  mm;  w  =  42-45  //;  esopha- 
gus 350  jji.  Yu  =  769f .  Tail,  in  9  ,  67  /x 
long  (2.6  a.b.d.).  Head  18-20  ,j.  wide, 
bearing  10  short,  subequal  cephalic  setae, 
the  longest  4  /x  long.  Buccal  cavity  22-24  X 
12  fx  anned  with  3  strong  teeth  of  which 
the  left  subventral  is  the  more  prominent. 
Amphids  in  male  7-9  jj.  wide  ( =  37-459f  of 
c.b.d.).  Anterior  end  of  esophagus  with  a 
pair  of  dark  pigmented  masses  (cf.  de  Man, 
1878).  Excretory  pore  45-50  ^  behind  an- 
terior end.  Renette  cell  prominent,  90-100 
fj.  long,  located  approximately  200  ^  poste- 
rior to  base  of  esophagus.  Spicula  27-31  /j. 
long,  proximally  cephalated.  Gubernacu- 
lum  absent,  although  a  dorsal  thickening  of 
the  cloacal  lining  may  give  the  impression 
of  the  presence  of  such  a  structure.  The 
thickened  area  appears  to  be  a  point  of 
attachment  for  muscles  associated  with  the 
spicula.  Male  tail  40  /x  long,  ventrally 
curved;  with  12-16  stout,  circumcloacal 
setae.  Distal  extremity  of  male  tail  slightly 
enlarged  and  bearing  two  pairs  of  short, 
stout,  subventral  setae  and  a  pair  of  slen- 
der, subdorsal  setae. 

Representation  in  samples  studied. — M- 
1,  Key  Biscayne. 

Geographieal  disiribution. — Cosmopoli- 
tan. 

Discussion. — While  our  species  appears 
to  be  O.  dujardinii  de  Man,  1878,  there  re- 


mains some  doubt  as  to  the  identity  of  this 
species.  SchuuiTnans-Stekhoven  (1950)  and 
Inglis  (1962)  reason  that  the  typical  O. 
dujardinii  is  devoid  of  a  gubernaculum, 
whereas  O.  dujardinii  de  Man  sensu  Steiner, 
1915  ( and  other  authors )  possesses  such  an 
organ  and  thus  represents  a  different  spe- 
cies. However,  we  are  of  the  opinion  that 
the  dorsal  thickening  of  the  cloacal  lining 
described  in  our  specimens  has  been  oc- 
casionally misinterpreted  as  a  gubernacu- 
lum and  we  regard  the  questioned  accounts 
of  O.  dujardinii  as  representing  de  Man's 
species. 

ONCHOLAIMIUM  Cobb,  1930 

Type  species.— Oncfio/a/m/um  oppendicula- 

tum  Cobb,  1930:    227,  figs.  2,  3,  6,  v, 

8,  9. 

A  discussion  of  this  genus,  with  a  key  to 
species,  was  given  recently  by  Chitwood 
(1960).  We  feel  that  separation  of  this 
genus  from  Oncholaimus  should  be  based 
mainly  on  the  presence  of  a  Demanian 
system  and  not  so  much  on  that  of  the  pre- 
anal  papillae  (not  postanal,  as  erroneously 
stressed  by  Wieser  [1953]  and  subsequent 
authors),  as  rudiments  of  preanal  papillae 
can  also  be  found  in  representatives  of  the 
latter  genus  as  well  as  in  other  oncholaimid 
genera.  This  would  suggest  that  the  two 
short-spiculed  species  of  MetoneJioJaimus 
described  by  Mawson,  i.e.,  M.  brevispicu- 
Jum  Mawson,  1957,  and  M.  thysanouraios 
Mawson,  1958,  actually  ought  to  be  re- 
ferred to  Oncholaimium.  The  diagnosis  of 
Cobb  ( 1930 )  has  to  be  emended  so  as  to 
include,  in  this  genus,  species  in  which  the 
Demanian  organ  is  provided  with  exit 
pores. 

Both  species  found  in  Florida  have  been 
previous!}'  reported  from  the  east  coast  of 
the  United  States. 

Oncholaimium  appendiculatum  Cobb,  1930 
Plate  VIII,  fig.  15,  a-c 

Oncholaimiuni   appcndicuhiluin   Cobb,    1930:   227, 
figs.  2,  3,  6,  V,  8,  9. 


256         BuUctin  Museum  of  Conipanitive  Zoology,  Vol  135,  No.  5 


L  =  2.27  mm;  \v  =  44  /x;  esophagus  =  360 
/x;  nerxc  ring  2(S()  //  l^'hind  anterior  end. 
Head  diameter  26  ji.  Lal)ial  eapsule  lobed, 
eonspicuous.  Labial  papillae  conical.  Ce- 
phalic setae  9  /x,  equal.  Buccal  cavity  30  X 
18  /..  Teeth  25  +  15-16  /x,  the  longest  one 
reaching  to  about  the  level  of  implantation 
of  the  cephalic  setae.  Amphids  11  /x  =  40% 
of  c.b.d.  wide.  Scattered  cervical  setae. 
ExcretoiA  pore  30  fx  in  front  of  nerve  ring. 
Spicula  65  fx  (one  tail  length).  No  guber- 
naculum.  Anal  diameter  25  /x.  One  large, 
"prehensile"  preanal  papilla.  Twenty-four 
to  26  circum-  and  postanal  setae,  about  13 
/x  long.  Tail  with  small  xentral  papillae  at 
the  beginning  of  the  distal  third. 

Representation  in  samples  studied. — M- 
1,  Key  Biscayne. 

Geo<iraphieal  distribution. — Woods  Hole, 
Massachusetts,  ?  Beaufort,  North  Carolina 
(Pearse,  Humm  and  Wliarton,  1942). 

Oncholaimium    domesficum    Chitwood    and 
Chitwood,   1938 

Plate  VIM,  fig.  16a-d 

Oncholaimium  di>iucsiiciitii  Chitwood  and  Chit- 
wood, 1938. 

Oncholaimium  oxijuiis  var.  domcsticus  Chitwood 
and  Chitwood,  193S:  458,  459,  fig.  1,  f-h:  nee 
Timm,   1952. 

L  =  3.3-3.4  mm;  w  =  <i  ,  65  /x,  9  ,  85  /x; 
esophagus  =  525-550  /x.  Tail  =  70  /x.  Vu  = 
64%.  Head  diameter  37  /x.  Labial  capsule 
lobed,  inconspicuous.  Six  labial  papillae. 
Cephalic  setae  subequal,  8  jx.  Buccal  cavity 
37-38  X  25-26  p..  Teeth  25  +  21  ix.  Am- 
phids 7-8  /x.  =  20%  wide  ( both  sexes ) .  Ex- 
cretory pore  70-75  /x  behind  buccal  cavity. 
Demanian  organ  well  developed,  uvette 
about  400  fx  posterior  to  vulva,  1-2  adanal 
openings  on  each  side.  Spicula  45  jx  (one 
a.b.d.).  No  gubernaculum.  Preanal  papilla 
with  two  setae.  Postanal  papillae  at  the 
beginning  of  the  distal  third  of  the  tail. 
Circumanal  setae  present.  Tail  60-70  /x, 
with  terminal  swelling. 

Representation  in  samples  studied. — M- 
1,  Key  Biscayne. 

Geoiiraphieal     distribution. — New     York 


(Chitwood  and  Chitwood,  1938),  California 
(Chitwood,  1960). 

Reinarks. — The  characters  distinguishing 
this  species  from  O.  oxyure  (Ditlevsen, 
1911)  are  the  short  teeth,  the  well-devel- 
oped preanal  papilla  with  setae  and  the 
stout  spicula.  The  specimens  described  by 
Timm  ( 1952 )  deviate  in  all  these  charac- 
teristics from  O.  domesticum  and  seem  to 
be  representatives  of  O.  oxyure. 

METONCHOLAIMUS  Filipjev,  1918 
Type     species.— Oncfio/a/mus     demani     Zur 
Strassen,  1894:    460,  pi.  29,  fig.  2. 

A  key  to  this  genus  was  given  recently 
by  Chitwood  ( 1960 ) .  We  are  of  the  opin- 
ion that,  because  of  the  shortness  of  their 
spicula,  M.  brevlspieulum  Mawson,  1957, 
and  M.  thysanouraios  Mawson,  1958,  be- 
long to  Oncholaimium.  M.  haplotretos 
Mawson,  1958,  is  considered  as  doubtful 
since  only  females  are  known. 

Our  material  contained  three  new  spe- 
cies, all  of  which  are  characterized  by  rela- 
tively short  spicula  ( <  180  fx  as  against 
250-750  IX  in  other  species)  and  by  the 
absence  of  a  gubernaculum  [which  is  also 
lacking  in  M.  albidus  ( Bastian,  1865)  but 
is  present  in  all  other  species].  Moreover, 
M.  intermedius  and  M.  simplex  are  dis- 
tinguished by  short  and  plump  tails,  M. 
simplex  by  the  presence  of  only  one  De- 
manian exit  pore,  xA/.  scissus  by  two  ventral 
papillae  on  the  6  tail  and  by  the  slit-like 
openings  of  the  Demanian  organ.  The  rela- 
tive position  of  osmosium  and  uvette  serves 
as  a  further  character  distinguishing  M. 
simplex  and  M.  intermedius. 

Mefoncholaimus  intermedius  new  species 
Plate  IX,  fig.  17  a,  d,  e;  Plate  X,  fig.  17  b,  c 

L  =  2.2-2.7  mm;  w  =  38  /x;  esophagus  = 
350-400  /x;  Vu  =  66-74%.  Head  diameter 
26  /x.  Labial  papillae  small  but  distinct. 
Labial  capsule  deeply  lobed.  Cephalic 
capsule  weak.  Cephalic  setae  8  /x,  sub- 
en  [ual.  Amphids  6  ^  in  9  ,  8-9  /x  =  30-33% 
of  c.b.d.  in   6  .  Buccal  cavitv  25-28  X  16-18 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        257 


fj..  Longest  tooth  19-20  /x.  Ventral  gland 
150-200  /x  behind  esophagus;  excretory 
pore  80-90  /x  behind  anterior  end.  In  some 
specimens  large  coelomocytes.  Demanian 
organ  well  developed,  uvette  325  /x  behind 
vuKa,  osmosium  anterior  to  uvette,  two 
exit  pores  70-100  /x  preanal.  Eggs  90  X  40 
/x.  Spicnla  70-77  ix  long,  no  gubernaculum. 
Preanal  elevation  with  short,  stout  seta. 
Fourteen  circumanal  setae.  One  pair  of 
postanal  elevations  with  setae,  near  be- 
ginning of  distal  third  of  tail.  Anal  diam- 
eter 25  /x.  Tail  in  9 ,  30  /x,  in  S  48  /x 
long.  Caudal  glands  350,  490,  and  560  /i 
preanal. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4070,  Type  slide  No.  66. 
Type  locality,  M-5,  Virginia  Key. 

Representation  in  samples  studied. — M- 
5,  Virginia  Key. 

Mefoncholaimus  simplex  new  species 

Plate  IX,  fig.   18  a;  Plate  X,  fig.  18  b-e 

L  =  1.95-2.32  mm;  w  =  6  ,  40,  9  ,  43-53 
/x;  esophagus  =  330-360  /x;  Vu  =  63-70%. 
Head  diameter  24-27  /x.  Labial  papillae 
small.  Labial  capsule  indistinct.  Cephalic 
capsule  relatively  (for  this  family)  well 
developed.  Cephalic  setae  8-9  fx,  subequal. 
Amphids  9  /x  in  9  ,  10  /x  =  40%  of  c.b.d.  in 
S .  Buccal  cavity  29-30  X  16-18  /x.  Excre- 
tory pore  90-100  /x  behind  anterior  end. 
Demanian  organ  well  developed,  uvette 
360-430  /x  posterior  to  vulva  and  200-280 
ji  anterior  to  anus;  osmosium  posterior  to 
uvette;  one  exit  pore,  opening  on  level  of 
anus  at  dorsal  side  of  body.  Moniliform 
glands  of  varying  grades  of  distinctness  in 
different  specimens.  Eggs  107-120  x  40-45 
fx.  Spicula  120  /x  long;  no  gubernaculum. 
Twelve  to  14  circumanal  setae  and  perhaps 
two  preanal,  indistinct  pores.  Tail  conical, 
37  fx  long  in  6  ,  40-46  ^a  in  9  .  Anal  diam- 
eter 26  IX  in  6  ,  30-32  /x  in  9  .  lu  the  single 
male    a    subterminal   dorsal   structure   was 


seen  on  the  tail  which  might  represent  the 
opening  of  two  of  the  three  caudal  glands. 
In  the  female  all  caudal  glands  definitely 
open  through  the  spinneret. 

HoJotiipc  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion ^Number  4070,  Type  slide  No.  67. 
Type  locality,  M-5,  Virginia  Key. 

Representation  in  samples  studied. — M- 
5,  Virginia  Key. 

Mefoncholaimus  scissus  new  species 

Plate  IX,  fig.  19  a,  b;  Plate  X,  fig.  19  c 

L  =  4.3-4.5  mm;  w  =  c5  ,  64,  9  ,  72  /x; 
diameter  at  base  of  esophagus  <^  ,  59,  9  , 
69  IX.  Esophagus  600-640  ^  long.  Vu  = 
65%.  Head  diameter  36-40  /x.  Head  with 
six  small  labial  papillae.  The  nerves  in- 
nervating the  labial  papillae  and  the  points 
at  which  they  pass  through  the  cuticle  are 
more  prominent  that  the  papillae  them- 
selves. Cephalic  setae  13-15  /x,  subequal. 
Amphid  10-11  p.  wide.  Buccal  cavity  45- 
48  x  25-27  IX.  Longest  tooth  35-38  fx, 
shorter  teeth  25-28  ju.  Excretory  pore  105- 
125  IX  from  anterior  end.  The  most  con- 
spicuous structures  of  the  Demanian  organ 
are  the  moniliform  glands.  These  glands 
are  approximately  225  /x.  long  and  open  to 
the  exterior  via  17  /j.  wide  slits  (not  pores), 
which  are  located  157-172  /x  preanal.  Eggs 
115-140  X  55-60  IX,  six  seen  in  the  uterus  of 
one  specimen.  Spicules  175-180  /x  long. 
Dorsal  wall  of  cloaca  thickened,  but  not 
fomiing  a  distinct  gubernaculum.  Tail  in 
9  ,  200-215  IX,  in  S  ,  220-230  ,x  long.  Male 
with  5  pairs  of  circumanal  setae,  6-7  pairs 
of  subventral  setae  and  two  prominent  ven- 
tral papillae. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4070,  Type  slide  No.  68. 
Type  locality,  M-5,  Virginia  Key. 

Representation  in  samples  studied. — M- 
5,  Virginia  Key. 


258        Bulletin  Mim'imt  of  Comparative  Zoology,  Vol.  135,  No.  5 


PROONCHOLAIMUS  Micoletzky,  1924 
Type      species.— Oncho/a/mus      megasfoma 
Eberth,  1863:   26,  pi.  1,  figs.  18-20. 

This  genus  is  easily  recognized  by  the 
large  bubble-like  cells  ( "Trabekula-Struk- 
tur,"  in  Gennan  literature)  that  occur  in 
the  pseudocoeloniic  caxity  between  the 
longitudinal  chords  and  the  intestine.  P. 
armi^cr  Gerlach.  1955,  does  not  possess 
these  cells  and  is,  therefore,  of  uncertain 
status.  Moreover,  since  the  spicular  appa- 
ratus turns  out  to  be  of  prime  importance 
for  classification,  a  number  of  species 
known  only  as  females  or  juveniles  have  to 
be  considered  species  inqidrendae.  These 
are:  P.  keiemis,  P.  longisetosus,  and  P. 
ohtusicaiidutus,  all  Kreis,   1932. 

The  remaining  species  fonn  an  extremely 
closely  related  group  and  are  difficult  to 
separate  on  the  basis  of  existing  informa- 
tion. We  agree  with  Inglis  (1962)  that  the 
shape  of  the  distal  end  of  the  spicula  might 
serve  as  an  important  taxonomic  character, 
and  use  it  as  such  in  the  differentiation  of 
our  species,  although  we  realize  that  the 
data  in  the  literature  on  which  some  of  our 
conclusions  are  based  probably  are  insuf- 
ficient. 

L/.sY  of  species  of  Prooxciiolaimus  {and 
their  synonyms) 

P.  meiiastoma  (Eberth,  1863)  [syn. 
Oncholaimus  mcii^astoma  Eberth,  1863,  Pro- 
oncholaimiis  mcditernineiis  Schuurmans- 
Stekhoven,  1943,  nom.  iiov.  for  P.  mcga- 
.itoma  Micoletzky,  1924  nee  Eberth,  P. 
me<^astovui  var.  neapoUtanus  Micoletzky, 
1924,  P.  neapoUtanus  (Micoletzky)  Kreis, 
1934];  P.  eherthi  (Fihpjev,  1918)  [Syn. 
Metoncholaimus  ehertlii  Filipjev,  1918]; 
P.  ornatus  Kreis,  1932;  P.  aran.sas  Chit- 
wood,  1951;  P.  hanijidensis  Inglis,  1962; 
P.  hastatus  n.  sp. 

Our  new  species,  P.  hastatus,  is  most 
closely  related  to  P.  hamjulensis,  in  that  the 
distal  extremity  of  the  spicula  possesses  a 
distinct  barb  which  is  separated  from  the 
subterminal   swelling   of   the   spicula   by   a 


"handle."    The   two    can   be    separated    as 

follows: 

Handle  between  subterminal  swelling  of  spieula 
and  barb  very  short;   length  of  spieula   119- 

135  m;  tail  with  long  terminal  setae  

P.    hanijidensis  Inglis,    1962 

Handle  between  subterminal  swelling  and  barb 
about  the  same  length  as  the  barb  itself; 
length  of  spicula  90-95  p.-  terminal  setae  very 
short  P-  hasiatiis  n.  sp. 

Prooncholaimus  hasfoius  new  species 
Plate  XI,  fig.  20  a-e 

L  =  2.10-2.53  mm;  w  =  80-100  ^;  esoph- 
agus: 6  ,  310-375  11,  9  ,  425  /x;  tail:  6  , 
120-150  /.,  9,  175  ,x;  Vu  =  73%.  Head 
diameter  28  /x.  Labial  capsule  well  devel- 
oped, lolled.  Cephalic  setae  5  ^,  subequal. 
Amphid  in  i  ,  10-11  ^  =  307^  of  c.b.d.,  in 
juvenile  9  =  7  /x.  Buccal  cavity  40  X  21  /x, 
teeth  31  +  23  /x.  Excretory  pore  on  level 
of  cephalic  setae  in  adults.  Esophagus  with 
posterior  pyriform  swelling.  Spicula  90-95 
/x  long,  funnel-shaped  proximally,  distally 
with  subterminal  swelling,  handle  and 
well-developed  barb.  Gubernaculum  15  /j. 
long,  rather  stout.  There  is  an  indication 
of  a  bursa  and  there  are  four  pairs  of 
adanal  setae.  More  setae  can  be  found 
anterior  to  the  anus  and  subventrally  along 
the  tail.    Anal  body  diameter  33  /x. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4073,  Type  slide  No.  69. 
Type  locality,  M-8,  Biscayne  Bay. 

Representation  in  samples  studied. — M- 
8,  Biscayne  Bay. 

VISCOSIA  deMan,  1880 
Type    species.— Oncho/a/Vnus    viscosus    Bas- 
tian,  1865:    136,  pi.  11,  figs.  131-133. 

There  is  a  group  of  species  in  this  genus 
in  which  the  cephalic  setae  are  reduced  to 
papillae  or  even  to  barely  visible  shallow 
pits  in  the  cuticle.  This  morphological 
feature,  in  addition  to  the  fact  that  in  this 
genus  the  spicula  offer  hardly  any  dis- 
tinguishing characters,  renders  classifica- 
tion particularK'  difficult.    Stress  has  to  be 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        259 


laid  on  size  of  aniphids,  shape  of  buccal 
cavity  and  teeth,  arrangement  of  male  cir- 
cumanal organs  like  papillae  and  bursa, 
and  shape  of  tail. 

Key  to  Species  of  Viscosia  Hamng 
Cephalic  Papillae 

1.  Buccal    ca\  it>'    dixided    In'    strong    cuticular 

ring  into  two  chamliers;  one  side  of  buc- 
cal ca\ity  weakh'  cuticularized  ( resem- 
bling the  condition  in  Oncholaimellus); 
S  amphids  607r  of  c.b.d.;  pharyngeal 
valve    about    three    times    its    own    length 

behind    buccal    ca\ity    

V.    oucholaimclloides    n.    sp. 

Buccal  cavity  not  divided  by  cuticular  ring 
( at  most  a  faint  line  can  be  seen ) ;  buccal 
wall  well  de\eloped  all  around;  amphids 
not  more  than  50%  of  c.b.d.  wide; 
phar\Tigeal  valve  not  more  than  its  own 
length   l:)ehind   buccal   cavity   2 

2.  Male  with  bursa   (or  circumanal  "alar  mem- 

brane" sensu  Chitwood,  1960,  who  was 
the    first   to   point    out   the    importance    of 

this   character)    .3 

Male  without  bursa  4 

3.  Male  amphids   %  of  c.b.d.;   walls  of  buccal 

cavity    strongly    cuticularized;     S     tail    6 

a.b.d.  V.  papillata  Chitwood,  1951 

Male  amphids  40-45%  of  c.b.d.;  walls  of 
buccal  ca\itv  normally  developed;    i    tail 

3.4^.6    a.b.d.    

V.  papillatoicles  Chitwood,  1960 

4.  Nhile  amphids  at  most  33%  of  c.b.d.  wide  __  5 
Male  amphids  4.5-50%   of  c.b.d.  wide  7 

5.  Cephalic    papillae    distinct;    long    tooth    not 

quite  reaching  to  anterior  end  of  buccal 
cavity;    i    with  3  preanal  and  3  postanal, 

small,    setose    papillae    ..^ 

V.   keiensis  Kreis,   1932 

Cephalic  papillae  indistinct,  long  tooth 
reaching  to  anterior  end  of  buccal  cavity; 
c^  with  not  more  than  1  preanal  and  1 
postanal   papilla    6 

6.  Male    amphids    33%     of    c.b.d.;     1    preanal 

papilla;  spicula  open  proximally  

V.    nuda    Kreis,    1932 

Male  amphids  25%  of  c.b.d.;  1  preanal  and 

1   postanal  papilla;   spicula  knobbed  

V .  mcridionalis  Kreis,  19.32 

7.  Cephalic  papillae  distinct;   longest  tooth  not 

quite   reaching  to  anterior  end  of  buccal 

cavity     8 

Cephalic  papillae  indistinct  to  seemingly  ab- 
sent;   longest    tooth    reaching    to    anterior 

end   of   buccal   ca\ity   

V^.  cdinleyensis  Kreis,  1932 

8.  Male  tail  cylindrical,  5  a.b.d.  long  


- V.  nicaraguensis  (Gerlach,  1957) 

Syn.  V .  papillata  var.  nicaraguensis 
Gerlach,   19.57 

Male  tail   filiform,  8-13  a.b.d.   long   _... 9 

9.    Male  with   onh'  traces  of  circumanal  papil- 
lae    V.  glabra  (Bastian,  1865) 

Male  with  6  pairs  of  setose  circumanal  papil- 
lae    V.  macramphida  Chitwood,  1951 

Remarks. — The  relationship  of  the  two 
latter  species  is  vm certain  because  no  good 
figures  of  the  male  amphids  in  V.  gjahra 
have  been  published.  However,  the  figures 
of  female  heads  as  given  by  De  Coninck 
(1944)  and  Schuunnans-Stekhoven  (1950) 
show  the  amphids  to  be  V?,  to  %  of  the  c.b.d. 
in  width,  or  nearly  as  wide  as  the  stoma, 
from  \\'hich  it  may  be  concluded  that  the 
male  amphids  should  be  just  as  wide  as  those 
of  V.  macramphida.  The  only  good  differ- 
ence we  could  find  between  the  two  species 
in  question  seems  to  be  the  size  and  arrange- 
ment of  the  genital  papillae  in  the  male. 
We  consider  this  difference  to  be  of  spe- 
cific value  for  the  time  being  since  figures 
of  the  male  genital  region  in  V.  glabra,  as 
given  bv  two  such  excellent  obsen'ers  as 
de  Man  (1890)  and  Micoletzky  (1924a), 
fail  to  show  anything  that  can  be  compared 
with  the  distinct  setose  papillae  that  Chit- 
wood ( 1951 )  and  we  ( see  below )  found 
in  V.  macramphida. 

Doubtful  species  are:  V.  linstowi  (de 
Man,  1904),  V.  pscudogjahra  Kreis,  1932, 
V.  duhiosa  Kreis,  1932,  V .  fatigans  Filipjev, 
1946,  all  of  which  are  known  only  as  juve- 
niles or  females,  and  V.  pelliicida  (Cobb, 
1898)  nee  Allgen,  1959,  of  which  no  figures 
were  given.  Tlie  statement  in  the  key  by 
^^'ieser  ( 1953 ) ,  copied  from  Kreis  ( 1934 ) , 
to  the  effect  that  in  V.  linsfoici  the  excre- 
tory pore  lies  only  two  stomatal  lengths 
from  the  anterior  end,  was  based  on  an 
erroneous  translation  of  de  Man's  text. 

Viscosia   oncholoimelloides  new  species 
Plate  XI,  fig.  21  a-c 

L  =  1.95  mm;  w  =  22  /x;  esophagus  =  300 
/x;  tail:  6  ,  140  p..  Head  diameter  12-13  jx. 
Lips  distinct;  cephalic  papillae  distinct. 
Buccal  cavity  separated  into  two  chambers 


260        Bulletin  Museum  of  Compomtive  Zoology,  Vol.  135,  No.  5 


by  cuticular  ring;  anterior  portion  5  /x,  pos- 
terior portion  S-9  /x  deep.  The  large  sub- 
ventral  tooth  is  anchored  to  the  buccal 
wall  from  the  cuticular  ring  back  to  about 
the  middle  of  the  posterior  chamber;  from 
there  on  the  buccal  \\'all  is  weakly  devel- 
oped, resembling  the  condition  in  Oncho- 
luimcUus.  The  two  smaller  teeth  are  nor- 
mally developed  as  in  other  species  of 
Viscosia  and  not  reduced  as  in  Onchohimcl- 
Iiis.  Large  subventral  tooth  10  /i  long. 
Pharyngeal  valve  7-8  /x  behind  buccal  cav- 
ity. Scattered  papillae  in  cervical  region. 
Excretory  pore  just  behind  nerve  ring,  165 
Ij.  behind  anterior  end.  Male:  spicula  dag- 
ger-shaped, 17  IX  long  (1  a.b.d.).  There 
are  five  pairs  of  circumanal  papillae  and 
one  more  pair  of  preanal  and  postanal 
papillae,  respectively.  Scattered  papillae 
on  tail. 

HoJotijpc  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4073,  Type  slide  No.  70. 
Type  locality,  M-8,  Biscayne  Bay. 

Representation  in  samples  studied. — M- 
8,  Biscayne  Bay. 

Viscosia  popillafa  Chitwood,  1951 
Plate  XII,  fig.  22  a-f 


Viscosia    })(i]uU(it(i    Chitwood,    1951:    627,    fij 
k-ii. 


1, 


L=  6,  1.18-1.24  mm,  9,  1.04  mm;  w  = 
39-47  /x;  diameter  at  base  of  esophagus  39- 
40  p..  Esophagus  235-250  p.  long.  Tail  120- 
12.5  fji  long.  Vu  =  67%.  Head  diameter  16  /x. 
Labial  capsule  well  developed,  cephalic 
papillae  distinct.  Amphid  in  male  6  /x,  in 
female  5  jj.  wide,  %  of  corresponding  body 
diameter.  Buccal  cavity  21-22  /x  deep,  its 
walls  particularly  strong.  Excretory  pore 
130-145  /x  from  anterior  end,  1.5-20  /x  behind 
ner\'e  ring.  Excretory  cell  30  /x  long,  40  /x 
behind  base  of  esophagus.  Spicula  24-25 
IL  long.  Male  with  circumanal  ala  contain- 
ing six  pairs  of  papillae,  the  posterior  pair 
the  mo.st  prominent.  In  addition  a  pair  of 
preanal  setae  is  present. 

Representation  in  samples  studied. — M-7, 


Everglades  National  Park;  V,  Vero  Beach. 
Geographical  distribution. — Copano  Bay, 
Texas  (Chitwood,  1951),  Chesapeake  Bay, 
Maryland  (Timm,  1952). 

Viscosia  macramphida   Chitwood,   1951 
Plate  XII,  fig.  23,  a-d 

Viscosia  macramphida  Chitwood,  1951:    627,  fig.  1, 

L  =  1.5-1.8  mm;  w  =  40  /x;  diameter  at 
base  of  esophagus  34-39  /x.  Esophagus  260- 
305  /x  long.  Tail  180-195  /x  long.  Head 
diameter  1.3-16  /x.  Labial  capsule  in- 
distinct, cephalic  papillae  distinct.  Amphid 
7  IX  wide.  Buccal  cavity  20-21  ix  deep. 
Excretory  pore  30  /x.  behind  nerve  ring. 
Spicula  23-25  fx  long.  Male  with  4-5  pairs 
of  setose  papillae  associated  with  circum- 
anal ala  as  in  V.  papillata. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne;  M-8,  Biscayne  Bay. 

Geographical  disiribution. — Aransas  Bay, 
Texas   (Chitwood,   1951). 

ENCHELIDIIDAE 
EURYSTOMININAE 
EURYSTOMINA  Filipjev,  1918 
Type  species.— Eurysfomo  specfobile  Marion, 
1870:  20,  21,  pi.  E,  figs.  1-1  b. 

This  genus  has  been  reviewed  by  Inglis 
( 1962 ) ,  who  bases  his  classification  almost 
entirely  on  the  shape  of  the  gubernaculum. 
While  we  agree  that  quite  generally  in 
marine  nematodes  more  emphasis  should 
be  placed  on  the  structure  of  the  male 
genital  armature,  we  feel  this  feature  should 
not  dominate  to  such  an  extent  that  other 
characters  are  ignored.  Thus,  Inglis  con- 
siders E.  americana  Chitwood,  1936,  and 
E.  minutisculae  Chitwood,  sen.su  Timm, 
1952,  to  be  conspecific  because  of  the  simi- 
larity in  the  shape  of  their  gubemacula,  al- 
though (in  Inglis'  own  words):  "Chitwood 
reports  only  one  row  of  denticles  in  the 
buccal  cavity  and  the  absence  of  ocelli 
while  Timm,  in  describing  his  E.  minutis- 
culae, mentions  three  rows  of  denticles  and 
ocelli."    Moreover,  it  could  be  added,  Chit- 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        261 


wood  shows  an  almost  rectangularly  bent 
spiculum,  Timm  a  semiciicular  one,  Chit- 
wood  a  short  terminal  excretoiy  duct,  Timm 
a  long  one,  Chitwood  a  tail  provided  with 
setae,  Timm  a  naked  one,  etc.  On  the  other 
hand,  Inglis  considers  E.  americana  Chit- 
wood,  1936,  noi  to  be  conspecific  with  E. 
americana  of  Chitwood,  1951,  because  of  a 
slight  difference  in  shape  between  the  two 
gubernacula  in  question,  although  all  other 
characters  in  the  two  descriptions  seem  to 
agree. 

This  approach  ignores  the  possibility  of 
small  variations  in  structural  features  and 
overrates  differences  taken  from  the  illus- 
trations of  authors  of  different  reliability. 

Our  material  contained  one  species  which 
is  sufficiently  close  to  £.  miniitiscidae  as 
described  by  Chitwood,  1951,  and  by  Timm, 
1952,  to  identify  it  with  tliis  species.  The 
gubernacula  of  our  male  specimens  are 
very  similar  to  that  figured  by  Timm, 
whereas  the  spicula  more  closely  resemble 
Chitwood's  figure.  The  heads  of  the  speci- 
mens described  by  Chitwood,  by  Timm, 
and  by  us  are  so  similar  that  it  would  seem 
unwarranted  to  refer  them  to  different 
species,  although  we  feel  that  Chitwood 
and  not  Timm  was  correct  in  the  intei*preta- 
tion  of  the  excretory  gland. 

Eurysfomina    minutisculae    Chitwood,    1951 
PlateXIII,fig.  24,  a,  b 

Eunjstomina    minutisculae   Chitwood,    1951:     629, 
fig.  3,  d-g. 

L  =  3.1-3.3  mm;  w  =  45  /x;  diameter  at 
base  of  esophagus  40-45  jx.  Esophagus  660- 
670  /I  long.  Tail  114-122  /x  (3  a.b.d.)  long. 
Head  diameter  20  /x.  Cephalic  setae  9  +  5 
IX.  Buccal  cavity  17-18  /x  long,  separated 
into  two  chambers  by  three  rows  of  denti- 
cles. Amphid  aperture  transversely  oval, 
displaced  dorsally.  Excretory  pore  on  level 
of  amphids;  terminal  excretory  duct  short. 
Ocelli  58  jx  from  anterior  end.  Gland-like 
structures  ( ? )  present  in  anterior  neck  re- 
gion. Spicula  63-66  /x  long,  apophysis  of 
gubernaculum  26-31  /x  long.  Supplements 
70-75  IX  and  130-155  /x  preanal.  Three  pairs 


of  preanal  setae  present,  one  subventral  and 
longer  that  the  two  submedian  pairs;  cuticle 
in  vicinity  of  anterior  pair  thickened. 

Representation  in  samples  studied. — M-1, 
Kev  Biscayne;  M-7,  Everglades  National 
Park. 

Geographical  distribution. — Aransas  Bay, 
Texas  (Chitwood,  1951),  Chesapeake  Bay, 
Maryland  (Timm,  1952),  ?  San  Salvador 
(Gerlach,  1955:    E.  aff.  minutisculae). 

ENCHELIDIINAE 
ILUUM  Cobb,  1920 

Type    species.— ////um    ex/7e    Cobb,     1920: 
261,262,  fig.  45. 

In  1920  Cobb  described  a  genus,  Illium, 
which  has  never  been  found  again  until  noN\'. 
Our  material  contains  two  species  of  which 
one  seems  to  be  Cobb's  I.  exile,  while  the 
other  is  new.  So  far  no  males  are  known 
in  this  genus,  for  which  reason  we  have 
to  deviate  from  our  rule  and  base  the  fol- 
lowing two  descriptions  on  females  only.  The 
shape  of  the  buccal  cavity  is  very  much 
as  in  Symplocostoma  or  Fohjgastrophora 
except  that  there  are  no  teeth.  This  might 
also  be  the  reason  for  the  weaker  develop- 
ment of  the  cuticular  rings  surrounding  the 
buccal  cavity.  Our  two  species  are  easily 
distinguished  by  the  dimensions  of  head, 
amphids,  and  buccal  cavity,  as  well  as  by 
the  position  of  the  amphidial  "sensilla"  with 
respect  to  the  buccal  cavity. 

iWium  exile  Cobb,  1920 
PlateXIII,  fig.  25a-d 

lUium  exile  Colib,  1920:    261,  262,  fig.  45 

L  =  2.02-2.22  mm;  w  =  23-30  /x;  esopha- 
gus =  438—492  ix;  nerve  ring  =  240  /x  behind 
anterior  end;  tail  =  110-122  ,x;  Vu  =60%. 
Head  rounded,  with  six  lips  and,  perhaps, 
minute  labial  papillae.  There  might  be  a 
second  circle  of  cephalic  papillae,  but  all 
these  organs  are  extremely  difficult  to  see. 
Buccal  cavity  consisting  of  vestibulum  and 
two  chambers,  separated  by  two  cuticular 
rings;  the  two  chambers  measure  7  /x  in 
length,  greatest  width  is  3  /x.  Amphids  con- 


262         BuUctin  Museum  of  Compavalivc  Zoology,  Vol  135,  No.  5 


sisting  of  opening,  about  3  /x  wide,  pouch, 
duct  and  an  unusual  structure  which  we 
call  "sensilla."  There  is  no  indication  of  an 
ocellus  as  assumed  by  Cobb.  Distance  of 
sensilla  from  anterior  end  =  12  /x,  c.b.d.  = 
10  fx.  Excretory  pore  a  short  distance  be- 
hind nei-ve  ring.   Anal  body  diameter  17  /x. 

Representation  in  samples  studied. — M-S, 
Biscayne  Ra>'. 

Geo<iraphieal  distribution. — Jamaica 
(Gobi),  1920). 

////urn  libidinosum  new  species 
Plate  XIII,  fig.  26,  a,  b 

L  =  2.28  mm;  w  =  70  /x;  esophagus  =  570 
/x;  nerve  ring  =  258  p.  behind  anterior  end; 
tail  =  200  /-.;  Vn  =  60?r .  Head  as  in  the  fore- 
going species  but  much  larger.  Buccal 
cavity  10  X  5  /x,  walls  more  cur\ed  than  in 
the  foregoing  species.  Amphids  5  /x  wide, 
sensilla  15  fx  behind  anterior  end,  filled  with 
fragments  of  a  dense  material.  Excretory 
pore  25  fx  behind  nerve  ring.  Anal  body 
diameter  33  p.. 

Ilolotype  specimen. — Female;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4073,  Type  slide  No.  71.  Type 
locality,  M-8,  Biscayne  Bay. 

Representation  in  samples  studied. — M-8, 
Biscayne  Bay. 

POlYGASJROPhORA  de  Man,  1922 
Type    species.— Po/ygasfrop/iora    aiienua\a 
de  Man,  1922a:    131,  132. 

In  this  genus  four  species  each  possessing 
seven  esophageal  bulbs  have  been  described, 
to  which  we  shall  add  a  fifth  one.  Of  these 
five  species,  P.  tenuicoUis  (Allgen,  1951) 
is  best  considered  a  species  inquirenda  as 
advocated  by  Chitwood  ( 1960 ) ,  and  not 
a  synonym  of  P.  heptabulha  as  suggested 
by  Wieser  ( 1953 ) .  The  remaining  four 
species  are  separated  mainly  by  the  arrange- 
ment of  their  cephalic  setae,  but  we  are 
not  so  sure  whether  this  character  will  turn 
out  to  be  as  stable  as  we  assume  at  present. 


List   and  Short  Characterization  of  Valid 

PoLVGASTROPHORA  spp.  Fosscssiug,  Seven 

Esop]uig,eal  Bulbs 

P.  maior  Sehulz,  1932:  Submedian  ce- 
phalic setae  in  both  sexes  very  unequal  in 
length  (1-2  and  ^f>  of  c.b.d.,  respectively). 
Excretory  pore  130-150  /x  from  anterior  ex- 
tremity. Spicula  length  =  %  of  tail.  Guber- 
naculum  conical. 

P.  heptabulba  Timm,  1952:  Submedian 
cephalic  setae  reduced  in  £  ,  one  head 
diameter  long  and  subequal  in  9  .  Excretory 
pore  44  /x  behind  anterior  end.  Spicula 
length  =  Vj  of  tail.    Gubernaculum  absent. 

P.  septemhulba  Gerlach,  1954:  Subme- 
dian cephalic  setae  subequal  ( 7-6  /x )  in  S  , 
very  unequal  in  5  (shorter  ones  about  2  /x). 
Excretory  pore  41-52  fx  from  anterior  end. 
Spicula  length  =  %  to  %  of  tail.  Guber- 
naculum alxsent  (Gerlach,  1954)  or  small 
(Chitwood,  1960). 

P.  eda.x  n.  sp.:  Submedian  cephalic  setae 
very  unequal  in  length  in  both  sexes  (8- 
7  +  2  /x ) .  Excretory  pore  48-50  ^  behind 
anterior  end.  Spicula  length  =  ^-i  of  tail. 
Gubernaculum  plate-shaped. 

Our  new  species  P.  eda.x  is  related  to  P. 
septembulba  from  which  it  can  be  distin- 
guished by  the  characters  presented  in  the 
above  list. 

Polygostrophora  edax  new  species 

Plate  XIV,  fig.  27,  a-c 

L  =  2.65-2.90  mm;  w  =  105-110  ^;  esoph- 
agus =  550-600  p.;  tail:  <i  ,  175  /x,  9  ,  165 
fx  (3.5-4  a.b.d.);  Vu  =  53%.  Head  diam- 
eter 10  /<.  Mouth  opening  crenate.  No 
labial  papillae.  Cephalic  setae  in  6,7  +  2 
^,  in  9,8  +  ?,  indistinct.  Amphids  oval,  4.5 
IX  wide  in  S  ■  Buccal  cavity  14  X  7  /x,  with 
two  faint  cuticular  rings  at  about  its  middle 
and  one  strong  basal  band  that  is  resolvable 
into  elongated  cuticular  bodies.  Refractory 
bodies  (ocelli?)  18-20  ^a  behind  anterior 
end.  Excretory  pore  48-50  /x  behind  an- 
terior end,  ampulla  about  72  fx.  Male: 
Spicula  80  IX,  knobbed  proximally.  Guber- 
naculum faint,  plate-shaped.  Subventral 
circumanal   setae  and  along  the  tail. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        263 


, TIHI    -. 

b 


:^ 


V 


Figure    2.      Types    of    preanal    supplements    in    the    Cyatholaimidae.      a — Pomponemo     (type    A);     b — Longicyatholaimus     (type 
B);     c — Paracanthonchus     (type     C-1);     d — Paracyatholaimus     (type    C-2,    two    forms). 


Holotypc  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection 
Number  4075,  Type  slide  No.  72.  Type 
locality,  V,  Vero  Beach. 

Representation  in  samples  studied. — V, 
Vero  Beach. 

CYATHOLAIMIDAE 
CYATHOLAIMINAE 

The  genera  within  the  subfamily  Cy- 
atholaiminae  can  be  divided  into  t\vo  groups 
based  on  the  presence  or  absence,  in  the 
male,  of  preanal  supplements.  In  those 
genera  in  which  the  males  possess  preanal 
supplements,  a  further  division  is  possible 
utilizing  the  structure  and  arrangement  of 
these  organs.  \\'ieser  (1954)  presented  a 
key  to  the  genera  of  this  subfamily,  a  key 
that  still  remains  useful,  but  requiring  sev- 
eral additions  and  emendations. 

The  genera  with  preanal  supplements 
can  be  separated  into  the  following  three 
basic  types  (A,  B,  C)  of  which  the  last 
is  subdivided  into  forms  with  "tuboid"  sup- 
plements (C-1)  and  fonns  with  "setose" 
supplements    (C-2)    (see  Text-fig.  2). 

Type  A )  Supplements  large,  complicated, 
consisting  of  several  elements;  cuticle  be- 
tween supplements  lamellated.  Genera  in- 
cluded: Fomponema  Cobb  (syn.  Endolai- 
miis  Filipjev),  Niimmoceplialus  Filipjev 
(syn.  Haustrifera  Wieser),  Craspodema 
Gerlach  (syn.  Kraspedonema  Gerlach)  and 


Aiui.xuncJiiiim  Cobb.  Tentatively  included 
is  Dispira  Cobb. 

Type  B )  Supplements  cup-shaped,  large 
to  minute,  with  narrow  ducts  leading  to 
the  cups.  Genera  included:  Longicyatholai- 
mus  Micoletzky,  Xyzzors  Inglis,  and 
Biarmifer  Wieser. 

Type  C-1 )  Supplements  tubular,  large  to 
medium.  Genera  included:  Paracanthon- 
chus Micoletzkv,  Acanthonchus  Cobb, 
SeuratieJhi  Ditlevsen  and  ParaseuraticUa 
Schuurmans-Stekhoven. 

Type  C-2)  Supplements  "setose,"  small. 
This  type  can  probably  be  derived  from 
type  C-1,  the  "setose"  papillae  described  by 
some  authors  most  likely  being  minute 
tubuli  with  very  narro\\'  ducts.  Genera  in- 
cluded: Paracyatholaimus  Micoletzky  (in 
Text-fig.  2,  represented  by  two  forms),  and 
Paraeyatholaimoides  Gerlach. 

Those  genera  that  definitely  do  not 
possess  preanal  supplements  are:  Cyatholai- 
mus  Bastian,  XenocyatJwkiimus  Gerlach, 
MetaeyathoJuimus  Schuurmans-Stekhoven, 
and  PhyUohimus  Murphy. 

Other  genera  are  considered  doubtful. 

There  are  two  main  difficulties  in  classi- 
fying genera  of  this  subfamily:  first,  separa- 
tion of  type  C-1  from  type  C-2,  since  the 
difference  between  small  tubuli  of  the 
Paracanthonclius-type  and  the  so-called 
"setose  papillae"  of  Paracyatholaimus  may 
be  only  slight;  second,  deciding  whether 
a  species  is  devoid  of  supplements  or  pos- 


264 


Bulletin  Muscuin  of  Couiixiiativc  Zoology,  Vol  135.  No.  5 


sesses  small,  cup-shaped  papillae  of  the 
type  B.  A  case  in  point  is  Lon<i,icycit1ioIaimiis, 
in  which  species  with  and  species  without 
supplements  ha^•e  been  deserib(>d  (see  be- 
low ) . 

The  genus  Choniokiimus,  referred  to  this 
subfamily  by  many  authors  and  included  in 
his  key  by  Wieser  ( 1954 ) ,  will  have  to  be 
transferred  to  the  Choanolaiminae  since  the 
redescription  of  the  type  species,  CJi.  papil- 
latiis  Ditle\'sen,  bv  Cerlach  ( 1964),  and  the 
description  of  Ch.  jHuiiciis-  Gerlach,  1957, 
have  made  it  clear  that  in  this  genus  the 
buccal  cavity  is  of  a  shape  (luite  different 
from  oth(n-  cyatholaimids.  Consequently,  the 
species  referred  to  Clioniolaimus  on  the 
strength  of  their  preanal  supplements  but 
with  a  typical  cyatholaimid  buccal  cavity, 
i.e.,  Ch.  macrodcntatus  Wieser,  1959,  and 
Ch.  iciescri  Inglis,  1963,  are  to  be  trans- 
ferred to  Lon0cijathoJ(iimus  (see  below, 
p.  265).  CyathoJaimu.s  fautraensis  Allgen, 
referred  to  Chonioloimus  by  Wieser  ( 1954) 
is  better  placed  with  Niimmocephalm  ( see 
Gerlach,  195Sb). 

POMPONEMA  Cobb,  1917 
Type  species.— Pomponema  mirobile  Cobb, 
1917:    118,  fig.  3. 

This  genus  is  characterized  by  the  strong 
developmcMit  of  the  l:)uccal  cavity,  in  which 
the  vestibular  ribs  seem  to  function  as 
particularly  mobile  clasping  organs;  the 
dorsal  tooth  is  powerful  and  opposed  either 
b\  two  strong  subventral  teeth  or  by  a 
great  number  of  denticles.  Further  char- 
acteristics are  the  heterogeneous  cuticular 
ornamentation,  the  lateral  differentiation 
of  the  latter,  and  the  peculiar  male  supple- 
ments. Closely  related  to  Fomponeina  is 
the  genus  Nt(mmocej)haIus  Filipjev  (syn. 
Housirifera  Wieser)  which  has  less  well 
developed  vestibular  ribs  and  teeth,  and 
in  which  the  cuticular  markings  are  more 
uniformly  dot-like.  Lon^icijatholmmus  line- 
dins  Gerlach,  1952,  which  was  referred  to 
Pomponema  by  Wieser  (1959),  is  perhaps 
better  placed   with   Nummoceplialus. 


Key  to  Species  of  Pomponema 

1.  Lateral    differentiation    beginning    on    level 

with  buccal  ca\ity;  in  the  cervical  region 
the  2  longitudinal  rows  are  half  the  c.b.d. 
apart  __.-  P.  multipapillatum  (Filipjev,  1922) 
Lateral  differentiation  beginning  approxi- 
mately at  end  of  esophagus;  2-4  longitu- 
dinal rows  which  are  far  less  apart  than 
half   the   c.b.d.    2 

2.  Cephalic  setae  in  two  circles,  4  -f  6  

P.   segregatuin    Wieser,    1959 
Cephalic  setae  in  one  circle  3 

3.  Suliniedian  pairs  of  cephalic  setae  very  un- 

equal in  length.    Dorsal  tooth  opposed  by 
groups     or     rows     of     smaller     teeth     or 

denticles  4 

Submedian  pairs  of  cephalic  setae  subequal. 
Dorsal  tooth  opposed  by  two  subventral 
teeth  5 

4.  Male   amphids    6-6.5    turns,    50%    of    c.b.d. 

wide  P.  stomachor  Wieser,  1954 

Male  amphids  4.5  turns,  35%  of  c.b.d.  wide 
P.  pohjdonta  Murphy,  1963 

5.  Male    amphids    6-6.5    turns,    90%    of    c.b.d. 

wide P-  mirahilc  Cobb,  1917 

Male  amphids  4.5  turns,  65%  of  c.b.d.  wide  .. 
P.   tcsscldtuin   n.   sp. 

Pomponema  tesselatum  new  species 
Plate  XIV,  fig.  28,  a-d 

L  =  1.53  mm;  w  =  42  /x;  esophagus  =  360 
/x.  Head  diameter  25  /x.  Labial  setae  14  ^. 
Cephalic  setae  16  +  13  ^.  Buccal  cavity 
spacious,  vestibulum  protrusible,  its  ribs 
giving  the  appearance  of  prehensile  clasp- 
ing organs  which  are  linked  by  joints  to 
the  buccal  wall;  dorsal  tooth  large,  hollow, 
opposed  by  two  similar  though  smaller 
subventral  teeth.  Amphids  19  ^i  =  68%  of 
c.li.d.  wide,  describing  4.5  turns.  Cuticular 
ornamentation  complex  and  heterogeneous; 
in  cer\'ical  region  each  annule  at  high  focus 
with  slit-like  markings,  at  low  focus  of 
tesselated  appearance;  what  are  seen  as 
dots  in  lateral  view  are  actually  short 
columns  between  the  outer  and  the  inner 
layer  of  the  cuticle;  these  columns  become 
rather  thin  from  the  mid-cervical  region  on, 
and  the  dots,  consequently,  smaller.  There 
are  two  rows  of  dots  per  annule.  A  lateral 
differentiation  in  the  form  of  four  longitu- 
dinal rows  of  larger  and  more  widely  spaced 
dots  begins  at  about  the  end  of  the  esopha- 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        265 


gus.  Many  pores  are  arranged  in  more  or 
less  regular  longitudinal  rows  all  along  the 
body.  Spicula  45  /x,  gubernaculum  36  /x, 
consisting  of  two  portions,  the  distal  one 
with  lateral  projections.  There  are  15  pre- 
anal  supplements  of  characteristic  shape. 
Between  the  supplements  the  cuticle  gives 
a  lamellated  appearance.  One  preanal  seta 
and  many  setae  on  the  tail  in  four  longitu- 
dinal rows.    Tail  132  /x,  a.b.d.  36  /x. 

Holofype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4068,  Type  slide  No.  73.  Type  locality, 
M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M-3, 
Key  Biscayne. 

LONGICYATHOLAIMUS  Micoletzky,  1924 
Type  species.— Cyafholaimus  longicaudatus 

de  Man,  1878:    111,  112,  pi.   10,  fig. 

16,  a-c. 

This  genus  is  characterized  by  the  spicu- 
lar  apparatus,  which  is  of  a  general  shape 
found  in  some  species  of  Faracanihonchus 
and  Paracyatholaimiis,  in  combination  with 
the  fact  that  the  preanal  supplements  are 
never  setose  or  tubular.  In  the  type  species, 
L.  longicaudatus  (de  Man),  indistinct  papil- 
lae of  type  B  (see  above,  p.  263  were  ap- 
parently seen  by  de  Man  ( 1878 )  and  by 
Kreis  ( 1928 )  but  not  by  Schuurmans-Stek- 
hoven  (1943).  Since  then  other  species  have 
been  described  in  which  supplements  were 
not  mentioned.  It  would  simplify  matters 
if  one  were  permitted  to  assume  that  in  all 
these  cases  the  small  cup-shaped  supple- 
ments were  so  indistinct  as  to  have  been 
overlooked,  but  the  possibility  that  there 
exist  species  truly  devoid  of  supplements 
must  be  considered.  In  such  cases  dif- 
ferentiation from  other  genera  without  sup- 
plements, particularly  from  Cijatholaimus, 
would  have  to  be  based  on  the  shape  of  the 
spicular  apparatus  and,  to  a  lesser  extent, 
on  the  shape  of  the  tail. 

Further  characteristics  mentioned  by 
Micoletzkv  are  the  lateral  differentiation  of 


the  cuticular  ornamentation  and  the  long, 
filiform  tail.  However,  species  with  fairly 
short,  though  always  slender,  tails  have  been 
described,  which  in  other  respects  fit  the 
generic  diagnosis. 

Classification  of  the  species  is  rendered 
difficult  by  the  uncertainty  as  to  how  the 
reputed  absence  of  preanal  supplements  is 
to  be  judged.  For  the  present,  we  shall 
accept  this  feature  at  its  face  value.  This 
provides  for  an  immediate  separation  of 
the  species  into  two  groups.  Group  A  con- 
taining species  described  with  supplements, 
and  Group  B,  containing  species  in  which 
no  supplements  were  reported.  The  species 
contained  within  these  groups  are  as  fol- 
lows : 

Longicyatliolaimus  species  group  A:  L. 
longicaudatus  (de  Man,  1878);  L.  minor 
(Cobb,  1898);  L.  effilatus  (Schuurmans- 
Stekhoven,  1946);  L.  .stekhoveni  Wieser, 
1954  (syn.  L.  effikitus  Schuurmans-Stek- 
hoven,  1950  nee  1946);  L.  quadriseta  Wieser, 
1954  nee  1959;^  L.  macrodentatus  (Wieser, 
1959)  new  combination  (syn.  Choniohimus 
macrodentatus)  and  L.  tvieseri  (Inglis, 
1963)  new  combination  (syn.  Choniolaimus 
uieseri). 

Longicyatholaimus  species  group  B:  L. 
trichurus  (Cobb,  1898);  L.  zo.sterae  Allgen, 
1933;  L.  ehoanolaimoides  (Schuunnans- 
Stekhoven,  1942);  L.  continus  Filipjev, 
1946;  L.  duhius  Filipjev,  1946;  L.  trichocauda 
Gerlach,  1955,  and  L.  dayi  Inglis,  1963. 

Species  of  doubtful  status  are:  L.  heteru- 
/•j/.s  (Cobb,  1898);  L.  tenuicaudatus  (Saveljev, 
1912)  and  L.  fiJicaudatus  Schuurmans-Stek- 
hoven,  1950. 

As  the  species  we  found  belongs  to  group 
A,  a  key  to  the  species  of  this  group  is 
provided. 


^  Note:  The  species  described  as  L.  quadriseta 
Wieser  liy  Wieser,  1959,  is  equipped  with  setose 
and  not  cup-shaped  supplements.  Since  there  are 
other  differences  from  the  type  (smaller  amphids, 
shape  of  gubernaculum)  we  consider  L.  quadriseta 
Wieser  sensu  Wieser,  1959,  to  be  a  different  species 
and  refer  it  to  Paracijatholaimus  under  the  name 
Paracyatholaitniis  pugetteiisis  new  name  and 
new   combination. 


266         Bulletin  MimiiDi  of  Comparative  Zoology.  Vol.  135,  No.  5 


Key  to  Species  of  Longicyatholaimus  Group  A 

1.  Lateral  differentiation  of  cuticle  in  4  longitu- 

dinal rows  L.  minor  (Cobb,  1898) 

Lateral  differentiation  of  cuticle  irregular  .    2 

2.  Posterior  portion  of  tail  filifomi,  much  longer 

than  conical  portion  

L.  lon^icaudattis  (de  Man,  1878) 

Posterior  portion  of  tail  cylindrical,  not  longer 
than    conical    portion    3 

3.  Gubernaculuni   distall>    truncate  or  notched: 

3  preanal  supplements   — L.   effilatus 

(Schuunnans-Stckhovcn,  1946)  and  L.  stck- 
hoveni  Wieser,  1954 
Gubernaculum  distally  with   well   dexeloped 
teeth    or    processes;    6-7    preanal    supple- 
ments     4 

4.  Male    amphids   about    60%    of   c.b.d.    wide; 

gubernaculum    distally    witli    three    digiti- 
form  processes  of  rather  unequal  shape  „___ 

L.   wieseri    (Inglis,    1963)    n.   comb. 

Male  amphids  30-40%  of  c.b.d.  wide:  guber- 
naculum distally  widi  3-4  equal-sh:iped 
teeth  5 

5.  Cephalic  .setae  digitiform,  8-10  +  5-6  /n  long. 

Preanal  supplements  5  +  2,  the  5  anterior 
ones  regularly  spaced:  gubernaculum  with 

3  distal  teeth  

___   L.  macrodentatus  (Wieser,  1959)  n.  comb. 
Cephalic    setae    conical,    slender,    subequal, 
12  -f  13  II  long;  preanal  supplements  4  +  2, 
the    first    one    almost    three    times    as    far 
from  the  2nd  as  the  latter  from  the  3rd; 

gubernaculum  with  4  distal  teeth  

L.  annac  n.  s-p. 

Longicyatholaimus  annae  new  species 
Plate  XV,  fig.  29,  a-c 

L  =  1.96  ami;  w  =  70  /x;  esophagus  =  300 
/J..  Head  diameter  33  /x.  Labial  papillae 
setose,  stout,  4.5  /x  long.  Cephalic  setae 
13  +  9  ix.  Buccal  cavity  deep,  with  one  large 
dorsal  tooth,  two  small  subventral  teeth 
and  cuticular  ridges.  Amphids  in  i  13 
/x  =  31%  of  c.b.d.  wide,  4.5  turns.  A  group 
of  dorsolateral  cervical  setae  a  short  dis- 
tance behind  amphids.  Cuticle  with  annules 
and  homogeneous  rows  of  dots;  no  lateral 
differentiation  except  on  tail;  there  is  one 
row  of  dots  per  annule  in  the  anterior  cervi- 
cal region,  but  two  rows  can  be  found  on 
the  remainder  of  the  body.  Many  pores  ar- 
ranged in  more  or  less  longitudinal  rows. 
Spicula  70  /x,  somewhat  S-shaped,  with 
velum   in   distal  half,  knobbed  proximally. 


Gubernaculum  64  ^,  stout,  distally  with  4 
equal-sized  teeth.  There  are  6  small  but 
distinct  cup-shaped  preanal  supplements, 
the  anteriormost  one  at  a  distance  of  210- 
230  fj.  preanally;  the  distance  between  the 
first  and  the  second  papillae  is  about  as 
great  as  that  between  the  latter  and  the 
anus.  Tlie  two  posteriormost  papillae  are 
more  closelv  spaced  than  all  the  others. 
Tail  420  /x,  a.b.d.  60  /x. 

Holotype specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4068,  Type  slide  No.  74.  Type  locality, 
M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M-3, 
Key  Biscayne. 

Remarks. — This  species  is  named  after 
Mrs.  Ann  Hopper,  friendly  hostess  to  wan- 
dering nematologists. 

XYZZORS  Inglis,  1963 

Type  species.— Xyzzors  fifzgeraldae   Inglis, 
1963:  544-546,  figs.  25-29. 

According  to  Inglis  ( 1963 )  Xyzzors  is 
characterized  by  irregular  lateral  differen- 
tiation of  the  cuticle,  cup-shaped  preanal 
papillae,  large  and  well-developed  buccal 
armature  and  some  structural  peculiarities 
of  the  spicular  apparatus.  However,  none 
of  these  characters  separates  Xyzzors  un- 
equivocally from  Longicyatholaimus.  The 
cuticle  and  the  supplements  are  of  the  same 
type  as  found  in  the  latter  genus,  the 
gubernaculum  is  somewhat  larger  than  in 
most  species  of  Lon<^icyathoJaimus  but, 
e.g.,  L.  dubiiis  Filipjev,  1946,  has  a  guber- 
naculum of  exactly  the  same  shape.  More- 
over, the  figures  given  by  Inglis  do  not 
justify  his  statement  that  "the  spicules  are 
much  more  elaborate  than  is  usual  in 
species  of  the  Cyatholaimidae."  The  two 
featiues  which  could  perhaps  serve  as  char- 
acters distinguishing  Xyzzors  from  Longi- 
cyatJiolaimus  are  the  nearly  conical  tail 
and  the  buccal  annature  which  indeed 
seems  to  be  somewhat  more  elaborate  than 
observed  in  the  latter  genus. 

In  oiu"  new  species  the  spicular  apparatus 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        267 


is  simpler  than  in  X.  fitz^i^eraldac  Inglis, 
and  the  proximal  ends  of  the  spicula  are 
not  doubled.  There  are  3  cup-like  supple- 
ments followed  posteriorly  b\'  two  indistinct 
ducts,  whereas  Inglis  reports  6  cup-like  sup- 
plements. The  amphids  describe  4  turns  in 
our  species  as  against  6.75  in  A',  fitzgeraldae. 

Xyzzors  inglisi  new  species 
Plate  XV,  fig.  30,  a-c 

L  =  1.25  mm;  w  =  54  ^;  esophagus  =  240 
/x.  Head  diameter  28  /x.  Labial  papillae 
stout.  Cephalic  setae  12  +  10  /x.  Buccal 
cavity  deep,  with  one  large  dorsal  tooth,  2 
pairs  of  subventral  teeth  and  one  conspicu- 
ous ridge  that  surrounds  the  ventral  half  of 
the  buccal  cavity.  Amphids  16  /i  =  487c  of 
c.b.d.  wide,  4  turns.  Cuticle  as  in  the  fore- 
going species.  Spicula  52  /x,  with  xelum; 
proximally  the  inner  edges  are  more  strongly 
cuticularized,  distally  the  outer  edges. 
Gubernaculum  38  fx,  distally  expanded  and 
dentate.  There  are  3  preanal  supplements 
that  consist  of  a  cup-shaped  portion  ( pro- 
truded in  Fig.  30,  b,  c)  and  a  duct  leading 
to  the  latter.  The  distances  are:  25  ix  from 
anus  to  posterior  papilla,  45  ix  from  this  to 
the  next  one,  30  /x  to  the  anterionnost  one. 
Between  the  last  papilla  and  the  anus,  two 
minute,  indistinct  ducts  can  be  discerned. 
Tail  conical,  108  /x,  a.b.d.  44  /x. 

Uolotype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4075,  Type  slide  No.  75.  Type  locality, 
V,  Vero  Beach. 

Representation  in  samples  studied. — V, 
Vero  Beach. 

PARACAN7HONCHL/S  Micoletzky,  1924 
Type  species.— Cyafho/a/mus  coecus  Bastion 

sensu   de  Man,    1889b:     204-207,   pi. 

7,  fig.  10,  a-g. 

The  species  of  this  genus  were  grouped 
by  W'ieser  ( 1954 )  who  used  the  shape  of 
the  gubernaculum  as  the  main  distinguish- 
ing feature.  The  shapes  of  gubernacula 
representing    species   belonging   to    groups 


A,  B,  and  C,  respectively,  in  Wieser's  key, 
are  shown  in  Text-figure  3.  There  are,  how- 
ever, species  which  do  not  fit  readily  into 
these  three  categories,  for  example,  those 
that  have  a  large  gubernaculum  subtermi- 
nally  dilated  and  tapering  towards  a  pointed 
or  spoon-shaped  distal  tip.  This  type  of 
gubernaculum  represents  a  transition  be- 
tween groups  A  or  B  and  C,  and  is  also 
shown  in  Text-figure  3.  Species  with  such 
a  gubernaculum  are  mainly  P.  mens  Wieser, 
1954,  and  the  new  species  to  be  described 
below,  but  P.  anii^idatiis  ( Schuurmans-Stek- 
hoven,  1950),  P.hatidus  Gerlach,  1957,  and 
P.  miitatus  \\'ieser,  1959,  come  rather  close. 
P.  platypus  n.  sp.  is  separated  from  its 
closest  relative,  P.  mens  \\^ieser,  1954, 
mainly  by  the  number,  size,  and  arrange- 
ment of  the  preanal  tubuli,  and  by  the  size 
of  the  spicular  apparatus. 

Paracanthonchus  platypus   new   species 
Plate  XVI,  fig.  31,  a-c 

L  =  1.18-132  mm;  w  =  44-48  jx;  esopha- 
gus 190  /.;  tail:  6  ,  135  /x,  9  ,  105  ^;  Vu  =  48%. 
Head  diameter  21-23  jx.  Labial  papillae 
conical.  Cephalic  setae  4  -h  5  /x.  A  short 
cephalic  capsule  present.  Buccal  cavity 
with  well-developed  vestibular  ribs  and 
with  medium-sized  triangular  tooth.  Am- 
phids in  £,llix  —  40%  of  c.b.d.,  in  $  ,  9 
/x  =  32%  of  c.b.d.  wide.  Excretory  pore  27- 
32  jx  behind  anterior  end.  Ocelli  47-50  fx 
behind  anterior  end,  with  fibrils  running 
from  the  pigment  spot  forward  and  back- 
ward, a  bit  reminiscent  of  the  structures 
described  for  Acanthonchus  rostratus  by 
Murphy  ( 1963 ) ,  but  a  proper  lens  was  not 
seen.  Cuticular  ornamentation  homoge- 
neous. Spicula  36  ^i,  gubernaculum  35  /x, 
spoon-shaped,  with  subterminal  dilation  and 
tapering  towards  the  tip;  there  is  a  lateral 
projection.  Preanally  there  are  4  tubuli, 
22-23  /x  long,  one  large  spine  (Fig.  31,  c), 
and  two  subventral  rows  of  slender  setae. 
On  the  tail  there  are  short  setae  and  a 
characteristic  ventral  pair  of  long  setae. 
Anal  body  diameter  43  /x. 

Holotype  specimen. — Male;  Canadian  Na- 


268 


Bulletin  Museum  of  Coiujxirdiivt'  Zoology,  Vol.  135,  No.  5 


Figure    3.      Types    of    gubernacula     in    Paracanthonchus.      a — P.    strandensis    (after    Schuiz,    1932);    b — P.    coecus    (after  Timm, 
1952);    c— P.    cochlearis    (after    Gerloch,    1957);    d— P.    platypus  n.  sp. 


tioiial  Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4073,  Type  slide  No.  76.  Type  locality, 
M-8,  Biscayne  Bay. 

Representation  in  samples  studied. — M-8, 
Bisca>ne  Bay. 

PARACYATHOLAIMUS  Micoletzky,  1921 
Type   species.— Cyaf/io/a/mus   dubiosus   BUt- 
schli,  1874:    284,  pi.  7,  fig.  31,  a,  b. 

This  genus  is  characterized  by  the  simple 
giibernaculum  which  is  supposed  to  hardly 
expand  distally,  and  in  the  diagnosis  by 
Micoletzky  ( 1924b )  was  described  as  being 
"ohne  Dornen,"  and  by  the  male  supple- 
ments which  Micoletzky  called  "Borsten- 
papillen."  These  supplements  probably  are 
not  true  setae  l)ut  represent  ducts  of  the 
ParacanthoncJiiis-type  with  the  lumina  so 
narrow  as  to  give  the  impression  of  setae. 

Gerlach  ( 1955 )  described  a  species,  P. 
paucipapilhitus,  in  which  the  gubernaculum 
expands  distally  to  form  a  plate  covered 
with  rasp-like  denticles.  The  supplements 
consist  of  strongly  protruding  conical  papil- 
lae with  cuticularized,  narrow  ducts.  The 
first  character  links  Gerlach's  species  to 
ParacantJioncliiis,  the  second  character  sets 
it  apart  from  all  other  species  of  Paracan- 
tli())icliiis  and  ParacyalJiokiiuiiis  with  the 
exception,  perhaps,  of  P.  diii^itatus  Gerlach, 
1957,  in  which  similar  supplements  have 
been  described  and  the  gubernaculum  is 
also  rather  strongly  dentate.  Our  material 
contained   representatives  of  what  at  first 


we  held  to  be  P.  paiicipapillatus.  However, 
on  closer  examination  we  noted  the  follow- 
ing differences:  1)  distal  end  of  guber- 
naculum with  distinct  teeth,  numbering 
from  4-8,  rather  than  with  a  rasp-like  field 
of  denticles,  2)  ducts  of  supplements  more 
elaborate,  and  3)  buccal  cavity  with  one 
large  dorsal  tooth  and  three  small  sub- 
ventral  teeth,  whereas  Gerlach  speaks  only 
of  one  dorsal  tooth. 

We  consider  these  differences  important 
enough  to  establish  a  new  species,  P.  pesams 
n.  sp.,  named  after  the  appearance  of  the 
distal  end  of  the  gubernaculum  which  re- 
sembles a  spread  bird's  foot.  This  species 
and  the  two  described  by  Gerlach  men- 
tioned above  form  a  rather  distinct  group 
within  the  two  genera  Paracanthonchus  and 
Parac\iatholaimus . 

Paracyafholaimus  pesavis  new  species 
Plate  XVI,  fig.  32,  a-e 

L  =  1.08-1.17  mm;  w  =  35-44  ^u;  esopha- 
gus =  195-200  /x.  Head  diameter  18-20  {x. 
I^abial  papillae  distinct.  Cephalic  setae 
7-8  +  5-6  /x.  Short  cephalic  capsule.  Buc- 
cal cavity  with  one  large  pointed  dorsal 
tooth  and  three  small  subventral  teeth. 
Amphids  11  ^a  =  44%  of  c.b.d.  wide,  ap- 
proximately 5  turns.  Cuticular  ornamenta- 
tion with  slight  lateral  differentiation  in 
cervical  and  anal  region.  Spicula  29-31  ix, 
with  velum.  Gubernaculum  22-23  jj.,  ex- 
panding distally  to  a  plate  which  in  one 
specimen    carried    8,    in    another    4    small 


Florida  Marine  Nematodes  •  Wieser  and  Hopper 


269 


teeth  (PL  XVI,  fig.  32,  c,  d ).  There  are  two 
large  conical  papillae  at  a  distance  of  25- 
28  and  50-53  /x,  respectively,  from  the  anus. 
The  papillae  are  penetrated  by  cuticular- 
ized  ducts  which  show  some  distal  elabo- 
rations; their  openings  are  posteriorly  di- 
rected. There  is  a  very  faint  third  papilla 
( 10-13  fx  in  front  of  the  anus )  which  might 
actually  consist  of  two  closely  spaced 
minute  tuliuli  as  found  in  other  species  of 
Favucijatholaimus.  Tail  80-100  jx  long,  nar- 
rowing abruptly  in  distal  third;  a.b.d.  35- 
40  IX.   Spinneret  6-7  ^  long. 

Holotijpe  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4071,  Type  slide  No.  77. 
Type  locality,  M-6,  Everglades  National 
Park. 

Representation  in  samples  studied. — M- 
6,  Everglades  National  Park;  V,  Vero 
Beach. 

Remarks. — The  closely  related  species,  P. 
pancipapiUatus,  was  originally  described 
from  the  Pacific  coast  of  San  Salvador,  and 
later  from  the  Congo  estuary,  and  from 
Brazil  ( Gerlach,  1957a,  b,  c).  The  possi- 
bility that  the  Atlantic  specimens  are  ac- 
tually representatives  of  our  new  species 
is  intriguing  but  remains  to  be  proven. 

HAL/CHOANOM/MUS  de  Man,  1888 
Type   species.— Sp///pf)era   robusta    Bastion, 
1865:   166,  pi.   13,  figs.  226,  227. 

Our  material  contained  two  closely  allied 
species  which  appear  to  represent  H.  qiiat- 
tuordccimpapillatus  Chitwood,  1951,  and 
//.  duodecimpapiUatiis  Timm,  1952.  As 
neither  author  specifically  stated  the  tail 
length  for  their  respective  species,  it  was 
necessary  to  make  use  of  the  de  Man  c 
value  in  identifying  the  two  populations. 
The  two  species  can  be  separated  on  the 
basis  of  the  tail  length  as  well  as  the  shape 
of  the  spicules  and  gubernaculum.  As  the 
heads  of  both  species  are  very  similar,  only 
one  has  been  figured,  H.  duodecimpapiUa- 
tiis. 


Halichoanolaimus     quattuordecimpapillatus 
Chitwood  1951 
Plate  XVII,  fig.  33,  a-c 

Halichoanolaimus     quattuordecimpapillatus     Chit- 
wood, 1951:  639,  fig.  7  c. 

L  =  6  ,  1.7-2.0,  9  ,  2.0-2.4  mm;  w  =  6  , 
60-65,  9  ,  70-90  fx;  diameter  at  base  of 
esophagus,  6  ,  55-70,  9  ,  70-83  p..  Vu  = 
43-45%.  Esophagus  in  6  ,  270-300  p,  in 
9  ,  310-340  p  long.  Excretoiy  pore  in  S  , 
150-155  p,  in  9  ,  162-180  p  from  anterior 
end.  Head  35-40  p  wide,  with  an  internal 
circle  of  6  labial  papillae  and  an  external 
circle  of  10  cephalic  papillae  of  which  the 
laterodorsal  and  lateroventral  pairs  are 
setose.  Amphid  15  p  wide,  16-19  p  from 
anterior  end,  spiral  with  3.75-4  turns.  Buc- 
cal cavity  typical,  containing  3  posterior 
apophyses,  the  anterior  margins  of  which 
bear  comb-like  ribs  and  a  medial,  retrorse 
tooth.  Spicules  88-90  p  long,  weakly  ceph- 
alated  proximally  and  narrowing  gradually 
distally.  Gubernaculum  paired,  40-45  p 
long,  with  characteristic  proximal  cephali- 
zation  (bottle-cap-opener).  Male  with  11- 
14  papilloid  supplements.  Tail  initially 
truncate-conoid,  then  filifomi,  in  6  ,  265- 
300  p  long,  in  9  ,  315-370  /x  long.  Spinneret 
10-12  p  long.  The  male  has  a  postanal  ven- 
tral depression  just  prior  to  the  filiform 
portion  of  the  tail.  The  de  Man  c  value  for 
both  sexes  is  5.8-6.6. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne,  Everglades  National  Park. 

Gcog^raphical  distribution. — Aransas  Bay, 
Texas  (Chitwood,  1951),  Congo  estuary, 
^^'est  Africa  (Gerlach,  1957b),  Cananeia, 
Brazil  (Gerlach,  1957c). 

Remarks. — An  entire  specimen  of  Spi- 
lophorella  paradoxa  was  found  within  the 
gut  of  one  female. 

Halichoanolaimus  duodecimpopillafus  Timm, 
1952 
Plate  XVII,  fig.  34,  a-d 

Halichoanolaimus  dtioclecimpapillatus  Timm,  1952: 
26-28,  pi.  5,  fiff.  44. 

L  =  ^  ,  1.9-2.2.    9  .  2.1-2.5  mm;  w  =  $, 


270         BuUciin  Mii.scuiii  of  Cotuparatwc  Zoology,  Vol.  135,  No.  5 


68-72,  9 ,  80-97  /x;  diameter  at  base  of 
esophagus,  6  ,  62  /a,  9  ,  65-87  fj..  Vu  =  40- 
47%.  Esophagus  in  i  ,  250-280  /x,  in  $  , 
270-310  ,x.  ExcretoiN'  pore  130-147  /x  from 
anterior  end.  Head,  eephahc  sense  organs 
and  buccal  cavity  as  described  above  for 
H.  quattuordecimpapiUatus.  Spicules  90- 
100  IX  long,  proximal  cephalization,  broad 
and  flat.  Distally  the  spicules  narrow  more 
abruptly  than  in  the  foregoing  species. 
Gubernaculum  about  50  jx  long,  of  char- 
acteristic shape.  Male  with  11-13  papilliod 
supplements.  Tail  elongate-conoid,  then 
filifonn,  in  i  ,  180-190  //long,  in  ?,  165- 
230  )x  long.  Spinneret  9-10  ix  long.  The 
de  Man  c  value  for  the  i  is  10.1-11.3  and 
for  the   9  ,  9.1-14.3. 

Representation  in  .'samples  studied. — V, 
Vero  Beach. 

Geo<i,r(iphical  distribution. — Chesapeake 
Bay,  Maryland   ( Tinim,  1952 ) . 

NEOTONCHINAE 
NEOTONCHUS  Cobb,  1933 
Type  species.    Neotonchus  punctatus  Cobb, 
1933:    87. 

Comesa  Cerlach,   1956:   94. 

Neotonchus    lufosus    Wieser    and    Hopper, 
1966 
Plate   XXVI,   fig.   56,   a-d 

L  =  0.87-1.02  mm;  w  =  42  /x;  esopha- 
gus =  120  IX.  Head  diameter  20  /x.  One 
circle  of  six  minute  labial  papillae,  one 
circle  of  six  short  cephalic  setae,  3  ix  long, 
and  four  submedian  setae  of  about  ecjual 
length.  Scattered  short  cervical  setae.  Am- 
phids  10  IX  =  507c  of  c.b.d.  wide,  4-4.2 
turns.  Cuticle  with  homogeneous  puncta- 
tion,  the  lateral  dots  somewhat  larger  and 
more  widely  spaced  than  the  submedian 
ones.  Buccal  cavity  14  /x  long,  with  one 
large  hollow  dorsal  tooth  and  two  small 
subventral  teeth.  Esophageal  bulb  pyri- 
form,  30  X  25  /x,  with  two  weak  interrup- 
tions. Excretory  pore  40  fx  from  anterior 
end.  Spicula  of  shape  typical  for  genus, 
bent  at  beginning  of  distal  third,  29  /x  long. 
Gubernaculum    plate-shaped,    dilated    dis- 


tally. One  stiff  preanal  seta  and  20  large, 
complicated  supplements.  Tail  75  /x  long, 
a.b.d.  28  /x. 

Representation  in  samples  studied. — M- 
4,  Rickenbacker  Causeway. 

Geographical  distribution. — Restricted  to 
above  locality. 

DESMODORIDAE 

The  Desmodoridae  was  first  subdivided 
by  Chitwood  ( 1936 ) .  The  original  group- 
ings, however,  do  not  appear  to  be  entirely 
satisfactory,  especially  with  regard  to  their 
generic  composition  and  systematic  posi- 
tion. Uncertainties  regarding  the  shape  of 
amphids  in  the  Stilbonematinae  present 
further  difficulties.  A  key  position  within 
the  family  is  assumed  by  S})irinia  (syn. 
Spirina )  which  not  only  links  the  Meta- 
chromadorinae  with  the  Desmodorinae — 
perhaps  via  Chromaspirina  ( see  Gerlach, 
1963) — but  also  shows  close  affinities  to 
the  Microlaimidae  and  the  Linhomoeidae 
(Wieser,  1954;  Timm,  1962).  The  discus- 
sion of  many  genera  belonging  to  the 
Desmodoridae  by  Gerlach  ( 1951b,  1963b ) 
has  been  a  valuable  aid  in  the  classification 
of  the  family.  The  characters  of  each  sub- 
family are  briefly  outlined  below: 

Brief  Characterization  of  Subfamilies  of 
Desmodoridae 

A)  Metachromadorinae  (Chitwood,  1936): 
Cuticle  always  finely  striated  (striation 
sometimes  so  indistinct  as  to  impart  a 
smooth  appearance  to  the  cuticle);  head 
not  sharply  set  off  from  striation,  non- 
rigid;  striation  always  surrounding  am- 
phids; cuticle  not  tiled;  esophageal  bulb 
always  present,  either  roimd  or  elongated; 
buccal  cavity  typically  with  well-developed 
annature,  except  in  Spirinia  where  there 
are  only  minute  teeth.  (  For  further  classi- 
fication, see  below. ) 

B)  Richtersiinae  Cobb,  1933:  Cuticle 
striated,  with  many  longitudinal  rows  of 
spines  or  hooks;  head  non-rigid;  buccal 
cavity  small  or  wide,  unanned;  esophagus 


Florida  Marine  Nematodes  •   Wiescr  and  Hopper        271 


cylindrical,  without  bulb.  With  the  genera: 
Richtersia  Steiner,  1916  (syn.  RichtersicUa 
Kreis,  1929),  and  Ptcwniiim  Cobb,  1933. 

C)  Desmodorinae  Micoletzky,  1924: 
Cuticle  heavily  annulated;  head  rigid, 
sharply  set  off  from  annulation;  amphids 
not  surrounded  by  annulation  (exceptions 
are  Paradesmodora  and  Metadesmodom  in 
which  the  heavy  annulation  serves  as  dis- 
tinguishing characters  from  the  Metachro- 
madorinae);  amphids  loop-shaped  or  spiral; 
cuticle  not  tiled  (except,  occasionally,  on 
the  head);  esophageal  bulb  round  or 
elongated;  buccal  cavitv  alwavs  amied 
with  distinct  teeth.  (For  further  classifica- 
tion, see  below.) 

D)  Stilbonematinae  Chitwood,  1936: 
Cuticle  striated  or  annulated,  not  tiled  or 
longitudinally  broken;  buccal  cavity  shal- 
low-conical or  absent,  unarmed  or  with 
minute  teeth;  head  always  well  defined, 
amphids  wholly  outside  striation,  spiral- 
shaped,  but  apparently  sometimes  sunk 
into  the  cuticle  so  that  only  the  slit-like 
opening  is  visible;  esophageal  bulb  round 
to  pyriform.  With  the  genera:  Eiibosfii- 
chus  Greeff,  1869  (syn.  Catancma  Cobb, 
1920,  ?Lcixm  Cobb,  1893),  Laxonema  Cobb, 
1920,  Lcptonemella  Cobb,  1920,  Stilboncma 
Cobb,  1920,  Robbed  Gerlach,  1956,  and 
Sqiianenui  Gerlach,  1963. 

E)  Ceramonematinae  Cobb,  1933':  Cuti- 
cle heavily  annulated,  tiled  or  longitudinally 
broken  by  spined  alae;  head  well  set  off 
from  annulation;  buccal  cavity  minute  or 
absent,  unarmed;  amphids  obscurely  spiral 
to  shepherd's  crook;  esophagus.  With  the 
genera:  Cemmonema  Cobb,  1920,  Xenclla 
Cobb,  1920,  DasynemeUa  Cobb,  1933  (syn. 
Dasiincma  Cobb,  1920),  Pristioncmo  Cobb. 
1933,  PseJkmcma  Cobb,  1933,  Dosync- 
moidcs  Chitwood,  1936,  MetodasyncmcUa 
de  Coninck,  1942,  and  Pteriji;oncma  Ger- 
lach, 1954. 

F)  Monoposthiinae  Filipjev,  1934:  Cuti- 
cle  coarselv    annulated,   broken    longitudi- 


^  This  subfamily  may  not  belong  to  the  Des- 
modoridae.  Gerlach  (1957)  considers  it  related 
to  the  Axonolaimidae  and  the  Halaphanolaimidae. 


nally  by  alae;  head  well  set  off  from  annu- 
lation, rigid;  amphids  circular,  surrounded 
by  annulation;  buccal  cavity  well  armed; 
esophageal  bulb  barrel-shaped,  ^^'ith  the 
genera:  Monoposfliia  de  Man,  1889, 
Niidora  Cobb,  1920,  Rhinema,  Cobb,  1920, 
and  Monoposiliioides  Hopper,  1963. 

Key  to  Genera  of  Metachromadorixae 
(  Based  on  classification  of  Gerlach,  1951 ) 

1.  Male      supplements      heavih'      cuticularized, 

large  and  tubular  2 

Male  supplements  indistinct  or  conical  or 
consisting  of  narrow  ducts,  not  large  and 
tul)ular;  only  in  M.  vivipara  and  M. 
quadrihuiha  are  the  supplements  heavily 
cuticularized  but  not  tubular  4 

2.  Supplements      strongly      S-shaped,      heavily 

cuticularized    _    3 

Supplements  faintly  S-shaped,  cuticulariza- 
tion  light  ..— Onyx  Cobb,   1891 

3.  Cephalic   and   subcephalic   setae   present 

Sigmophora   Cobb,   1933 

Cephalic  setae  only  present  

Polysigrna  Cobb,  1920 

4.  Teeth  absent   or  minute   

..„ Spirinia  Gerlach,  1963 

Teeth  well  developed  5 

5.  Esophageal    bulb    weakly    developed,    round 

to  pyriform;   cuticular  lining  faint  

CJiromaspirina   Filipjev,   1918 

Esophageal  bulb  well  de\eloped,  usually 
elongate,  sometimes  "barrel-shaped,"  rarely 
cla\ate  and  indistinctb'  set  off  from  esoph- 
agus; cuticular  lining  usually  distinct  

Metachromadora  Filipjev,   1918 

Fseudomeiachioinadora  Timm,  1952,  is  a  genus 
of  doubtful  position.  The  esophagus  is  barely 
enlarged  posteriorly,  the  cuticular  striation  is  all 
but  absent,  the  amphids  are  situated  near  the  lips 
and  the  buccal  cavity  is  cylindrical  and  strongly 
armed. 

SP/R/N/A  Gerlach,  1963 
Type    species.— Sp/ra    parasitifera    Bastion, 
1865:    159,  160,  pi.  13,  figs.  201-203. 

Spira  Bastian,  1865,  nee  Brown,  1844,  and  Sp/c/no 
Filipjev,   1918,  nee  Kayser,  1889. 

Our  material  contained  two  species,  one 
of  which  undoubtedly  is  S.  parasitifera 
( Bastian,  1865 ) .  A  redescription  of  this 
cosmopolitan  species  with  discussion  of  its 
synonymy  was  recently  given  by  Gerlach 
( 1963b  ) .    Our  second  species  is  closely  re- 


272 


Bulletin  Mu.scuin  of  Conijxirative  Zoology,  Vol.  135,  No.  5 


lated  to  S.  .striaiicaudata  (Timm,  1962) 
from  which  it  can  be  distinguished  b\'  the 
hook-shaped  proximal  end  of  the  spicula 
and  the  presence  of  small  but  distinct  teeth 
in  the  conical  buccal  cavity.  These  two 
species  are  separated  from  all  other  mem- 
bers of  the  genus  by  the  flagellate  tail 
which  in  both  sexes  is  much  more  distinctly 
and  coarseK'  striated  than  the  rest  of  the 
body.  This  is  such  a  conspicuous  and  char- 
acteristic feature  that  we  feel  justified  in 
establishing  a  new  subgenus  on  it. 

Spirinia    (S.)    parasitifero    (Gerlach,    1963) 
Plate  XVII,  fig.  35,  a-e 

S))irinia  (S.)  paiasififcia  ( Bastian,  1865)  Cer- 
lath,  1963b:   67. 

Spira  pcirasififcm  Bastian,  1865:  159-160,  pi.  13, 
figs.  201-203;  Spirilla  parasitifcra  auct.;  Spi- 
lophora  oxyccphala  ButsL-lili,  1874:  Spirina 
nidro.siensis  Allgen,  1933;  S.  zosterae  Filipjev, 
1918;  and  S.  rotivillei  Schnnrmans-Stekho\en, 
1950. 

L  =  1.57  mm;  w  =  55  /x;  esophagus  =  140 
/x;  nerve  ring  77  fi  from  anterior  end;  tail  = 
140-155  /x.  Head  diameter  22  /j,;  cephalic 
setae  5  /x,  on  level  of  amphids.  Cervical 
setae  beginning  at  short  distance  behind 
amphids,  rather  scattered.  Amphids  6  /x 
wide.  Buccal  cavity  small,  with  3  minute 
teeth.  Esophageal  bulb  40  X  36  /x.  Spicula 
50  IX,  knobbed  proximally,  with  velum,  tail 
conical,  a.b.d.  24-28  /x. 

Rcprcseniaiion  in  samples  .studied. — M- 
4,  Rickenbacker  Causeway,  V,  Vero  Beach. 

Geographicol  distribution. — Baltic,  North 
Sea,  North  Atlantic,  Black  Sea,  Mediterra- 
nean, Barents  Sea,  Indian  Ocean  (Mal- 
dives ) . 

Spirinia  iPerspiria)   new  subgenus 
Type   species.— Sp/r/n/a    iPerspiria)    hamafa 
new  species. 

Diffcrcnlifd  diaiino.sis. — Perspiria  n.  subg. 
is  distinguished  from  S})irinia  sensu  siricto 
by  the  more  prominently  striated  and 
flagellate  tail.  In  Spirinia  sensu  stricto  the 
tail  is  conoid  and  the  striations  are  fine — no 
coarser  than  those  on  the  remainder  of  the 
bodv. 


In  addition  to  the  new  species  described 
below,  Spirinia  striaticaudata  (Timm,  1962) 
(syn.  Spirina  striaticaudata)  is  also  in- 
cluded in  the  new  subgenus. 

Key  to  Species  of  SpiRhyiA  (  Perspiria  ) 

1.    Stoma  without  teeth;  proximal  end  of  spicula 

knobbed,    rounded    

S.  (P.)  striaticaudata  (Timm,  1962) 

Stoma  with   minute   teeth;   proximal   end   of 

spicula    hook-shaped    

____   S.    (P.)    hamata  new  species 

Spirinia  [Perspiria)  hamata  new  species 
Plate  XVIII,  fig.  36,  o-c 

L  =  2.04  mm;  esophagus  =  160  /x.  Head 
diameter  16-21  /i.  Cephalic  setae  4-5  jx. 
Very  few  short,  scattered  cervical  setae. 
Amphids  5-6  p..  Buccal  cavity  small,  with 
three  minute  teeth.  Esophageal  bulb  45  X 
36  fx.  Excretory  pore  between  bulb  and 
nerve  ring.  Cuticular  striation  distinct  but 
weak.  Spicula  37-43  /x,  gubemaculum  16 
fx.  Tail  230-340  /x,  from  beginning  of  sec- 
ond fifth  on  coarsely  striated,  tip  unstriated. 
Anal  body  diameter  32-43  /x. 

Holotypc  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4068,  Type  slide  No.  78. 
Type  locality,  M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
3,  Key  Biscayne. 

CHROMASPIRINA  Filipjev,  1918 

Type  species.— C/iromosp/r/na  pontica  Filip- 
jev, 1918:  229,  230-234,  pi.  7,  fig. 
45,  a-c. 

Mesodoni.s  Cobb,    1920:    325. 

As  previously  suggested  by  Gerlach 
( 1963b ) ,  this  genus  probably  links  the 
Metachromadorinae  with  the  Desmodori- 
nae.  We  do  not  agree,  however,  with  Ger- 
lach that  the  intermediate  position  of  this 
genus  is  sufficient  justification  for  merging 
the  two  subfamilies  together.  The  position 
of  Chromaspirina  becomes  less  problemati- 
cal if  some  of  Gerlach's  new  combinations 
are  returned  to  thcnr  former  status.    Thus 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        273 


we  cannot  accept  the  transfer  of  Dcsmo- 
dora  inflexa  Wieser,  1954,  and  D.  dimorpha 
Hopper,  1961,  to  Chromaspirina.  Both  spe- 
cies are  characterized  by  heavy  anniilation 
and  a  rigid  head  with  the  amphids  situated 
outside  the  annulation  (although  in  the 
latter  species,  admittedly,  the  position  of 
the  amphids  with  respect  to  the  cuticular 
annulation  is  a  bit  doubtful )  and  thus  most 
likely  represent  true  members  of  Desmo- 
dorci.  The  same  would  seem  to  apply  to 
D.  rahosa  Gerlach,  1956,  which  was  also 
tranferred  to  Chromaspirina  by  Gerlach. 
However,  this  species  is  known  only  from 
one  female  and  thus  remains  doubtful. 

Other  doubtful  species  are  C.  paiicispira 
Schuurmans-Stekhoven,  1950  (  $  only)  and 
C.  robusta  Wieser,  1954  (juv.  only).  Two 
more  species  are  known  only  as  females, 
i.e.,  C.  crinita  Gerlach,  1952,  and  C.  pellita 
Gerlach,  1954,  but  in  their  cases  the  pilosity 
of  either  head  (crinita)  or  body  (pellita) 
is  probably  sufficient  to  recognize  them  as 
good  species.  The  following  are  regarded 
as  good  species  of  CJiromaspirina:  C.  cy- 
lindricollis  (Cobb,  1920)  (syn.  Mesodorus 
ci/lindricoUis),  C.  indica  Gerlach,  1963,  C. 
madagascaricnsis  Gerlach,  1953,  C.  para- 
pontica  Luc  and  De  Goninck,  1959,  C. 
pontica  Filipjev,  1918,  and  C.  thicnji  De 
Goninck,   1943. 

The  conspecificity  of  C.  pontica  Filipjev 
sensu  Gerlach,  1951,  with  Filipjev's  species 
is  questionable,  as,  in  Gerlach's  specimens, 
the  cephalic  setae  are  9-11  ^u,  long  (as 
against  5  //),  the  amphids  are  relatively 
larger,  and  the  gubernaculum  is  shorter 
and  of  a  somewhat  different  shape. 

From  the  above  mentioned  species,  C. 
inaiirita  n.  sp.,  is  separated  by  the  indis- 
tinct cephalic  papillae,  the  large  and  oval 
amphids  which  are  about  as  long  as  the 
head  is  wide,  the  shape  of  the  gubernacu- 
lum, and  the  presence  of  ventral  papillae 
on  the  tail. 

Chromaspirina    inaurifa    new   species 
Plate  XVIII,  fig.  36,  a-c 

L  =  1.18-1.35  mm;  w  =  29-35  /x;  esopha- 


gus =  105-115  p.;  tail:  c^  ,  90-95  /x,  9  ,  90 
ix;  Vu  =  49%.  Eggs  30-33  x  60-70  /x.  Head 
diameter  14  /x.  Six  minute  cephalic  papil- 
lae. Four  cephalic  setae,  5  /j.  long.  In  $ 
two  ( dorsal  and  ventral )  subcephalic  setae, 
in  both  sexes  scattered  short  cervical  setae. 
Amphids  in  6  ,  12-14  +  10-12  ,i,  loop- 
shaped,  inner  circle  more  heavily  cuticular- 
ized,  6  X  5.5  //,  more  narrowly  coiled.  Gu- 
ticular  annulation  distinct.  Buccal  cavity 
with  three  small  teeth  (these  teeth  are 
smaller  than  in  most  other  species  of 
Chromaspirina  and  stress  the  proximity  of 
the  genus  to  Spirinia).  Esophageal  bulb 
pyriform,  25  X  20  /x.  Spicula  27-28  /x,  with 
faint  velum,  hooked  proximal  end.  Guber- 
naculum sickle-shaped,  15  fx.  The  preanal 
ventral  cuticle  is  crenate  over  a  long  dis- 
tance. Immediatelv  in  front  of  the  anus 
there  is  a  stout  spine.  On  the  tail  there  are 
four  ventral  supplements  each  accompanied 
by  a  pair  of  setae.  Subventrally  and  sub- 
dorsallv  more  setae  can  be  seen.  A.b.d. 
23^. 

Holotijpe  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4068,  Type  slide  No.  79. 
Type  locality,  M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
3,  Key  Biscayne. 

METACHROMADORA  Filipjev,  1918 

Type  species.— Mefoc/iromoc/oro  macroufera 

Filipjev,    1918:    218,   219-225,  pi.   6, 

fig.  42a;  pi.  7,  fig.  42,  b-h. 

Key  to  Species  of  Metachromadora 

1.  Male  supplements  conical,  each  consisting 

of  three  ( 1  central,  2  lateral )  cuticular- 

ized    pieces     subgenus     Chroma- 

doropsis  Filipjev,    1918   2 

Male  supplements  indistinct  or  narrow 
ducts  the  openings  of  which  are  differ- 
entiated into  button-shaped  or  conical 
bodies     3 

2.  Esophageal  bulb,  2-sectioned  

M.  vivii)cira  (de  Man,  1907) 

(syn.  Chrornadora  civipaia  de  Man,  1907) 

Esophageal  bulb,  4-sectioned  

M.  quadrihidha  Gerlach,  1956 

3.  Somatic  setae  arranged  in  10  dense  longi- 

tudinal   rows    subgenus    Metomjx 


274         Bulletin  Miiscmii  of  Coin])(initwe  Zoology.  Vol.  135,  No.  5 


Chitwood,    1936   

M.   horrida  Chitwood,    1936 

Somatic  setae  not  arranged  in  10  dense 
longitudinal    rows   4 

4.  Head   with   pronounced  longitudinal   stria- 

tion  subgenus     Mctachrumadora 

Filipjev,    1918'    ...-     5 

Head  without  pronounced  longitudinal 
striation    6 

5.  Length  2.4-2.6  mm;  male  with  26-48  pre- 

anal    supplements    

M.    luacrotitem   Filipjev,    1918 

Length  1.0-1.4  mm;  male  with  12-14  pre- 

anal     supplement    

M.  chandkri  (Chitwood,  1951) 

(syn.  Ichthiiodesmodora  chandleri 

Chitwood,  1951. 
and  MetacJiroiuadora 
parasitifera  Timm,  1952 ) 

6.  Posterior    portion     of    esophagus     clavate; 

bulb  not  well  set  off,  partitions  absent 
or  indistinct;  cuticle  smooth  to  indis- 
tinctly striated.  (Subgenus  doubtful, 
perhaps  a  new  one  to  be  established )  7 

Posterior  portion  of  esophagus  bulbular; 
bulb  well  set  off,  partitions  distinct: 
striations  of  cuticle  distinct  9 

7.  Male    without   preanal    supplements    

M.  .s/)//y;/(.v  Ccrlach,   1955 

Male  with  distinct  preanal  supplements  _       8 

8.  Male    with    3    knob-like    preanal    supple- 

ments      M.   chivata   C.erlach,    1957 

Male  with  19-21  conoid  preanal  supple- 
ments   M.  serrata  Gerlach,    1963 

9.  Posterior    cephalic    setae    absent    or    stout 

and  short;  amphids  (at  least  in  male) 
on     thick     cuticularized     plates;     cuticle 

with  lateral  wings  subgenus  Mcta- 

chromadoroklcs  Timm,    1961    10 

Posterior  cephalic  setae  slender;  amphids 
not  on  thick  cuticularized  plates — .   13 

10.  I'^sophageal  bullj,  2-sectioned  _.-. 

M.   remanei  Gerlach,    1951 

Esophageal   bulb,   3-sectioned   11 

11.  Male    without    preanal    supplements    and 

without     ventral     caudal     "bumps"     or 

"warts"  M.  vuh^arls  Tinnn,   1961 

Male  with  17-23  preanal  supplements  and 
with  2  ventral  caudal  "bumps"  or 
"warts"     12 

12.  Cephalic   .setae   present   

M.  pidvinata  new  species 

Cephalic  setae  absent   

.,  M.   complexa   Timm,    1961 

13.  Lateral     wings     present     subgenus 


'  M.  ctjfito.'ieirae  Filipjev,  1918,  also  belongs  to 
this  subgenus.  However,  it  appears  to  have  Iieen 
described  on  the  basis  of  a  single  female,  and,  as 
such,   its  systematic  position   is  doulitful. 


Neonyx  Cobb,  1933  14 

Lateral  wings  absent  _  _  _  subgenus  Bradij- 
laimus  Schuurmans-Stekhoven,    1931     ___   18 

14.  Circles    of    cephalic    setae    in    typical    ar- 

rangement   (6  +  6-f  4)    15 

Circles  of  cephalic  setae  in  atypical  ar- 
rangement, 6  +  6  -(-  8,  through  fusion 
with  subcephalic  setae  ( Cobb  mentions 
this  condition  in  rather  vague  fashion 
for    M.    cancellata)    17 

15.  Esophageal    bulb,    barrel-shaped;    preanal 

supplements,   8;   body  shape,  obese,   de 
Man  a  value  :=  i,   16-24,    9,  9.5-11.5 

M.  ohcsa  Chitwood,   1936 

Esophageal  bulb,  elongate;  preanal  sup- 
plements, 12;  body  shape  more  slender, 
de  Man  a  value  =  <^ ,  33-44;    9,  34  ____   16 

16.  Subcephalic     and     cervical     setae     shorter 

than  the  longest  cephalic  setae  

M.  pscudocampijcoma  Hopper,  1961 

Sulxephalic  and  cervical  setae  longer  than 

the   longest   cephalic   setae   

M.    canipijcoma    (Cobb,    1933) 

( doubtful  species ) 
( syn.  Neonijx  campijcoma  Cobb,  1933 ) 

17.  Buccal    ca\  itv   with    denticles    

'.._.  M.   cancdlaia   (Cobb,   1933) 

(syn.  Neomjx  cancellata  Cobb,  1933) 

Buccal    caxity   without      denticles    

M.    tncridiana  new  species 

18.  Esophageal  bulb,  3-sectioned  19 

Esophageal  bulb,  2-sectioned  21 

19.  Buccal    cavity   with    denticles;    head    with 

several  circles  of  long  subcephalic  and 
cervical  setae  posterior  to  cephalic  setae 

M.  setosa  Hopper,   1961 

Buccal  cax'ity  without  denticles;  head  with 
only  a  single  circle  of  long  subcephalic 
setae  posterior  to  cephalic  setae  20 

20.  Male  with  9-10  preanal  supplements  

M.   onyxoidcs  Chitwood,   1936 

Male   without  supplements    

M.  asiipplementa  (Crites,  1961)" 

(syn.  Neotiyx  a.sti))j)lcmi'nta  Crites,  1961) 

21.  Esophageal    bulb    elongate,    length    more 

than  twice  width  — 22 

Esophageal  bull)  ovate,  length  less  than 
twice    width    23 

22.  Amphid    7.5    /u   wide,    less    than    one-third 

c.b.d.   M.  ^crlachi  new  name 

( syn.  M.  onyxoide.s  .scu.sii  Gerlach,  1955, 
nee  Chitwood,  1936) 
Amphid    15    ^j,    wide,    more    than    two- 
thirds  c.b.d.   __     M.  spectaii.s  Gerlach,  1957 

23.  Amphid    12    fx    wide,    more    than    one-half 

c.b.d.    M.    pncuniaticd    Gerlach,    1954 

Amphid    7-8    /j.    wide,    about    one-fourth 

c.b.d.   M.  .^ticcica    (Allgen,   1929) 

(syn.  Oi.'itolaimti.s  .succiciis  Allgen,  1929) 

"Possiblv  a  svnon\ni  of  M .  oiujxoidcs. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        275 


Metachromadora         {Mefachromadoroides) 
pulv'mafa  new  species 

Plate  XIX,  fig.  38,  a-c 

L  =  1.72  (juv.  9  =1.11)  mm;  w  =  95 
(juw  9  =65)  /jl;  esophagus  =  315  (juv. 
9  =220)  /x;  tail:  juv.  9  =  80  /x;  Vu  647^. 
Head  diameter  on  level  of  amphids  40  fi. 
Lips  cushion-like,  demarcated  from  head 
by  distinct  groove.  Labial  papillae  conical. 
Cephalic  setae  stout,  8  /x  long  in  male. 
Short  cervical  setae.  Somatic  setae  up  to 
18  jji  in  posterior  part  of  body.  Amphids 
in  6  ,  22  X  18  /x.  on  cuticularized  plates, 
ring-shaped,  in  9,8x6  /j,,  loop-shaped. 
Cuticular  striation  distinct,  reaching  to 
base  of  lips.  Lateral  differentiation  not 
very  pronounced,  beginning  around  end  of 
esophagus,  ending  at  some  distance  in  front 
of  anus.  Buccal  cavity  strongly  cuticular- 
ized, in  6  ,  50  /t  long,  with  large  dorsal 
tooth,  2  small  subxentral  teeth  and  an  ad- 
ditional tooth  at  the  bottom  of  the  buccal 
cavity;  vestibulum  with  ribs.  Esophageal 
bulb  tripartite,  \\  ith  heavy  cuticular  lining, 
about  105  X  62  ^.  Spicula  55  /x  long,  15  /x 
broad.  Gubemaculum  27  fj..  Preanally  the 
\entral  cuticle  is  extended  and  forms  a 
striated  membrane  which  is  traversed  by 
about  23  narrow  ducts,  each  duct  ending 
in  a  button-shaped  body.  Extended  cuticle 
reaching  540  fx  preanal.  Immediately  in 
front  of  anus  one  strong  spine.  Tail  110 
/x  =  2  a.b.d.  long,  in  6  with  two  ventral, 
conical,  cuticularized  warts.  Many  long 
spines  in  longitudinal  rows. 

Ilolofype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4075,  Type  slide  No.  80. 
Type  locality,  V,  Vero  Beach. 

Representation  in  sampJes  studied. — V, 
Vero  Beach. 

Remarks. — The  only  other  species  in  this 
subgenus  with  postanal  ventral  warts  is  M. 
complexa  Timm,  1961,  which,  however,  is 
devoid  of  cephalic  setae. 


Metachromadora    {Bradylaimus)    onyxoides 
Chitwood,  1936 

Metachromadora    (Bradylaimus)    onyxoides   Chit- 
wood,  1936:  5,  Hr.  1,  v-x. 

Our  specimens  appear  to  be  typical  rep- 
resentatives of  M.  onyxoides  as  described 
by  Chitwood  (1936)  and  Hopper  (1961a). 
The  amphids  in  our  male  measured  7-8  /x. 

Representation  in  samples  studied. — M- 
5,  Virginia  Key. 

Geogra  ph  ica  1  distrib  ution . — B  eauf  ort , 
North  Carolina  (Chitwood,  1936),  Gulf 
Shores,  Alabama  (Hopper,  1961a)  and  ? 
Pemambuco,  Brazil  ( Gerlach,  1956 ) . 

Remarks. — Gerlach's  (1955)  identification 
of  specimens  from  San  Salvador  as  M. 
onyxoides  is  doubtful.  In  the  table  below 
some  dimensions  of  our  own  specimens  are 
compared  \\'ith  those  of  the  animals  de- 
scribed by  Hopper  ( 1961a )  and  by  Gerlach 
( 1955 ) .  It  follows  that  Gerlach's  speci- 
mens belong  to  a  different  species  for 
which  the  name  M.  fs^erlachi  new  name  is 
proposed. 


Present 

Hopper, 

Gerlach, 

material 

1961 

1955 

(/t) 

(/") 

Length  of 

first 

cephalic 

setae 

2.5-3 

"short" 

6-7 

Length  of 

second 

cephaHc 

setae 

10 

11m 

15-17 

Length  of 

spicula 

50 

60  m 

27 

Length  of 

gul^er- 

nacu- 

lum 

35 

40  m 

15 

Esophageal 

bulb. 

length 

of  par- 

titions 

25+25+25 

20+25+25  M 

29+35 

Metachromadora      (Bradylaimus)      gerlachi 

new   name 

Metachromadora   onyxoides   Chitwood   .sen.su    Ger- 
lach,  1955. 


276        BiiUctiii  Mtisctint  of  Couipaidlivc  Zoology,  Vol.  135,  No.  5 


Mefacliromadora  ( BradyJciimus)  '^crhchi 
new  name  is  related  to  M.  omjxoidcs  Chit- 
wood,  1936,  from  which  it  is  distinguished 
by  ha\ini:;  a  two-sectioned  esophageal  bulb, 
the  bulb  in  M.  omjxoidcs  being  three-sec- 
tioned. 

hAeiochromadora   {Neonyx)  meridiana  new 
species 
Plate  XIX,  fig.  39,  a-d 

L  =  0.95-1.05  mm;  w  =  ^  ,  47,  9  ,  55-62 
fj.;  esophagus  175-185  //.  Lips  prominent. 
There  are  three  distinct  circles  of  cephalic 
sense  organs:  in  front  two  circles,  each 
composed  of  6  conical,  setose  papillae, 
followed  by  one  circle  of  8  slender  setae, 
each  measuring  6-7  /x.  It  is  assumed  that 
this  circle  consists  of  the  t\'pical  4  cephalic 
plus  4  subcephalic  setae.  Close  to  the  am- 
phids  there  are  two  more  subcephalic  setae 
on  each  side  of  the  body,  measuring  8  /x. 
Cervical  setae  in  the  anterior  region  up  to 
10  II,  in  posterior  region  up  to  15  /<.  long. 
Amphids  spiral,  7  /x  =  30%  of  c.b.d.  wide. 
Cuticular  striation  reaching  to  anterior  end 
of  amphids;  lateral  alae  starting  around 
middle  of  cervical  region.  Buccal  cavity 
with  rather  thin  \\'alls,  very  strong  dorsal 
tooth  and  small  subventral  projections. 
Esophageal  bulb  tripartite,  with  heavy  cu- 
ticular lining,  60-70  X  28-32  /x  in  ^  ,  80  X 
35  /x  in  9  .  Eggs  70-75  X  40-50  /x.  Spicules 
arcuate,  58  /x  long,  gubernaculum  29  /x  long. 
Male  with  9-10  thin  preanal  supplements, 
traversing  the  raised  ventral  cuticle.  Tail 
80-95  /x  long.  Female  a.b.d.  =  28  /x,  male  = 
35  /I.  Caudal  setae  on  male  tail  arranged 
as  illustrated  in  Plate  XIX,  figure  39c. 

Il()l()ti/))c  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4075,  Type  slide  No.  81. 
T\\)v  locality,  V,  Vero  Beach. 

Representation  in  samples  studied. — V, 
Vero   Beach. 

Remarks. — At  first  we  thought  we  had 
found  M.  ohesa.  However,  Chitwood's 
original  description  is  rather  poor  and  all 
other  authors  who  subs(Yjuently  described 
this  species  (Timm,  1952;  Hopper,  1961b; 


Crites,  1961)  mention  the  typical  arrange- 
ment of  6  short  and  4  long  cephalic  setae, 
whereas  in  our  specimens  there  is  definitely 
a  circle  of  S  long  cephalic  setae. 

DESMODORINAE 

Key  to  Genera  of  Desmodorinae 

1.  Esophageal  bulb  elongate,  tripartite  2 

Esophageal  bulb  round  to  pyrifonn ..._  3 

2.  Head  with   large  plates   in  posterior  portion 

(head   "jointed"  or  tiled)    

Acanthophoryngoides  Chitwood,   1936^ 

Head   simple,   without   plates   

AcantJiopIwrynx    Marion,    1870 

syn.  Xanthudora  Cobb,  1920 

3.  Amphids   half  or  completely  surrounded  by 

annulation 4 

Amphids  not  surrounded  by  annulation  

Desmodora  de   Man,    1889 

(see  discussion  by  Gerlach,  1963b) 

4.  Amphids  half  surrounded  by  annulation,  not 

situated  on   cuticularized  plates   

Paradcsmodora  Schuurmans-Stekhoven,  1950 
Amphids  completely  surrounded  by  annula- 
tion, situated  on  cuticularized  plates  

Metadesmodora  Schuurmans-Stekhoven,  1942 

PARADESMODORA  Schuurmans-Stekhoven, 

1950 
Type    species.— Poroc/esmoc/ora     cephalafa 

Schuurmans-Stekhoven,  1950:  117,  fig. 

67,  a-e. 

This  genus  contains  the  following  spe- 
cies: P.  campheUi  (Allgen,  1932)  Gerlach, 
1963  (syn.  Spirina  campheUi),  P.  immersa 
Wieser,  1954,  P.  punctata  Gerlach,  1963, 
and  P.  toretites  n.  sp.  The  type  species,  P. 
ccpJwlata  Schuin-mans-Stekhoven,  1950,  is 
known  from  a  juvenile  female  only  and  is 
considered  a  species  inquirenda. 

Our  new  species  can  be  distinguished 
from  all  other  species  of  the  genus  by  the 
shape  and  arrangement  of  male  supple- 
ments, the  hook-shaped  spicula  and  the 
cuticular  differentiations  of  the  head. 

Paradesmodora  toreutes  new  species 
Plate  XX,  fig.  40,  a-f 

L  =  1.62  mm;  w  =  21  /x;  esophagus  =  114 


^  In  this  genus  the  cuticular  annulation  is  not 
quite  as  pronounced  as  in  other  genera  of  this 
subfamily. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        277 


fx;  head  diameter  14  /x.  Lips  distinct,  papil- 
lae minute.  Cephalic  setae  6  /x.  Cuticle  in 
posterior  portion  of  (adult)  head  very 
much  enlarged  and  forming  plates  which 
surround  the  anterior  portions  of  the  am- 
phids.  Amphids  6  /x  ==  40%  of  c.b.d.  wide, 
one  circular  loop.  There  are  a  few  short 
cervical  setae.  Cuticular  annulation  coarse. 
Esophageal  bulb  pyriform,  22  X  15  fx.  Spic- 
ula  semicircular,  20  /x,  proximal  end  hooked. 
There  are  10  preanal  supplements  each 
consisting  of  a  ventral  bump  and  a  thicken- 
ing of  the  cuticle.  The  row  of  supplements 
extends  220  fx  preanally.  Bet\veen  two  sup- 
plements there  are  13-18  cuticular  annules. 
Tail  95  /J.  long,  a.b.d.  20  /j..  Juvenile  tail  7 
a.b.d.  long. 

Holofype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4068,  Type  slide  No.  82.  Type  locality, 
M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M-3, 
Key  Biscayne. 

DEShAODORA  de  Man,  1889 

Type  species.  — Sp//opfiora  comments 

Butschli,  1874:   282,  283,  pi.  5,  fig.  27, 

a,  b;  pi.  7,  fig.  27,  c,  d. 

Gerlach  (1963)  has  recently  reviewed  this 
genus,  relegating  several  kno\\'n  genera  to 
subgeneric  rank  (i.e.,  Pseudochromadora, 
Xenodesmodoia,  Croconema,  Bolhonema, 
Desmodorella  and  Zalonema).  Excellent 
discussions,  with  keys,  for  both  the  sub- 
genera and  the  species  within  each  sug- 
genus  are  presented  in  Gerlach's  paper. 

Gerlach's  (1963b,  p.  84)  discussion  of 
Xenodesmodora  makes  synonymization  of 
BJa  Inglis,  1963,  with  the  former  subgenus 
inevitable.  The  type  species,  Bla  nini  Inglis, 
1963,  therefore,  is  transfen^ed  to  the 
genus  Desmodora  and  becomes  Desmodora 
(Xenodesmodora)  nini  (Inglis,  1963)  new 
combination. 

The  genus  Desmodora  was  represented 
in  our  material,  in  samples  M-2  and  M-3 
from  Key  Biscayne,  by  a  single,  well-known 
species,    Desmodora    {Vseudochromadora) 


quadripapillata  (Daday,  1899)  Gerlach,  1963 
(synonyms  Fseudocliromadora  quadripapil- 
lata Daday,  1899,  Micromicron  cephalata 
Cobb,  1920,  and  M.  luticola  Timm,  1952). 

Pertinent  measurements  from  a  male 
specimen  are  as  follows:  L  =  0.74  mm. 
Head  diameter  14  /x.  Amphids  6  /x.  Spicula 
28  /x. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne,  M-3,  Key  Biscayne. 

Geographical  distribution. — New  Guinea 
(Daday,  1899),  Costa  Rica  (Cobb,  1920), 
Chesapeake  Bay,  Maryland  (Timm,  1952), 
and  Cananeia,  Brazil  ( Gerlach,  1957 ) . 

MONOPOSTHIINAE 
MONOPOSTHIA  de  Man,  1889 
Type   species.— Sp///p/iora    cosfofo    Bastion, 
1865:    166,  167,  pi.  13,  figs.  228,  229. 

We  prefer  not  to  follow  Gerlach  ( 1963 ) 
in  his  synonymization  of  Monoposthia  and 
Nudora.  Thus  in  Monoposthia  only  species 
are  retained  in  which  the  spicula  are  absent 
and  the  single  gubernaculum  is  not  con- 
spicuously enlarged  proximally.  \\'ithin  the 
genus  there  is  a  group  of  species  character- 
ized by  the  enlarged  second  annule.  A  key 
to  this  group  reads  as  follows: 

Key  to  Species  of  Monoposthia 

1.  Botli   first  and  second  annule  enlarged;   am- 

pliids  between  the  two  annules  

M.  thorakista  Schulz,  1935 

Only   second   annule   enlarged;    amphids   on 
this  annule  2 

2.  Cuticle    with    12    longitudinal    rows    of    V- 

shaped    markings    ____ 

M.  dtiodecimalata  Chitwood,  1936 

Cuticle  with  6  longitudinal  rows  of  V-shaped 

markings  3 

3.  Cephalic    setae    measuring   less    than    %    of 

head  diameter;  amphids  %  of  c.b.d.  wide   _ 

M.  mielcki  Steiner,  1916 

Cephalic  setae  approximately  one  head  diam- 
eter long;  amphids  about  ^f,  of  c.b.d.  wide  . 

M.   mirabilis  Schulz,    1932 

syn.  M.  longiseta  Allgen,  1935 

Our  material  is  representative  of  M. 
mirabilis  Schulz,  1932,  agreeing  in  all  es- 
sential points  with  the  excellent  redescrip- 
tion  given  of  this  species  by  Luc  and  De 


278         Bulletin  Museum  of  Coiiipdidfive  Zoology,  Vol.  135,  No.  5 


Coninck  (1959).  M.  ovnata  Tinini,  1952, 
described  on  [he  basis  of  one  jiuenile,  is 
either  a  synonym  of  M .  iniidhilis  or  else  a 
■sjiccics  i)Hjiiiicnchi. 

Monoposfhia  mirabilis  Schuiz,  1932 
PlateXX,  fig.  41,a,  b 

Moiii>i)().sflii(i  iiiirahilis  Sclmlz,  1932:    3S0-382,  fiji. 
2fi,   a-g 

Male. — L  =  1.6  mm;  w  =  57  /x.  Diameter 
at  base  of  esophagns,  54  ^.  Esophagns  210 
/L  long.  Head  diameter  19  /x,  bearing  six 
labial  papillae  and  four,  19  /x  long,  cephalic 
setae.  Cuticle  with  6  longitudinal  rows  of 
\'-like  markings.  Reversal  of  V's  at  mid- 
bod).  Bod\-  bearing  4  rows  of  somatic 
setae  which  carry  on  to  the  tail.  Buccal 
cavity  armed  with  dorsal  tooth  and  opposed 
by  a  number  of  denticles.  Gonad  single, 
outstretched.  Gubernaculum  38  /x  long.  Tail 
110  /x  long,  a.b.d.  39  /x.  A  prominent  double 
pair  of  fleshy  papillae  are  positioned  pre- 
anally,  surrounded  by  a  cuticularized  ridge. 

Representation  in  samples  studied. — M-2, 
Key  Biscay ne. 

Gen<iraphical  distiihtition. — European  At- 
lantic coasts,  Mediterranean  Sea  (Gerlach, 
1952). 

MONOPOSTHIOIDES  Hopper,  1963 

Type   species.— A^onoposf/i;o/c/es   anonopos- 

ihia     Hopper,     1963:     850-852,    figs. 

n-18. 

This  genus  is  differentiated  from  Mo)U)- 
posthia  by  the  shape  of  the  gubernaculum 
("spiculum"  in  the  interpretation  of  other 
authors),  the  long  spine  attached  to  the 
latter,  the  presence  of  two  testes  and  the 
absence  of  V4ike  markings  in  the  anterior 
portion  of  the  body. 

Our  specimens  agree  with  this  diagnosis 
except  for  the  absence  of  the  gubernacular 
spine.  No  trace  of  it  could  be  detected  in 
the  four  males  examined.  Further  dif- 
ferences from  the  type,  M.  anonoposthia, 
are:    the  cephalic  setae  measure  only  about 


one  head  diameter  as  against  1.5,  and  in 
the  gubernaculum  it  is  the  dorsal  rather 
than  the  ventral  arm  of  the  proximal  exten- 
sion which  is  the  longer,  ^^'e  thus  consider 
our  specimens  to  represent  a  new  species 
which  we  call  M.  mayri  n.  sp.  in  honor  of 
Dr.  Ernst  Mayr  of  the  Museum  of  Com- 
parative Zoology,    Harvard    University. 

Monoposthioides  mayri  new  species 
Plate  XX,  fig.  42,  a,  b 

L  =  6  ,  1.72,  9  ,  1.4-1.5  mm;  w  =  40  /x; 
esophagus  =  1S5  /x;  Vu  =  90-92%.  Head 
diameter  16  /x.  Lips  distinct,  with  6  setose 
papillae.  Cephalic  setae  17  jx.  Cuticle  with 
broad  rings  in  the  anterior  cervical  region 
whence  they  gradually  become  narrower. 
The  second  annule  is  even  more  enlarged 
than  the  adjacent  ones  and  measures  7  /x 
in  width.  There  are  at  least  12,  probably 
14,  longitudinal  rows  of  V-like  markings 
which  start  about  50  ^u  from  the  anterior 
end.  The  reversal  of  the  V's  takes  place 
in  the  male  at  140  /x  behind  the  posterior 
end  of  the  esophageal  bulb,  in  the  female  at 
240-250  1^1.  (in  M.  anonoposthia  the  V-shaped 
markings  are  reversed  in  the  male  at  a  point 
opposite  the  anterior  margin  of  the  bulb 
and  in  the  female  at  a  point  approximately 
one  and  one-half  tail  lengths  anterior  to 
the  anus ) .  Buccal  cavity  cylindrical,  mea- 
suring 22  /x  from  tip  of  tooth  to  base,  with 
one  large  dorsal  tooth  and  small  subventral 
projections.  Amphids  more  or  less  pocket- 
shaped  and  sunk  into  the  cuticle,  somewhat 
irregular  in  outline.  Esophageal  bulb  47  X  25 
/x.  Single  gubernaculum  42  /x  long,  axe-like, 
proximal  end  23  ^  wide,  gliding  in  an  anal 
sheath  as  reported  for  M.  anonoposthia. 
Tail:  i  ,  140  /x,  9  ,  100-105  /x  long;  a.b.d.  = 
30  /x. 

Jlolotijpe  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4067,  Type  slide  No.  83.  Type  locality, 
M-3,  Key  Biscay  ne. 

Representation  in  samples  studied. — M-3, 
Kev  Biscavne. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        279 


MICROLAIMIDAE 
PARAMICROLAIMUS  Wieser,  1954 
Type       species.— Param/cro/a/mus       primus 
Wieser,    1954:    64,   fig.    135,   a-c. 

In  this  genus  the  first  circle  of  (6) 
cephahc  sense  organs  is  not  papilloid  but 
consists  of  slender  setae  which,  in  the  three 
species  so  far  described,  are  more  than  half 
as  long  as  the  (4)  cephahc  setae  of  the 
second  circle.  In  our  new  species,  the  setae 
of  the  first  circle  are  considerably  shorter 
than  those  of  the  second.  Thus  the  arrange- 
ment of  cephalic  organs  is  very  much  like 
that  of  several  species  of  Microlaimiis.  How- 
ever, further  distinguishing  features  of 
Poromicroloimus  are  the  transversely  oval 
amphids  and  the  distinct  preanal  papillae  in 
the  male.  The  following  species  are  known: 
P.  primus  \\'ieser,  1954  (9  only!),  P. 
papiUatus  (Gerlach,  1954)  (syn.  Micro- 
laimiis papilhtus),  P.  spindifer  Wieser, 
1959,  and  P.  htnatus  n.  sp.  In  addition  to 
the  shorter  cephalic  setae,  P.  lunatiis  is 
characterized  by  the  semicircular  spicula. 

Paramicroloimus  lunafus  new  species 
Plate  XXI,  fig.  43,  a-c 

L  =  1.2-1.3  mm;  w  ^  29-33  ,x.  Head 
diameter  11-13  ^.  Labial  papillae  distinct. 
First  circle  of  cephalic  setae  short,  at  best 
3  fx  long.  Second  circle  of  cephalic  setae, 
7  IX  long  in  (^  ,  11  /x  long  in  9  ■  Amphids 
transversely  oval,  10  fx  wide  (%  c.b.d.)  in 
S  ,8  IX  wide  ( -.^  c.b.d. )  in  9  ;  located  10-13 
IX  behind  anterior  end.  Cuticular  striation 
distinct,  extending  anteriorly  only  to  base 
of  cephalic  setae.  Buccal  cavity  wide,  coni- 
cal, \\'ith  one  large  dorsal  tooth,  2  sub  ventral 
teeth  and  projections  at  the  base.  Esopha- 
gus 145-155  IX  long,  temiinated  by  esopha- 
geal bulb  28  X  22  ix.  Tlie  musculature  of 
the  esophageal  bulb  is  indistinctU-  divided 
into  two  parts  by  a  weak  transverse  break. 
Spicula  45  ix,  regularly  bent.  Gubernaculum 
simple,  19  /x  long.  The  male  bears  5  prom- 
inent preanal  supplements  which  appear  to 
be  tubular  in  structure.  The  cuticle  in  the 
area  of  each  supplement  is  somewhat  swol- 


len. The  3rd  and  4th  preanal  supplements 
are  always  closer  together  than  are  any  of 
the  remaining  ones.  Tail  conical,  85-90  jx 
long,  bearing  6  pairs  of  subventral  setae; 
a.b.d.  =  25-27  ,x. 

Holotype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Xematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4066,  Type  slide  No.  84.  Type  locality, 
M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne. 

CHROMADORIDAE 
CHROMADORINAE 

The  genera  in  this  subfamily  are  arranged 
as  in  ^Vieser  ( 1954 ) ;  that  is,  the  primary 
division  is  into  genera  with  hollow  or  with 
solid  teeth.  \\''ithin  each  of  these  two 
groups  further  separation  is  based  on  the 
structure  of  the  cuticle  (homogeneous  or 
heterogeneous  and  with  or  without  lateral 
differentiation ) . 

HYPODONTOLAIMUS  de  Man,   1888 
Type    species.— Sp///pfiera    inaequalis    Bas- 
tian,  1865:  166,  pi.  13,  figs.  223-225. 

Group  A  in  W'ieser's  key  (1954)  contains 
species  with  very  long  somatic  setae  and 
up  to  now  is  represented  by  six  species. 
Since  classification  of  these  species  leans 
heavily  on  features  of  the  spicular  ap- 
paratus, //.  lieymonsi  (Steiner,  1922),  known 
from  1  9  only,  is  considered  a  species 
inquirenda.  A  key  to  the  species  reads  as 
follows : 

Key  to  Species  of  HypoDO-\TOLAiMUS 
(Group  A  of  \\'ieser,  1954) 

1.  Cephalic    setae    twice    the    head    diameter. 

Pharyngeal  bull")  not  \'ery  powerful,  dorsal 
tooth  weakh'  S-sl:aped,  not  pushed  into 
\entral  Isuccal  wall.  Indistinct  preanal 
papillae  present  .  H.  seto.ms  (Biitschli,  1874) 
Cephalic  setae  measurintj;  about  1  head  diam- 
eter. Phar>ngeal  bulb  powerful,  dorsal 
tooth  strongly  S-shaped,  pushed  into  ven- 
tral buccal  wall.    \o  preanal  papillae  2 

2.  Distal  portion  of  tail  with  ventral  cur\ature. 

First  circle  of   ( 6 )    cephalic  sense  organs 


280 


Bulletin  Miisciiiii  of  Coiiipdidlivc  Zoology,  Vol.  135,  No.  5 


papilloid.      Esophageal     bulb     iiiclistinctb' 

set  off  from  esophagus  — 

H.  colesi  Inglis,  1962 

Distal  portion  of  tail  with  dorsal  curvature. 
First  circle  of  (6)  cephalic  sense  organs 
setose.    Esophageal  bulb  well  set  off  3 

3.  Gubernaculum  with  hook-shaped  apophysis 

H.  steineri  Wieser,  1954 

Cubemaculum   without   apophysis   4 

4.  Gubernaculum  well  de\elopcd,  spoon-shaped, 

half  the  length  of  the  spicula  — - 

....    H.   scJtuunnansstc'khovcni   Cerlach,    1951 
Gubernaculum  reduced  to  a  short  plate  with 
thin  lamella  between  the  2  spicula.    Free 
portion  of  gubernaculum  much  shorter  than 
half  the  length  of  the  spicula  5 

5.  Ciubernacular  lamella  with  dorsal  extension. 

Spicula    evenly    cur\ed,    of    nearly    equal 
width  throughout.    Tail  without  break  .... 

H.  soliv(ip.u.s  Hopper,  1963 

Gubernacular  lamella  minute,  without  ex- 
tension. Spicula  tapering  unevenly  and  of 
characteristic    shape     (see    Fig.    44,    d). 

Tail  with  distinct  break  in  cuticle  _._ 

H.  iiilcrnipttis  n.  sp. 

Group  H  of  Wieser's  key,  comprising 
species  without  elongated  somatic  setae, 
includes  a  few  species  characterized  by  the 
possession  of  a  double  bull).  This  group, 
which  has  been  referred  to  as  a  distinct 
subgenus,  PtijcholaimelJus  Cobb,  1920,  by 
Gerlach  ( 1955 ) ,  is  represented  by  the  fol- 
lowing 4  species: 

H.  carinotufi  (Cobb,  1920),  distinguished 
1)\  long  cephalic  setae  but  imperfectly 
known.  The  specimens  described  under 
this  name  by  Timm  ( 1952 )  almost  certainly 
do  not  belong  to  //.  carinatiis  but  more 
likely  to  one  of  the  following  two  species: 

H.  ponticiis-  Filipjev,  1922.  (For  synonyms 
and  description  see  Gerlach,  1951a.) 

//.  pandispicuhius  Hopper,  1961.  On  the 
basis  of  our  own  material  we  can  confirm 
Hopper's  conclusion  that  this  species  is 
separated  from  //.  ponticus  by  the  .shape 
of  spicula  and  gubernaculum  and  by  the 
shorter  cephalic  setae. 

H.  macro  dent  at  us  Timm,  1961,  character- 
ized by  knob-like  swellings  at  the  base  of 
the  dorsal  tooth,  and  by  tlie  shape  of  the 
spicula. 


Hypodontolaimus  (H.)  interruptus  new  spe- 
cies 
Plate  XXI,  fig.  44,  a-d 

L  =  0.58-0.59  mm;  w  =  23-25  ^;  esopha- 
gus =  95-100  /x.  Head  diameter  15  /x.  Lips 
large.  Cephalic  setae:  first  circle  of  six  = 
2.5  IX,  second  circle  of  four  =  15  fx.  Cervi- 
cal and  somatic  setae  up  to  35  fx  long.  The 
somatic  setae  are  arranged  in  two  sub- 
lateral  rows  on  each  side  of  the  body.  The 
same  holds  for  the  cervical  setae  but  some 
submedian  setae  can  also  be  seen.  Cuticle 
annulated,  with  dots  between  the  annules; 
lateral  differentiation  consisting  of  two 
longitudinal  rows  of  larger  dots  and  faint 
wings.  Distance  of  longitudinal  rows  =  4  ^ 
in  mid-body,  5  ^  in  cervical  region.  Buccal 
cavity  with  powerful  dorsal  tooth.  Pharyn- 
geal bulb  well  developed.  Esophageal  bulb 
20  X  16  IX.  Amphids  elliptical.  Spicula  21 
/(.  long,  tapering  unevenly,  tip  obliquely 
truncate.  Gubernaculum  reduced  to  a  small 
plate  with  minute  lamella  between  the 
spicula.  Tail  85-87  jx  long,  with  break  in 
cuticle  shortly  before  the  middle;  a.b.d.  22  /x. 

Holotijpe  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4070,  Type  slide  No.  85.  Type  locality, 
M-5,  Virginia  Key. 

Representation  in  samples  studied. — M-5, 
Virginia  Key. 

Hypodontolaimus    (Pfycholaimellus)    pondi- 
spiculatus  Hopper,  1961 
Plate  XXI,  fig.  45,  d,  e;  Plate  XXII,  fig.  45, 
a-c,  f 

H t/podontolaimus  {Ptijcholaimellus)  pandispictdatus 
'Hopper,  1961:    360,  361,  figs.  1-4. 

L  =  1.14  mm;  w  =  28-30  /x;  esophagus  = 
ISO  II.  Head  diameter  15  ix.  Cephalic  setae 
(only  second  circle  of  fom-  seen)  5-6  fx. 
Lips  distinct.  Cervical  and  somatic  setae 
short,  in  two  sublateral  rows.  Cuticular 
ornamentation  typical;  longitudinal  rows  of 
dots  5  IX  apart  in  mid-body,  4  /x  on  level 
of  esophageal  bulb.  Buccal  cavity  with 
medium-sized  dorsal  tooth  and  subventral 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        281 


pro|ections.  Pharyngeal  bulb  not  veiy  large. 
Esophageal  bulb  40  X  28  fi,  barrel-shaped, 
double,  the  anterior  portion  smaller  than 
the  posterior  one.  Excretory  pore  22  /x  be- 
hind anterior  end.  Spicula  35  /x  long,  with 
xelum.  Two  gubernacula,  12  fj.  long,  proxi- 
mally  expanded.  Tail  110  /x  long,  a.b.d.  35  fi. 

Representation  in  samples  studied. — M-3, 
Key  Biscayne. 

Geop-aphical  distribution. — Gulf  Shores, 
Alabama  ( Hopper,  1961 ) . 

Remarks. — Our  specimens  differ  from  the 
type  in  the  more  posterior  position  of  the 
excretory  pore,  but  there  is  agreement  on 
all  other  essential  points. 

RHIPS  Cobb,  1920 

Type    species.— Rfi/ps    ornata    Cobb,    1920: 
339,  340,  fig.  118,  a-c. 

This  genus  comprises  two  species,  R. 
ornata  Cobb,  1920,  and  R.  Jon^cauda  Timm, 
1961,  the  latter  being  characterized  by  the 
elongated  tail.  W^e  found  Cobbs  species 
in  our  material  and  shall  give  a  figure  of 
the  head  end,  the  spicular  apparatus  being 
well  described  in  Cobb's  original  publica- 
tion. 

Rhips  ornata  Cobb,  1920 
Plate  XXII,  fig.  46,  a,  b 

Rhips  ornata  Cobb,  1920:    339,  340,  fig.  118,  a-c 

L  =  1.46  mm.  Head  diameter  12  /x.  Lips 
distinct,  with  6  setose  labial  papillae.  Ce- 
phalic setae  5  +  3  ^u,,  in  two  circles.  Amphids 
10  /x  wide.  Head  with  sLx  triangular  plates 
that  probably  serve  as  supports.  Cuticular 
annules  resolvable  into  basketwork-like 
structvires  in  the  cervical  region,  into  elong- 
ated, more  or  less  hexagonal  structures  in 
remainder  of  body.  V-shaped  lateral  dif- 
ferentiations in  posterior  portion  of  body. 
A  few  cervical  setae,  measuring  up  to  15  /x. 
Buccal  cavity  with  large  dorsal  tooth  and 
two  small  subventral  teeth.  Spicula  43  +  28 
/J.  long.  Gubemaculum  (lateral  pieces)  19 
/x.    Tail  130  ,M  long,  a.b.d.  23. 

Representation  in  samples  studied. — M-5, 
Virginia  Key. 

Geographical  distribution. — Ocean  Beach, 


Florida  (Cobb,  1920),  Brazil  (Gerlach, 
1957),  and  ?  Campbell  Islands  (Allgen, 
1932). 

CHROMADORA  Bastion,  1865 

Type     species.— C/iromac/ora     nudicapitata 

Bastian,  1865:  168,  pi.  13,  figs.  230- 

232. 

This  genus  was  discussed  by  Wieser 
( 1954,  1955 ) .  We  found  what  is  probably 
C.  mocrolaimoides  Steiner,  1915,  as  in  our 
specimens  the  distal  end  of  the  guber- 
naculum  seems  to  be  more  strongly  ex- 
panded than  that  figured  by  Steiner  or 
Chitwood  ( 1951 ) .  However,  this  portion 
is  weakly  cuticularized  and  difficult  to  see. 
The  species  is  characterized  by  the  large 
esophageal  bulb,  the  two  small  and  rather 
faint  preanal  papillae,  the  weak  cui"vature 
of  the  spicula  and  the  long  spinneret. 

Chromadora  macrolaimoides  Steiner,  1915 
Plate  XXIi,  fig.  47 

Chromadora  macrolaimoides  Steiner,  1915:  234- 
237,  fiffs.  23-27;  Wieser,  1955  (nee  Steiner, 
1922:  Allien,  1927). 

Chromadorella  macrolaimoides, — Filipjev,  1918; 
Chitwood,  1951. 

Spicula  21  /x  long,  weakly  curved,  with 
velum.  Gubernaculum  15  /x  long,  strongly 
expanded  distally,  edge  slightly  serrated. 
There  are  two  preanal  supplements,  rather 
faint  and  of  the  usual  cup-like  shape  but 
surrounded  by  an  additional  cuticular  dif- 
ferentiation. The  distance  of  the  posterior 
supplement  from  the  anus  is  20  /x.  Spin- 
neret 6  p.  long. 

Representation  in  samples  studied. — M-1, 
Key  Biscayne,  M-4,  Rickenbacker  Cause- 
way. 

Geographical  distribution. — Sumatra 
(Steiner,  1915),  Texas,  Culf  Coast  (Chit- 
wood, 1951),  and  Japan  (Wieser,  1955). 

TIMMIA  Hopper,  1961 

Type    species.— Parac/iromac/ora    parva 
Timm,  1952:  24,  25,  figs.  38,  39. 

Parachromadora  Tinini,  1952,  nee  Micoletzky,  1914, 
nee  Schulz,  1939. 

This  genus  is  distinguished  from  Chro- 


282         Bulletin  Musntin  of  Conipmatwc  Zoology,  Vol.  135,  No.  5 


madorina  Filipjev,  1918,  solely  by  the  oc- 
currence of  a  tubular  supplement  in  addi- 
tion to  the  usual  preanal  papillae  in  the 
male.  Our  material,  from  \'ero  Beach,  con- 
tained representatives  of  the  type  species, 
Timmia  parva  (Timm,  1952)  Hopper,  1961. 
In  addition  to  the  present  locality,  the 
species  is  known  from  Chesapeake  Bay, 
Maryland  (Timm,  1952),  and  Gulf  Shores, 
Alabama  (Hopper,  1961). 

SPILOPHORELLA  Filipjev,  1918 
Type     species.— Sp/7op/iora     paradoxa     de 
Man,   1888:  45-47,  pi.  4,  fig.   19. 

We  seem  to  have  typical  repiesentati\'es 
of  the  cosmopolitan  species,  SpilopliorcUa 
paradoxa,  in  our  material. 

Spilophorelia    paradoxa    (de    Man,    1888) 
Filipjev,   1918 

Spilo])hora  paradoxa  de  Man,  1888:    45-47,  pi.  4, 

fiti.   19. 
Spilophorelia   paradoxa    (de   Man,    1888)    Filipjev, 

1918:    2,59. 

L  =  0.73-0.80  mm;  w  =  31  /-,;  esopha- 
gus =  145  /x;  Vu  =  447c.  Head  diameter  11 
/x.  Cephalic  setae  5  fx.  Esophageal  bulb 
double,  typical.  Spicula  36-43  ^;  guber- 
naculum  30-36  ^.  Tail  120-130  p.  long, 
spinneret  20-22  /x. 

Representation  in  samples  siudied. — M-2, 
Key  Biscayne,  M-6,  Everglades  National 
Park. 

Geographical  distribution. — Cosmo- 
politan. 

PROCHROMADORELLA  Micoletzky,  1924 
Type  species.— Cfiromoc/ora  neopolitana  de 
Man,   1878:    113,   114,  pi.  9,  fig.   17, 
o-c. 
Prochromadorella       mediterranea       (Mico- 
letzky,   1922) 
Plate  XXII,  fig.  48,  a-c;  Plate  XXIII,  fig. 
48,  d,  e 

Chromadora  mediterranea  Micoletzky,  1922b; 
Chromadorella  pontica  Filipjev,  1922;  and  ? 
Hypodontolaimus  arahieiis  Cnlib,   1891. 

L  =  0.5-0.67  mm;  w  =  17-21  ^a;  esopha- 
gus =  100-105  /x;   Vu  =  46%.    Head  diam- 


eter 10-11  fx.  Lips  and  labial  papillae 
distinct.  Cephalic  setae  in  two  circles,  6 
short  ones  (about  1-1.5  /x),  4  longer  ones 
(5  /x).  A  few  cei-vical  setae  up  to  10  /x, 
amongst  which  one  characteristic  circle  of 
four  sublateral  pairs,  two  on  each  side,  at 
about  20  jj.  from  the  anterior  end.  Cuticle 
typical,  with  at  first  dots,  then  elongated 
hexagonal  bodies  and  rods  between  annules. 
Amphids  faint  though  large,  oval.  Excretory 
pore  on  level  of  cephalic  setae.  Buccal 
cavity  with  three  solid,  subequal  teeth. 
Esophagus  enlarged  posteriorly.  Spicula 
semicircular,  chord  18  /x  long.  Guber- 
naculum  10  /x,  distally  slightly  expanded, 
with  two  or  three  teeth.  Two  faint  preanal 
supplements  15  and  27  /x  from  anus,  respec- 
tively. Tail  in  c^  ,  75-100  /.  ( =  6-6.7  a.b.d.), 
in   2,  110  /x  (9  a.b.d.). 

Representation  in  .samples  studied. — M-1, 
Key  Biscayne. 

Geographical  distribution. — Mediter- 
ranean Sea,  Black  Sea,  Red  Sea  (Gerlach, 
1958),  Bay  of  Bengal  (Timm,  1961). 

Remarks. — Our  specimens  are  in  perfect 
agreement  with  the  type  and  with  material 
from  the  Mediterranean. 

CHROMADORELLA  Filipjev,  1918 
Type   species.— C/iromac/ora    filiformis    Bas- 
tian,  1865:  169,  pi.  13,  figs.  242-244. 

Since  our  material  contained  three 
species,  two  of  which  are  new,  we  shall 
provide  a  new  key  to  the  genus  (see  also 
Wieser,  1954). 

Key  to  Species  of  Chromadorella 

1 .  Cuticular    ornamentation     always     consisting 

of  dots  and  rod-like  markings  2 

Cnticular  ornamentation  in  anterior  cervical 
region  consisting  of  solid  liands  with 
crenate  contour,  or  of  fused  hexagonal 
bodies  5 

2.  Foiu"  longitudinal  rows  of  dots  on  each  side 

of  body  C.  circa  tnflexa  Wieser,  1954 

Two  longitudinal  rows  of  dots  on  each  side 
of  body   3 

3.  Lateral  differentiation  beginning  with  cutic- 

ular annulation,   there  measuring  %  to  % 
of  c.b.d.  


Florida  Marine  Nematodes  •  Wieser  and  Hopper        283 


C.  parapoecilosoma  ( Micoletzky,  1922)^ 

Lateral  differentiation  beginning  a  short  dis- 
tance behind  cuticular  annulation,  there 
measuring  not  more  than  ^ik  of  c.b.d.   ____  4 

4.  Longitudinal  rows  Yiq  to  M2  of  c.b.d.  apart; 

membrane  present 

C   memhranata   Micoletzky,    1924 

Longitudinal  rows  i(;  to   Vis  of  c.li.d.   apart; 

membrane  absent  

C.  filiformis   (Bastian,   1865) 

5.  Twelve   preanal   supplements   6 

Five  to  6  preanal  supplements  7 

6.  Lateral   differentiation   irregular   

C.  edmondsoni  Wieser,   1959 

Lateral  differentiation  two  longitudinal  rows, 
widely  spaced  in  anterior  cer\ical  region  ._ 
C.  galeata  Wieser,   1959 

7.  Esophageal    bulb    distinctly   set    off,    barrel- 

shaped,  short.   Spicula  nearly  semicircularly 

curved  C.  parabolica  Wieser,  1954 

Esophageal  bulb  not  so  well  set  off,  more 
elongated.     Spicula   rectangularly  bent   -—  8 

8.  Three  longitudinal  rows,  in  anterior  cervical 

region  Yn)  of  c.b.d.  apart  C.  trilix  n.  sp. 

Two  longitudinal  rows,  in  anterior  cervical  re- 
gion %  of  c.b.d.  apart      C.  vanmeterae  n.  sp. 
We    consider    C    iinjtilicoJa    Filipjev,    1918,    as 
doubtful  since   only   females   are   known. 

Chromadorella  filiformis   (Bastion,    1865) 
Plate  XXIII,  fig.  49,  a,  b;  Plate  XXIV,  fig. 
49,  c,  d 

Chromadora  filiformis  Bastian,  1865;  DicJiroinadora 
tcnuicauda  Schuurmans-Stekhoven,  1950;  CJiru- 
madorella  filiformoides   Chitwood,    1951. 

L  =  0.77  mm;  w  =  26  /x;  esophagus  =  115 
fx.  Head  diameter  14-15  /x.  Lips  distinct. 
Six  short  cephahc  setae  and  4  long  ones, 
measuring  10-11  p..  Ocelli  and  pairs  of 
cervical  setae  20  /x  behind  anterior  end 
( ocelli  sometimes  indistinct ) .  Cuticle  an- 
nulated,  with  transverse  rows  of  dots  be- 
tween annules  which  become  elongated  in 
the  posterior  cen'ical  region.  Lateral  dif- 
ferentiation through  larger  dots  \\'hich  ar- 
range themselves  into  distinct  longitudinal 
rows  at  about  the  level  of  the  ocelli.  Be- 
tween esopliageal  bulb  and  anus  the  longi- 


^  This  species  is  not  well  known.  There  are 
doul^ts  concerning  the  cuticular  ornamentation 
since  Micoletzky  (1922)  and  Schuurmans-Stek- 
hoven (1943)  mention  only  dots  and  rod-like 
markings,  whereas  Wieser  (1951)  figures  solid 
liands  with  crenate  contoiu'. 


tudinal  rows  are  from  3.5-4.5  fx  apart,  that 
is,  Vc,  to  Vk  of  c.b.d.  Buccal  cavity  with  three 
solid,  subequal  teeth.  Esophageal  bulb 
elongated,  30-34  X  15-16  fi.  E.xcretory  pore 
on  level  of  nerve  ring,  60  /x  behind  anterior 
end.  Spicula  5  /x  wide,  strongly  curved; 
true  length  =  30  fx,  chord  =  22  /x.  Guber- 
naculum  19  /x  long,  consisting  of  a  piece  be- 
tween the  two  spicula  and  a  caudal  p)late 
which  distally  ends  in  a  three-pronged  pro- 
jection. There  are  5  large  preanal  supple- 
ments.   Tail  110  /x  long,  a.b.d.  2,5  fx. 

Representation  in  samples  studied. — M-1, 
Key  Biscayne. 

Gcoii,raphica]  distribution. — Atlantic, 
Mediterranean,  Black  Sea,  Red  Sea,  Suma- 
tra, Japan,  Sargasso  Sea,  Texas. 

Remarks. — As  has  been  noted  in  other 
species,  the  appearance  of  the  ocelli  in  dif- 
ferent specimens  is  variable,  probably  due 
to  the  action  of  the  preserving  fluid.  Con- 
sequently, Chit  wood's  species,  C.  filifor- 
moides, cannot  be  maintained. 

A  further  fact  that  should  be  mentioned 
is  that  in  our  specimens  the  longitudinal 
rows  of  dots  seem  to  be  more  widely  spaced 
than  indicated  by  de  Man  (1890)  in  his 
excellent  description  of  European  repre- 
sentatives of  this  species.  Comparative  data 
on  this  point  would  be  desirable. 

Chromadorella   trilix   new  species 

Plate  XXIII,  fig.  50,  a-c;  Plate  XXIV,  fig. 
50,  d,  e 

L  =  0.95  mm;  w  =  28-30  ix\  esophagus  = 
135  /x.  Head  diameter  13  /x.  Lips  distinct. 
Labial  papillae  and  first  circle  of  cephalic 
setae  not  seen.  Four  cephalic  setae  5-6  jx 
long.  Cuticular  ornamentation  consisting  of 
solid  bands  with  crenate  contour  in  anterior 
cervical  region  which  further  posterior  be- 
come resolved  into  rod-like  markings.  Lat- 
eral differentiation  beginning  with  annula- 
tion, at  first  consisting  of  round  markings, 
1.5  fj.  apart,  then  of  two  rows  of  larger  dots. 
About  30  fx  behind  the  anterior  end,  one  of 
the  two  longitudinal  rows  of  dots  moves 
into  the  middle  of  the  lateral  fields  and  is 
replaced  sublaterally  by  a  new  row  of  dots. 


284        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


The  result  is  a  lateral  differentiation  con- 
sisting of  three  longitudinal  rows  which 
run  to  approximately  the  level  of  the  anus 
where  the  middle  row  drops  out.  In  mid- 
body  the  outer  rows  are  3.5-4  /x  apart. 
Buccal  cavity  with  three  solid,  subequal 
teeth.  Esophagus  elongated,  with  three 
fairh-  distinct  interruptions.  Spicula  26  /x 
long.  Gubernaculum  expanded  distally. 
Five  large  supplements,  extending  to  90  /x 
preanal.    Tail  91  ix,  a.b.d.  22  /x. 

1 1  olotype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4067,  Type  slide  No.  86.  Type  locality^ 
M-2,  Key  Biscayne. 

Representation  in  .samjyh's  studied. — M-2, 
Key  Biscayne. 

Cbromadorello  vanmeterae  new  species 
Plate  XXIII,  fig.  51,  a-c;  Plate  XXIV,  fig. 
51,  d,  e 

L  =  1.25  mm;  w  =^  35  ^u;  esophagus  =  160 
IX.  Head  diameter  20  /x.  Head  slightly  swol- 
len. Lips  and  labial  papillae  distinct.  First 
circle  of  cephalic  setae  not  seen,  second 
circle  of  four  setae,  5-6  /x.  Cervical  setae 
about  35  fx,  behind  anterior  end,  somatic 
setae  along  lateral  fields.  Cuticular  orna- 
mentation consisting  of  solid  bands  with 
erenate  contour  in  anterior  cervical  region, 
resolving  into  elongated  markings  further 
posterior.  Lateral  differentiation  by  larger 
dots,  forming  two  longitudinal  rows,  4-6  /x 
apart.  Buccal  cavity  typical.  Esophageal 
bulb  elongated,  with  distinct  plasmatic  in- 
terruptions. Spicula  31  jx  long.  Guber- 
naculum simple.  Six  preanal  supplements, 
extending  to  113  /x  preanal.  Tail  150  /x,  a.b.d. 
23  IX. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  87. 
Type  locality,  M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

Remarks. — This  species  is  dedicated  to 
Miss    Nancy    Van    Meter   who    guided    us 


safely  to  the  muds  of  Florida  Bay,  Ever- 
glades National  Park. 

EUCHROMADORA   de  Man,    1886 

Type    species.— Cfiromadora    vulgaris    Bas- 

tian,  1865:  167,  168,  pi.  13,  figs.  233- 

235. 

This  genus  is  rather  difficult  because  it 
seems  to  contain  a  great  number  of  species 
distinguishable  by  subtle  characters  of  the 
cuticle  and  the  spicular  apparatus.  Most 
descriptions,  however,  are  too  vague  to 
permit  comparison  on  a  sufficiently  de- 
tailed level.  We  agree  with  Inglis  (1962) 
that  the  highly  developed  cuticle  possesses 
a  number  of  features  that  could  be  used  for 
taxonomic  purposes.  Probably  the  most 
suitable  is  the  structure  of  the  annules 
underlying  the  variously  shaped  blocks, 
rods,  and  "basketwork"-types  of  differen- 
tiations. These  annules  have,  along  the 
lateral  line  of  the  body,  anterior  or  poste- 
rior projections  which  may  be  straight, 
notched,  fenestrated,  or  even  entirely  sepa- 
rated from  the  annules,  thus  fomiing  small 
cuticular  pieces  between  the  latter.  All 
these  differentiations  may  be  specific  but 
more   comparative  studies   are   required. 

A  grouping  of  the  genus  seems  to  be 
possible  along  the  lines  indicated  by  Wieser 
( 1954 ) .  Thus,  there  is  a  group  of  species 
in  which  the  dorsal  tooth  is  relatively  small, 
poorly  cuticularized,  forward  pointing  and 
not  embedded  in  pharyngeal  tissue.  This 
group  contains  the  following  species:  E. 
amokurae  (Ditlevsen,  1921),  (syn.  Spi- 
Io))Jiora  amokurae  Ditlevsen,  1921);  E. 
arctica  Filipjev,  1946;  E.  hiederitzi  Steiner, 
1918  (somewhat  doubtful);  and  E.  medi- 
terranea  Allgen,  1942.  All  other  species 
referred  to  this  group  by  Wieser  ( 1954, 
group  B )  are  insufficiently  described  and 
are  considered  species  inquircndac. 

In  the  second  group,  the  dorsal  tooth  is 
large  and  heavily  cuticularized,  its  base 
embedded  in  pharyngeal  tissue.  Two  of 
the  included  species  are  characterized  by 
a  well-developed,  oval  bulb,  distinctly  set 
off  from   the   esophagus,   \'iz.:     E.  Joricata 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        285 


(Steiner,  1916)  (synonyms:  Spilophora 
loricata  Steiner,  1916,  E.  archaica  Steiner 
and  Hoppli,  1926),  and  E.  tijrrlienica 
Brunetti,   1951. 

The  remaining  species  are  difficult  to 
separate.  They  are  Hsted  below,  together 
with  a  short  characterization  of  their  dis- 
tinguishing features,  but  the  original  de- 
scriptions should  be  consulted  in  each  case. 

E.  vulgaris  (Bastion,  1865) 

Chromadom  vulgaris  Bastian,   1865;   E.   frUlcntata 
Allgen,   1929. 

Spicula  asymmetrical. 

E.  striata   (Eberth,   1863) 

Odonfobius  striatiis  Eberth,    1863,   non   E.   striata 
of  Chitwood,    1951    (=E.  gaidica). 

Spicula  70-90  /x,  stout,  central  portion 
enlarged;  lateral  pieces  of  gubernaculum 
about  50%  of  spicula. 

E.  gaulica  Inglis,   1962 

E.  striata  of  Chitwood,   1951,   non   Eberth,   1863; 
E.  chitwoodi  Coles,  1965. 

Spicula  42-53  /x,  proximal  portion  slen- 
der, proximal  and  "crinkled,"  distal  portion 
enlarged;  gubernaculum  50-66%  of  spicula. 

E.  parafricana  Gerlach,  1958 

Spicula  25-27  /x,  stout;  gubernaculum  of 
even  thickness,  boomerang-shaped.  Ce- 
phalic setae  shorter  than  in  above  species. 

E.  pectinata  n.  sp.  (See  below,  p.  286.) 

Closely  related  to  foregoing  species. 
Spicula  just  as  stout  but  more  heavily 
cuticularized,  37-40  /x;  gubernaculum 
sharply  bent,  with  acute  tip.  Comb-like 
ridges  in  buccal  cavity.  Cephalic  setae  even 
shorter.  Cuticular  annules  transversely 
split. 

E.  permutabilis  Wieser,  1954 

Spicula  104-133  /x.  Tail  plump  (4  a.b.d. 
in  9  ).  Gubernaculum  sharply  curved  dis- 
tallv. 


E.  tokiokai  Wieser,   1955 

Spicula  41-50  ^u,  slender  throughout. 
Gubernaculum  66%  of  spicula,  slightly 
cui-ved  distally. 

E.  meadi  n.  sp.  (See  below,  p.  286.) 

First  circle  of  6  cephalic  setae  elongated, 
second  circle — at  least  in  adults — absent. 
The  absence  of  the  second  circle  of  ce- 
phalic setae  distinguishes  E.  meadi  from 
the  remaining  species  of  this  group. 

In  addition  to  these  species  there  are  two 
more  with  unusually  shaped  spicular  ap- 
paratus which  might  not  belong  to  the 
genus  at  all,  viz.,  E.  inflatispicuhtm  Schuur- 
mans-Stekhoven,  1943,  and  E.  kryptospicu- 
hirn  Allgen,  1951. 

All  the  other  species  described  are  here 
considered  as  doubtful,  or  have  been  trans- 
ferred to  other  genera  by  \\'ieser  ( 1954 ) . 

Euchromadora   gaulica   Inglis,    1962 

Plate   XXIV,   fig.   52,   a;    Plate   XXV,   fig. 
52,  b-d 

Euchromadora  gaulica  IngHs,  1962:   260. 

E.  striata  of  Chitwood,   1951,   non  Eberth,   1863. 

E.  chituoodi  Coles,   1965   (  xew  synonymy). 

L  =  1.11-1.57  mm;  w  =  40-52  fx;  esopha- 
gus =  245  /x;  Vu  =  50%.  Head  diameter 
19  /x.  First  circle  of  six  cephalic  setae  short, 
second  circle  of  four,  7-9  /x.  Cuticular 
ornamentation  consisting,  in  front,  of  three 
trans\'erse  rows  of  dots,  followed  by  the 
usual  annules  covered  with  hexagonal 
bodies  which  themselves  are  interconnected 
by  a  meshwork  of  lines.  Further  posterior, 
these  bodies  are  more  elongated  and  form  a 
sort  of  grid  that  links  the  annules.  The 
latter  are  solid  and  possess  lateral  projec- 
tions ( "lateral  plates"  of  Inglis,  1962)  which 
are  directed  anteriorly  in  the  anterior  half 
of  the  body,  posteriorly  in  the  posterior 
half.  Each  projection  fits  into  a  notch  of 
the  following  annule,  thus  forming  a  series 
of  joints.  This  condition  has  already  been 
described  by  Steiner  (1918).  Occasionally 
the  lateral  projections  are  separated  from 
the  annules  and  then  can  be  seen  as  cuticu- 


286        BuIIclin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


lar  i)ieces  between  the  latter.  Traces  of 
pigment  spots  occur  in  the  anterior  cervical 
region.  Buccal  cavity  with  large  dorsal 
tooth  and  sexeral  subventral  and  \entral 
denticles  which  form  a  comb-like  ridge. 
Pharyngeal  bulb  well  developed,  esopha- 
geal bulb  not  set  off,  indistinct.  Spicula 
42-47  fj.  long,  proximal  end  giving  a 
"crinkled"  impression,  distal  two-thirds  di- 
lated. Lateral  plates  of  gubernaculum  25- 
28  /x  long,  proximal  end  straight  or  round 
depending  on  focus,  distal  end  acute,  with 
two  minute  subterminal  denticles;  dorsal 
plate  26-33  ,m.  long,  with  lateral  projections. 
Tail  135-150  jj.  long,  a.b.d.  30  /x  in   6  ■ 

Representation  in  samples  studied. — M- 
1,  Key  Biscayne. 

Geographical  distribution. — Texas,  Gulf 
Coast  (Chitwood,  1951),  Mediterranean 
Sea  (Inglis,  1962),  coast  of  England 
(Coles,  1965). 

Euchromadora  pectinata   new  species 
Plate  XXIV,  fig.  53,  a;  Plate  XXV,  fig.  53, 
b-d 

L  =  1.65-1.97  mm;  w  =  40-50  [x;  esopha- 
gus =  25S-290  im;  \u  =  487c.  Head  diam- 
eter 19  fji.  Labial  papillae  indistinct.  Ce- 
phalic setae:  first  circle  of  six,  papillose; 
second  circle  of  four,  3.5-4  i^.  Cuticular 
ornamentation  beginning,  a  short  distance 
behind  the  cephahc  setae,  with  two  fused 
annules  on  which  faint  longitudinal  stria- 
tion  can  be  seen,  followed  by  single  annules 
which,  however,  from  about  the  middle  of 
the  cervical  region  to  the  middle  of  the 
tail,  are  transversely  split.  Conseciuently 
each  annul(>  consists — at  least  in  the  lateral 
portion  of  the  body — of  two  parts  of  which 
the  posterior  one  is  the  larger.  This  poste- 
rior part  possesses  the  same  lateral  projec- 
tions as  described  for  the  foregoing  species, 
only  less  pronounced.  Moreover,  there  are 
transverse  "lacunae"  on  the  annules  of  the 
mid-body.  The  differentiation  superim- 
posed upon  the  annules  consists  of  the 
usual  hexagonal  blocks  in  the  anterior 
cervical  region,  rod-like  markings  in  the 
remainder  o\  the  body.  These  markings  are 


thinner  than  in  the  foregoing  species.  The 
cuticle  shows  ventral  or  subventral  differ- 
entiations in  the  vulvar  region,  in  the  anal 
region,  and  in  males  at  a  distance  of  120 
fj.  preanal  and  70  /x  postanal,  respectively. 
Buccal  cavity  with  medium-sized  tooth,  not 
as  strongly  cuticularized  as  in  other  species 
of  this  group,  and  a  series  of  comb-like  struc- 
tures. Esophagus  dilated,  no  true  bulb. 
Spicula  6-7  /x  wide,  strongly  cuticularized, 
without  velum,  37-40  jj.  long.  Lateral  plates 
of  gubernaculum  20-22  ix  long,  sharply 
bent  distallv  and  with  acute  tip.  Tail  in 
6  ,  160-180'  ij.  long,  a.b.d.  35  /x,  in  9  ,  200 
IX  long,  a.b.d.  27  /x. 

Ilolotijpc  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4073,  Type  slide  No.  88. 
Type  locality,  M-8,  Biscayne  Bay. 

Representation  in  samples  studied. — M- 
8,  Biscayne  Bay. 

Euchromadora  meadi  new  species 

Plate  XXIV,  fig.  54,  a,  b;  Plate  XXV,  fig. 
54,  c-f 

L  =  1.50  mm;  w  =  40  /x;  esophagus  =  200 
/x.  Head  diameter  19  fx.  Lips  and  labial 
papillae  distinct.  First  circle  of  cephalic 
setae  2  /x,  no  second  circle  in  adults.  In 
juveniles  four  short  setae  could  be  seen 
immediately  behind  the  first  circle  of  six 
setae.  Amphids  fairly  distinct,  12-13  /x 
wide.  Cuticular  ornamentation  beginning 
with  large  annule,  longitudinally  striated, 
followed  by  narrower  annules  which  are 
all  solid  and  show  no  secondary  develop- 
ment except  the  lateral  projections  de- 
scribed in  the  species  above.  Hexagonal 
blocks  verv  thin  even  in  anterior  cei-vical 
region.  Buccal  cavity  with  large  dorsal 
tooth  and  at  least  two  subventral  projec- 
tions, no  comb-like  ridges.  No  distinct 
esophageal  bulb.  Spicula  45—47  /x  long, 
slender,  with  velum.  Lateral  plates  of 
gubernaculum  22-24  /x  long,  hammer- 
shaped;  dorsal  plate  18  /x  long,  with  char- 
acteristic proximal  projection.  In  the  anal 
area    the     cuticular    annules     are    weaklv 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        287 


cuticularized  subventrally  to  form  a  sort 
of  "window."  Tail  135  /x  long,  a.b.d.  35  fx, 
in  juveniles  5-6  a.b.d.  long. 

Holotypc  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  89. 
Type  locality,  M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

ATROCHROMADORA  Wieser,  1959 
Type  species.— Sp//op/iora   parva   de  Man, 
1893:  89-91,  pi.  5,  fig.  5. 

We  found  what  appeared  at  first  to  be 
A.  parca  (de  Man,  1893),  but  closer  study 
revealed  the  following  differences: 

parva  dcniiculata 
Spacing  of  longitudinal 

rows    (mid-body)                     3.5  ^  6 /^ 

Length  of  spicula                       17-18  ix  34  ^ 

Distal  end  of  gubernaculuni     straight  denticulated 

Length  of  spinneret                      12  ^  7-8  ^ 

=  V-  tail  =  lis  tail 

Afrochromadora  denficulafa  new  species 
Plate  XXVI,  fig.  55,  a-f 

L  =  0.79-0.80  mm;  w  =  31-32  /x;  esopha- 
gus =  96-109  ix;  Vu  =  467f .  Head  diameter 
11  fx.  Lips  distinct,  labial  papillae  and  first 
circle  of  cephalic  sense  organs  not  seen. 
Cephalic  setae  5-6  fx.  Cervical  setae  as  in 
A.  parva.  Amphids  distinctly  spiral,  but 
small.  Cuticular  ornamentation  beginning 
with  transverse  rows  of  dots,  followed  by 
annules.  In  the  lateral  region  of  the  body 
a  few  dots  or  rod-like  markings  can  be  seen 
between  the  annules,  particularly  two 
longitudinal  rows  of  larger  dots,  the  spac- 
ing of  \\'hich  is  5-6  fx  between  cervical  re- 
gion and  anus.  Buccal  cavity  with  three 
solid,  subequal  teeth.  Esophageal  bulb 
round.  Excretory  pore  37  fx  in  front  of  end 
of  esophagus,  \'entral  gland  reaching  to  48 
fx  behind  the  esophagus.  Spicula  nearly 
semicircular,  3  /x  wide,  34  ^  long.  Guber- 
naculum  23  fx  long,  strongly  dilated  distally, 
with  denticulated  end  plate.  Tail  124-128 
IX  long,  a.b.d.  28  /x  in    6  ,  20  /x  in    9  .    Spin- 


neret 7-8  fx.  In  6  there  is  a  slight  ventral 
swelling,  45  fx  postanally. 

HoJotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  90. 
Type  locality,  M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

COMESOMATIDAE 

MESONCHIUM  Cobb,  1920 
Type  species.— /Vlesoncfi/um  poriferum  Cobb, 
1920:  294,  295,  fig.  76. 

Pepsonema  Cobb,  1920:  295. 

The  following  fixe  species  are  known: 
M.  poriferum  Cobb,  1920;  M.  pcllucidum 
(Cobb,  1920)  syn.  Pepsonema  pcllucidum; 
M.  nini  Inglis,  1961;  M.  punctatum  Timm, 
1961;  M.  janetae  Inglis,  1963.  They  can  be 
separated  mainly  by  the  shape  of  their 
spicula  and  by  the  pattern  of  their  lateral 
cuticular  differentiations.  Moreover,  M. 
janetae  has  shorter  cephalic  setae,  and  M. 
nini  amphids  with  more  turns  than  the  rest 
of  the  species. 

Our  species  agree  with  the  description  of 
M.  pcllucidum  by  Cobb  (1920)  except  that 
in  the  posterior  portion  of  the  body  we 
observed  four  longitudinal  rows  of  dots, 
whereas  Cobb  states  that  near  the  tail 
"there  are  sometimes  six  or  possibly  eight 
rows."  We  cannot  agree  with  Timm's 
(1961)  synonymization  of  M.  pcllucidum 
with  M.  poriferum  since  in  the  tvvo  species 
the  longitudinal  rows  are  differently  ar- 
ranged, and  in  M.  pcllucidum  the  spicula 
are  retrorsely  barbed  (a  fact  already  men- 
tioned by  Cobb),  whereas  in  M.  poriferum 
they  are  acute  distally.  The  description  of 
M.  punctatum  Timm  is  confusing.  In  the 
text  the  presence  of  the  usual  number  of 
four  cephalic  setae  is  mentioned,  whereas 
the  figure  shows  the  head  equipped  with 
what  seems  to  be  eight  or  ten  setae.  In  the 
text  the  amphids  are  said  to  describe  2.5 
turns;  in  tlie  figure  they  definitely  have 
three  turns.    Even  without  these  confusing 


288        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


discrepancies  it  would  be  difficult  to  give 
unequivocal  reasons  %\'hy  M.  punctatum 
and  M.  porifcrum  .should  be  distinct  spe- 
cies. 

hAesonchium  pellucidum    (Cobb,    1920) 
Plate  XXVII,  fig.  57,  a-d 

Pepsonema    pvUucidiim    Col^b,     1920:     295,    296, 

fig.  77. 

L  =  1.77-2.00  mm;  w  =  73-75  ju;  esopha- 
gus =  250  IK  \'u  =  477< .  Head  diameter 
15  fx.  Cephalic  setae  12  /j..  Amphids  10  p., 
2.5  turns.  Buccal  cavity  cylindrical,  23  ij. 
long,  with  three  teeth.  (The  heads  of  our 
male  specimens  agree  in  eveiy  respect  with 
Cobb's  figures.)  Cuticular  differentiation: 
In  both  sexes  there  are  four  longititdinal 
rows  of  coarser  dots,  running  from  behind 
the  amphids  to  behind  the  end  of  the 
esophagus.  From  there  on  the  two  outer 
rows  gradually  disappear  and  the  whole 
lateral  field  is  raised  to  form  a  lateral  wing. 
In  mid-body  there  are  only  two  longitudi- 
nal rows  of  coarser  dots.  In  the  posterior 
portion  of  the  body  the  wings  are  flattened 
again  and  the  two  outer  rows  of  dots  re- 
appear. The  spaces  between  the  coarser 
dots  are  resolvable  with  difficulty  into 
transverse  rows  of  minute  and  closely 
spaced  dots  which  are  continuations  of  the 
usual  transverse  rows  of  dots  into  the 
lateral  fields.  Esophagus  with  a  barrel- 
shaped  posterior  bulb  and  a  small  cardia. 
Ventral  gland  opposite  end  of  esophagus, 
excretory  pore  behind  nerve  ring.  Spicula 
110  ;u,  long,  proximally  cephalate,  distally 
retrorsely  barbed.  Cubernaculum  heavy, 
with  30  IX  long  apophysis.  There  are  about 
16  indistinct  tubular  supplements.  Tail  200 
ij.  long,  a.b.d.  45-50  /x. 

Representation  in  .samples  studied. — M- 
8,  Biscayne  Bay. 

Geographical  distribution. — Kingson,  Ja- 
maica (Cobb,  1920). 

SABAIIERIA  de  Rouville,  1903 
Type  species.— Soba/ier/a  cettensis  de  Rou- 
ville, 1903:    11. 

Two  closely  related,  but  distinct,  species 
of    Sabatieria   were   found    in    the    sample 


from  Vero  Beach.  Following  the  key 
given  by  W^ieser,  1954,  they  belong  to 
the  group  encompassing  S.  cupida  Bresslau 
and  Schuurmans-Stekhoven  in  Schuurmans- 
Stekhoven,  1935,  S.  heterura  (Cobb,  1898), 
S.  rugosa  Schuurmans-Stekhoven,  1950,  S. 
siinilis  (Allgen,  1933),  and  S.  tcnuicaudata 
(Bastian,  1865). 

The  two  new  species,  S.  paradoxa  and  S. 
paracupida,  can  be  separated  from  the 
above-mentioned  related  species  by  use  of 
the  following  key. 

Key  to  Group  Based  on  Male  Characteristics 

1.  Supplements   28-32    

S.  teniiicaudata  (Bastian,  1865) 

( doubtful  species ) 
Supplements    15-22   2 

2.  Cephalic  setae  more  than  %  of  head  diam- 

eter in  length   3 

Cephalic   setae   less   than   %   of  head   diam- 
eter in  length   4 

3.  Spicules     46-50     ^     long,     1.2     anal     body 

diameters     

S.   cupula  Bresslau  and   Schuurmans-Stek- 
hoven in  Schuurmans-Stekhoven,  1935 
Spicules    63-68   ix   long,    1.7-1.8    anal    body 
diameters   S.   paracupida  n.   sp. 

4.  Spicules  1.6  or  more  anal  body  diameters  in 

length     5 

Spicules   1.3  or  less  anal  body  diameters  in 
length     6 

5.  Spicules  85  |t  long,  vv'ithout  proximal  cepha- 

lation  S.  heterura  (Cobb,  1898) 

Spicules  60-62  m  long,  proximally  distinctly 
cephalated  S.  paradoxa  n.  sp. 

6.  Amphid    with    2.5    turns    

S.    rugosa   Schuurmans-Stekhoven,    1950 

Amphid   with   2   turns   

S.  similis  (Allgen,  1933) 

SabotieriQ  paradoxa  new  species 
Plate  XXVII,  fig.  58,  a-d 

L  =  1.46-1.66  mm;  w  =  42-44  /..  Diam- 
eter at  base  of  esophagus  3^39  /x.  Esopha- 
gus 135-140  IX.  Tail  140-142  /x  (4.0-4.1 
a.b.d.).  Head  diameter  13  ^.  Labial  and 
cephalic  papillae  distinct.  Ceplialic  setae 
5-6  IX  long,  38-46%  of  head  diameter.  Cer- 
vical setae,  short,  widely  dispersed,  somatic 
setae  shorter  and  more  widely  spaced. 
Amphids  spiral,  with  2.25  turns;  7  jx.  wide 
(54%  of  head  diameter).    Lateral  differen- 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper 


289 


tiation  prominent,  the  transverse  rows  of 
punctations  overlying  the  lateral  chords 
being  coarser  and  more  widely  spaced  than 
on  the  remainder  of  the  bodv.  Spicules 
60-62  /x  long  (1.7-1.8  a.b.d.),'  proximally 
cephalated,  distally  quite  narrowed.  Gu- 
bernaculum  with  lateral  guiding  pieces. 
Apophyses  of  gubernaculum  25-27  /x  long. 
Preanal  supplements  minute,  17-19  in  num- 
ber. Caudal  setae  present,  arranged  as 
illustrated  (PI.  XXVII,  fig.  58,  b). 

HoJohipc  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4075,  Type  slide  No.  92. 
Type  locality,  V,  Vero  Beach. 

Representation  in  samples  studied. — M- 
7,  Everglades  National  Park,  V,  Vero 
Beach. 


SabaWer'ia  paracupida   new  species 
Plate  XXVIII,  fig.  59,  a-c 

L  =  1.7-1.85  mm;  w  =  47-52  /x.  Diam- 
eter at  base  of  esophagus  44  fi.  Esophagus 
180-184  fx.  Tail  148-150  /x  (4  a.b.d.).  Head 
diameter  13-14  fi.  Labial  and  cephalic 
papillae  distinct.  Cephalic  setae  10  fx  long, 
71-77%  of  head  diameter.  Cervical  and 
somatic  setae  similar  to  preceding  species. 
Amphids  spiral,  with  2.25  turns;  9  /x  wide 
( 64-70%  of  head  diameter ) .  Lateral  differ- 
entiation as  in  preceding  species.  Spicules 
63-68  IX  long  (1.7-1.8  a.b.d.),  lacking  proxi- 
mal cephalation.  Gubernaculum  with  lat- 
eral guiding  pieces.  Apophyses  of  guber- 
naculum 20  fx  long.  Preanal  supplements 
minute,  19-22  in  number.  Caudal  setae 
present,  arranged  as  illustrated  (Pi.  XXVIII, 
fig.  59,  c). 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4075,  Type  slide  No.  93. 
Type  locality,  V,  Vero  Beach. 

Representation  in  samples  studied. — V, 
Vero  Beach. 


AXONOLAIMIDAE 

AXONOLAIMUS  de  Man,  1889 

Type    species.— Anop/osfoma    spinosum 

BiJtschli,    1874   sensu    de   Man,    1888: 

19-21,  pi.  2,  fig.  n. 
Axono/a/Vnus  hexapiius  new  species 
Plate  XXVIII,  fig.  60,  a-e 

L  =  6  ,  1.8,  ?  ,  1.9  mm;  w  =  i  ,  28,  9  , 
35  /x;  diameter  at  base  of  esophagus  28,  31 
ya.  Esophagus  172,  137  /x.  Vu  =  54%.  Tail 
128,  138  fx  (4.3-4.6  a.b.d.).  Head  diameter 
13,  14  IX.  Head  with  six  labial  papillae  and 
four  cephalic  setae  (21-24  /x  long).  An 
additional  21-24  /x  long  seta  is  located 
about  5  n  behind  each  amphid.  Amphid 
12-13.5  IX  long  by  6-7  /x  wide,  slightly  to 
distinctly  open  loop-shaped,  8  /x  from 
anterior  end.  Buccal  cavity  14-16  /x  deep, 
of  which  the  posterior  conoid  portion 
makes  up  about  10  /x.  Anterior  part  of 
buccal  cavity  with  six  weak  odontia. 
Esophagus  clavate,  10  /x  wide  at  base  of 
buccal  cavity,  12  /x  at  the  nerve  ring,  ex- 
panding in  the  posterior  %  to  20  /x  at  the 
base.  Excretory  pore  slightly  behind 
middle  of  conoid  portion  of  the  buccal 
cavity.  Excretory  pore  ampulla  40  ^  from 
head  end,  at  level  of  "break"  in  esophageal 
musculature.  Renette  cell  large,  posterior 
to  base  of  esophagus.  Pseudocoelomocyte 
immediately  behind  renette  cell.  Cuticle 
finely  striated.  Somatic  setae  about  5  /x 
long. 

Female  didelphic,  amphidelphic,  ovaries 
outstretched;  with  two  eggs  observed  in 
the  posterior  uterus  (eggs  75-110  /x  by  30 
ix).  Male  diorchic,  testes  opposed,  out- 
stretched. Spicules  39  ix  long.  Gubernacu- 
lum short,  with  12  ix  long  apophyses.  Sup- 
plementary organs  glandular,  with  minute 
exit  pores.  Male  and  female  tail  obtusely 
conoid.  Caudal  setae  arranged  as  illus- 
trated (PI.  XXVIII,  fig.  60,  c),  those  near 
terminus  longer  than  the  remainder.  Spin- 
neret three-lobed,  the  associated  glands 
located  in  the  tail. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes.  Ento- 
mology Research  Institute,  Ottawa,  Collec- 


290         Bulletin  Mu.scui)}  of  Coiiijxinitivc  Zoology,  Vol.  135,  No.  5 


tion  Number  -4070,  Type  slide  No.  94. 
Type  loeality,  M-5,  Virginia  Key. 

Representation  in  samples  studied. — M- 
5,  X'irginia  Ke\-. 

Remarks. — A.  Jie.xapihis  n.  sp.  can  be 
distinguished  from  all  other  A.X(»iokiimus 
species  by  the  presence  of  a  long  lateral 
seta  associated  near  the  base  of  each  am- 
phid.  In  possessing  this  character,  along 
with  the  tri-lobed  spinneret,  this  animal 
has  some  relationship  to  the  monodelphic 
genus  Synodontium  Cobb,  1920. 

ODONTOPHORA  Butschli,  1874 

Type   species.  —  Oc/onfopfiora   morina 

Butschli,  1874:    285,  pi.  3,  fig.  13. 
Odontophora  voriabilis  new  species 

Plate  XXVIII,  fig.   61,   a,   b;   Plate  XXIX, 

fig.  61,  c-e 

L  =  1.3-2.2;  w  =  36-50  /.-;  Vu  =  53-56%. 
Diameter  at  base  of  esophagus  29-33  /x. 
Esophagus  112-135  jx  long.  Head  diameter 
11-15  /x.  Head  with  cephalic  papillae,  4 
cephalic  setae  and  3  circles  of  subcephalic 
setae.  Cephalic  setae  14-17  /x  long  in  6  , 
11-14  IX  in  9  .  First  circle  of  subcephalic 
setae  (paramphidial )  11-13  fx,  second 
circle,  S-9  /x,  third  4-5  /x  long.  The  sub- 
cephalic setae  all  occur  within  the  range 
of  the  buccal  cavity.  Cervical  setae  located 
1.5-2.5  /x  posterior  to  base  of  buccal  cavity, 
the  first  circle  the  longest,  7-8  /x,  the  next 
two  following  closely  and  being  progres- 
sively shorter.  Amphid  (S-10  /x  long.  Buccal 
cavity  23-2.5  /x  long,  the  posterior  conoid 
portion  being  17  /x  long.  Buccal  cavity  with 
6  odontia,  the  odontia  with  prominent 
apophyses  to  which  are  attached  longi- 
tudinally directed  muscles,  an  arrangement 
that  undoubtedly  serves  to  evert  the  odon- 
tia. I^xcretory  pore  at  base  of  odontia. 
Eggs  100-110  /x  long  by  45  /x  wide,  two  per 
uterus.  Spicula  sickle-shaped,  47-49  /x  long, 
the  chord  being  32-33  /x.  Gubernaculum 
with  12-13  /x  long  apophysis.  Preanal  sup- 
plements not  observed.  Tail  length  vari- 
able. 70-117  /x  long.   Male  tail  with  numer- 


ous caudal  setae  and  with  tenninal  setae 
17-18  IX  long.  Female  tail  with  fewer 
caudal  setae  and  devoid  of  terminal  setae. 

Holofype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4075,  Type  slide  No.  95. 
Type  locality,  V,  Vero  Beach. 

Representation  in  samples  studied. — V, 
Vero  Beach. 

Remarks. — The  specific  name  variabilis 
ser\es  to  indicate  the  variation  found  within 
the  specimens  recovered  from  the  single 
sample  taken  from  the  Vero  Beach  area. 
Body  dimensions,  such  as  total  length, 
maximum  width  and  tail  length,  are  quite 
variable.  Morphological  stioictures,  how- 
ever, i.e.,  spicula,  buccal  cavity,  amphid, 
distribution  and  size  of  cephalic  and  sub- 
cephalic setae,  were  conspicuously  in  close 
agreement.  In  particular,  the  shapes  of  the 
spicula  were  nearly  identical  despite  the 
fact  that  the  tails  in  the  three  male  speci- 
mens observed  measured  70  ^,  85  ^i,  and 
117  /x  in  length. 

O.  variabilis,  in  possessing,  in  addition  to 
the  cephalic  setae,  three  circles  of  subce- 
phalic setae  which  could  be  interpreted  as 
being  arranged  in  four  submedian  rows,  is 
related  to  O.  aniiustilaimoides  Chitwood, 
1951,  and  perhaps  may  be  that  species. 
However,  no  mention  is  made  of  the  pres- 
ence of  cervical  setae  in  O.  angiistilaimoides, 
these  being  a  prominent  feature  in  O.  vari- 
abilis. In  addition,  the  size  of  the  eggs  and 
the  number  of  eggs  per  uterus  differs. 
Moreover,  since  no  males  have  been  de- 
scribed of  O.  angustilaimoides,  this  species 
is  ot  doubtful  status. 

PARODONTOPHORA  Timm,  1963 

Type  species.— Pseuc/o/e//a  paragranulifera 

Timm,  1952:  45,  pi.  9,  fig.  78. 
Parodontophora  brevomphida  (Timm,  1952) 

Timm,  1963 

Pscudoldla  brcvanipliida  Timm,   1952:   44,  45,  pi. 

9,  fi.us.  76,  77. 
Parodontophora    ])acifica     ( Allgen )     sensu    Timm, 

1963:  35,  36,  fig.  1,  y,  h  (nf;av  synonymy). 


Florida  Marine  Nematodes  •   Wicscr  and  Hopper        291 


L  =  1.1-1.4  mm;  ^^'  =  36-lS  /x;  Vu  =  47- 
50%.  Body  bearing  four  rows  of  somatic 
setae,  two  on  either  side  of  each  lateral 
chord.  Anteriorly  the  somatic  setae  are 
about  4  fx  long  in  the  region  of  the  nerve 
ring.  At  the  vuKa  their  length  is  reduced 
to  less  than  2  /x.  Except  for  the  anterior- 
most  14  /A,  the  cuticle  is  finely  striated.  One 
dorsal  and  two  ventral  setae  are  located  in 
the  region  of  the  amphid.  Amphid  short 
shepherd's  crook,  with  the  ventral  ann 
longer  than  the  dorsal.  Dorsal  arm  about 
10  11  long,  ventral  arm  14-17  ^  long.  Am- 
phid located  on  a  level  with  and  distinctly 
shorter  than  the  19  /x  long  posterior  cylin- 
drical portion  of  the  stoma.  Head  rounded, 
with  six  lips,  six  labial  papillae,  and  four 
7-8  IX  long,  cephalic  setae  (75%  of  corre- 
sponding body  diameter).  There  are  no 
subcephalic  setae.  Anterior  part  of  stoma 
with  six  prominent  odontia,  posteriorly 
cylindrical,  25-28  /x  deep.  Esophageal 
diameter  increasing  posteriorly;  without 
basal  bulb.  Approximately  15  ^a  posterior 
to  the  base  of  the  buccal  cavity  the  tuboid 
esophageal  marginal  rays  commence.  Nerve 
ring  encircling  esophagus  at  approximately 
65%  of  its  length.  Excretory  pore  not  de- 
tected, the  ampulla,  however,  located  at 
the  base  of  buccal  cavity.  Renette  cell  65 
/u.  long  (407c  of  esophageal  length),  located 
posterior  to  base  of  esophagus.  A  promi- 
nent pseudocoelomocyte  occurs  immedi- 
ately posterior  to  the  renette  cell. 

Female  didelphic,  amphidelphic,  ovaries 
outstretched.  Eggs  not  observed.  Male 
diorchic,  testes  opposed,  outstretched. 
Spicules  arcuate,  31-36  ^  long  (according 
to  angle  of  view).  Gubernaculum  8-10  ^ 
long,  arcuate.  Male  and  female  tail  elon- 
gate-conoid, 135-155  fj-  long,  the  terminal 
Vs  nearly  cylindrical.  Subterminal  setae 
present  on  some  specimens.  Male  with  a 
short  preanal  seta  and  two  subventral  rows 
of  7-8  setae.  Spinneret  present,  the  gland 
cells  located  in  the  anterior  portion  of  the 
tail. 


Representation  in  samples  studied. — V, 
Vero  Beach. 

Geographical  distribution . — Chesapeake 
Bay,  Maryland  (Timm,  1952),  Bay  of 
Bengal  (timm,  1961),  Arabian  Sea  at 
Karachi  (Timm,  1962),  Maldives  (Ger- 
lach,  1962). 

Remarks. — At  the  present  time  consider- 
able confusion  exists  in  the  taxa  that  have 
been  identified  as  P.  paeifica  (Allgen, 
1947)  (syn.  Odontophora  p.  Allgen).  Ger- 
lach  ( 1962 )  considers  a  variety  of  species 
as  synonyms  of  Allgen's  species.  In  his  list 
of  synonyms,  Gerlach  lumps  species  in 
\\'hich  the  amphids  are  18  ^  long  together 
with  species  in  \\'hich  the  amphids  are  80- 
150  /x  long.  This  action  seems  a  bit  prema- 
ture and  is  considered  doubtful.  Timm 
(1963)  in  creating  the  genus  Farodonto- 
phora  for  species  of  Odontophora  with 
parallel  stomatal  walls,  intimates  that  P. 
paeifica  might  best  be  considered  as  a 
species  inquirenda,  validating  P.  quadri- 
sticha  ( Schuurmans-Stekhoven,  1950).  We 
agree  with  Timm's  statement  and  herein 
consider  P.  paeifica  (Allgen,  1947)  as  a 
species  inquirenda.  However,  in  support- 
ing such  a  move  we  do  not  feel  that  all 
subsequent  P.  paeifica  descriptions  refer 
to  P.  quadristicha  and  feel  that  P.  brevam- 
phida  can  be  suitably  differentiated  from 
Schuurmans-Stekhoven's    species. 

For  the  present  the  following  differences 
can  be  used  to  separate  the  two  species 
under   consideration: 

P.  quadristicha  [syn.  O.  paeifica  Allgen 
of  Wieser,  1956]  has  amphids  24-32  fx  long, 
a  buccal  cavity  40  /x  long,  and  a  renette 
cell  90-100  ^'long  (app.  50-55%  of  the 
esophageal  length ) . 

P.  brevamphida  [syn.  Pseudolella  pa- 
eifica (Allgen,  1947)  of  Timm,  1961, 
Odontophora  paeifica  Allgen  of  Gerlach, 
1962,  and  Parodontophora  paeifica  (Allgen, 
1947)  of  Timm,  1963]  has  amphids  12-18 
/x  long,  a  buccal  cavity  25-31  jj,  long,  and  a 
renette  cell  32-65  ,i  long  (app.  30-40%  of 
the  esophageal  length). 


292        Bidhiin  Mii.scuiii  of  Comparative  Zoology,  Vol  135.  No.  5 


LEPTOLAIMIDAE 

ALAIMELLA  Cobb,  1920 

Type    species.— A/a/me//a     fruncafa     Cobb, 

1920:  234,  fig.  7b. 
Alaimella  cincto  Cobb,  1920 

Plate  XXIX,  fig.  63,  a-c 
AlaimelU  cincta  Colih,  1920:  233-234,  fig.  7a. 

Mu]c,—L  =  1.3  mm;  w  =  16  /x;  diameter 
at  base  of  esophagus  13  ix.  Esophagus  250 
ja  long,  \\  itli  conoid  cardia.  Head  7  /x  wide, 
bearing  six  papillae  and  four  12-13  /x  long 
cephalic  setae.  Aniphid  7  p.  wide,  with 
central  raised  "fleck."  Cuticle  coarsely 
annulated,  the  annules  bearing  prominent 
longitudinal  markings.  Spicules  27  /x  long 
(chord  22  /x),  proximally  cephalated.  Gu- 
bernaculum  9  /x  long,  fail  100  /x  long  (7 
a.b.d. ),  with  at  least  one  caudal  seta. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

Geographical  distribution.  — Bis- 
cayne Bay,  Florida  (Cobb,  1920),  Aransas 
Bay.  Texas  (Chitwood,  1951). 

Remarks. — The  central  raised  portion  of 
the  amphid  in  our  specimens,  while  illus- 
trated by  Chitwood  for  his  example,  was 
not  originally  depicted  by  Cobb.  This 
feature  is  somewhat  suggestive  of  A.  trun- 
cata  Cobb,  1920.  However,  regarding  the 
cuticle  of  A.  truncata,  Cobb  states,  "Sec- 
ondary markings  of  the  cuticle  faint,  if 
any."  The  Miami  specimens,  by  manifest- 
ing prominent  longitudinal  markings,  can- 
not be  regarded  as  representative  of  A. 
fruneata  in  view  of  this  strong  statement 
questioning  the  presence  of  such  markings. 
Thus,  even  though  the  amphid  on  our 
specimens  has  a  central  fleck,  we  feel  the 
remainder  of  the  characters  support  our 
view. 

CAMACOLAIMIDAE 
CAMACOLAIMUS  de  Man,  1889 

Type     species.— Camoco/a/mus     tardus     de 

Man,  1889a:  8. 
Comaco/a/mus  prytherchi  Chitwood,   1935 
Plate  XXIX,  fig.  64,  a-c 

CamucoUiimus    prytherchi    Chitwood,     1935:     49, 
50,  fig.  7,  a-c. 


L  =  5  ,  1.8,    9  ,   1.9;  w  =:  5  ,  33,    $  ,  39; 
diameter   at   base    of   esophagus    .30-32   /x; 
esophagus  240-250 /x  long.  Vu  =  54%.  Tail 
90-105  ^  (3.1-3.4  a.b.d.).    Head  13-14  ix 
wide,  bearing  six  labial  papillae  and  four, 
8-10  /x  long,  cephalic  setae.    Somatic  setae, 
short  and  thick,  staggered  in  region  over- 
lying lateral   chords,   widespaced.    Cuticle 
finely  striated.    Lateral  surfaces  with  slight 
"bulges"    anteriorly,    progressively    getting 
higher  posteriorly  until  finally  lateral  alae 
are  formed  at  a  point  approximately  mid- 
way   from    vulva    to    anus,    fading    away 
quickly  on  the  tail.    Dorsal  tooth  and  its 
base,   15  /x  long.    Esophagus   divided  into 
three  distinct  regions — corpus,  isthmus,  and 
swollen    terminal   region.     Nerve   ring   en- 
circling   isthmus    immediately    behind    the 
coipus.      Tenninal    region     gradually    en- 
larged.   Cardia  8  /x  long  and  12  /x  wide. 
Spicules  53  /x  long.    Male  tail  with  caudal 
alae  and  a  pair  of  postanal  setae. 

Representation  in  samples  studied. — V, 
Vero  Beach. 

Geographical  distribution. — North  Caro- 
lina (Chitwood,  1933);  Chile  (Wieser, 
1956);  Maldives  (Gerlach,  1962). 

Remarks. — An  examination  of  the  type 
specimen  of  C  prytherchi  discloses  the 
presence  of  both  the  caudal  alae  and  the 
postanal  setae.  We  wish  to  thank  Mr.  Cur- 
tis Sabrosky  for  making  the  specimens 
available  for  study. 

DIPLOPELTIDAE 

Some  genera  of  this  family  have  recently 
been  transferred  to  the  family  Linhomoei- 
dae  by  Gerlach  (1963a).  We  postpone 
judgment  on  these  changes  until  the  ma- 
terial for  our  monograph  is  complete  and 
shall  adhere,  in  this  paper,  to  the  old  clas- 
sification. 

PARATARVAIA  new  genus 

Type  species.— Paraforvo/a  sefa  n.  sp. 

Definition. — Diplopeltidae  (or  Linho- 
moeidae?).  Head  with  6  labial  papillae 
and  10  cephalic  setae  in  two  circles,  6  +  4. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        293 


Amphid,  doulile  spiral,  situated  on  punc- 
tated plaque.  Esophagus  cylindroid  with 
small  basal  bulb,  without  valves.  Cardia 
small.  Female  didelphic,  amphidelphic. 
Male  without  supplements.  Spicules  arcu- 
ate, gubernaculum  with  medial  piece, 
lateral  guiding  pieces  and  posteriorly  di- 
rected apophyses. 

Remarks. — Parotorvoia  possesses  charac- 
ters that  show  affinity  with  both  the 
Diplopeltidae  (spiral  amphid,  amphid  on 
plaque )  and  the  Linhomoeidae  ( reduced 
buccal  cavity  and  spicular  apparatus  as  in 
Terschellingio,  number  of  cephalic  sensory 
organs ) .  However,  as  we  are  going  to  post- 
pone our  judgment  of  the  recent  proposals 
by  Gerlach  (1963),  we  provisionally  assign 
Paratorvaia  to  the  Diplopeltidae.  Within 
the  Diplopeltidae  Parafarvoia  is  related  to 
both  Tarvaia  Allgen,  1934,  and  Disconema 
Filipjev,  1918.  It  is  distinguished  from 
Tarvaia  by  possessing  10  cephalic  setae,  as 
opposed  to  four.  From  Disconema,  which 
lacks  a  plaque,  Paratarvaia  is  separated  by 
possessing  a  punctated  plaque  as  well  as 
by  having  the  10  cephalic  setae  separated 
into  two  circles  of  6  and  4. 

Paratarvaia  sefa  new  species 

Plate  XXX,  fig.  65,  a-c 

L  =:  l.;3-1.4  mm;  \\-  =  23-24  /x;  width  at 
base  of  esophagus  20-21  fx.  Head  13  ^. 
wide,  bearing  an  internal  circle  of  six  setose 
labial  papillae  and  a  divided  external  circle 
of  10  (6  +  4)  setae,  16  +  18  /x  in  length. 
Amphid  a  double  spiral,  21-23  /x  long  and 
16-17  fx  wide,  situated  on  a  punctated 
plaque  25-27  /jl  long  and  17  {x  wide.  In 
dorsal-ventral  view  the  plaques  are  seen 
to  be  separated  by  a  distance  of  2-3  /x. 
Cuticle  coarsely  annulated,  the  annules 
about  1.5  IX  wide  just  posterior  to  the  am- 
phids,  about  2.0  ix  at  mid-body  and  about 
1  IX  wide  on  the  tail.  Body  with  fine  so- 
matic setae  which  extend  onto  the  tail. 
Buccal  cavity  extremely  reduced,  \\'ith 
minute  lips  protruding  from  the  head  con- 
tour. Esophagus  165-170  ^  long,  cylindroid 
to  a  small,  non-valvular,  terminal  bulb. 
Cardia   present,    flattened,   not   prominent. 


Excretory  pore  not  observed.  Vu  =  54%, 
female  with  two  opposed  ovaries.  Number 
of  testes  in  male  not  determined.  Spicules 
arcuate,  40  /x  long  (chord  25  /x).  Guber- 
naculum complicated,  bearing  a  medial 
piece  between  the  spicules  and  two  sigmoid 
lateral  guiding  pieces.  Apophyses  to  guber- 
naculum 10  /x  long.  Male  without  supple- 
ments. Tail  150-180  /x  long,  the  anterior  % 
conoid,  the  remainder  cylindroid  to  the 
slightly  swollen  terminus.  Terminus  with 
spinneret  and  two  8  /x  long,  terminal  setae. 

HoJotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  96. 
Type  locality,  M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

DIDELT A  Cobb,  1920 

Type    species.— D/c/e/fa    maculatum    Cobb, 
1920:  252,  253,  fig.  33. 

We  found  in  our  material  one  juvenile 
specimen  which  in  every  respect  resembles 
Cobb's  Didelta  maculatum.  Since  Cobb's 
figure  of  the  head  is  very  good  we  supply 
only  a  figure  of  the  tail  of  our  single  speci- 
men. 

Didelta  maculatum  Cobb,  1920 
Plate  XXX,  fig.  66 

Didelta  maculatum  Cobb,  1920:  252,  253,  fig.  33. 

Head  diameter  23  fx.  Cephalic  setae  15  + 
3  II.  Amphids  ( with  plaque )  25  X  17  /x. 
Esophagus  enlarged  posteriorly,  short  car- 
dia (25  11  long,  30  jx  wide)  present.  Tail 
410  /x,  a.b.d.  37  p..  No  spinneret  or  caudal 
glands. 

Representation  in  samples  studied. — M- 
8,  Biscayne  Bay. 

Geographical  distribution. — Off  Key 
West,  Florida  (Cobb,  1920). 

LINHOMOEIDAE 

TERSCHELUNGIA  de  Man,  1888 
Type  species.— Tersche///ng/a   communis  de 
Man,  1888:  12,  pi.  I,  fig.  7. 

The    material    from    Florida    contained 


294        Bulletin  Museum  of  Compamiive  Zoology,  Vol.  135,  No.  5 


three  species  of  TerschcUiuiiia.  Excellent 
examples  of  T.  Jongicaudata  were  found 
both  in  Biscayne  Bay  and  in  Florida  Bay, 
with  a  variant  found  at  Vero  Beach.  The 
distribution  of  the  cervical  setae  in  the 
specimens  from  the  former  habitats  is  ex- 
actly as  depicted  by  de  Man  in  his  original 
figures  of  the  species.  The  variant  from 
Vero  Beach  has  a  slightly  different  arrange- 
ment of  the  cerx'ical  setae,  viz.,  a  reduction 
in  number  and  a  more  posterior  location  of 
the  first  circle  (compare  Pi.  XXX,  figs.  67a,  b, 
and  6Sa,  b).  These  differences  are  stable 
\\  ithin  Florida  habitats.  Subsequent  collec- 
tions from  South  Carolina  contain  speci- 
mens, not  onlv  with  the  above  two  diver- 
gent  patterns  of  cervical  setae,  but  with 
intermediary  distributions  as  well. 

A  second  species,  T.  monohystem  n.  sp. 
is  unique  in  that  only  the  anterior  ovary 
appears  to  be  well  developed,  the  posterior 
gonad  being  quite  rudimentary.  The  third 
species,  T.  lon^spiculata  n.  sp.,  can  be 
distinguished  by  the  arrangement  of  the 
cephalic  and  cervical  setae  and  also  by  the 
long  spicules,  which  are  about  2.5  anal  body 
diameters  in  length. 

Terschellingio   longicaudata  de  Man,   1907 
Plate  XXX,  figs.  67,  a-c,  68,  a,  b;  Plate 
XXXI,  figs.  67,  d,  68,  c-e. 

Terschellin^id  lon^icaiuhitd  de  Man,  1907:   230. 

L  =  1.1-1.3  mm;  w  =  29-32  /x;  diameter 
at  base  of  esophagus  27-29  jj,.  Esophagus 
90-125  II-  nerve  ring  at  50%.  Head  with 
four  cephalic  setae,  4-5  fi  long.  Cervical 
setae  4-5  p.  long  and  arranged  in  distinctive 
patterns.  The  Biscayne  Bay  and  Florida 
Bay  specimens  have  two  paramphidial  pairs, 
two  postamphidial  pairs  and  a  somewhat 
broken  circle  of  eight  situated  about  mid- 
wav  between  the  amphid  and  the  nerve 
ring  (PI.  XXX,  fig.  67a).  The  Vero  Beach 
specimens  have  a  circle  of  four  postamphid- 
ial, two  sublateral  and  a  dorsal-ventral  pair 
(PI.  XXX,  fig.  6(Sa,  b).  Amphids  circular, 
7-8  fx  wide.  Terminal  esophageal  bulb  25 
/x  long  and  nearl\'  of  equal  width.    Cardia 


elongate,  14  /x  long,  5  /x  wide.  Excretory 
pore  located  at  a  level  slightly  anterior  to 
the  anterior  end  of  the  esophageal  bulb. 
Spicules  38-46  /x  long  (about  1.4-1.8  a.b.d), 
proximally  cephalated.  Apophyses  of  guber- 
naculum  10-12  /x  long.  Tail  325-365  /x  long 
(about  14  a.b.d.). 

Representation  in  samples  studied. — M-4, 
Rickenbacker  Causeway;  M-6,  M-7,  Ever- 
glades National  Park;  M-8,  Biscayne  Bay; 
V,  Vero  Beach. 

Geofiraphical  distribution. — Cosmopoli- 
tan. 

Jerschellingia  monohystera  new  species 
Plate  XXXI,  fig.  69,  a-f 

L  =  0.97-1.06  mm;  w  =  33-35  fx;  Vu  = 
44%.  Diameter  at  base  of  esophagus  31  /x. 
Esophagus  83-93  [x.  Head  diameter  13  /x. 
Head  with  four  cephalic  setae,  4  /x  long. 
Four  somatic  setae  occur  in  the  cervical 
region.  The  anteriormost  lies  8-12  /x  behind 
the  base  of  the  amphid  and  is  placed  ventral 
to  the  lateral  surface  of  the  body.  The  three 
others  are  staggered  dorsal  and  ventral  in 
respect  to  the  lateral  surface.  Amphid  cir- 
cular, 5  ij.  wide,  3-5  /x  from  anterior  end; 
corresponding  body  diameter  14-15  jj..  Fe- 
male with  only  the  anterior  ovary  developed 
and  producing  eggs  (195  /m  long),  posterior 
branch  rudimentary  (70  ^a  long).  Spicules 
40  fx  long  (about  1.3  a.b.d.),  without  proxi- 
mal cephalization.  Apophyses  of  guber- 
naculum  about  13  fx  long.  Tail  220-225  //. 
long  (7.5-10  a.b.d.),  of  which  the  posterior 
half  is  filiform.  Caudal  setae  arranged  as 
illustrated  (PI.  XXXI,  fig.  69e). 

JloJotiipe  speeimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4075,  Type  slide  No.  99.  Type  locality, 
V,  Vero  Beach. 

Representation  in  samples  studied. — V, 
Vero  Beach. 

Terschelimgia  longispiculafa  new  species 
Plate   XXXI,   fig.   70,    c,   d;    Plate   XXXII, 
fig.  70,  a,  b 

L  =  2.2-2.3  mm;  w  =  63-65  /x;  Vu  =  about 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        295 


40%.  Diameter  at  l)ase  of  esophagus  55- 
57  fj..  Esophagus  with  six  cephalic  papillae 
and  four  cephalic  setae,  4-5  ^  long.  Two 
circles  of  prominent  cervical  setae  present, 
four  setae  in  each  circle  and  all  setae  4  /x 
long.  The  first  circle  occurs  at  the  level 
of  the  posterior  edge  of  the  amphids.  In 
addition  to  the  prominent  cervical  setae, 
setae  of  lesser  stature  also  occur  in  the 
anterior  neck  region.  While  these  could  be 
arranged  in  transverse  circles  of  eight 
(sometimes  incomplete),  they  could  also 
be  said  to  be  arranged  in  eight  longitudinal 
rows  in  which  the  prominent  cervical  setae 
might  represent  the  anteriormost  seta  of 
each  ro\\'.  Amphid  circular,  10  ^  wide,  11- 
14  fj.  from  anterior  end;  corresponding  body 
diameter  3.3-36  fj..  Terminal  esophageal 
bulb  50  fx  long  and  42  ^  wide.  Spicules  122 
fx  long  (about  2.5  a.b.d. );  proximally  cepha- 
lated.  Apophyses  of  gubemaculum  25-30 
/A  long.  Tail  435  ^  long  ( about  10  a.b.d. ) . 
Caudal  setae  arranged  as  illustrated  (PI. 
XXXI,  fig.  70,  c). 

HoJoUjpc  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4072,  Type  slide  No.  98.  Type  locality, 
M-7,   Everglades   National  Park. 

Representation  in  samples  siiidied. — M-7, 
Everglades  National  Park,  and  M-8,  Bis- 
cayne  Bay. 

MONHYSTERIDAE 

MONHYSTERINAE 

PARAMONHYSTERA  Steiner,  1916 

Type  species.— Monohystera  (Paramonohys- 

fera)  megacephala  Steiner,  1916:  639- 

641,  pi.  32,  fig.  37,  a-f. 

A  key  to  the  subgenera  and  species  of 
Paramonhystera  is  provided  by  Wieser 
(1956).  Following  this  author,  the  sub- 
genus Paramonhystera  sensti  stricto  con- 
tains the  following  species:  P.  megacephala 
Steiner,  1916,  P.  micramphis  Schuurmans- 
Stekhoven,  1950,  P.  biformis  Wieser,  1956, 
and  P.  proteus  Wieser,  1956.  Of  these,  P. 
micramphis  is  known  from  females  and 
juveniles  only  and  is  considered  a  species 


inquircnda.  Our  new  species,  P.  canicula 
n.  sp.,  is  distinguished  from  the  remaining 
species  by  the  long  cephalic  setae,  the 
round  amphids  and  the  characteristic  shape 
of  the  gubernaculum. 

Paramonhystera  canicula   new  species 
Plate  XXXII,  fig.  71,  a-d 

L  =  1.60  mm;  w  =  42  /a;  esophagus  =  250 
IX.  Head  diameter  25  /x.  Lips  large,  rounded, 
each  with  two  labial  setae  5  fx  long.  Ce- 
phalic setae  23  +  18  /x.  Amphids  with  very 
faint  contour,  convex,  in  S  ,  14  ix  =  52%  of 
c.b.d.  wide.  Cervical  and  somatic  setae  in 
irregular  longitudinal  rows,  the  former  10- 
12  fx  long.  One  short,  asymmetrical  (dorso- 
lateral) seta  between  amphid  and  cephalic 
setae  on  each  side  of  body.  Cuticular  an- 
nulation  coarse.  Spicula  135  fx  long, 
knobbed  proximally.  Gubemaculum  rather 
complicated,  consisting  of  a  proximal  por- 
tion, cylindrical  and  20  ix  long,  and  a  distal 
portion,  IS  /x  long,  dilated,  with  ventrolateral 
projections,  a  deep  notch  and  a  serrated 
distal  edge.    Tail  175  /x  long,  a.b.d.  35  /x. 

HoJotype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4068.  Type  slide  No.  100.  Type  locality, 
M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M-3, 
Key  Biscayne. 

STEINERIA  Micoletzky,  1922 

Type      species.— A^ono/iysfera      polychaeta 

Steiner,  1915:  224-226,  figs.  1-3  (new 

subsequent  designation). 

MonJiystera  setosissima  Cobb,  1893,  trans- 
ferred to  Steineria  and  regarded  as  the  type 
species  by  Schuunnans-Stekhoven  and  De 
Coninck  (1933:  10),  is  herein  rejected  as 
the  type  species  of  the  genus  Steineria  on 
the  grounds  that  it  was  not  one  of  the  in- 
cluded species  brought  to  the  subgenus 
Steineria  when  first  established  by  Micolet- 
zky (1922a:  168).  This  action  is  in  accord 
with  Article  69a  (ii)  of  the  International 
Code  of  Zoological  Nomenclature  adopted 


296        Bulletin  Museum  of  Coniixiidtivc  Zoology,  Vol.  135,  No.  5 


by  the  XV  International  Congress  of  Zo- 
ology, London,  July,  1958. 

In  this  genus,  two  species  with  punctate 
cuticle  have  been  known  so  far,  i.e.,  S. 
punctata  Gerlach,  1955,  and  S.  gcrhchi 
\Meser,  1959.  P'rom  our  Florida  material  a 
third  species  can  be  added  that  is  rather 
closely  related  to  S.  punctata  but  can  be 
separated  on  the  following  counts:  sub- 
cephalic  setae  in  eight  groups  of  5  and  4 
(instead  of  6  and  3),  cephalic  setae  rela- 
tively longer,  gubernaculum  of  different 
shape. 

Sfeineria  ampullacea  new  species 

Plate  XXXII,  fig.  72,  c;   Plate  XXXIII,  fig. 
72,   a,   b 

L  =  1.44  mm;  w  =  68  [x;  esophagus  =  210 
/J..  Head  diameter  24  ^.  Lips  round,  with 
small  labial  papillae.  Cephalic  setae  jointed, 
15  -I-  12  p.  Subccphalic  setae  in  eight 
groups,  the  sublateral  ones  with  5  setae, 
measuring  20  +  27  +  30  +  40  +  62  ^,  the 
submedian  ones  with  4  setae.  Many  cervi- 
cal and  somatic  setae  in  eight  longitudinal 
rows.  Cuticle  annulated,  the  annulcs  re- 
solvable into  dots.  Amphids  10  /x  in  i  ,  16 
fi  behind  anterior  end.  Vestibulum  of  buc- 
cal cavity  vertically  striated  which  might 
be  a  characteristic  separating  this  genus 
from  Thcii.stus,  in  which  the  supporting 
structures  of  the  vestibulum  give  the  im- 
pression of  a  more  oblique  striation.  Ex- 
cretory pore  just  posterior  to  nerve  ring,  the 
whole  gland  very  short,  its  posterior  end  30 
/ji  anterior  to  the  end  of  the  esophagus. 
Spicula  55  //  long,  gubernaculum  with 
curved  apophysis,  23  /x  long.  Tail  175  /x 
long,  a.b.d.  55  /x,  terminal  setae  35  fx. 

IloJoti/ix'  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4075,  Type  slide  No.  101.  Type  locality, 
V,  Vero  Beach. 

Representation  in  samples  ."itudied. — V, 
\^ero  Beach. 


THERISTUS  Bastian,  1865 
Type  species.— Tfier/sfus  ocer  Bastian,  1865: 
156,  157,  pi.  13,  figs.  187,  188. 

Keys  to  the  subgenera  and  their  species 
are  provided  by  Wieser,  1956.  Species  de- 
scribed subseqvient  to  this  work  are  con- 
sidered in  a  more  recent  paper  (Wieser, 
1959). 

Subgenus  Penzancia  de  Man,   1889 
Type      species.— T/ier/sfus      ve/ox      Bastian, 
1865:  157,  pi.  13,  figs.  189-191. 

The  species  of  this  subgenus  may  be 
separated  into  three  groups  according  to 
the  presence  or  absence  of  a  gubernaculum 
and  to  the  shape  of  the  gubernaculum,  as 
follows : 

A)  Gubernaculum  with  distal  hook  or 
triangular  plate  of  characteristic  shape 
( see  Text- fig.  4 ) : 

T.  hipunctatus  (G.  Schneider,  1906) 
(syn.  Monhijstcm  hipunctata);  T. 
flevensis  Schuurmans-Stekhoven,  1935 
(syn.  Monhijstera  velox  Biitschli, 
1874,  de  Man,  1922  nee  Bastian  [see 
Gerlach,  1951c]);  T.  omhvonensis 
Schulz,  1935  (see  Gerlach,  1951);  T. 
paramhroncnsis  Timm,  1952;  T. 
macwflevensis  Gerlach,  1953;  T. 
metaflevensis  Gerlach,  1955;  T.  horosi 
Andrassy,  1958;  and  T.  calx  n.  sp. 

B )  Gubernaculum  conical  or  blunt: 

T.  inermis  Gerlach,  1952;  T.  paivulus 
Timm,  1952  (doubtful  species);  T. 
hiarcospiculum  Timm,  1952;  T.  tersus 
Gerlach,  1954;  T.  acrihus  Gerlach, 
19.54;  T.  megalaimoicles  Wieser,  1956; 
and  T.  stranus  Gerlach,  1957. 

C)  Gubernaculum  absent: 

T.  megalainui  Stewart,  1914  (doubt- 
ful species);  T.  aeulcatus  Schulz, 
1935;  r.  hetewseanicus  Wieser,  1955; 
and  T.  hamatus  Gerlach,  1956. 
All  the  other  species  listed  by  Wieser 
( 1956 )  are  here  considered  doubtful. 

Group  A,  which  was  represented  by  two 
species  in  our  material,  is  very  homogeneous. 
The  shape  of  the  spicular  apparatus   (see 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        297 


Figure  4.  Spicular  apparatus  of  species  of  Ther;stus  flevensis 
group.  a — T.  ombronensis  (after  Gerlach,  1951);  b — T. 
flevensis  (after  Gerlacfi,  1951);  c — T.  bipunctatus  (after 
Skwarra,  1924);  d — T.  macroflevensis  (after  Gerlach,  1953); 
e — r.  borosi  (after  Andrassy,  1958);  f — T.  parambronensis 
(after  Timm,  1952);  g — T.  metaflevensis  (after  Gerlacfi,  1955); 
fi — same  species,  present  material;  i — T.  calx  (present 
material). 

Text-fig.  4 )  immediately  separates  T.  horosi 
and  T.  calx  from  all  other  species.  As  for 
the  remaining  species,  it  could  be  argued 
that  T.  flevensis,  T.  ambronensis,  and  T. 
bipunctatus  represent  three  subspecies  of 
T.  flevensis,  characterized  by  the  position 
of  the  vulva  ( 65-72% ) ,  the  more  posterior 
position  of  the  amphids,  and  the  slight 
subterminal  protuberance  at  the  inner  edge 
of  the  gubernacular  hook  (shown  by  Gerlach 
in  1951,  but  not  in  1957).  On  the  other 
hand,  T.  metaflevensis  and  T.  parambronen- 
sis have  the  vulva  at  83-877^  and  the  guber- 
naculum  ends  distally  either  in  a  triangular 
plate  or  a  smooth  hook.  However,  the  latter 


species  is  insufficiently  described  and  the 
figure  of  the  spicular  apparatus  is  obviously 
rather  stylized.  T.  macroflevensis  is  said 
to  be  characterized  mainly  by  the  long 
somatic  setae.  In  other  respects  it  links 
the  flevensis-group  with  metaflevensis  and 
parambronensis. 

We  have  identified  our  second  species 
with  T.  metaflevensis  although  we  saw  the 
gubernaculum  to  be  slightly  different  from 
that  described  by  Gerlach  and  found  only 
one  lateral  seta  instead  of  three.  It  is  im- 
possible at  this  time  to  judge  the  systematic 
value  of  these  differences. 


metaflevensis    Ger- 


Therisfus     iPenzancia] 
lach,  1955 
Plate  XXXIII,  fig.  73 

Theristus  (Peuzducia)  metaflevensis  Gerlach,  1955: 
291-293,  fig.  25,  a-tl 

L  =  2  ,  1.36,  S  ,  1.8  mm;  w  =  34-40  fx- 
esophagus  =  230-275  /x;  Vu  =  83%.  Head 
diameter  2  ,  19,  S  ,  26  /i.  Cephalic  setae  9 
/x  in  female,  14  +  12  ^  in  male,  only  the 
usual  ten  setae  seen,  that  is,  no  additional 
lateral  setae.  Amphids  in  male  7  /m  wide, 
19  fx  behind  anterior  end,  in  female  6  fx 
and  13  fx,  respectively.  Spicula  53  fx  long, 
typical.  Gubernaculum  rather  large,  distally 
with  triangular  plate  in  which  a  hook- 
shaped  contour  can  be  discerned.  Tail  in 
2  ,  170  IX,  in  S  ,  240  /x  long;  a.b.d.  in  the 
latter  42  jx. 

Representation  in  samples  studied. — M-4, 
Rickenbacker  Causeway. 

Geograph ical  distribution . — San  Salvador, 
Brazil  (Gerlach,  1955). 

Theristus   (Penzancia)   calx   new  species 
Plate  XXXIII,  fig.  74,  a,  b 

L  =  3.25  mm;  w  =  70  /x;  esophagus  =  350 
fx.  Head  diameter  31  fx.  Six  setose  labial 
papillae,  10  cephalic  setae,  16  +  13  /x.  Am- 
phids 7  fx  =  20%  of  c.b.d.  wide,  23  fx  beliind 
anterior  end.  No  cervical  setae.  Spicula 
118  fx  long,  of  characteristic  shape;  the 
diagonal  list  that  can  be  observed  in  all 
species  of  this  group  is  here  developed  into 


298 


Bulletin  Mit.sciiin  of  Comparative  Zoology,  Vol  135,  No.  5 


a  lateral  plate  connecting  proximal  and 
distal  end  of  the  spiculum.  The  guber- 
naculum  is  reduced  except  for  the  triangular 
plate  that  also  in  this  species  is  \er>'  prom- 
inent.  Tail  300  /x  long;  a.b.d.  60  /<.. 

Holotijpc  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4069,  Type  slide  No.  102.  Type  locality, 
M-4,  Rickenbacker  Causeway. 

Representation  in  samples  studied. — M-4, 
Rickenbacker  Causeway. 

Subgenus  Dapfonemo  Cobb,  1920 
Type  species.— Dapfonema  {issertdens  Cobb, 
1920:  281,  282,  fig.  66a. 

In  this  subgenus  there  is  a  group  of 
species  distinguished  by  a  characteristically 
shaped  gubernaculum  which  gradually  en- 
larges in  its  distal  half,  has  a  subterminal 
constriction  and  ends  in  a  prominent  hook 
(see  PI.  XXXIV,  fig.  75,  e).  Previously  this 
group  consisted  of  the  following  closely 
related  species:  T.  huetschJii  Bresslau  and 
Schuurmans-Stekhoven,  1940;  T.  huet.schU- 
oides  Chitwood,  1951,  and  T.  parobuetschlii 
Timm,  1961.  To  this  group  we  add  a 
fourth  species,  T.  ostentator  n.  sp.,  which 
is  separated  from  the  other  three  by  the 
inuch  larger  male  amphids,  the  shorter 
spicula,  the  longer  cephalic  setae  and  some 
other  minor  characters. 

Theristus  {Dopfonema)  ostentafor  new  spe- 
cies 
Plate  XXXIII,  fig.  75,  o,   b;  Plate  XXXIV, 
fig.  75,  c-e 

L  =  1.76-1.77  mm;  w  =  30-32  ju;  esopha- 
gus =  .300-310  m  Vu  =  63%.  Head  diameter 
20-23  /x.  Lips  large,  with  strongly  de- 
veloped framework  (labial  capsule).  Labial 
setae  3  p..  Cephalic  setae  20  +  1.5-16  fx, 
with  additional  lateral  setae  (1-3).  Cervi- 
cal setae  short.  Amphids,  in  male,  15  /x  = 
66/f  of  c.b.d.  wide,  in  female,  8  /x  and  30%, 
respectively.  Cuticular  annulation  coarse. 
Buccal  cavity  spacious,  with  cuticularized 
walls.  Spicula  38-42  /x  long,  cephalate  prox- 


iniall\ .  Gubernaculum  17-18  ix  long,  plate- 
shaped  in  its  distal  half,  with  subterminal 
constriction  and  terminal  hook;  there  are 
also  lateral  projections.  Tail  in  male,  130- 
155  /x  long,  a.b.d.  25-26  /x;  in  female,  175 
IX  long,  a.b.d.  26  /x.  Tenninal  setae,  $  ,  20 
/x  long. 

Holotype  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4068,  Type  slide  No.  103.  Type  locahty, 
M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M-3, 
Key  Biscayne. 

Subgenus   Trichotherisfus   Wieser,    1956 
Type    species.— S/e;ner/a    mirabilis    Schuur- 
mans-Stekhoven     and      De      Coninck, 
1933:  10,  11,  pl.  4,  fig.  5;  pi.  5,  figs. 
1-3. 

Mesothcri.stus  Wieser,   1956:     80,  91,  NEW  SYN- 
ONYMY. 

We  consider  it  inadvisable  to  base  sub- 
generic  division  entirely  on  differences  in 
the  length  of  somatic  setae  and  we  there- 
fore merge  the  two  subgenera  mentioned 
above.  The  important  feature  of  this  sub- 
genus is  the  occurrence  of  long  somatic 
setae.  These  setae  may  be  confined  to  the 
anterior  region  of  the  body  but  they  should 
not  be  confused  with  the  single  circle  of 
eight  groups  of  elongated  cervical  setae 
about  on  level  with  the  amphids,  character- 
istic of  the  subgenus  Fseudostcineria. 

Further  classification  of  the  subgenus 
may  be  based  on  the  shape  of  the  guber- 
naculum, the  length  and  arrangement  of 
the  somatic  setae,  etc.,  as  set  out  in  the 
following  key. 

Key  to  Species  of  Subgenus  Trichotheristus 

1.  Gubernaculum  without  apophysis  2 

Gubernaculum  with  apophysis  3 

2.  Len^'th   of  spicula    12.5   m-     Setae   all   over 

body    --    T.    loniiisc'tosus    Schuurmans-Stek- 
hoven and  De  Coninck,  1933 
Length  of  spicula  64  /x.    Setae  in  anterior 

half  of  cervical  region  only  

T.  circumscriptus  Wieser,  1959 

3.  Apophysis  large,  set  at  an  angle  to  the  distal 


Florida  Marine  Nematodes  •  Wieser  and  Hopper 


299 


shaft     of     the     gubernacukim,     pointing 

dorsocaudally  4 

Apophysis  small,  in  direct  continuation  of 
distal  shaft,  pointing  dorsally  8 

4.  Somatic   setae,   at   least   in   cervical  region, 

measuring  2  c.b.d.  or  more  5 

Somatic  setae  not  surpassing  one  c.b.d.  in 
length  7 

5.  Spicula  strongly  cephalate  proximally,   dis- 

tally    with    large    lateral    hook.     Longest 

setae  in  mid-cervical  region  

- -__  T.  sanctimarteni  Timm,  1957 

Spicula  not  cephalate,  without  lateral  hook; 
longest  setae  in  anterior  cervical  region  ,    6 

6.  Distal  half  of  spicula  S-shaped.    Additional 

circle  of  six  subcephalic  setae  in   i   

r.  floridanus  n.  sp. 

Distal  half  of  spicula  more  or  less  straight. 

No  subcephalic  setae  T.  mirahilis 

( Schuurmans-Stekhoven  and  De 
Coninck,  1933) 

7.  Cephalic  setae  23—26  m  long,  male  amphids 

13  fi  wide,   apophysis   of   gubemaculum 

rod-shaped  T.  laxus  Wieser,  1956 

Cephalic  setae  13-16  fj.,  amphids  5-8  /j., 
apophysis  of  gubemaculum  plate-shaped  .. 

T.  setosus  (Butschh,  1874); 

T.  hirtus  Gerlach,  1951 
(for  differences  see  Gerlach,  1951) 

8.  Somatic   setae,   at  least   in   cervical   region, 

measuring   1.5  c.b.d.   or  more   9 

Somatic  setae  not  surpassing  one  c.b.d.  in 
length  T.  erecius  n.  sp. 

9.  Amphids  distinctly  spiral.    Cephalic  capsule 

well  developed.    One  circle  of  four  sub- 
cephalic setae  in  male  ..  _  T.  galeatus  n.  sp. 

Amphids  circular.    No  cephalic  capsule.    No 

subcephalic   setae   10 

10.  Lateral  cephalic  seta  in  c?  elongated.  Am- 
phids 0.5  head  diameters  behind  anterior 
end  T.  heterus  Gerlach,   1957 

Lateral  cephalic  seta  not  particularly  elon- 
gated. Amphids  1.5  head  diameters  be- 
hind   anterior   end    

T.  setifer  Gerlach,  1952 

Therisfus    (Trichotheristus)    floridanus    new 
species 
Plate  XXXIV,  fig.  76,  a-d 

L  =  1.50-1.57  mm;  \v  =  50-60  ^;  esopha- 
gus =  325-350  fx;  Vu  =  659f .  Head  diameter 
20-23  /JL.  Lips  round,  labial  papillae  setose, 
short.  Head  with  12  cephalic  setae,  20  +  16 
jx  long;  in  6  there  is  an  additional  circle  of 
6  subcephalic  setae.  Long  and  short  somatic 
setae,  the  longest  ones  measuring  60-70  ^ 
which  is  nearly  3  c.b.d.  in  the  cervical  re- 


gion, about  1.2  c.b.d.  in  mid-body.  Amphids 
in  <i  ,  7  /x,  in  9  ,  6  /j.  wide,  18-23  /a  behind 
anterior  end.  Spicula  37  p.  long,  distal  half 
S-shaped.  Gubemaculum  complicated,  with 
plate-shaped  apophysis,  10  /x  long.  Tail  in 
6  225  iJL  long  =  5  a.b.d.,  in  9  250  ix  long  = 
5.5  a.b.d.  Terminal  setae  60  /x  long. 

Holotijpe  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4067,  Type  shde  No.  104.  Type  locality, 
M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M-2, 
Key  Biscayne. 

Therisfus  [Trichotheristus)  erecfus  new  spe- 
cies 
Plate  XXXIV,  fig.  77,  b;  Plate  XXXV,  fig. 
77,  a,  c,  d 

L  =  1.3  mm;  w  =  55  ^a;  esophagus  =  275 
fx.  Head  diameter  16-24  /x.  Lips  round, 
labial  papillae  setose,  short.  Cephalic  setae 
10-17  +  S-15  fx  long.  Short  and  long 
somatic  setae,  the  longest  ones  measuring 
34  fx  which  is  about  one  body  diameter  in 
the  cervical  region.  There  are  four  char- 
acteristic pairs  of  sublateral  setae  behind 
the  amphids.  Amphids  in  A  6-7  jx  =  about 
23%  of  c.b.d.  in  width,  16-22  ^x  behind 
anterior  end.  Cuticle  with  lateral  alae 
( incisures? )  which  occur,  for  the  most  part, 
as  t\\o  parallel  refractive  lines  7-8  /x  apart 
(about  ^7— s  c.b.d.).  In  the  posterior  neck 
and  preanal  regions,  there  frequently  occurs 
a  third  refracti\e  line  between  those  nor- 
mally present.  Faint  cuticular  striations 
can  be  traced  over  the  lateral  alae.  Spicula 
35-38  fi  long,  proximal  end  cephalate,  distal 
half  slightly  S-shaped.  distal  end  with  a 
lateral  tooth.  Gubemaculum  23  ^,  with 
small,  plate-shaped  dorsal  apophysis.  Tail 
200-210  ,x  long,  a.b.d.  40  /x. 

Holotijpe  specimen. — Male;  Canadian  Na- 
tional Collection  of  Nematodes,  Entomology 
Research  Institute,  Ottawa,  Collection  Num- 
ber 4067,  Type  slide  No.  105.  Type  locality, 
M-2.  Ke\-  Biscayne. 


300        Bulletin  Mimtim  of  Comparative  Zoology,  Vol.  135,  No.  5 


Representation  in  samples  studied.— M-2,  time.   We  follow  Wieser  (1956,  1959)  and, 

Kev  Biscayne,  M-3,  Key  Biscayne,  and  M-4,  on  the   basis   of  the   shape   of  the   guber- 

Rickenbacker  Causeway.          '  naculum,  distinguish  three  groups  of  species, 

Remarks.— The    closest    relative    of    this  viz.,  A)   gubernaculum  without  apophysis, 

species  is  T.  sefosus  from  which  it  can  be  B)   with  small  dorsal  apophysis,   C)    with 

separated  by  the  small  and  dorsally  directed  distinct  caudal  apophysis.    No  new  species 

apophysis  of  the  gubernaculum.  have  been  described  since  Wieser's  paper 

of  1959,  but  probably  more  of  the  species 

T,      •  .      .T  •  L   xL     •  .     \       ;     i              r^^  included  in  the  grouping  by  Wieser  have 

Thenstus  {Tnchofhensfus)  galeatus  new  spe-  ,               . ,        ?       ^     '='   .  -^     .        .  ^     ,^^ 

^  to   be   considered   as   species   inqimendae, 

ni  *^'^^x/N,x/  L-      -TO  e.g.,   T.   lon^ieaudatus  Filipjev,   1922,   and 

Pate  XXXV,  fig.  78,  a-c  „^ '     .     ,.  ^       ^,,    iqqa 

^  T.  naviculworus  Cobb,  1930. 

L  =  1.02  mm;  vv  =  32  p.;  esophagus  =  205  Group  A,  species  without  gubernacular 
p..  Head  diameter  15  p..  Labial  papillae  apophysis,  contains  a  number  of  species  in 
setose,  short.  Cephalic  setae  jointed,  10  in  x\'hich  the  spicula  in  their  distal  fifth  or 
number,  15  +  12  /x  long.  Male  with  4  sub-  sixth  show  a  very  characteristic  outward 
cephalic  setae.  Short  and  long  somatic  bend.  This  additional  curvature  of  the 
setae,  the  longest  ones  measuring  65  /x  which  spicula  appears  to  give,  as  it  were,  more 
is  about  three  times  the  c.b.d.  in  the  cervical  room  to  the  development  of  the  distal  por- 
region.  Cuticular  annulation  coarse,  one  i-jon  of  the  gubernaculum  which  in  these 
annule  1.4  /x  wide.  The  head  seems  to  be  species  shows  some  complex  features  that 
strengthened  by  a  cephalic  capsule  which  a,-e  difficult  to  analyze.  The  situation  is 
consists  of  two  portions,  an  anterior  and  illustrated  by  figures  79,  c,  and  80,  c,  d  on 
a  posterior  one.  Amphids  spiral,  6  x  7  /x,  plate  XXXVI  in  this  paper.  To  this  sub- 
32-36  /x  behind  anterior  end.  Spicula  36  /x  group  belong  with  certainty  T.  kornocensis 
long,  cephalate  proximally,  distally  with  a  (Allgen,  1929)  sensu  Wieser,  1959,  T. 
lateral  tooth.  Gubernaculum  with  dorsal  oxyuroidcs  ( Schuurmans-Stekhoven,  1931) 
apophysis  and  a  distal  three-pronged  piece,  and  T.  fistulatus  n.  sp.,  but  it  cannot  be 
In  the  postanal  region  there  are  indications  excluded  that  some  more  species  described 
of  'breaks'  in  the  cuticular  annulation.  j^  the  literature  show  the  same  differentia- 
These  breaks,  however,  seem  to  be  rather  t^on  of  the  spicular  apparatus.  T.  trecuspi- 
irregular  and  are  differently  arranged  in  datus  Wieser,  1959,  displays  the  same  char- 
different  specimens.  Tail  170  /x  long,  a.b.d.  acteristic  but,  by  possessing  a  small  dorsal 
36  /x.  apophysis  of  the  gubernaculum,  has  been 

Holotype  specimen. — Male;  Canadian  Na-  assigned  to  the  next  group, 

tional  Collection  of  Nematodes,  Entomology  Group   B   of   this   subgenus,    comprising 

Research  Institute,  Ottawa,  Collection  Num-  species    with    small    dorsal     gubeniacular 

ber  4070,  Type  sHde  No.  106.  Type  locality,  apophysis,   is   represented  in   our   material 

M-5,  Virginia  Key.  by  two  well  characterized  new  species,  T. 

Representation  in  samples  .studied. — M-5,  toifus  n.  sp.,  and  T.  xyaliformis  n.  sp. 
Virginia  Key. 

Theristus       {Cylindrofheristus)       oxyuroides 

Subgenus  Cy//ndroffiensfus  Wieser,  1956  (Schuurmans-Stekhoven,    1931) 

Type  species.— Mono/iysfera  normandica  de  ,         vvv\/i    i-       to 

..         in/->rt    iJLr,    1-71       1    o    I-        1    ij  Pate  XXXVI,  tig.   /V,   a-c 

Man,  1890:  169-171,  pi.  3,  figs.  1-ld.  ^ 

rri  .         1                              4-    4.1      ,r      .4-,...4-  ,..-,^K  Monkt/stera      oxi/uroidcs      Schuurmans-Stekhoven, 

This  subgenus  presents  the  greatest  piob-  ^yg^/g^^    656,  fig.  8,  a-c. 

lems  of  classification,  and,  due  to  the  in- 
sufficiency of  many  descriptions,  no  satis-  L  =  0.9  mm;  w  =  43  p.;  esophagus  =  140 
factory  treatment  is  possible  at  the  present  p.    Head  diameter  14  p.    Lips  round,  with 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        301 


short  setose  papillae.  Twelve  subequal 
cephalic  setae,  9  fj,  long.  Rows  of  short 
cer\ical  and  somatic  setae.  Amphids  7  fi 
wide,  12  IX  behind  anterior  end.  Spicula  28 
fx  long,  shaipK'  bent  in  middle  and  with  an 
additional  uutnard  curvature  in  its  distal 
sixth.  Gubeniaculum  sleeve-like,  without 
apophysis,  distal  half  with  curved  pieces. 
Tail  160  IX  long,  a.b.d.  29  /x. 

Representation  in  samples  studied. — M-4, 
Rickenbacker  Causeway. 

Geograplueal  distribution. — Baltic,  North 
Sea,  Zuiderzee;  Chesapeake  Bay,  Marvland 
(Timm,  1952). 

Remarks. — Our  material  contained  what 
we  consider  to  be  fairly  typical  representa- 
tives of  T.  oxyuroides  ( Schuurmans-Stek- 
hoven,  1931).  The  characteristic  distal 
curvature  of  the  spicula  is  apparent  in 
Schuurmans-Stekhoven's  original  descrip- 
tion, but  both  Gerlach  (1951c)  and  Timm 
(1952),  as  far  as  one  can  infer  from  their 
figures,  seem  to  hold  that  it  is  the  guber- 
naculum  and  not  the  spicula  that  is  curved 
distally.  T.  kornoeensis  (Allgen,  1929)  and 
T.  fistulatus  n.  sp.  can  be  separated  from 
T.  oxyuroides  by  the  more  backward  posi- 
tion of  the  amphids  and  by  differences  in 
the  finer  structure  of  the  spicular  appara- 
tus. 

Theristus    {Cylindrofherisfus)    fistulatus    new 
species 
Plate  XXXVI,  fig.  80,  a-d 

L  =  0.86-1.04  mm;  w  =  30-.36  p.;  esopha- 
gus =  144-180  IX.  Head  diameter  15  /x.  Lips 
round,  bearing  short  setose  papillae.  Ten 
to  12  cephalic  setae  13-1-11  ix  long.  Scat- 
tered cervical  and  somatic  setae.  Amphids 
8-9  IX  =  40-45  per  cent  of  c.b.d.  wide,  IS- 
IS IX  behind  anterior  end.  Spicula  25-27 
IX  long,  cephalate  proximall\%  outwardly 
curved  in  its  distal  sixth.  Gubeniaculum 
sleeve-like,  without  apophysis,  distalh'  with 
curved,  tubular  piece.  Tail  210-228  ^  long, 
a.b.d.  2.3-26  ix. 

Holotype  specimen. — Male;  Canadian 
National   Collection    of   Nematodes,   Ento- 


mology Research  Institute,  Ottawa,  Collec- 
tion Number  4068,  Type  sHde  No.  107. 
Type  locality,  M-3,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
3,  Key  Biscayne. 

Remarks. — T.  fistulatus  is  characterized 
by  the  tubular  element  in  the  distal  portion 
of  the  gubeniaculum,  whereas  T.  kornoeensis 
is  equipped  with  two  lateral  projections  at 
the  distal  tip  of  the  gubemaculum. 

Theristus  {Cylindrofherisfus)  tortus  new  spe- 
cies 
Plate  XXXVI,  fig.  81,  a-d 

L  =  0.74-0.76  mm;  w  =  25-30  ix;  esopha- 
gus =  168-192  IX.  Head  diameter  10-12  ix. 
Lips  round,  bearing  short  setose  papillae. 
Ten  to  12  cephalic  setae,  10-12  +  7-9  /x 
long.  No  cervical  or  somatic  setae.  Am- 
phids in  9  ,  8  ix  =  47%  of  c.b.d.  wide,  24  ix 
behind  anterior  end,  in  ^  ,  10  /x  =  60%  of 
c.b.d.  wide,  23  ix  behind  anterior  end. 
Thirteen  to  15  cuticular  annules  between 
anterior  end  and  amphids.  Spicula  slender, 
of  even  thickness,  27  ^  long,  cephalate 
proximally.  Gubeniaculum  15-16  ^  long, 
^^'ith  short  dorsal  apophysis,  oblique  grooves 
in  distal  half  and  two  conspicuous,  laterally 
pointing  projections  at  its  distal  end.  Tail 
in  9  ,  175  /x,  in  6  ,  132-156  ^  long,  a.b.d. 
20^,. 

Holotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  108. 
Type  locality,  M-2,  Key  Biscayne. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

Remarks. — T.  tortus  n.  sp.  is  distinguish- 
able from  all  other  species  of  this  group  by 
the  large  and  posteriorly  situated  amphids, 
and  by  the  shape  of  the  gubeniaculum.  Its 
closest  relative  is  T.  resimus  Wieser,  1959, 
which  has  a  similar  gubeniaculum,  although 
without  the  oblique  grooves  that  occupy 
the  anterior  part  of  the  gubeniaculum  in 
T.  tortus. 


302        B 


ulletin  Museum  of  Comparative  Zoology,  Vol  135,  No.  5 


Therisfus  {Cylindrofherisfus)  xyaliformis  new 
species 
Plate  XXXVI,  fig.  82,  a-d 

L  =  0.6:3-0.73  mm;  w  =  21-22  /x;  esopha- 
gus =  125-150  iJ.  Head  round,  diameter 
6  /x.  Labial  papillae  indistinet.  Ten  ce- 
phalic setae,  subequal,  5  /x  long.  No^erv^ 
cal  and  somatic  setae.  Amphids  5  ^  -  507^ 
of  c.b.d.  wide,  20-22  /j.  behind  anterior  end, 
about  30  cuticular  annules  between  ante- 
rior end  and  amphids.  Buccal  cavity 
unusuallv  deep  bv  enlargement  of  the  "buc- 
cal ring."  Spicula  19-20  ,x  long,  cephalate 
proximally.  Gubernaculum  distally  with 
small  lateral  projection,  proximally  with 
hook-shaped  dorsally  pointing  apophysis. 
Tail  127-145  /x  long,  a.b.d.  16-19  /x. 

Ilolotype  specimen. — Male;  Canadian 
National  Collection  of  Nematodes,  Ento- 
mology Research  Institute,  Ottawa,  Collec- 
tion Number  4067,  Type  slide  No.  109. 
Type  locality,  M-2,  Key  Biscayne. 

Repvesentotion  in  samples  studied. — M- 
2,  Key  Biscayne. 

Remarks.— In    T.   xyaliformis  n.   sp.    the 
buccal  cavity  differs  from  that  of  all  other 
species  of  Theristus  by  its  elongation.   The 
elongation  has  come  alSont  by  the  widening 
of  the  "buccal  ring"  which  is  in  fact  the 
prostome    (in  the   sense   of  Osche,    1952), 
i.e.,  that  part  of  the  buccal  cavity  that  lies 
between  the  base  of  the  lips  and  the  an- 
terior end  of  the  esophagus.  Since  the  same 
type  of  elongation  is  a  characteristic  feature 
of  the   subfamily  Xyalinae,   T.   xyaliformis 
can   be   considered   to   link   this   subfamily 
with    the    Monhysterinae.      Moreover,     T. 
xyaliformis    is    characterized    by    the    far 
posteriorly    situated    amphids    and   by   the 
hook-shaped  apophysis  of  the   gubernacu- 
lum. 

MONHYSTERA  Bastion,  1865 
Type    species.— Mon/iysfera    stagnalis    Bas- 
tian,  1865:  97,  pi.  9,  figs.  9-11. 

The  proper  status  of  our  species  is  im- 
possible to  determine.  It  is  undoubtedly 
closely  related  to  M.  parva   (Bastian)   but 


the  differences  of  opinion  concerning  this 
species  cannot  be  resolved  at  present.    De 
Man  (1888)  figures  the  spicula  of  M.  parva 
without  any  teeth  or  projections  but  sub- 
sequent  authors    all    show    the    spicula    in 
their    proximal     third     or     fourth    to     be 
ecjuipped  with  a  ventral  projection   (actu- 
ally the  point  of  attachment  of  the  muscu- 
lar "velum").   De  Coninck  and  Schuurmans- 
Stekhoven     (1933)     claim     that    de    Man 
overlooked  this  projection  and  consider  M. 
heteroparva    Micoletzky,     1924,    to    be    a 
synonym.   Timm  ( 1952 )  rejected  this  claim 
and    revived    M.    heteroparva    as    a    valid 
species,  to  be  separated  from  M.  parva  by 
the   possession   of  this   ventral   tooth.    An- 
drassy  ( 1958 )  figured  the  spicular  appara- 
tus of  M.  parva  not  only  with  the  proximal 
projection  but  also  with  a  distal  "nose"  and 
hook.     Our    species    closely    agrees    with 
Andrassy  s  description  and  figure  and  we 
tend   to   think   that   indeed   many    authors 
overlooked,    or    misinteipreted,    either    the 
proximal  or  the  distal  projection,  or  both, 
and  that  M.  parva  is  a  cosmopolitan  species 
which   needs   to  be   restudied   in   its   type 
habitat.   We  observed  faint  preanal  supple- 
ments but  again  have  our  doubts  whether 
this    can   be    considered   a   real   difference 
from  previous  descriptions. 

Monhystero  porva  (Bastian,  1865) 
Plate  XXXVII,  fig.  83,  a-d 

TachijhocUtcs  parvus  Bastian,  1865:  156,  pi.  13, 
fifts.  185,  186;  Monhystcra  parva  var.  meridUina 
Micoletzky,  1922;  M.  heteroparva  Micoletzky, 
1924;  M.  kossncmis  Paramonon,  1929;  M.  ant- 
arctica    Col^b,    1914. 

L  =  0.58-0.72  mm;  w  =  18-22  /x;  esopha- 
gus =  102-115  /x;  Vu  =  65%.  Head  diam- 
eter 8-10  /x.  No  labial  papillae  seen.  Ten 
cephalic  setae,  4  +  3  /x  long.  Amphids  in 
6  ,  3.5  IX  =  36%  of  c.b.d.  wide,  10  ^  behind 
anterior  end.  Two  pairs  of  submedian 
cervical  setae,  25  /x  behind  anterior  end, 
excretory  pore  37  /x,  nerve  ring  63  /x  behind 
anterior  end.  Spicula  26-27  /x  long,  with 
proximal  "handle"  and  projection  on  which 
the  velum  attaches,  distally  with  a  triangu- 


Florida  Marine  Nematodes  •  Wieser  and  Hopper 


303 


lar  cuticularized  piece  which  forms  a  sort 
of  recurved  hook.  Gubernacular  apophysis 
13  /ji.  About  15  indistinct  preanal  and  two 
postanal  supplements.  In  some  specimens 
these  supplements  are  hardly  visible  and 
appear  only  as  darker  and  lighter  portions 
of  the  cuticle.  Tail  95-100  /x  long,  a.b.d. 
16-18  IX. 
Representation   in  samples  studied. — M- 

1,  Key  Biscayne. 

Gcoiiraphical  distribution. — Cosmopoli- 
tan. 

XYALINAE 

SCAPTRELLA  Cobb,  1917 

Type      species.— Scapfrella      cincfa      Cobb, 

1917:  119,  fig.  4. 
Scaptrella  cincta  Cobb,  1917 
Plate  XXXVII,  fig.  84,  a-c 

Scaptrella  cincta  Cobb,  1917:   119,  fig.  4. 

L  =  1.7  mm;  w  =  41  /x;  diameter  at  base 
of  esophagus  33  fx.  Esophagus  220  /x.  Tail 
360  IX.  Head  diameter  20  ix.  Labial  setae 
6-7  IX  long.  Cephalic  setae  12  ( 6  +  6 ) ,  the 
lateral  pairs  30  +  18  ^,  the  submedian  50  + 
30  IX  long.  Amphid  circular,  \\ith  internal 
spiral,  4  ,  10  fi,  $  ,  8  ^u,  wide.  Buccal  cavity 
cylindrical,  30  ix  deep.  Anteriorly  armed 
with  6  jointed  odontia.  In  addition,  a  blunt, 
weakly-sclerotized,  dorsally-positioned  on- 
chium  is  present.  Cuticle  coarsely  striated, 
about  3  n  wide  in  neck  region  and  about  2 
IX  in  mid-body.  Somatic  setae  numerous, 
very  fine,  averaging  17  /x  long  on  most  of 
the  body  length.  Tail  with  numerous 
caudal  setae  and  a  pair  of  20  /j.  long  termi- 
nal setae.  Spicula  34  /x  long,  proximal!) 
cephalated.  Gubernaculum  with  bidentated 
lateral  guiding  pieces. 

Representation   in   samples  studied. — M- 

2,  Key  Biscayne. 

Geofirapliical  distribution. — Atlantic 
Coast  from  Massachusetts  to  North  Caro- 
lina (according  to  Chitwood,  1951),  Per- 
nambuco,  Brazil  ( Gerlach,   1956 ) . 

Remarks. — Scaptrella  eincta  is  separated 
from  S.  brevicaudata  Gerlach,  1952,  bv  the 


longer  and  more  filiform  tail  and  by  the 
longer  cephalic  setae. 

XENOLAIMUS  Cobb,  1920 

Type    species.— Xeno/a/mus    striafus    Cobb, 

1920:  250,  251,  fig.  30. 
Xenolaimus  striafus  Cobb,  1920 

Plate  XXXVII,  fig.  85,  a-d 
Xenolaimus  striatus  Cobb,  1920:  250,  251,  fig.  30. 

L  =  1.06  mm;  w  =  33  /x;  esophagus  =  300 
IX.  Head  diameter  16-17  ^.  Lips  7  /x  high, 
flap-like.  Labial  setae  5.5  ix  long.  Ten 
cephalic  setae,  the  longer  ones  measuring 
19  /x;  the  two  setae  of  each  submedian  pair 
stick  together.  Buccal  cavity  wide  and 
deep,  with  two  weakly  cuticularized  teeth 
or  cuticular  folds  projecting  from  the  base 
of  the  mouth  to  the  base  of  the  lips.  Am- 
phids  in  an  enlarged  portion  of  the  fifth 
cuticular  annule.  This  enlargement  about 
6  IX.  First  cuticular  annule  wider  than  the 
follo\\'ing  ones.  The  head,  in  the  words  of 
Cobb  (1920:  250),  "is  protrusile  and  ap- 
pears as  if  surrounded  by  a  balustrade 
composed  of  the  anterior  annules  of  the 
cervical  cuticle."  Cuticular  annulation 
coarse,  with  an  unspecified  number  ( 12? ) 
of  longitudinal  rows  of  V-shaped  struc- 
tures. Spicula  asymmetrical,  28-30  +  32- 
34  /x  long.  Apophyses  of  gubemacula  also 
asymmetrical,  17  +  21  ^  long.  Tail  conical, 
145  IX  long,  a.b.d.  27  /x. 

Representation  in  samples  studied. — M- 
2,  Key  Biscayne. 

Geographical  distribution. — Biscavne 
Bay,  Florida  (Cobb,  1920). 

Remarks. — The  present  record  represents 
the  first  time  this  species  has  been  found 
since  Cobb's  original  description.  An  un- 
identified example  of  Xenolaimus  was  re- 
corded in  a  list  of  nematodes  from  the  Gulf 
Coast  of  Florida  by  King,  1962  (Xenolaimus 
sp.). 

ACKNOWLEDGMENTS 

The  collecting  trip  during  the  period 
from   May   7   to  June   8,   1963,   was   made 


304        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135.  No.  5 


possible  through  NSF  grant  GB  498  to  one 
of  us  ( WAV. ) .  In  this  connection,  we  wish 
to  thank  Professor  Ernst  Mayr  and  Dr. 
Giles  Mead  of  the  Museum  of  Comparative 
Zoology  of  Harvard  University  for  their 
support  and  encouragement.  Particular 
thanks  are  due  to  Mr.  R.  H.  Mulvey,  Chief, 
Nematology  Section,  Entomology  Research 
Institute,  Canada  Department  of  Agricul- 
ture, Ottawa,  in  whose  Section  the  slides  of 
our  collections  were  mounted  and  who 
supported  our  project  in  many  other  ways. 
We  wish  to  thank  Mr.  Curtis  W.  Sabrosky, 
U.S.D.A.,  \\'ashington,  for  making  avail- 
able for  study  several  type  specimens  from 
the  United  States  National  Museum.  We 
are  grateful  to  Dr.  S.  P.  Meyers,  of  the  In- 
stitute of  Marine  Science,  University  of 
Miami,  Miami,  Florida,  who  arranged  for 
our  stay  at  the  Institute  and,  in  addition, 
provided  valuable  assistance.  Miss  Nancy 
Van  Meter  served  as  an  efficient  guide  to 
collecting  sites  in  the  Everglades  National 
Park. 

Many  more  people  were  very  helpful  to 
us  during  our  trip  and  we  shall  acknowl- 
edge their  hospitality  and  cooperation  as 
we  proceed  with  publication  of  our  reports. 

REFERENCES  CITED 

Allgex,  C.  a.  1927.  Freilebende  marine  Nem- 
atoden  von  der  Kiiste  Tasnianiens.  Zool.  Anz., 
73:    197-217. 

.      1933.      Fieilel:)endt'  Neniatoden  avis  dem 

Trondheimsfjord.  Capitata  Zoologica,  Deel  IV, 
Afl.  2:     1-162. 

.      1959.     Frceliving      marine      nematodes. 

Further  results  of  the  Swedish  Antarctic 
Exped.  1901-1903,  Stockhobi.  Zoolosv,  5: 
1-293. 

Andrassy,  I.  1958.  Ergebnisse  der  zoologischen 
Aufsammlungen  des  ungarischen  naturwissen- 
schaftHchcn  Museums  in  Agypten  im  Jahre 
19.57.  Ann.  Hist.  Nat.  Mus.  Nat.  Hung.^  50: 
135-150. 

Bastian,  C.  H.  1865.  Monograph  of  the  An- 
guilluh'dae  or  free  nematoids,  marine,  land, 
and  fresh-water;  with  descriptions  of  100  new 
species.    Proc.  Linn.  Soc.  London,  25:  73-184. 

Brunetti,  B.  1949.  Contributo  alia  conoscenza 
dei  Nematodi  del  M.  Tirreno.  II.  Monit. 
Zool.    Ital.,   57:     41-59. 

.      1950.      Osservazioni    critiche    sul    genere 


Enoploides  Saveljev,  1912  (Nematoda,  Eno- 
plata).    Monit.  Zool.  Ital.,  58:  45-48. 

BiJTSCHLi,  O.  1874.  Zur  Kenntnis  der  freile- 
benden  Neniatoden  insbesondere  der  des 
Kieler  Hafens.  Abh.  Senckenberg.  Naturf. 
Ces.,  9:  2.37-292. 

Chitwood,  B.  C;.  1935.  A  new  nematode, 
Ccimacolaimus  pnjtheicJii,  n.  sp.  ( Camaco- 
laimidae).  Proc.  Helminthol.  Soc.  Washing- 
ton, D.C.,  2:  49-50. 

.      1936.      Some     marine     nematodes     from 

North  Carolina.  Proc.  Helminthol.  Soc. 
Washington,  D.C.,  3:   1-16. 

.  1951.  North  American  marine  nema- 
todes.   Texas  J.  Sci.,  3:   617-672. 

.      1956.     A  revision  of  the  genus  Haliplec- 

tiis  Cobb,  1913.  Proc.  Helminthol.  Soc. 
Washington,   D.C.,   23:    78-87. 

.      1960.     A    preliminary    contribution    on 

the  marine  nemas  ( Adenophorea )  of  north- 
ern California.  Trans.  Am.  Microscop.  Soc, 
79:  .347-384. 

Chitwood,  B.  C,  and  M.  B.  Chitwood.  1938. 
Notes  on  the  "culture"  of  aquatic  nematodes. 
J.  Washington  Acad.  Sci.,  28:  455-460. 

.      1950.     An    introduction    to    nematology. 

Sect.  1.  Monumental  Printing  Co.,  Baltimore, 
Md.,  pp.  viii  +  213. 

CoBii,  N.  A.  1893.  Tricoma  and  other  nema- 
tode genera.  Proc.  Linn.  Soc.  N.S.  Wales, 
8:    389-421. 

.      1898.     Australian       free-living       marine 

nematodes.  Proc.  Linn.  Soc.  N.S.  Wales, 
23:    383-407. 

.      1917.      Notes    on    nemas.     Contrib.    Sci. 

Nematol.,   5:    117-128. 

.  1920.  One  hundred  new  nemas.  Con- 
trib. Sci.  Nematol.,  9:  217-343. 

.      1922.     Greeffiella.    ( Tnchodcrma  Greef, 

1869;    not    Trichoderma    Steph.,     18,35).     J. 


Washington  Acad.  Sci.,   12:  229-303. 

.      1930.     The   demanian   vessels   in   nemas 

of  the  genus  OncJwIaimtis;  with  notes  on  four 
new  oncholaims.  T-  Washington  Acad.  Sci., 
20:  22.5-241. 

.      19'33.     New   nemic   genera   and   species, 

with  taxonomic  notes.  Edited  by  Margaret 
V.  Cobb.    J.  Parasitol,  20:  81-94. 

Coles,  J.  W.  1965.  A  critical  review  of  the 
marine  nematode  genus  Euchromadora  de 
Man,  1886.  Bull.  Brit.  Mus.  (Nat.  Hist.) 
Zool.,   12:    159-194. 

De  Coninck,  L.  A.  1944.  Scientific  results  of 
Prof.  Dr.  P.  Van  Dye's  Expedition  in  Ice- 
land. XVI.  Les  Nematodes  libres  des  eaux 
et  des  terres  saumatres.  Biol.  Jaarboek,  Cent, 
11:   165-220. 

De  Coninck,  L.  A.,  and  J.  H.  Schuurmans-Stek- 
HOVEN.  1933.  The  free  living  marine  nemas 
of  the   Belgian   coast.     Part  II.    With   general 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        305 


remarks  on  the  structure  and  the  system  of 
nemas.  Mem.  Mus.  Hist.  Nat.  Belc;.,  S8: 
1-163. 

Crites,  J.  L.  1961.  Some  free-Hving  marine 
nematodes  from  tlie  sand  beaches  of  Fiver's 
Island,  North  Carohna.  J.  Elisha  Mitchell 
Sci.    Soc,    77:    75-80. 

DiTLEvsEX,  H.  1911.  Danish  free-living  nema- 
todes. Vid.  Medd.  Naturh.  Forening  Kj0ben- 
haxn   (1912),  63:   213-256. 

.       1926.      Free-living      nematodes.       The 

Danish   Ingolf  Exp.,  4:    1-42. 

Eberth,  G.  1863.  Untersuchungen  liber  Nem- 
atoden.    Leipzig,  77  p. 

FiLiPjEV,  I.  N.  1918.  (Free-living  marine  nema- 
todes of  the  vicinity  of  Sevastopol.  Part  I.) 
(In  Russian.)  Tr.  Osob.  Zool.  Lab.  Sevastopol 
Biol.  Stantsiya  Ross.  Akad.  Nauk  s.,  2:  1-352. 

.      1927.     Les    nematodes    libres    des    mers 

septentrionales  appartenant  a  la  famille  des 
Enoplidae.    Arch.  Naturgesch.,  91,  A:   1-216. 

.      1934.     The    classification    of    the    free- 


living  nemas  and  their  relation  to  the  i^ara- 
sitic  nemas.  Smithsonian  Inst.,  Misc.  Collect., 
89:    1-63. 

C.ERLACH,  S.  A.  1951a.  Nematoden  aus  der 
Familie  der  Chiomadoridae  von  den  deutschen 
Kiisten.    Kieler  Meeresforsch.,  8:    106-132. 

.      1951b.     Revision    der    Metachromadora- 

cea,  einer  Gruppe  freilebender  mariner  Nem- 
atoden.   Kieler  Meeresforsch.,  8:   59-75. 

.      1951c.     Freilebende  Nematoden  aus  der 

\'erwandtschaft  der  Gattung  Theristus.  Zool. 
Jahrb.    (Syst.),  80:   .379-406. 

.      1952.     Die     Nematodenbesiedlung     des 

Sandstrandes  und  des  Kiistengrundwassers  an 
der  italienischen  Kiiste.  I.  Systematischer 
Teil.    Archivio  Zool.  Italiano,  37:    517-640. 

.      1954.     Freilebende   Nematoden  aus   der 

Lagoa  de  Freitas  (Rio  de  Janeiro).  Zool. 
Anz.,   153:    135-143. 

.      1955.     Zur    Kenntnis    der    freilebenden 

marinen  Nematoden  von  San  Salvador.  Z. 
Wiss.   Zool.,   158:   249-303. 

.      1956.     Diagnosen  neuer  Nematoden  aus 

der  Kieler  Bucht.  Kieler  Meeresforsch.,  12: 
85-109. 

.      1957a.     Marine     Nematoden     von     der 

Kongomiindung.  Bull.  Inst.  Ro\ .  Sci.  Nat. 
Belgique,  33:    1-16. 

.      1957b.     Marine     Nematoden     aus     dem 

Mangrove-Gebiet  von  Gananeia.  ( Brasilian- 
ische  Meeres-Nematoden  III.)  Abhandl. 
Akad.  Wiss.  Lit.  Mainz,  Math.-Naturvv.  Kl. 
Jg.   1957,  Nr.,  5:    131-176. 

.      1957c.     Die  Nematodenfauna  des  Sand- 


litoralen  Region   in  der  Kieler  Bucht.    Kieler 

Meeresforsch.,  14:   64-90. 
.      1958b.     Deuxieme     contribution     a     la 

faune   des   nematodes   des   eaux   interstitielles 

littorales    de    Madagascar.     Mem.    Inst.    Sci. 

Madagascar,  Ser.   F,  2:   343-365. 
.      1958c.     Freilebenden     Nematoden     von 

den  Korallenriffen  des  Roten  Meeres.    Kieler 

Meeresforsch.,   14:   241-246. 
.       1962.      Freilebende      Meeresnematoden 

von  den  Malediven.    I.    Kieler  Meeresforsch., 

18:    81-108. 

.      1963a.     iJber    freilebende    Meeresnema- 
toden.     Revision     der     Linhomoeidae.      Zool. 

Jahrl).   (Syst),  90:  599-658. 
— .      1963b.     Freilebende      Meeresnematoden 


von  den  Malediven  II.  Kieler  Meeresforsch., 
19:   67-103. 

.      1964.     Revision  der  Choniolaiminae  luid 

Selachinematinae  ( freilebende  Meeresnema- 
toden). Mitt.  Hamburg.  Zool.  Mus.  Int.,  40: 
23-50. 

Hopper,  B.  E.  1961a.  Marine  nematodes  from 
the  coast  line  of  the  Gulf  of  Mexico.  Gan.  J. 
Zool.,    39:    183-199. 

.      1961b.      Marine     nematodes     from     the 

coast  line  of  the  Gulf  of  Mexico,  II.  Can.  J. 
Zool.,  39:   359-365. 

.      1962.     Free-living  marine  nematodes  of 

Rhode  Island  waters.  Can.  J.  Zool.,  40:  41- 
52. 

.      1963.     Marine      nematodes      from      the 

coast  line  of  the  Gulf  of  Mexico.  III.  Addi- 
tional species  from  Gulf  Shores,  Alabama. 
Can.  J.  Zool.,  41:   841-863. 

.      1966.      Theristus  polychaetophihis  n.  sp. 

( Nematoda ) ,  an  external  parasite  of  the 
spionid  polychaete  Scolelepis  (Scolelcpis) 
squamata  (Miiller,  1806).  Can.  J.  Zool.,  44: 
787-791. 

Ixc;lis,  W.  G.  1962.  Marine  nematodes  from 
Banynls-sur-Mer:  With  a  review  of  the 
genus  Eunjstomina.  Bull.  Brit.  Museum  (Nat. 
Hist.),  Zool.,  8:    211-283. 

.     1963.     New  marine  nematodes  from  off 


strandes    an    der    Kiiste    von    Mittelbrasilien. 
(  Brasilianische  Meeres-Nematoden  I\^  )   Mitt. 
Zool.   Mus.,   Berlin,  33:   411-459. 
.      1958a.     Die    Nematodenfauna    der   sub- 


the  coast  of  South  Africa.  Bull.  Brit.  Museum 
(Nat.   Hist.)    Zool.,    10:     531-552. 

KixG,  C.  E.  1962.  Some  aspects  of  the  ecology 
of  psammolittoral  nematodes  in  the  north- 
eastern Gulf  of  Mexico.  Ecologv,  43:  515- 
523. 

Kreis,  H.  a.  1928.  Weiterer  Beitrag  zur  Ken- 
ntnis der  freilebenden  marinen  Nematoden. 
Arch.    Naturgesch.,   92,    A:     1-29. 

.      1932.     Papers  from  Dr.  Th.  Mortensen's 

Pacific  expedition  1914-16.  LXI.  Freilebende 
marine  Nematoden  von  den  Sunda-Inseln. 
II.  Oncholaiminae.  Vidensk.  Medd.  Dansk 
Naturh.  Foren.  K0benhavn,  93:  23-69. 

.      1934.     Oncholaiminae      Filipjev,      1916. 


306        Bulletin  Museum  of  Comparative  Zoology,  Vol  135,  No.  5 


Eine  monographische  Studie.  Capita  Zoologica, 

4:    270  pp. 
Luc,  M.,  AND  L.  A.  P.  De  Coxixck.      1959.     Tra- 

vau.\  de  la  Station  Biologique  de  Ro.scoff,  LII. 

Nematodes    libres    marins    de    la    region    de 

Roseoff.   Arch.  Zool.  Exp.  Cen.,  98:    103-165. 
Man,    J.    G.    de.      1878.     Contribution    a   la    con- 

naissance  des  nematodes  du  Colfe  de  Naples. 

Tijdschr.  Ned.   Dierk.  Ver.,  3:    88-118. 
.      1888.     Sur  fiuelques  nematodes  libres  de 

la   Mer  du   Nord,   nouveaux  ou   peu   connus. 

Mem.  Soc.  Zool.  France,  1:     1-51. 
.      1889a.     Especes  et  genres  nouveaux  de 

nematodes  libres  de  la  Mer  du  Nord  et  de  la 

Manche.    Mem.  Soc.  Zool.   France,  2:     1-10. 
.      18891).     Troisieme  note  sur  les  nematodes 

libres  de  la  Mer  du  Nord  de  la  Manche.    Mem. 

Soc.   Zool.   France.  2:     182-216. 
.      1890.     Quatrieme  note  sur  les  nematodes 

libres  de  la  Mer  du  Nord  et  de  la  Manche. 

Mem.  Soc.  Zool.  France,  3:    169-194. 

1907.     Sur    quelfjues    especes    nouvelles 


ou  peu  connues  de  nematodes  libres  vivant 
sur  les  cotes  de  la  Zelande.  Tijdschr.  Ned. 
Dierk.  Ver.,  2:    227-244. 

.      1922a.     Neue     freilebende     Nematoden 

aus  der  Znidersee.  Tijdschr.  Ned.  Dierk. 
Ver.,  18:    124-134. 

.      1922b.     tJber  einige  marine  Nematoden 

von  der  Kiiste  von  Walcheren,  neu  fiir  die 
Eissenschaft  und  fiir  unsere  Fauna,  unter 
welchen  der  sehr  merkwiirdige  CataJaimus 
Max  Weberi  n.  sp.  Bijdr.  Dierk.  K.  Zool. 
Genootsch.  Natura  Artis  Magistra.  Amsterdam, 
22:    117-124. 

MAmoN,  A.  F.  1870.  Recherches  zoologiques  et 
anatomiques  sur  des  nematoides  non  parasites, 
marins.  Ann.  Sci.  Nat.,  Paris,  Zool.,  13: 
1-100. 

Mawson,  p.  M.  1958a.  Free-living  nematodes. 
Section  2:  Additional  Enoploidea  from  Ant- 
arctic stations.  B.A.N.Z.  Antarctic  Res.  Exp. 
1929-1931,   Rep.    Ser.   B,   6:     291-305. 

.      1958b.     Free-living   nematodes.     Section 

3:  Enoploidea  from  subantarctic  stations. 
B.A.N.Z.  Antarctic  Res.  Exp.  1929-1931,  Rep. 
Ser.  B,  6:    307-358. 

Micoletzky,  H.  1922a.  Die  freilebenden  Erd- 
Nematoden  mit  besonderer  Beriicksichtigung 
der  Steiermark  und  der  Bukowina,  zugleich 
mit  einer  Revision  samtlicher  nicht  mariner, 
freilebender  Nematoden  in  Form  von  Genus- 
Beschreibungen  und  Bestimmungsschlus.seln. 
Arch.  Naturg.,  Berlin  (1921).  87,  Abt.  A, 
1-650. 

.      1922b.     Neue  freilebende  Nematoden  aus 

Suez.  Sitzungsber.  Akad.  Wiss.  Wien,  Math.- 
Naturw.   Kl.,    131:     77-103. 

.      1924a.     Weitere    Beitriige   zur   Kenntnis 

freilebender   Nematoden   aus   Suez.     Sitzungs- 


ber.   Akad.    Wiss.    Wien,    Math.-Naturw.    Kl., 

132:    225-262. 
.      1924b.      Letzter  Bericht  iiber  freilebende 

Nematoden     aus     Suez.      Sitzungsber.     Akad. 

Wiss.   Wien,   Math.-Naturw.   Kl.,    133:     137- 

179. 
— .      1930.     Freilebende    marine    Nematoden 


von  den  Sunda-Inseln.  I.  Enoplidae  ( Papers 
from  Dr.  Th.  Mortensen's  Pacific  Expedition 
1914-16,  LIII).  Vidensk.  Medd.  Dansk 
Naturh.  Foren.  K0benhavn  (1929-1930),  87: 
243-339. 

Murphy,  D.  G.  1963.  A  note  on  the  structure 
of  nematode  ocelli.  Proc.  Helminthol.  Soc. 
Wa.shington,    D.   C,    30:     25-26. 

OscHE,  G.  1952.  Systematik  und  Phylogenie  der 
Gattung  RJwhditis  (Nematoda).  Zool.  Jahrb. 
(Syst),    81:     190-280. 

Pearse,  a.  S.,  H.  J.  HuMM,  AND  G.  Walton. 
1942.  Ecology  of  sand  beaches  at  Beaufort. 
North  Carolina.    Ecol.  Monogr.,  12:    135-190. 

Saveljev,  S.  1912.  Zur  Kenntnis  der  freile- 
benden Nematoden  des  Kolafjords  und  des 
Relictensee  Mogilnoje.  Trans.  Soc.  Imp.  Nat. 
St.  Petersburg  (Zoologie),  43:    108-126. 

Schuurmans-Stekhoven,  J.  H.  1931.  Okolo- 
gische  und  moiphologische  Notizen  iiber 
Zuiderseenematoden  I.  Die  westliche  Halfte 
der  Zuidersee.    Morphol.  Okol.,  20:    613-678. 

.      1935.     Nematoda    errantia.     In    Grinipe 

and  Wagler,  Tiervvelt  der  Nort-  u.  Ostsee. 
Lief.,  28  V.  b,  1-173. 

.      1942.     The    free     living    nematodes     of 

the  Mediterranean.  III.  The  Balearic  Islands. 
Zool.  Mededeel.  Rijksmus.  Nat.  Hist.,  Leiden, 
23:    229-262. 

.      1943.     Freilebende    marine    Nematoden 

des  Mittelmeeres.  IV.  Freilebende  marine 
Nematoden  der  Fishereigriinde  bei  Alexandrien. 
Zool.  Jahrb.  (Syst.),  76:    323-380. 

.      1950.     Mediterranean  free-living  marine 


nematodes.  I.  The  Bay  of  Villefranche.  Mem. 
Mus.  Roy.  Hist.  Nat.  Belg.,  37:    1-220. 

Schuurmans-Stekhoven,  J.  H.,  and  L.  A.  De 
CoNiNCK.  1933.  Diagnosis  of  new  Belgian 
marine  nemas.  Bull.  Mus.  Rov.  Hist.  Nat. 
Belg.,  9:    1-15. 

Seinhorst,  J.  W.  1959.  A  rapid  method  for 
the  transfer  of  nematodes  from  fixative  to 
anhydrous  glycerin.    Nematologica,  4:    67-69. 

Skwari^a,  E.  1924.  Ergiinzung  zur  Kenntnis 
eines  freilebenden  Nematoden  Monohxjstcra 
hipiinctata  G.  Schneider.  Arch.  Hydrobiol., 
15:    225-226. 

Southern,  R.  1914.  Nemathelmia,  Kinorhyncha 
and  Chaetognatha.  Proc.  Roy.  Irish  Acad., 
31:    sect.  3,  Glare  Island  Surv.,  pt.  54,  80  pp. 

Steixer,  C;.  1915.  Freilebende  marine  Nema- 
toden von  der  Kiiste  Simiatras.  Zool.  Jahrb. 
(Svst.),  38:    224-244. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        307 


— .      1916.  Freilebende   Nematoden   aus   der 

Barentssee.  Zool.  Jahrb.  (Syst. ),  39:  511- 
676. 

.      1918.  Neue  und  wenig  bekannte  Nema- 


toden von  der  Westkiiste  Afrikas.  Zool.  Anz., 

50:   4-18. 
.      1922.     Beitrage    zur    Kenntnis    mariner 

Nematoden.    Zool.  Jahrb.    (Syst.),  44:     1-68. 
TiMM,    R.    W.      1952.     A    survey    of    the    marine 

nematodes    of    Chesapeake     Bay,     Maryland. 

Chesapeake     Biol.     Lab.,     Solomons     Islands, 

Maryland,   Publ.   No.  95:     1-70. 
.      1961.     The    marine    nematodes    of    the 

Bay  of  Bengal.    Proc.  Acad.  Sci.,  Pakistan,  1: 

1-88. 
.      1962.     Marine  nematodes  of  the  family 

Linhomoeidae  from  the  Arabian  Sea  at  Karachi. 

Can.  J.  Zool.,  40:     165-178. 
.      1963.     Marine  nematodes  of  the  suborder 


Monhysterina  from  the  Arabian  Sea  at  Karachi. 
Proc.  Helminthol.  Soc.  Washington,  D.  C, 
30:  .34-49. 
\\'iESER,  W.  1951.  Untersuchungen  iiber  die 
algenbewohnende  Mikrofauna  mariner  Hart- 
boden,   I.    Zur   Okologie   und   Systematik   der 


Nematoden-fauna  \  on  Plvmouth.  Osterr.  Zool., 

Z.,  3 :    425-480. 
.      1953.     Free-living  marine  nematodes.    I. 

Enoploidea.    Chile   Reports   10.    Lunds   Univ. 

Arsskr.,  N.F.  Avd.  2,  49:    1-155. 
.      1954.     Free-living  marine  nematodes.  II. 

Chromadoroidea.     Chile    Reports    17.     Lunds 

Univ.  Arsskr.,  N.F.  Avd.  2,   50:     1-148. 
.      1955.     A  collection  of  marine  nematodes 

from    Japan.     Publ.    Seto   Marine    Biol.    Lab., 

4:    159-181. 
.      1956.     Free-living  marine  nematodes.  III. 

Axonolaimoidea   and   Monhysteroidea.     Lunds 

Univ.   Arsskr.   Avd.  2,   52:     1-115. 
— ■ .      1959.     Free-living  nematodes  and  other 

small   invertebrates   of  Puget  Sound  beaches. 

Univ.   Washington    Pubis.    Biol.,    19:     1-179. 
WiESER,    W.,    AND    B.    E.    Hopper.      1966.     The 

Neotonchinae,  new  subfamily  (Cyatholaimidae: 

Nematoda)    with    an    analysis    of    its    genera, 

Neotonchus    Cobb,    1933    and    Gomphionema 

new  genus.    Can.  J.  Zool.,  44:    519-532. 
Zur     Strassen,     O.    "k.      L.     1894.     Uber     das 

rohrenformige      Organ      von      Oncholaimus. 

Zeitschr.    Wissensch.    Zool.,    58:     460-474. 

(Received  12  January  1966) 


308        Biillcliii  Museum  of  Cotnparative  Zoology,  Vol  135,  No.  5 


Plate    i 
Antlcoma    lata    Fig.    1 ,    a-e:      a — anterior    end    of    male;    b — posterior   end   of  female;   c — posterior  end  of   male;    d — sup- 
plement;   e — spicular  apparatus.     Anticoma   trichura   Fig.   2,    a-d;     a — anterior  end  of  male;  b — anterior  neck  region  of  male; 
c — posterior    end    of    mole;    d — spicular    apparatus. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        309 


Plate   II 
Halalaimus  cf.   fletchen   Fig.   3,  a-c:     a — anterior  end  of  male;  b — posterior  end  of  male;  c — splcular  apparatus.     Halalaimus 
meyeni  Fig.   4,    a-c:     a — anterior  end   of  male;    b — posterior   end  of  male;  c — spicular  apparatus. 


310        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


5a 


Plate    III 

Halanonchus     macrurus     Fig.     5,     a-c:      a — anterior    end     of     male;     b — posterior  end     of    male;     c — spicular     apparatus. 

Cytolaimium    exile    Fig.    6,    a-d:      o — spicular    apparatus;     b-d — posterior    ends    of  tfiree    different    specimens.      Enoploides 
gryphuz    Fig.    1  1 ,    c:     en    foce   view. 


Florida  Marine  Nematodes  •   Wicscr  and  Hopper        311 


8a 


/      7b 


Plate    IV 
Bathylaimus    australis    Fig.    7,    a-e:     a — anterior    end    of    male;     b — posterior    end    of    mole;    c — posterior    end    of    another 
male;    d — posterior    end    of    female;    e — spicular    apparatus.     6of/iy/oimus     orfhropoppus     Fig.     8,     a-f:      a — anterior    end     of 
male,    dorsoventro!    view;    b — omphid    of    mole;    c — anterior    end   of  male,    lateral  view;   d — spicular   apparatus  and    proximal 
spicular    cephalization    from    two    other    specimens;    e — posterior    end    of    female;    f — posterior    end    of   male. 


312         Bulletin  Museum  of  Couiixirative  Zoology,  Vol.  135,  No.  5 


Plate  V 

Phanodermopsis  longisetae  Fig.  9,  a-C:  a — anterior  end  of  mole;  b — posterior  end  of  male;  c — spicular  apparatus. 
Fnop/o/des  b/su/cus  Fig.  10,  a-d:  a — anterior  end  of  male;  b — posterior  end  of  mole;  c — spicular  apparatus  and  supple- 
ment of  another  male;  d — gubernaculum  of  tfiird  male  in  lateroventral  view.  fnop/o/des  gryphui  Fig.  11,  a,  b: 
a — spicular    apparatus;    b — gubernaculum    in    lateroventral    view. 


Florida  Marine  Nematodes  •  Wiescr  (tud  Hopper        313 


Plate  VI 
A/lesacontho/'des    psitiacus    Fig.    13,    a-e:     a — anterior    end    of    male    (cephalic    setae    cut    off);     b — anterior    end     of    same 
male;    c — posterior    end    of    mole;    d — spicular    apparatus;    e — supplement. 


314        BiiUctin  Museum  of  Comparatitc  Zoology,  Vol.  135,  A'o.  5 


Plate   VII 

Mesacanthoides    fibulatus    Fig.    12,    a-c:     a — anterior    end    of     mole;     b — posterior     end     of     mole;     c — spiculcr     apparatus. 
Onc/io/a/mus    dujardinii    Fig.    14,   o,    b;     a — anterior   end   of   mole;    b — posterior    end    of    male. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        315 


16a 


Plate  VIII 
Oncholaimium    appendiculatum    Fig.    15,    a-c:     a — anterior  end    of    male;    b — region    of    nerve    ring;    c — posterior    end    of 
male.      Oncholaimium     domesticum     Fig.     16,     a-d:      a — anterior    end    of    male;    b — demanian  organ,    region    of    uvette;    c — 
posterior     end     of     male,     lateroventral     view     to     show     pairedness    of    pre-    and     postanal     papillae;    d — posterior    end     of 
female. 


316        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


Plate    IX 
Metoncholaimus    intermedius    Fig.     17,    a,    d,    e-.      a — anterior  end   of   male;    d — posterior  end   of  female;    e — tail    of    juve- 
nile   female.     Mefoncholoimus    simplex    Fig.    18,    a — anterior    end  of  mole.    Metoncho/o/mus  sc/ssus  Fig.   19,  a,  b:    a — anterior 
end    of    male;    b — posterior    end    of    female. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        317 


17b 


.m.g. 


18d 


18e 


X 


19< 


Plate  X 

Metoncholaimus  intermedius  Fig.  17,  b,  c;  b — demanian  organ,  region  of  uvette  and  opening  of  enterine  efferent  Into 
intestine;  c — posterior  end  of  male.  Metoncho/oimus  simplex  Fig.  18,  b-e:  b — posterior  end  of  male;  c — part  of 
demanian  organ  (uv.,  uvette;  os.,  osmosium);  d,  e — posterior  ends  of  two  females  {m.g.,  moniliform  glands).  Meton- 
choloimus    scissus    Fig.    19,    c — posterior    end    of    male. 


318         Bulletin  Museum  of  Coin])aralive  Zoulogij,  Vol.  135,  No.  5 


20e 


Plate  XI 

Prooncholaimus  hastatus  Fig.  20,  a-e:  a — anterior  end  of  male;  b — anterior  end  of  young  female;  c — posterior  end 
of  male;  d — distal  portion  of  spiculum;  e — posterior  end  of  female.  Viscosio  oncholaimelloides  Fig.  21 ,  a-c:  a — anterior 
end   of  male;    b — anal    region    of   male;   c — posterior   end   of    mole. 


Florida  Marine  Nematodes  •  Wic.scr  and  Hopper        319 


22b 


23a 


23b 


22d 


22e 


23d 


Plate  XII 

Viscosia  paptllata  Fig.  22,  a-f:  a — anterior  end  of  female;  b — anterior  end  of  mole;  c — spicular  region,  lateral  view; 
d — posterior  end  of  male;  e — posterior  end  of  male,  ventral  view;  f — spicular  region,  ventral  view.  V/scosio  macramphida 
Fig.   23,   o-d:     a,   b — anterior  ends  of  two  males;   c — spicular  region;  d — posterior  end  of  male. 


320         Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  5 


Plate  XIII 

Euryitomina  minutisculae  Fig.  24,  a,  b:  a — anterior  end  of  male;  b — posterior  end  of  male.  Illium  exile  Fig.  25,  a-d: 
a — anterior  end  of  female,  lateral  view;  b — anterior  end  of  anotfier  female,  dorsal  view;  c — region  of  nerve  ring;  d — 
posterior   end    of   fe-nale.     Illium    Itbidinosum    Fig.    26,    a,    b:    a — anterior    end    of    female;    b — posterior    end    of    female. 


Florida  Marine  Nematodes  •  Wicscr  and  Hopper        321 


Plate   XIV 

Polygastrophora  edax  Fig.  27,  a-c:  a — anterior  end  of  mole;  b — posterior  end  of  female;  c — posterior  end  of  male. 
Pomponemo  tesselatum  Fig.  28,  a-d:  a — anterior  end  of  male;  b — cuticulor  differentiation  in  lateral  region  {h.f.,  fiigfi 
focus;   /./.,    low  focus;   m.e.,   mid-esophagus;   m.b.,   mid-body);    c — posterior   end   of   male;   d — anal    region    of   male. 


322         Bulletin  Museum  of  Coniparatitc  Zoology,  Vol.  135,  No.  5 


Plate  XV 

Longicyatholaimus    annae    Fig.    29,    a-c:     a — anterior    end    of  male;  b — posterior   end   of   male;    c — spicular   apparatus    and 

row    of   supplements.     Xyzzors    ing/isi    Fig.    30,    a-c:     a — anterior    end  of    male;    b — posterior    end    of    mole;    c — spicular    ap- 
paratus. 


Florida  Marine  Nematodes  •  Wicscr  and  Hopper        323 


Plate  XVI 

Paroconfhonchus   platypus    Fig.    31,    a-c:     a — anterior   end    of    male;    b — posterior    end    of    female;  c — spicular    apparatus. 

Paracyatholaimus    pesavis    Fig.    32,     a-e:      a — anterior    end    of     male;     b — posterior     end     of     male;  c — spicular     apparatus; 
d — distal    end    of    gubernaculum    of    another    male;     e — row    of  supplements. 


324         Bullcliii  Museum  of  Coinpiudtwc  Zoology,  Vol.  135,  \o.  5 


Plate   XVII 

Halichoanolaimus  quattuordecimpapillatui  Fig.  33,  a-C:  a — posterior  end  of  male;  b — spinneret;  c — spicular  apparatus. 
Halichoanolaimui  duodecimpapillatus  Fig.  34,  a-d:  a — onterior  end  of  male;  b — posterior  end  of  male;  c — spicular  ap- 
paratus; d — spinneret.  Spirinia  paraiitifera  Fig.  35,  a-e:  a — anterior  region  of  male;  b — posterior  end  of  male;  c, 
d — spicular  apparatus  of  two  more  males;   e — posterior  end  of  female. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        325 


37a 


Spirinia     hamata     Fig.      36,      a-c: 
Chromaspirina    inaurita    Fig.    37,    a-f: 
d — amphid    of    another    male;    e — posterior    end    of    male;    f — anal    region 


Plate  XVlll 

0 — anterior     region      of    male;     b — posterior    end     of     male;     c — spiculor    apparatus, 
a — anterior    region    of  male;    b — anterior    end    of    mole;    c — anterior   end   of    female; 


326        Bulletin  Museum  of  Coiuparatitc  Zoologij.  Vol.  135,  No.  5 


Plate  XIX 

Metachromadora  pulvinata  Fig.   38,    a-c:     a — anterior  end   of  male;  b — anterior  end  of  juvenile;  c — posterior  end  of  male. 

Metachromadora  meridiana   Fig.    39,    o-d:     a — anterior   end    of  male;  b — anterior  region  of  male;  c — posterior  end  of  male; 
d — spicular  apparatus. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        327 


Plate  XX 

Porodesmodora  toreutes  Fig.  40,  a-f:  a — anterior  end  of  male;  b — anterior  end  of  juvenile;  c — region  of  esophageal 
bulb;  d — tail  of  juvenile;  e — posterior  end  of  mole;  f — spiculor  apparatus.  Monopostfi;a  mirabilis  Fig.  41,  a,  b:  a — 
anterior  end  of  mole;  b — posterior  end  of  male.  Monoposthioides  mayri  Fig.  42,  a,  b:  a — anterior  end  of  male; 
b — posterior  end  of  male. 


328         Bulletin  Mu.'icuni  uf  Contparativc  Zuolug,y,  Vul.  135,  No.  5 


Plate  XXI 
Paramicrolaimus    lunofus    Fig.    43,    a-c:      a — anterior    end    of    male;     b — spicular    apparatus;     c — posterior  end     of     male. 
Hypodonto/aimus     inferruptus     Fig.     44,     a-d:      a — anterior     end    of    male;     b — region    of    esopfiageai     bulb;  c — lateral     dif- 
ferentiation   of    cuticle    in    mid-body;    d — posterior    end    of    male.     Hypodontolaimus    pandnpiculatus    Fig.    45,  d,    e:     spicular 
apparatus  of  two  males. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        329 


45b 


_i  U  ■  I .  r  «  .  I  LAJ- 


-ri  I  1 1  1  i-rp-l- 


46b 


48c 


Plate  XXII 
Hypodontolaimus  pandispiculatus  Fig.  45,  a-c,  f:  a — anterior  end  of  male;  b — anterior  region  of  male;  c — lateral  dif- 
ferentiation of  cuticle,  1:  bulbar  region,  2:  mid-body,  3;  anal  region;  f — posterior  end  of  male.  Rhips  ornata  Fig.  46, 
a,  b:  a — anterior  end  of  male;  b — cuticular  differentiation  in  mid-esophagus  region.  Chromadoro  macrolaimoides  Fig.  47: 
anal  region  of  mole.  Prochromodore//a  mediterranea  Fig.  48,  a-c:  a — anterior  end  of  male;  b — cuticular  differentiation: 
mid-esopfiagus    (above)    and    mid-body;    c — region    of    esopfiageol   bulb. 


330         Bulletin  Museum  of  CotnpaniCwe  Zoology,  Vol.  135,  No.  5 


49a 


50a 


51a 


Plate  XXIII 
Prochromadorella  medilerranea  Fig.  48,  d,  e:  d — posterior  end  of  mole;  e — posterior  end  of  female.  Chromadorella 
liliformis  Fig.  49,  a,  b:  a — anterior  end  of  mole;  b — esopfiageal  bulb.  Chromadorella  tnlix  Fig.  50,  o-c:  a — anterior 
end  of  mole;  b — esopfiageal  bulb;  c — cuticular  differentiation:  anterior  cervical  region  (above)  and  mid-esopfiagus. 
Chromadorella  vanmeterae  Fig.  51,  a-c:  a — anterior  end  of  male;  b — esopfiageal  bulb;  c — cuticular  differentiation:  pos- 
terior cervical    region   (above)    and   mid-body. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        331 


54b 


52a  ■ "-"^  ■"  -*  53a 

Plate  XXIV 

Chromadorella  liliformis  Fig.  49,  c,  d:  c — anal  region  of  male;  d — posterior  end  of  male 
Fig.  50,  d,  e:  d — spicular  apparatus;  e — posterior  end  of  male.  Chromadorella  vonmeterae  Fig, 
apparatus;  e — posterior  end  of  male.  Euchromadora  gaulica  Fig.  52,  a — anterior  end  of  male.  Euchromadora  pectinata 
Fig.  53,  a — anterior  end  of  mole.  Euchromadora  meadi  Fig.  54,  a,  h:  a — anterior  end  of  male;  b — anterior  end  of 
juvenile. 


Chromadorella     trilex 
51,    d,    e:      d — spicular 


332         Bulletin  Miiscinu  of  Comparative  Zoolo<iy.  Vol.  135,  No.  5 


Plate  XXV 
Euchromadora  gaulica  Fig.  52,  b-d:  b — cuticular  differentiation,  1:  end  of  esophagus,  high  focus  (above)  and  low 
focus,  2:  mid-body,  low  focus,  3:  anal  region,  low  focus;  c — spicuiar  apparatus;  d — posterior  end  of  male,  fuchromo- 
doro  pectinata  Fig.  53,  b-d:  b — cuticular  differentiation,  mid-body;  c — spicuiar  apparatus;  d — posterior  end  of  male. 
Euchromadora  meadi  Fig.  54,  c-f:  c — cuticular  differentiation,  end  of  esophagus;  d — anal  region  of  male;  e — posterior 
end   of  male;   f — gubernaculum    of   another   male. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        333 


Plate  XXVI 
a — anterior    end    of    male;    b — distal    end    of    spicular    apparatus;    c — region 


Afrocfiromodoro    denticulate    Fig.    55,    a-f 
of  esophageal  bulb;  d — tail  of  female;  e — posterior  end  of  mole;  f — cuticular  differentiation,   end  of  esophagus.     Neotonchus 
lutosus  Fig.  56,  a-d:    a — anterior  end  of  male;  b — anterior  region  of  male;   c — spicular   apparatus;  d — posterior  end   of   male. 


334         BuUcfi)i  Museum  of  Coiuparatwc  Zoology,  Vol.  135,  Xo.  5 


Plate  XXVII 
Mesonc/i/um    pe/lucidum    Fig.    57,    a-d:      a — lateral    differentiation  of  cuticle  in   lateral  view  and  in  cross  section,    1:     mid- 
body,    2:     anal    region   of   male,    3:     anal    region    of   female;    b — tail    of    female;    c — posterior    end  of    male;    d — esophageal 
bulb.     Sabatierla    parodoxo    Fig.    58,    a-d:     a — anterior    end    of    male;    b — posterior    end    of    male;  c,    d — spicular    apparatus 
of  two  different  males. 


Florida  Marine  Nematodes  •  Wiescr  and  Hopper        335 


Plate  XXVIII 

Sabatieno  paracupida  Fig.  59,  a-c:  a — anterior  end  of  male;  b — spicular  apparatus;  c — posterior  end  of  male. 
Axonolaimus  hexapilus  Fig.  60,  a-e-.  a — anterior  end  of  male;  b — amphid  of  female;  c — posterior  end  of  male;  d — 
spicular  apparatus;  e — tail  terminus.  Odonlophora  variabilis  Fig.  61,  a,  b:  a — anterior  end  of  male;  b — spicular  ap- 
paratus   (lower    right,    mislabeled    60b). 


336        Bulletin  Mu.scinu  of  Conipanttivc  Zoology,  Vol.  135,  No.  5 


Plate  XXIX 

Odontophora  vor/obil/s  Fig.  61,  c-e:  c,  d — posterior  ends  of  two  males;  e — posterior  end  of  female.  Parodontophora 
brevamphida  Fig.  62,  a-e:  a — anterior  end  of  male;  b — amphids  of  two  other  specimens;  c — posterior  end  of  male;  d — 
spicular  apparatus;  e — posterior  end  of  female.  Alaimella  cincta  Fig.  63,  a-c:  a — anterior  end  of  male;  b — posterior 
end  of  male;  c — spicular  apparatus.  Camoco/oimus  prytherchi  Fig.  64,  a-c:  a — anterior  end  of  male;  b — posterior  end 
of    male;    c — spicular    apparatus. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        337 


Plate  XXX 

a — anterior   end  of  mole;  b — spicular    apparatus;    c — posterior    end    of    male;    d — anterior 

Didelta    maculatum    Fig.  66 — posterior     end     of     juvenile.       Terschellingia     longicaudata 

(Bisccyne    Bay-Florida    Bay)    Fig.    67,    a-c:      a — anterior    end  of  male;   b — anterior  region  of  male;  c — posterior  end  of  male; 

(Vero    Beach)     Fig.    68,     a,     b:      a — anterior    end    of    male;  b — onterlor    region    of    male. 


Parafarvoia   seta    Fig.   65,   a-e-. 
region    of    male;    e — tail    terminus 


338         Bitllcfin  Museum  of  Comparative  ZooIoa,tj.  Vol.  135,  No.  5 


Plate  XXXI 
Tencbelhngia    longicaudata     (Biscayne     Bay-Florida     Bay)     Fig.     67,     d — spicular    apparatus;     (Vero     Beach)     Fig.     68,     c-e: 
c,    d — spicular    apparatus    of    two    males;    e — posterior    end    of  male.     Tenchellingia  monohystera   Fig.   69,    a-f:     a — anterior 
end     of     female;     b — onterior     end     of     male;     c — anterior     region    of    female;    d — fema'e    tail;    e — posterior    end    of    male; 
f — spicular     apparatus.      TerscheUingia     longispiculata     Fig.     70,    c,    d:      posterior    end    of    male;    d — spicular    apparatus. 


Florida  Marine  Nematodes  •  Wicscr  and  Hopper        339 


Plate  XXXll 
Terschellingia    longispiculata    Fig.    70,    a,    b:     a — anterior    end     of    female;     b — anterior    end     of     male.      Paromonhysfera 
con/cu/o    Fig.    71,    a-C:     a — anterior    end    of    male;    b — posterior     end     of    male;     c — distal     portion     of    spicular    apparatus. 
Steineria     ampullacea     Fig.     72,     c — posterior     esophageal     region. 


340        Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  5 


Plate  XXXIII 
Steineria    ampullacea    Fig.     72,     a,     b:      a — anterior    end    of    male;     b — posterior    end    of    male.      T/ier/stus    metoflevensis 
Fig.   73 — posterior  end  of  male.     Theristus  calx   Fig.    74,   a,   b:     a — anterior    end    of    male;     b — spiculor    apparatus.      Theristus 
oitentator   Fig.    75,   o,    b:     a — anterior  end   of   male;    b — anterior    end    of    female. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        341 


Plate  XXXIV 

Jheristus    ostenfator    Fig.    75,    c-e:      c — tail    of    female;    d — posterior     end     of  male;     e — spicular     apparatus.       Theristus 

floridanus    Fig.    76,    a-d:     a — anterior  end   of   male;    b — posterior     end    of     male;  c — cepfialic     seta;     d — spicular     apparatus. 
Tberisfus    erectus    Fig.    77,    h — posterior   end    of    male. 


342         Bulletin  Museum  of  Coiupumtitc  Zoology,  Vol.  135,  No.  5 


Plate  XXXV 
Theristus    erectus    Fig.    77,    a,    c,    d:     a — anterior    end    of    male;    c,     d — spicular    apparatus    of    two    specimens.      Theristus 
galeatus    Fig.    78,    a-c:     a — anterior    end    of    male;    b — posterior  end  of  male;   c — anal    region   of   male. 


Florida  Marine  Nematodes  •  Wieser  and  Hopper        343 


Plate  XXXVI 

Theristus  oxyuroides  Fig.  79,  a-C:  a — anterior  end  of  male;  b — posterior  end  of  male;  c — spiculor  apparatus. 
Theristus  fistulatus  Fig.  80,  a-d:  a — anterior  end  of  male;  b — posterior  end  of  male;  c,  d — spiculor  apparatus  of  two 
specimens.  Theristus  tortus  Fig.  81,  o-d:  a — anterior  end  of  male;  b — anterior  end  of  female;  c — posterior  end  of  male; 
d — spiculor  apparatus.  Theristus  xyaliformis  Fig.  82,  a-d:  a — anterior  end  of  male;  b — posterior  end  of  mole;  c,  d — 
spiculor    apparatus    of    two    specimens. 


344         BiiUetin  Museum  of  Coinparalivc  Zoology,  Vol.  135,  No.  5 


Plate  XXXVII 
A/lonhysfero  parva  Fig.  83,  a-d:     a — esophageal   region  of  mole;     b — anterior     end    of     mole;     c — posterior     end    of     male; 
d — spicular    apparatus.      Scaptrella    cincta     Fig.     84,     a-c:      a — anterior    end    of    male;    b — spicular    apparatus;    c — posterior 
end  of  male.     Xeno/o;mus  striatus   Fig.  85,    a-d:     a — anterior  end    of    male;    b — posterior    esopfiageal    region;    c — anal    region 
of    male;    d — posterior    end   of   mole. 


,  ■.  ^6  ■.    ;;■■■■■-,'    '  ■      "  ■    (    ■    -. 


SuliQtln 

Museum   of 

Comparative 

Zoology 


J  ^V-J.  r^ L  iJj u'jS -f.^ 


The    Ameiva    (Lacertilia,    Teiidae]    of 
Hispaniola.    III.    Ameiva    taeniura  Cope 


ALBERT  SCHWARTZ 


HARVARD   UNIVERSITY  VOLUME    135,   NUMBER  6 

CAMBRIDGE,  MASSACHUSETTS,   U.S.A.  APRIL  24,    1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

Memoris  1864-1938 

JoHNSONLv,  Department  of  Mollusks,  1941- 

OccASioNAL  Papers  on  Mollusks,  1945- 

Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  hst  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).   $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.    $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.    ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©   The   President   and    Fellows   of   Harvard    College    1967. 


THE  AhAElVA  (LACERTILIA,  TEIIDAE)  OF  HISPANIOLA. 
ill.  ANiEWA  TAENIURA  COPE 


ALBERT  SCHWARTZ^ 

INTRODUCTION 

Ameiva  tucniura  Cope  was  described  in 
1862  on  the  basis  of  an  unknown  number 
of  s\'nt\'pes  from  near  Jeremie,  Department 
du  Sud,  Haiti.  Barbour  and  Noble  (1915: 
433  ct  scq.)  regarded  A.  taenium  as  a  syno- 
nym of  A.  lineolafa  Dumeril  and  Bibron; 
the  two  species  are  very  distinct  in  size, 
squamation  and  pattern,  as  Schmidt  ( 1921a: 
17)  later  demonstrated.  Barbour  and  Noble 
examined  15  specimens  of  A.  taenium;  with 
one  exception,  all  were  from  various  Hai- 
tian localities.  Schmidt  reported  twelve 
specimens  from  the  Republica  Dominicana, 
and  showed  that  the  species  was  wide- 
spread throughout  that  republic.  Mertens 
(1939:72-73)  collected  six  specimens  at 
three  localities  in  the  Republica  Domini- 
cana. Much  more  material  was  available  to 
Cochran  (1941:274),  who  noted  the  oc- 
currence of  the  species  at  various  Haitian 
and  Dominican  localities,  as  well  as  on  the 
islets  of  Ile-a-Vache,  Petite  Cayemite  and 
Grande  Cavemite.  Cochran  had  previouslv 
(1928:56  and  1934:179)  described  two  other 
Hispaniolan  Ameiva:  A.  baibouii  from  He 
de  la  Gonave,  and  A.  rosamondae  from  Isla 
Saona.  These  two  species  are  correctly  as- 
sociated with  A.  faeniiira,  as  Mertens  (loc. 
cit.)  has  pointed  out.  Finally,  Schmidt 
(1919:524)  had  described  Ameiva  navassae 
from  Navassa  Island  between  Hispaniola 
and  Jamaica;  later  (1921b),  he  regarded  A. 

1  lO.nnn  S.W.  84  St..  Miami.  Florida  33143. 


navassae  as  being  related  to  the  Cuban 
Ameiva  aiiberi  Cocteau.  In  actuality,  A. 
navassae  is  identical  with  A.  taeniura;  the 
precise  status  of  this  name  is  discussed  in 
detail  below. 

Before  proceeding,  the  status  of  the  sup- 
posed syntypes  of  A.  taeniiira  must  be  dis- 
cussed. Barbour  and  Loveridge  (1929:214) 
considered  that  three  specimens  in  the  Mu- 
seum of  Comparative  Zoology  at  Harvard 
University  (MCZ  3614)  were  the  original 
syntypes.  Examination  of  these  specimens 
shows  that,  instead  of  their  being  A.  taeni- 
nra  from  Hispaniola,  they  are  specimens  of 
Ameiva  thoraciea  Cope  from  the  Bahamas. 
Cope  gave  a  detailed  pattern  description 
in  his  original  work,  and  gave  as  well  mea- 
surements on  a  single  lizard,  which  was 
3  inches  in  snout-xent  length  and  10  inches 
6  lines  in  total  length.  None  of  the  three 
supposed  syntypes  has  a  snout-vent  length 
of  76  mm  (=3  inches),  the  snout-vent 
lengths  being  107,  92  and  70  mm.  The 
pattern  of  these  "synt\'pes"  likewise  does 
not  agree  with  Cope's  description.  There 
is  no  question  that  MCZ  3614  does  not  con- 
tain the  syntypes  of  taeniura;  on  the  other 
hand,  it  is  c^uite  clear  that  Cope  did  indeed 
have  specimens  of  the  Hispaniolan  lizard  in 
hand  when  he  wrote  the  original  diagnosis. 
There  are  three  other  specimens  of  A. 
taeniiira  in  the  Hai-\^ard  collection  (MCZ 
3608,  3609  [2  specimens] )  which  were  in- 
volved in  the  same  loan  to  Cope  that  re- 
sulted in  the  confusion  of  the  presumed  syn- 

Bull.  Nhis.  Cornp.  Zool.,  135(6):   345-375,  April,  1967         345 


346        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  6 


t>ix'.s.  TIksi'  li/.arcls  include  two  with 
siK)ut-\cnt  lengths  of  75  and  77  mm,  and 
thus  (luitc  close  to  the  snout-\ent  length 
given  l3y  Cope.  All  agree  also  ([uite  closely 
with  Cope's  pattern  description.  None, 
ho\\e\-er,  presentK'  measures  267  mm  (  = 
10  inches  6  lines)  in  total  length.  MCZ 
.3608.  which  has  a  snout-\ent  length  of  75 
mm,  is  the  onK  specimen  which  currently 
has  a  tail  (in  t\vo  pieces),  and  a  total  length 
of  ahout  240  mm  is  achieved  when  the 
specimen  is  assembled.  It  is  possible  that 
MCZ  3608  is  one  of  the  syntypes,  and  that 
Cope's  total  length  measurement  is  in  error, 
but  I  am  reluctant  to  designate  this  lizard 
as  a  lectotype;  it  seems  preferable  to  con- 
sider the  original  material,  on  which  the 
name  A.  tacniiini  was  based,  as  lost. 

A.  tacniura  is  now  represented  in  collec- 
tions ])\  adecjuate  series  from  the  Tiburon 
Peninsula  and  the  Peninsula  de  Barahona; 
both  of  these  regions  pertain  to  the  south 
island  of  Hispaniola  (Williams,  1961).  De- 
spite intensive  recent  collecting  in  the  Re- 
publica  Dominicana,  on  the  north  island,  A. 
taeniiiKi  remains  rather  poorly  known  in 
that  republic.  I  ha\e  examined  406  speci- 
mens of  A.  tacniura  from  Hispaniola,  lies 
de  Petite  Cayemite  and  Grande  Cayemite, 
He  de  la  Gonave,  Ile-a-Vache,  Isla  Saona, 
and  Isla  Carenero  in  the  Bahia  de  Samana. 
Most  of  the  material  from  the  Republica 
l^ominicana  and  much  of  the  that  from 
Haiti  has  been  collected  by  Miss  Patricia 
A.  Heinlein  and  Messrs.  Donald  W.  Buden, 
iliMiald  V.  Klinikowski,  David  C.  Leber, 
Drill  lis  I!.  Paulson,  Richard  Thomas  and 
the  aullior,  and  is  presently  denoted  as  the 
Albert  Schwart/  Field  Series  (ASFS).  To 
the  abo\e  companions  I  wish  to  express  my 
gratitude,  and  especially  to  Richard  Thomas 
for  \  isiting  Isla  Saona  on  my  behalf.  I  have 
borrowed  material  from  the  following  insti- 
tutions and  pri\ate  collections:  American 
Mu.seum  of  Natural  History  (AMNPI), 
Charles  M.  Bogert  and  George  \V.  Foley; 
Carnegie  Museum  (CM),  Neil  D.  Rich- 
mond and  Clarence  J.  McCoy;  Museum  of 
Comparatixc   Zoology    (MCZ),   Ernest    E. 


Williams;  Natur-Museum  und  Forschungs- 
Institut  Senckenberg  (SMF),  Konrad 
Klemmer;  University  of  Florida  collections 
(UF),  Walter  Auffenberg;  United  States 
National  Museum  (USNM),  Doris  M.  Coch- 
ran and  James  A.  Peters;  Peabody  Museum 
at  Yale  University  (YPM),  Charles  A.  Reed; 
Donald  W.  Buden  (DWB),  and  Richard 
Thomas  (RT).  I  am  grateful  to  the  above 
for  pcnnission  to  study  specimens  in  their 
care.  The  Harvard  collections  from  the 
Tiburon  Peninsula  recently  obtained  with 
the  aid  of  National  Science  Foundation 
grant  GB2444  to  Dr.  Ernest  Williams,  have 
proved  extremely  pertinent  and  valuable, 
especially  in  defining  the  parameters  of  A. 
tacniura  near  the  type  locaHty.  Paratypes 
of  new  forms  have  also  been  placed  in  the 
Museum  of  Natural  History,  University  of 
Kansas  (KU),  and  in  the  University  of  Illi- 
nois Museum  of  Natural  History  (UIMNH). 
The  main  body  of  A.  tacniura  on  the 
Hispaniolan  mainland  lies  to  the  south  of 
the  Cul  de  Sac-Valle  de  Neiba  plain.  Not 
only  is  the  species  abundant  in  this  region, 
but  it  also  occupies  three  satellite  islands 
( Ile-a-Vache,  the  Cayemites )  associated 
with  it.  The  species  has  not  been  taken  on 
Isla  Beata  and  Isla  Alto  Velo  which  are  also 
associated  with  the  south  island  of  Hispan- 
iola. On  the  north  island,  A.  tacniura  is 
known  from  the  Llanos  de  Azua  (where  it 
appears  to  be  quite  rare),  and  from  the  re- 
gion between  Santo  Domingo  and  Cabo 
Engaiio  (the  eastern  extremity  of  the  is- 
land ) .  There  is  an  adequate  series  from  the 
Peninsula  de  Samana  (where  the  species 
appears  to  be  abundant,  although  on  two 
visits  to  the  Peninsula  I  have  not  seen  it 
there).  To  the  north  in  the  Republica  Do- 
minicana, A.  tacniura  is  known  from  the 
northern  foothills  of  the  Cordillera  Central, 
from  the  Valle  de  Constanza  region,  Puerto 
Plata,  and  near  Loma  de  Cabrera  near  the 
Dominico-IIaitian  border.  In  northern  Haiti, 
there  are  specimens  only  from  Plaisance 
and  St.  Michel  de  I'Atalaye;  the  species 
occurs  as  well  near  Trou  Forban.  The  oc- 
currence of  A.  tacniura  on  the  north  island 


Ameiva    taeniura  in  Hispaniola  •  Schwartz        347 


satellites  Isla  Saona  and  He  de  la  Gonave 
( and  on  the  islet  Isla  Carenero )  has  already 
been  mentioned.  The  gaps  in  the  distribu- 
tion on  the  north  island  surely  are  not  all 
real.  However,  colleeting  by  ourselves  and 
others  in  intermediate  areas  has  not  re- 
vealed the  species. 

Ameiva  taeniura,  in  contrast  to  A.  lineo- 
lata  and  A.  chrysolaema,  is  a  denizen  of 
shady  and  mesic  situations.  It  is  common 
about  Camp  Perrin  in  southwestern  Haiti 
and  on  Ile-a-Vache  nearby.  Large  series 
from  the  Jeremie  area  attest  to  its  abun- 
dance in  that  region.  On  the  Peninsula  de 
Barahona,  which,  south  of  the  Sierra  de 
Baoruco,  is  extremely  arid,  A.  taeniura  oc- 
cupies more  shady  stands  of  dry  hardwood 
and  Acacia  forest;  its  interaction  with  A. 
chnjsolaema  near  Laguna  de  Oviedo  has 
ahead)'  been  discussed  (Schwartz  and  Kli- 
nikowski,  1966).  In  the  Llanos  de  Azua, 
the  only  two  specimens  secured  were  from 
a  shady  ravine;  the  remainder  of  the  habi- 
tat was  xeric  thorn  scrub.  Mertens  ( 1939 ) 
noted  the  occurrence  of  A.  taeniura  along 
the  coast  and  even  in  the  Avicennia  zone 
at  San  Pedro  de  Macoris;  the  same  situation 
occurs  at  Ile-a-Vache  and  near  El  Macao 
in  the  extreme  eastern  Republica  Domini- 
cana.  Neither  A.  chnjsolaema  nor  A.  lineo- 
lata  is  so  closely  associated  with  shady  and 
mesic  situations  as  is  A.  taeniura,  in  increas- 
ing dependence  upon  this  niche,  the  three 
species  may  be  ranked  as  lincoJata-chnjso- 
laema-taeniura  with  lineoJato  the  most  con- 
firmed denizen  of  hot  and  dry  habitats. 

The  altitudinal  range  of  A.  taeniura  is 
extensive.  It  occurs  from  sea  level  in  many 
areas  to  elevations  of  4250  feet  (1296  me- 
ters) in  the  Cordillera  Central  in  the  Re- 
publica Dominicana,  and  5600  feet  (1707 
meters)  in  the  Montague  Noire  in  Haiti. 
No  other  Hispaniolan  Ameiva  (nor  for  that 
matter  ^^'est  Indian  Ameiva)  has  such  a 
broad  altitudinal  range.  Doubtless  its  pref- 
erence for  more  mesic  and  cool  situations 
has  allowed  A.  taeniura  to  ascend  to  these 
greater  heights  in  the  uplands. 

There  are   12  subspecies  of  A.  taeniura 


discussed  in  the  present  paper;  there  is  good 
evidence  that  this  is  not  the  complete  roster 
of  races  on  this  species,  but  adequate  ma- 
terial is  lacking  from  the  northern  half  of 
Haiti  and  the  western  and  central  portions 
of  the  Republica  Dominicana.  I  have  once 
again  placed  most  emphasis  on  coloration 
and  pattern — two  features  which  vary  geo- 
graphically in  a  rational  manner.  Empha- 
sizing these  characters  at  the  expense  of 
scale  counts  demands  that  data  on  pigmen- 
tation and  pattern  must  be  taken  on  fresh 
specimens  in  the  field.  Without  these  data, 
some  well-characterized  races  might  be  com- 
pletely overlooked.  In  actuality,  one  must 
rely  primarily  on  these  two  attributes  in  A. 
taeniura,  since  in  most  cases  scale  charac- 
ters are  extremely  variable.  For  example, 
the  range  in  number  of  rows  of  longitudinal 
ventrals  for  the  entire  sample  of  A.  taeniura 
is  28  to  35;  the  range  of  this  character  in 
lizards  from  the  Jeremie  region  alone  is  29 
to  34,  and  most  other  samples  are  compar- 
able, usually  merely  lacking  one  or  the 
other  extreme,  or  both.  Counts  of  fourth 
toe  scales  (both  toes  combined),  femoral 
pores  (both  series  combined),  and  scales 
in  the  fifteenth  caudal  verticil  show  some- 
what more  differentiation,  but  in  hardly  any 
case  is  there  complete  separation  between 
subspecies  on  these  counts. 

The  number  of  transverse  rows  of  ventral 
plates  has  been  used  to  characterize  species 
of  Ameiva.  In  A.  taeniura,  these  rows  are 
either  8  or  10,  with  only  one  population 
(southwestern  shore  of  the  Tiburon  Penin- 
sula) having  a  modal  condition  of  8.  Any 
large  sample  (with  the  exception  of  18 
lizards  from  the  Cordillera  Central)  in- 
cludes both  8-  and  10-row  lizards.  Thus 
even  the  number  of  transverse  ventral  rows 
is  not  constant  in  most  cases.  Although 
scale  counts  are  given  for  all  subspecies, 
they  have  in  general  been  de-emphasized 
and  must  be  used  with  discretion. 

Ameiva  taeniura  may  be  defined  as  fol- 
lows: 1)  a  moderate  sized  species  of  the 
genus  Ameiva  with  snout-\'ent  length  to  102 
mm  in  males  and  103  in  females;  2)  dorsal 


348        BuUctUi  Musfinti  of  Comparative  ZooJo<i.y.  Vol.  135.  No.  6 


o 


-ff 


JT- 


O 
N 

1 


Q) 

. — . 

0) 

■ 

-n 

' — 

a> 

(« 

:S 

•p 

O 

X 

o 

c 

-O 

n 

o 

u 

n 

c 

-Q 

0) 

n 

O) 

VI 

c 

&.^ 


D 

0} 


-Q     _>-     <U 


o   in 
3   Lq 

D    .? 


Q. 


O 


-Q 


t    E 


o 

E 

D 


D 

X 


O 

G) 
D 


2    S^ 


CD 
E 


O 

c 

CD 


'^      X:      <U 


^  i  2  °  ^ 

O     .^-   1-     -    £ 


—     u    ■  - 


0    .;: 

-P    S 
O      O 


ro  -^   "o 


E 

>^ 

D 

u 


.2  O 


O 


O      oi  O  ^  "D 

Q.   .E  — .  'u  D 

to     —  CN  0)  N 

X     —  •  -  Q.  ■-' 

O  D 

c;  L_  (-)  t/i 

^     O  o  Ji  (1) 

■    •-  ^  -  '^ 


Ameiva    taeniura   in  Hispam()l\  •  Schicariz        349 


caudal  scales  keeled  and  oblique;  3)  ven- 
trals  in  8  or  10  transverse  rows  and  in  28 
to  35  longitudinal  ro\\'s;  4 )  fourth  toe  sub- 
digital  scales  from  61  to  91;  5)  femoral 
pores  24  to  41;  6)  fifteenth  caudal  verticil 
with  18  to  31  scales;  7 )  dorsal  pattern  con- 
sisting of  either  a )  a  series  of  five  pale 
longitudinal  lines  on  a  dark  back  or,  b)  a 
median  dorsal  pale  longitudinal  zone,  or  c) 
a  "combination"  of  the  two  conditions;  and 
8)  hemipenis  extending  to  about  the  sixth 
or  se\'enth  caudal  verticil,  sulcate  surface 
naked;  sulcus  bifurcates  slightly  apically, 
the  branches  ending  in  tw  o  weakly  bifid  ap- 
ical areas  on  each  side;  non-sulcate  surface 
entirely  flounced,  the  flounces  extending 
around  the  organ  onto  the  sulcate  surface  to 
near  the  sulcus  itself;  a  small  smooth  tri- 
angular area  on  the  non-sulcate  side  which 
divides  the  flounces  for  about  one-half  the 
length  of  the  organ  into  two  fields  corre- 
sponding to  the  apical  areas. 

SYSTEMATIC  ACCOUNT 

Ameiva  faeniura  faeniura   Cope,   1862 

Ameiva  taeniura  Cope,  1862,  Proc.  Acad.  Xat. 
Sci.  Pliiladelphia,  14:63  (type  locality — near 
Jeremie,   Dept.   dn    Sud,    Haiti) 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  moderate 
size  (males  to  88  mm,  females  to  76  mm, 
snout-vent  length),  usually  10  transverse 
rows  of  ventrals,  low  number  of  fourth  toe 
subdigital  scales  and  femoral  pores,  mod- 
erate number  of  scales  in  the  fifteenth 
caudal  verticil;  dorsal  pattern  consisting  of 
a  dorsal  zone  bordered  by  one  (parame- 
dian) or  two  (paramedian  and  dorsolateral) 
pairs  of  pale  longitudinal  lines  (the  upper- 
most of  which  separating  a  fairh'  broad 
longitudinal  dark  zone  between  itself  and 
the  middorsal  pale  zone),  lateral  fields 
black  with  scattered  pale  (rusty  in  life?) 
dots,  especially  posteriorly;  throat  orange. 

Distribution:  The  northern  and  western 
portions  of  the  Tiburon  Peninsula  of  Haiti 
from  Marfranc  east  to  Miragoane  and  vi- 
cinity; inland,  in  the  eastern  portion  of  its 


range,  to  the  vicinity  of  Fond  des  Negres 
and  St.  Michel  du  Sud  (Fig.  1). 

Discussio7i:  I  have  not  seen  living  exam- 
ples of  A.  t.  taeniura,  and  thus  am  unable 
to  treat  the  nominate  subspecies  in  the  same 
detailed  manner  as  most  of  the  other  races. 
However,  I  have  had  the  advantage  of  see- 
ing a  large  body  of  freshly  collected  and 
well  preserved  material  in  the  Museum  of 
Comparative  Zoology,  and  the  following 
notes  on  pattern  and  coloration  are  dra\\'n 
primarily  from  this  recently  taken  material. 

The  dorsal  pattern,  although  somewhat 
variable,    shows    the    following    situation. 
Tliere   is   usually  a   broad  middorsal   pale 
longitudinal  zone  from  the  head  onto  the 
dorsal  third  of  the  tail.    The  basic  pattern 
of  Tiburon  taeniura  (as  will  be  shown  in 
the    descriptions   following)    consists   of   a 
series  of  five  pale  longitudinal  lines  on  a 
dark  ground  color.    In  A.  t.  taeniura,  the 
median  dorsal  longitudinal  zone  embraces 
as  many  as  the  median  and  two  paramedian 
longitudinal  lines,   so  that  the  result  is   a 
longitudinal  pale  zone,  bordered  by  a  black 
longitudinal  zone  (the  interspace  between 
the   original    paramedian   and   dorsolateral 
light  lines),  which  in  turn  is  bordered  by 
the  dorsolateral  light  line.   In  many  lizards, 
the  extent  of  the  dorsal  pale  zone  is  vari- 
able, so  that  the  pale  longitudinal  parame- 
dian lines  may  still  be  visible  and  not  in- 
corporated into  the  dorsal  pale  zone.    The 
dorsolateral  light  lines  are  prominent,  and 
begin  above  the  eye  and  extend  onto  the 
proximal  half  of  the  tail,  \\here  they  are 
wdde  and  usually  blue-green.   On  the  head, 
the  dorsolateral  lines  are  rarely  bordered 
medially  b\'  black.    The  lateral  fields  are 
black  and  extend  from  the  temporal  regions 
along  the  sides  to  the  basal  t\\'o-thirds  of 
the  tail.    The  lateral  fields  contain  a  few 
tiny  pale  spots  (which  probably  were  dull 
red  or  brick  colored  in  life )  especially  pos- 
teriorly, and  are  bordered  ventrally  by  a 
pale  line  which  begins  at  the  upper  edge  of 
the  auricular  opening  and  continues  onto 
the  basal   quarter  of   the   tail.    This   pale 
lateral  line  is  regularly  represented   ante- 


350         Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  6 


rior  to  tlu"  auiicular  opciiiiiii  1)\'  one  or  two 
[lalc  dots  or  dashes  on  the  cheek  between 
the  eye  and  the  ear.  Below  tli(^  lateral  line 
the  lower  sides  are  dull  ura\ish  or  brown- 
ish stippled  with  pale  l)lue  or  blue-gray. 
The  \enter  is  blue-gray  (presumably  blue 
in  life),  and  the  tail  is  bright  blue  \entrally 
and  distally,  often  with  a  median  dorsal 
black  zone  (the  continuation  of  the  dark 
area  between  the  dorsolateral  pale  body 
lines),  a  black  lateral  zone  (the  continua- 
tion ot  the  lateral  fields),  and  a  \'entrolat- 
eral  dark  line  (the  continuation  of  the  dark 
area  below  the  lateral  line).  These  dark 
tail  areas  are  s(>parated  b\-  bold  blue  longi- 
tudinal lines  and  the  blue  underside  of  the 
tail.  Tlie  throats  are  presently  pale  pink- 
ish-orange, and  I  assume  that  in  life  they 
were  bright  orange.  There  is  no  striking 
sexual  or  ontogcMietic  difference  in  pattern, 
although  juNcniles  ha\-e  the  dorsal  pale 
lines  more  distinct  than  adults.  In  other 
races  which  I  have  seen  in  life,  generally 
tlic  orange  throats  of  females  are  less  bril- 
liant than  those  of  males. 

The  largest  male  ( from  Jeremie )  mea- 
sures 88  mm  in  snout-\ent  length,  and  the 
largest  female  ( from  Carrefour  Sanon,  near 
Jeremie)  measures  76  mm.  The  longitudi- 
nal \'entrals  vary  loetween  29  and  34  (mean 
32.0)  and  these  scales  are  most  often  ar- 
ranged in  10  transverse  rows  (74.1  per 
cent;,  with  25.9  per  cent  having  8  trans- 
verse rows  of  ventrals.  The  fourth  toe  sub- 
digital  scales  range  from  65  to  82  (mean 
73.2),  and  the  femoral  pores  range  from 
24  to  35  fm(>an  29.3).  The  scales  in  the  fif- 
teenth caudal  \erticil  vary  between  23  and 
29  (mean  2.5.8). 

Ol  the  two  specimens  from  Fond  des 
Negres  and  St.  Michel  du  Sud  (USNM 
7262.3  and  AMNM  49721),  the  latter  agrees 
well  in  iiattern  with  A.  t.  tacniura.  The  in- 
di\idual  from  Fond  des  Xegres  differs  from 
all  other  A.  /.  lueniura  in  having  the  five 
dorsal  hiics  well  expressed,  although  the 
iiKcliaii  one  is  fairly  broad  and  expanded. 
I  lonsidcr  this  individual  as  being  a  some- 
what aberrant  /\.  /.  tacniura. 


So  many  localities  for  A.  t.  tacniura  can 
not  be  found  on  any  map  that  it  is  difficult 
to  state  with  certainty  what  the  altitudinal 
limits  of  the  race  may  be.  There  are  speci- 
mens from  sea  level  along  the  coast  ( 10  mi. 
E  Baraderes,  Petit  Tron  de  Nippes,  Rose- 
aux )  to  about  600  feet  ( 183  meters )  at  St. 
Michel  du  Sud  and  Fond  des  Negres.  Field 
notes  ( MCZ )  by  Francois  Vuilleumier  state 
that  Amciva  'with  blue  tails"  were  regularly 
encountered  but  not  collected  on  a  trip  be- 
tween Lopino  and  Pourcine  in  the  lower 
northern  ranges  of  the  Massif  de  la  Hotte. 
Presumably  the  high  elevation  of  the  Massif 
de  la  Hotte  to  the  south  prevents  A.  t.  tac- 
niura from  meeting  the  subspecies  on  the 
south  coast. 

Specimens  examined:  Haiti,  Dcpf.  du 
Sud,  Jeremie,  6  (MCZ  3608-09,  3  specimens; 
USNM  59240-42);  Laye,  nr.  Jeremie  (not 
mapped),  1  (MCZ  65070);  Tiga,  nr.  Jeremie 
(not  mapped),  1  (MCZ  65071);  Carrefour 
Sanon,  nr.  Jeremie  (not  mapped),  18  (MCZ 
65072-84,  69994-98);  Perine,  nr.  Jeremie 
(not  mapped),  2  (MCZ  65085-86);  Bozor, 
nr.  Jeremie  (not  mapped),  4  (MCZ  65087- 
89,  65113);  Place  Negre,  2  (MCZ  65092- 
93);  Parotv,  nr.  Jeremie  (not  mapped),  1 
(MCZ  65094);  Bozo,  nr.  Jeremie  (not 
mapped),  5  (MCZ  69999-70003);  La  Source, 
nr.  Jeremie  (not  mapped),  1  (MCZ  70004); 
Marfranc,  2  (MCZ  74547-48);  Tessier,  nr. 
Marfranc  (not  mapped),  4  (74556-59); 
Troii  Bois  on  Jeremie  road  ( not  mapped ) ,  5 
(MCZ  74551-55);  Roseaux,  4  (MCZ  74549- 
50,  USNM  58245^6);  10  mi.  (16  km)  E 
Baraderes,  4  (USNM  80767-70);  Petit  Trou 
de  Nippes,  2  (USNM  80799-800);  Mirago- 
ane,  3  (  USNM  77070-71,  72635);  St.  Michel 
du  Sud,  1  (AMNH  48721);  Fond  des  Negres, 
1  (USNM  72623);  (the  following  localities, 
all  "near  Miragoane,"  are  unlocatable  and 
unmapped  and  may  be  in  either  the  De- 
partemcnt  du  Sud  or  the  Dcpartement  de 
VOucst):  Commune  Aquin,  1  (MCZ  66302); 
Risque,  2  (MCZ  66303-04);  Butete,  6  (MCZ 
6630.5-06,  CM  37926-29);  Nan  Carosse,  2 
(MCZ  66307-08);  Mingrette,  7  (MCZ 
66309-10,  66314-15,  CM  37930-32). 


Ameiva    taeniura   in   Hispamola  •   Schwartz        351 


POPULATIONS  ON  THE  ILES 

DE  GRANDE  AND   PETITE  CAYEMITE 

This  seems  an  appropriate  place  to  dis- 
cuss the  small  sample  of  A.  taeniura  from 
the  islands  of  Grande  and  Petite  Cayemite, 
which  lie  off  the  north  coast  of  the  Tiburon 
Peninsula  between  Jeremie  and  Baraderes. 
The  adjacent  mainland  of  Haiti  is  occupied 
by  A.  f.  taeniura. 

There  are  two  specimens  (MCZ  25535- 
36)  from  Grande  Cayemite  and  five  (USNM 
80819-23)  from  Petite  Cayemite.  The  for- 
mer consist  of  a  male  ( snout-vent  length  78 
mm )  and  a  female  ( 75  mm ) ,  and  the  latter 
four  males  (snout-vent  lengths  51  to  68 
mm )  and  a  female  ( 67  mm ) .  When  treated 
as  a  group,  the  following  scale  counts  are 
obtained:  longitudinal  ventrals  in  30  to  33 
rows  (mean  31.9)  and  usually  in  10  (one 
exception  with  8)  transverse  rows.  The 
fourth  toe  subdigital  scales  vary  between 
81  and  91  (mean  85.2),  the  femoral  pores 
between  28  and  32  (mean  29.9),  and  there 
are  from  26  to  30  (mean  28.4)  scales  in  the 
fifteenth  caudal  verticil.  The  population(s) 
is  thus  very  high  (and  almost  completely 
separable  from  A.  t.  taeniura — with  an  over- 
lap of  only  two  scales )  in  number  of  fourth 
toe  scales.  In  other  counts  they  are  com- 
parable, although  they  average  higher  in 
both  femoral  pores  and  fifteenth  verticil 
scales.  In  fact,  no  other  population  of  A. 
taeniura  approaches  the  very  high  number 
of  fourth  toe  scales  possessed  by  these 
Cayemite  specimens. 

Examination  of  the  patterns  shown  by 
these  specimens  leaves  much  to  be  desired. 
The  two  specimens  from  Grande  Cayemite 
are  presently  dark  and  have  five  bold  pale 
dorsal  lines  with  no  trace  of  any  middorsal 
pale  zone.  These  two  lizards  are  abun- 
dantly different  from  adjacent  A.  t.  taeniura. 
The  five  specimens  from  Petite  Cayemite 
present  a  different  aspect,  for  all  these 
lizards  are  patterned  \'ery  much  like  the 
mainland  specimens,  with  a  pale  middorsal 
zone  bordered  by  a  black  zone.  These  Pe- 
tite Cayemite  specimens  are  presently  not 


separable  in  pattern  from  those  from  the 
mainland.  Thus,  the  Grande  Cavemite 
specimens  are  different  from  A.  t.  taeniura 
in  both  pattern  and  fourth  toe  scales, 
whereas  the  Petite  Cayemite  individuals 
differ  only  in  fourth  toe  scales  and  are  com- 
parable in  pattern. 

It  is  tempting  to  name  the  Cayemite 
lizards  as  a  distinct  race,  which  I  have  little 
doubt  that  they  are.  However,  the  anoma- 
lous pattern  condition  makes  it  imperative 
that  additional  material  be  secured  before 
the  above  course  is  followed.  It  is  possible 
that  Petite  Cayemite  lizards  might  be  best 
interpreted  as  intergrades  or  intermediates 
between  A.  t.  taeniura  and  a  distinct  sub- 
species on  Grande  Cayemite,  although  of 
the  two  islands,  the  latter  is  closer  to  the 
mainland.  Another  possibility  is  that  each 
has  its  own  distinct  subspecies;  only  fresh 
material  will  clarify  the  situation. 

Ameiva  taeniura  regnotrix  new  subspecies^ 

Holotype:  MCZ  81072,  an  adult  male, 
from  Camp  Perrin,  Dept.  du  Sud,  Haiti,  one 
of  a  series  collected  26  July  1962  by  David 
C.  Leber.    Original  number  X2989. 

Parati/pes:  MCZ  81073-75,  same  data  as 
type;  UF  21318-19,  UIMNH  61602-03,  same 
locality  as  type,  22  July  1962,  native  collec- 
tor; ASFS  X2667-70,  X2686-88,  X2701-04, 
same  locality  as  type,  23  July  1962,  native 
collector;  ASFS  X2818-22,  same  locality  as 
type,  24  July  1962,  native  collector;  CM 
40555-58,  same  locality  as  type,  28  July 
1962,  native  collector;'  AMNH  94233-34, 
Carrefour  Canon,  500  feet  ( 152  meters ) , 
Dept.  du  Sud,  Haiti,  1  August  1962,  R.  F. 
Klinikowski;  AMNH  94232,  4  km  NE  Carre- 
four  Canon,  Dept.  du  Sud,  Haiti,  1  August 
1962,  D.  C.  Leber;  KU  93303-04,  Ra\ine 
Citronnier,  10  km  N,  2  km  E  Cavaillon, 
Dept.  du  Sud,  Haiti,  6  August  1962,  D.  R. 
Paulson;  UF  21320,  14  km  N  Cavaillon,  1800 
feet  (549  meters),  Dept.  du  Sud,  Haiti,  6 
August  1962,  D.  C.  Leber;  KU  93305-07, 

^  Latin,  ruling;  tliis  is  the  only  species  of  the 
genus  known  from  its  geographic  range. 


352 


Bulletin  Museum  of  Comparative  Zoology.  Vol.  135,  No.  6 


UIMNH  61604-06,  Cavaillon,  Dept.  du  Sud, 
Haiti,  6  August  1962,  native  collector. 

Diugnosis:  A  subspecies  of  A.  tacnium 
characterized  by  a  combination  of  moderate 
size  (males  to  (S8  mm,  females  to  81  mm 
snout- vent  length),  regularly  8  transverse 
rows  of  ventrals,  very  low  number  of  fourth 
toe  subdigital  scales,  low  number  of  femoral 
pores,  and  moderate  number  of  scales  in 
the  fifteenth  caudal  \erticil;  dorsal  pattern 
consisting  of  five  dorsal,  longitudinal,  yel- 
low to  tan  lines,  the  median  and  parame- 
dian lines  at  times  enclosed  in  a  reddish 
brown  middorsal  zone  ( although  these  lines 
still  maintain  their  distinctness  if  so  in- 
cluded), lateral  fields  black  and  without  in- 
cluded pale  dots;  throat  orange. 

DistriJ)iition:  Extreme  southwestern  por- 
tion of  the  Tiburon  Peninsula,  from  Carre- 
four  Canon  and  Camp  Perrin  in  the  west, 
east  to  the  vicinity  of  Cavaillon   (Fig.  1). 

Description  of  fijpc:  An  adult  male  with 
the  following  measurements  and  counts: 
snout- vent  length  83  mm,  tail  204  mm; 
ventrals  in  33  longitudinal  and  8  transverse 
rows;  fourth  toe  subdigital  scales  38  and  34 
(total  72);  femora]  pores  15  and  17  (total 
32);  26  scales  in  the  fifteenth  caudal  verti- 
cil. Dorsal  ground  color  reddish  brown  in 
life,  with  a  series  of  five  longitudinal  yel- 
lowish tan  lines,  the  median  and  parame- 
dian lines  slightly  wider  than  the  dorso- 
lateral lines  and  blending  with  the  reddish 
brown  head  color  on  the  occiput.  A  black 
longitudinal  zone  between  the  paramedian 
and  dorsolateral  lines  beginning  on  the 
supraoculars,  extending  medially  between 
the  supraoculars  and  median  head  shields, 
and  continuing  onto  the  base  of  the  tail, 
the  dorsolateral  lines  the  most  prominent 
and  extending  onto  the  proximal  three- 
quarters  of  the  tail  as  dorsolateral  pale 
yello\\'ish  lines.  The  paramedian  lines  ex- 
tend onto  the  base  of  the  tail  where  they 
fuse  with  one  another  at  about  the  eigh- 
teenth dorsal  caudal  verticil.  Lateral  fields 
black,  without  included  pale  dots.  Lateral 
pale  line  from  the  auricular  opening  to 
the  groin,  and  then  continuing  behind  the 


leg  as  a  broad,  blue  ventrolateral  caudal 
line.  Throat  vivid  orange  (pi.  2C12;  all 
color  designations  from  Maerz  and  Paul, 
1950),  this  color  extending  onto  chest;  re- 
mainder of  venter  grayish  orange  extending 
onto  underside  of  tail.  Lower  sides  below 
yellow  lateral  line  brown.  Limbs  vaguely 
marbled  black  and  dark  brown.  A  promi- 
nent yellow  dash  on  the  cheek  and  a  yello\\' 
reversed  C  bordering  the  anterior  edge  of 
the  auricular  opening. 

Voriofion:  The  series  of  38  A.  t.  regnatrix 
has  the  following  counts:  longitudinal  ven- 
trals 28-33  (mean  31.3);  rows  of  transverse 
ventrals  8  (94.7  per  cent)  or  10  (5.3  per 
cent);  fourth  toe  scales  61-74  (mean  66.7); 
femoral  pores  24-34  (mean  28.9);  fifteenth 
verticil  23-28  ( mean  24.8 ) . 

The  coloration  and  pattern  of  A.  f.  reg- 
natrix are  fairly  constant.  The  five  dorsal 
longitudinal  lines  and  the  black  lateral 
fields  without  included  pale  or  red  dots  are 
regular  features  of  the  series.  Some  speci- 
mens have  the  median  and  paramedian  dor- 
sal lines  widened,  at  times  so  much  so  as  to 
form  a  dorsal  pale  zone.  However,  even 
in  these  cases  the  integrity  of  the  longitudi- 
nal lines  is  quite  clear.  The  extension  of 
the  black  longitudinal  zone  onto  the  supra- 
oculars and  between  them  and  the  median 
head  scales  is  a  constant  pattern  feature. 
The  underside  of  the  tail  in  adults  is  or- 
ange-gray proximally  and  blue-gray  distally; 
in  juveniles  the  tail  is  vivid  blue  dorsally, 
whereas  in  adults  only  the  more  distal  por- 
tion is  blue,  the  proximal  region  being  black 
with  prominent  tan  to  yellowish  dorsal  and 
dorsolateral  lines. 

Comparisons:  The  major  scale  difference 
between  the  races  tacniura  and  regnatrix 
is  that  the  former  customarily  has  10  trans- 
verse rows  of  ventrals,  the  latter  8.  Al- 
though there  is  much  overlap,  the  mean 
(66.7)  of  fourth  toe  scales  in  regnatrix  is 
considerably  lower  than  that  (73.2)  in 
tacniura. 

In  pattern  the  two  are  quite  distinct.  The 
middorsal  zone  of  tacniura  contrasts  with 
the  five-lined  dorsum  of  regnatrix.    Taeni- 


Ameiva    taeniura   in  Hispaniola  •  Schioartz 


353 


lira  regularly  has  some  pale  flecks  in  the 
black  lateral  fields,  whereas  regnatrix  reg- 
ularly lacks  intrafield  markings.  The  ex- 
tension of  the  black  longitudinal  zones  onto 
the  head  shields  affords  a  rapid  means  of 
differentiating  the  two  races. 

Remarks:  A.  t.  regnatrix  occurs  from  sea 
level  to  elevations  of  1800  feet  ( 549  meters ) 
in  the  Massif  de  la  Hotte  north  of  Cavaillon. 
At  Camp  Perrin,  the  lizards  were  abundant 
among  rocks  along  the  edges  of  cultivated 
fields.  At  Carrefour  Canon  they  were  col- 
lected along  the  edge  of  a  canal  in  mesic 
coffee  canopy  forest,  and  14  km  N  of  Cavail- 
lon they  were  encountered  along  the  edge  of 
Coffca  in  the  uplands.  The  Ravine  Citron- 
nier  localitv  is  in  xeric  scrub,  but  is  within 
the  uplands  of  the  Massif  de  la  Hotte. 

Ameiva  taeniura  aequorea^  new  subspecies 

Holotype:  MCZ  81086,  an  adult  male, 
from  western  end,  Ile-a-Vache,  Dept.  du 
Sud,  Haiti,  one  of  a  series  collected  4  Au- 
gust 1962  by  Ronald  F.  Klinikowski,  Da\  id 
C.  Leber,  and  Dennis  R.  Paulson.  Original 
number  X3416. 

Parati/pes:  AMNH  94235-37,  CM  40559- 
61,  KU  93308-11,  UF  21321-23,  UIMNH 
61607-09,  same  data  as  type;  ASFS  X3570- 
76,  same  locality  as  type,  6  August  1962, 
R.  F.  Klinikowski. 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  small 
size  (males  to  80  mm,  females  to  73  mm 
snout-vent  length),  regularly  10  transverse 
rows  of  ventrals,  moderate  number  of 
fourth  toe  subdigital  scales  and  scales  in 
the  fifteenth  caudal  verticil,  and  low  num- 
ber of  femoral  pores;  dorsal  pattern  consist- 
ing of  five  longitudinal  yellow  dorsal  lines, 
of  which  the  median  is  often  obsolescent 
or  incomplete  on  the  neck,  on  a  black 
ground,  lateral  fields  of  black  without  or 
with  very  few  tin\'  pale  dots;  throat  dark 
orange. 

Distribution:  Ile-a-Vache,  Haiti  (Fig.  1). 

Description  of  type:    An  adult  male  with 

^  Latin,  sea-girt,  referring  to  Ile-a-Vache. 


the  following  measurements  and  counts: 
snout-vent  length  80  mm,  tail  146,  distal 
two-thirds  regenerated;  ventrals  in  33  longi- 
tudinal and  10  transverse  rows;  fourth  toe 
subdigital  scales  40  and  39  ( total  79 ) ;  fem- 
oral pores  15  and  13  (total  28);  21  scales  in 
the  fifteenth  caudal  verticil.  Dorsal  ground 
color  black,  with  five  longitudinal  yellow 
lines  which  are  distinctly  orange  anteriorly. 
The  median  line  is  obscure  and  broken  on 
the  neck,  whereas  the  paramedian  lines  are 
somewhat  brighter  and  the  dorsolateral 
lines  are  boldly  distinct.  The  longitudinal 
black  stripe  begins  on  the  supraoculars  and 
sends  a  short  branch  between  the  supra- 
oculars and  the  median  head  scales.  The 
lateral  fields  are  black  and  have  a  very 
few  tiny  scattered  pale  dots.  The  parame- 
dian lines  extend  onto  the  tail,  where  they 
join  at  about  the  twenty-third  dorsal  caudal 
verticil.  The  dorsolateral  lines  expand  upon 
the  base  of  the  tail  and  continue  ( becoming 
progressively  bluer)  to  the  point  of  regen- 
eration. There  is  an  orange  cheek  spot  and 
an  orange  preauricular  spot.  The  lateral 
line  begins  at  the  auricular  opening  and 
continues  to  the  groin;  posterior  to  the  hind- 
limb,  the  lateral  line  continues  as  a  bold 
white  ventrolateral  tail  stripe.  The  sides  be- 
low the  lateral  line  are  black,  somewhat 
stippled  with  gray.  The  throat,  chest,  and 
anterior  half  of  venter  were  dark  orange 
(pi.  1G12)  in  life,  the  remainder  of  the 
\  enter  being  duller  grayish  orange.  The 
limbs  are  coarsely  marbled  black  and  tan. 
The  underside  of  the  tail  is  blue. 

Variation:  The  series  of  24  A.  t.  aequorea 
has  the  following  counts:  longitudinal  ven- 
trals 30-32  (mean  31.5);  rows  of  transverse 
ventrals  10  (95.8  per  cent)  or  8  (4.2  per 
cent);  fourth  toe  scales  65-83  (mean  76.6); 
femoral  pores  26-31  (mean  28.3);  fifteenth 
verticil  21-26  (mean  23.2). 

The  coloration  and  pattern  of  the  para- 
types  are  close  to  that  of  the  type.  No  other 
specimen  has  any  included  pale  dots  in  the 
black  lateral  fields,  and  indeed  they  are  far 
from  conspicuous  in  the  type.  The  median 
pale  dorsal  line  is  always  visible,  regularly 


354         Biilliliii  Museum  of  Comparative  Zoology,  Vol.  135,  No.  6 


broken  on  tlic  neck,  and  may  be  somewliat 
obscured  b\-  a  tan  clouding  between  the 
paramedian  lines.  Even  in  the  latter  case, 
the  median  line  is  visible.  The  throats, 
chests,  and  anterior  portion  of  the  venter  are 
dark  orange  in  males;  females  have  the 
throats  somewhat  paler.  The  underside  of 
the  tail  may  be  blue  or  gray.  The  limbs  are 
heavily  marbled  with  black  and  tan. 

Comparisons:  The  most  pertinent  com- 
parison is  between  aequorea  and  the  ad- 
jacent re^natrix.  The  fonner  has  10  trans- 
verse rows  of  ventrals  and  the  latter  8.  The 
much  lower  mean  of  fourth  toe  scales 
(66.7)  in  regnafrix  differentiates  the  main- 
land race  from  aequorea  (with  a  mean  of 
76.6).  Aequorea  also  appears  not  to  reach 
so  large  a  size  as  regnatrix.  The  two  are 
comparable  in  pattern,  but  the  broken  me- 
dian line  on  the  neck  in  aequorea  will  dif- 
ferentiate them.  The  dark  orange  throat 
and  orange-tinted  dorsal  lines  of  aequorea 
differ  from  the  brighter  throat  and  yellow 
to  tan  lines  of  regnatrix.  From  taeniura,  the 
Ile-a-Vache  subspecies  differs  in  lacking  a 
middorsal  pale  zone  and  in  lacking  pale 
dots  in  the  lateral  fields.  Both  aequorea 
and  taeniura  have  10  transverse  rows  of 
ventrals.  The  northern  race  reaches  a 
slightly  larger  size  than  aequorea. 

Retnarks:  A.  t.  aecjuorea  is  close  to  reg- 
natrix in  pattern,  and  differs  primarily  in 
details  of  coloration  and  in  number  of 
transverse  rows  of  ventrals.  Its  origin  from 
regnatrix  on  the  adjacent  mainland  is  ob- 
\  ions.  On  Ile-a-Vache,  A.  t.  aequorea  was 
abundant  in  shady  situations  especially 
along  the  mangrove  border,  where  the  earth 
was  cool  even  at  midday.  Many  more 
lizards  were  observed  than  were  collected. 

Ameiva  taeniura  navassae  Schmidt,  1919 

Amc'iva  nava.ssac  Scliiiiiclt,  1919,  iSiill.  Amer.  Mus. 
Nat.  Hist.,  41(12):524  (type  locality — Navassa 
Island ) . 

It  seems  appropriate  here  to  comment 
upon  Ameiva  navassae  Schmidt.  This  spe- 
cies is  known  from  a  single  specimen 
(AMNH  12607;  collected  by  Hollo  II.  Beck 


on  Navassa  between  13  and  19  July  1917. 
It  is  an  adult  male,  with  the  following  mea- 
surements and  counts:  snout-vent  length 
85  mm;  ventrals  in  34  longitudinal  and  10 
transverse  rows;  fourth  toe  subdigital  scales 
43  and  42  ( total  85 ) ;  femoral  pores  15  and 
16  (total  31);  scales  in  fifteenth  caudal 
verticil  28.  The  dorsum  is  presently  dark 
\\'ith  four  pale  longitudinal  lines  and  a 
median  pale  middorsal  area;  the  dorso- 
lateral pale  lines  extend  anteriorly  over  the 
outer  edge  of  the  supraoculars  and  more 
or  less  onto  the  canthus.  The  lateral  fields 
are  black  and  without  included  pale  dots. 
The  lower  sides  are  vaguely  marbled  with 
light  and  dark.  The  tail  is  regenerated  for 
its  distal  half.  Four  dorsal  pale  lines  ex- 
tend onto  the  unregenerated  portion  of  the 
tail  and  there  is  a  broad  pale  ventrolateral 
line  as  well.  Both  fore-  and  hindlimbs  are 
vaguely  marbled  with  dark  and  light.  There 
is  a  pale  cheek  spot  and  a  pale  preauricular 
spot.  The  top  and  sides  of  the  head  are  dull 
tan,  and  the  chin  and  throat  are  now  pale, 
in  contrast  to  the  dull  grayish  blue  of  the 
venter. 

There  can  be  no  doubt  that  A.  navassae 
is  related  to  A.  taeniura,  rather  than  to  A. 
auheri  from  Cuba,  as  Schmidt  (1921b:  559) 
suggested.  It  may  seem  strange  that 
Schmidt,  who  had  collected  A.  taeniura  in 
the  Repiiblica  Dominicana  (1921a),  did 
not  recognize  the  similarity  of  the  two  "spe- 
cies." This  is,  however,  easily  attributable 
to  the  fact  that  Dominican  A.  taeniura  lack 
the  lined  pattern  of  the  Tiburon  races, 
which  Schmidt  had  not  observed. 

A.  navassae  resembles  most  closely  in 
details  of  pattern  specimens  of  A.  t.  regna- 
trix. Thomas  ( 1966 )  has  commented  on  the 
possibility  that  the  type  specimen  of  navassae 
may  have  originated  in  the  vicinity  of  Les 
Cayes  (from  which  port  Beck  set  out  for 
Navassa;  Wetmore  and  Swales,  1931:19) 
and  later  was  mislabeled  as  ha\ing  come 
from  Navassa.  This  supposition  reaches 
greater  importance,  for,  should  regnatrix 
and  navassae  be  identical,  the  southwestern 
Tiburon   race   would   then   take   the  name 


Ameiva   taeniura  in  Hispaniola  •  Schwartz        355 


navassae.  The  resemblance  of  navassac  and 
regnatrix  may  be  due  to  close  relationship, 
since  a  Hispaniolan  lizard  arriving  on  Na- 
vassa  might  be  most  logically  expected  to 
have  come  from  the  tip  of  the  Tiburon  Pen- 
insula, rather  than  elsewhere.  The  fact  that 
no  collector,  either  before  or  after  Beck  ( see 
Thomas,  1966,  for  details),  has  taken  an- 
other specimen  of  Ameiva  on  Navassa  may 
be  very  significant,  or  it  may  be  of  no  sig- 
nificance whatsoever,  if  the  supposed  de- 
structive effects  of  the  lighthouse-keeper's 
domestic  animals  can  be  blamed  for  the 
disappearance  of  some  of  the  Navassan 
fauna. 

In  any  event,  from  the  very  fact  that  I 
have  not  used  the  name  navassae  for  the 
Camp  Perrin-Cavaillon  lizards,  it  is  obvious 
that  I  am  not  convinced  of  the  identity  of 
navassae  with  regnatrix.  The  fourth  toe 
scales  in  navassae  number  85;  no  specimen 
from  the  Tiburon  Peninsula  itself  nor  from 
Ilc-a-Vache  (whence  the  type  of  navassac 
might  have  come )  has  so  high  a  count  ( 130 
examined),  the  highest  being  a  count  of 
83  for  aequorea.  The  high  count  for  reg- 
natrix is  74.  The  navassae  count  of  85  is, 
however,  included  by  the  counts  of  Grande- 
Petite  Cayemite  lizards  (81-91).  However, 
since  Beck  is  not  known  ever  to  have  visited 
these  islets,  or  even  to  have  visited  the 
north  coast  of  the  Tiburon,  it  does  not  seem 
likely  that  the  iy^e  of  navassae  originated 
on  the  Cayemites.  The  34  longitudinal  rows 
of  ventrals  in  navassac  are  greater  in  num- 
ber than  in  any  specimen  of  regnatrix,  but 
are  equalled  or  exceeded  by  three  speci- 
mens of  taeniura.  A.  navassae  has  10  trans- 
verse rows  of  dorsals  in  contrast  to  8  ro\\  s  in 
regnatrix. 

As  Thomas  ( 1966 )  has  suggested,  it  is 
possible  that  A.  navassae  came  from  a 
local  population  of  A.  t.  regnatrix  which 
has  as  yet  not  been  sampled,  a  population 
in  which  such  high  fourth  toe  counts  do 
occur.  I  hesitate  to  say  that  no  Ameiva 
occurred  in  the  recent  past  on  Navassa. 
What  can  be  said  is  that  A.  navassae  must 
be  regarded  as  a  subspecies  of  A.  taeniura. 


whatever   the   histoiy    and   provenance   of 
the  type  and  only  specimen. 

Ameiva  taeniura  var/ca'  new  subspecies 

Holotijpe:  MCZ  81076,  an  adult  female, 
from  Morne  Calvaire,  1  mi.  (1.6  km)  SW 
Petionville,  2300  feet  (701  meters),  Dept. 
de  rOuest,  Haiti,  taken  21  June  1962  by  na- 
tive collector.    Original  number  X1299. 

Paratypes:  ASFS  X1322,  same  data  as 
tvpe,  22  June  1962,  R.  F.  Klinikowski; 
USNM  59220,  Petionville,  Dept.  de  FOuest, 
Haiti,  23  March  1917,  J.  B.  Henderson 
and  P.  Bartsch;  ASFS  X2265-66,  Belle 
Fontaine,  north  base  of  Morne  la  Visite,  ± 
3000  feet  (915  meters),  Dept.  de  FOuest, 
Haiti,  13  Julv  1962,  D.  C.  Leber;  UIMNH 
61610,  Furcy,  5600  feet  (1707  meters),  Dept. 
de  FOuest,  Haiti,  6  July  1962,  E.  Cyphale; 
ASFS  X2351-52,  Furcy,  5600  feet,  Dept. 
de  FOuest,  Haiti,  15  July  1962,  native  col- 
lector; MCZ  58016-17,  Furcy,  5600  feet, 
Dept.  de  FOuest,  Haiti,  26  October  1950,  A. 
Curtiss;  MCZ  63606  (2  specimens),  YPM 
3657-58,  Furcy,  5600  feet,  Dept.  de  FOuest, 
Haiti,  May  1960,  L.  Whiteman;  MCZ  65.347, 
Furcy,  5600  feet,  Dept.  de  FOuest,  (no 
date),  A.  Curtiss;  AMNH  70144-45,  Furcy, 
5600  feet,  Dept.  de  FOuest,  Haiti,  Septem- 
ber 1949,  A.  Curtiss. 

Associated  specimens:  Haiti,  Dept.  de 
VOuest,  Petit  Goave,  1  (USNM  59244);  6.2 
mi.  (9.9  km)  W  Fauche,  1  (ASFS  X2047); 
1.3  mi.  (2.1  km)  NE  Fauche,  2  (ASFS 
X204.3-44);  5  km  S  Dufort,  1  MCZ  63339); 
Momance,  2  (MCZ  8637,  8643);  Canefour, 
1  (MCZ  59503);  Morne  de  Cayette,  1  (MCZ 
63605);  Diquini,  3  (MCZ  8691-92,  8695); 
Bas  Cap  Rouge,  5  (MCZ  65167-68,  CM 
37831-33);  Marbial,  21  km  XE  Jacmel,  6 
(MCZ  65163-66,  CM  37829-30);  1  to  2  mi. 
( 1.6  to  3.2  km)  E  Cayes  Jacmel,  1  (AMNH 
39899 ) ;  halfway  between  Cayes  Jacmel  and 
Marigot,  2  (MCZ  58105,  AMNH  49761). 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  moderate 

^  Latin,  straddling,  in  reference  to  its  occurrence 
on  both  sides  of  the  Massif  de  la  Selle. 


356         Bulletin  Museum  of  Comparative  Zoology.  Vol  135,  No.  6 


size  (males  to  90  mm,  females  to  82  mm 
snout-vent  length),  usually  10  transverse 
rows  of  ventrals,  lo\\'  number  of  fourth  toe 
subdigital  seales,  moderate  number  of  fe- 
moral pores  and  seales  in  the  fifteenth  cau- 
dal verticil;  dorsal  pattern  consisting  of  a 
broad  pale  to  medium  brown  zone  bordered 
by  bright  yellow  dorsolateral  lines,  lateral 
fields  black  with  scattered  red  to  buffy  dots; 
throat  orange. 

DistriJ)ution:  The  base  of  the  Tiburon 
Peninsula,  from  Petit  Goave  to  Petionville 
on  the  north,  and  into  the  uplands  as  far  as 
Furcy  and  Belle  Fontaine;  on  the  south  side 
of  the  Massif  de  la  Selle  from  Bas  Cap 
Rouge  and  Marbial  near  Jacmel,  west  to 
near  Marigot  (Fig.  1). 

Description  of  type:  An  adult  female 
with  the  following  counts  and  measure- 
ments: snout-vent  length  82  mm,  tail  bro- 
ken; ventrals  in  33  longitudinal  and  8  trans- 
verse rows;  fourth  toe  subdigital  scales  36 
and  37  ( total  73 ) ;  femoral  pores  15  and  15 
( total  30 ) ;  27  scales  in  the  fifteenth  caudal 
verticil.  A  pale  brown  dorsal  band  extend- 
ing from  the  occiput  onto  the  base  of  the 
tail,  \\'here  it  becomes  gradually  constricted 
and  disappears  on  about  the  nineteenth 
dorsal  caudal  \erticil;  dorsal  zone  bordered 
laterally  by  bright  yellow  dorsolateral  lines, 
and,  in  the  region  of  the  neck,  some  slightly 
darker  longitudinal  ai-eas  enclosed  within 
the  band  adjacent  to  the  dorsolateral  lines. 
Lateral  fields  black  with  a  few  widely  scat- 
tered red  dots  throughout  their  length,  al- 
though the  dots  are  more  concentrated  pos- 
teriorly. Lateral  yellow  line  from  auricu- 
lar opening  to  groin,  with  orange  cheek 
marking  which  is  confluent  with  an  orange 
auricular  marking.  Lateral  line  resumed  be- 
hind hindliml:)  and,  along  with  dorsolateral 
lines,  continued  onto  tail.  Lower  sides 
flecked  black  and  gray.  Throat  and  lower 
labials  orange  (Maerz  and  Paul,  1950:  pi. 
1119),  this  color  extending  onto  anterior 
abdomen;  posterior  venter  gray,  as  also  un- 
derside of  tail.  Limbs  tannish,  marbled 
with  dark  gray  and  black. 

Variation:    The  series  of  35  A.  t.  varica 


has  the  following  counts:  longitudinal  ven- 
trals 29-33  (mean  31.8);  rows  of  transverse 
ventrals  10  (75.0  per  cent)  or  8  (25.0  per 
cent);  fourth  toe  scales  64-80  (mean  71.5); 
femoral  pores  26-35  (mean  30.8);  fifteenth 
verticil  23-29  ( mean  25.8 ) . 

A.  t.  varica  is  a  somewhat  variable  race. 
Since  it  has  a  wide  altitudinal  range  ( from 
sea  level  to  5600  feet  [1707  meters]),  and 
since  it  occurs  on  both  sides  of  the  Massif 
de  la  Selle,  such  variation  is  not  surprising. 
On  the  other  hand,  I  am  unable  to  distin- 
guish specimens  from  the  region  between 
Petit  Goave  and  Momance,  or  from  Mar- 
bial and  Cayes  Jacmel,  from  upland  speci- 
mens from  Petionville  and  Furcy.  Although 
most  specimens  resemble  the  type  in  hav- 
ing a  broad  brown,  pale  brown,  or  reddish 
brown  middorsal  zone,  some  (UIMNH 
61610,  for  example)  have  buffy  indications 
of  the  median  and  paramedian  lines.  Oc- 
casional individuals  (CM  37830)  lack  dots 
in  the  black  lateral  fields.  Others,  rather 
than  having  the  middorsal  zone  bordered 
directly  by  the  dorsolateral  yellow  lines, 
have  an  inteiposed  black  stripe  between 
the  lines  and  the  zone  (MCZ  63605,  for 
example).  In  this  latter  condition,  the 
black  lines  stop  on  the  neck  and  do  not 
continue  anteriorly  onto  the  head.  In  fresh 
specimens  the  throats  are  orange,  and  the 
venters  vary  between  dull  orange  and  gray. 
The  dorsolateral  lines  may  be  bright  yellow, 
as  in  the  type,  buffy,  or  yellow  anteriorly 
and  grayish  yellow  posteriorly.  The  lower 
sides  may  be  dotted  with  blue,  and  this 
color  may  occur  also  on  the  lateralmost 
ventral  plates.  The  dots  on  the  lateral  fields 
vary  from  red  to  buffy,  and  there  may  be 
some  red  spotting  on  the  lower  sides.  The 
hindlimbs  mav  be  flecked  with  dull  red 
(brick). 

Co7n))arisons-:  A.  t.  varica  may  be  differ- 
entiated from  re^natrix  in  that  the  former 
has  usually  10  transverse  rows  of  ventrals, 
the  latter  regularly  8.  From  regnatrix  and 
acqiiorca,  varica  differs  in  usually  having 
prominent  red  dots  in  the  lateral  fields;  al- 
though   tacniura    may    have    lateral    dots, 


Ameiva    taeniura   in  Hispaniola  •  Schwartz 


OO  I 


they  are  most  often  restricted  to  the  pos- 
terior portion  of  the  lateral  fields.  The 
lined  dorsa  of  acquorco  and  rcgnatrix  also 
distinguish  these  races  from   vaiico. 

In  size,  varied  is  larger  than  acqiiorea 
and  equal  to  taeniura;  variea  and  regnutrix 
are  about  equal  in  size.  In  fourth  toe  scales, 
variea  is  most  strongly  different  from  reg- 
natrix  (means  of  71.5  in  the  former,  66.7  in 
the  latter). 

Remarks:  Both  the  altitudinal  and  geo- 
graphic distributions  of  A.  t.  variea  are 
extensive;  the  race  occurs  from  sea  level  to 
5600  feet  ( 1707  meters )  in  the  Mome 
THopital.  Geographically,  variea  occurs  on 
both  sides  of  the  Massif  de  la  Selle,  as  high 
as  about  ISOO  feet  ( 549  meters )  on  the  Pla- 
teau Cap  Rouge  and  on  the  coast.  I  doubt 
that  there  is  direct  contact  between  the 
northern  and  southern  populations,  since 
the  abrupt  northern  escaipment  of  the  Mas- 
sif de  la  Selle  interxenes  between  these 
two  regions.  The  way  of  contact  must  be 
devious.  It  is  interesting  that  specimens  of 
variea  occur  at  the  foot  of  the  scai-p  at  Belle 
Fontaine.  Considering  that  the  Massif  de 
la  Hotte  in  the  west  separates  the  races 
taeniura  and  regnatrix,  it  is  surprising  that 
apparently  the  Massif  de  la  Selle  does  not 
act  in  the  same  manner  in  the  east;  on  the 
other  hand,  there  are  no  specimens  of  the 
western  races  from  high  elevations  in  the 
La  Hotte  (although  this  may  well  be  an 
artifact  of  collecting).  Specimens  of  A.  /. 
variea  have  been  taken  in  a  wooded  thicket 
in  a  mesic  cultivated  area  (Mome  Calvaire), 
from  a  river  flood-plain  in  brush-covered 
rocks  and  in  cultivated  areas  (Belle  Fon- 
taine), and  along  the  inner  margin  of  man- 
groves and  on  a  rocky  hillside  near  a  Musa 
patch  (Fauche). 

The  only  subspecies  which  approaches 
variea  closely  on  the  west  is  taeniura;  the 
closest  localities  for  the  two  are  Mira- 
goane  (taeniura)  and  Petit  Goave  (variea), 
which  are  separated  by  about  23  kilometers. 
Possibly  some  of  the  specimens  from  local- 
ities "near  Miragoane'  would  bridge  this 
gap    slightly.     Specimens    from    the    Mira- 


goane area  show  no  approach  to  variea. 
The  relationships  between  variea  and  the 
more  northern  and  eastern  subspecies  will 
be  discussed  below. 

Ameiva  taeniura  barbouri  Cochran,  1928 

Articiva  harl>ouri  Cochran,  1928,  Proc.  Biol.  Soc. 
Washington,  41:56  (type  locality — La  Source, 
He  de  la  Gonave ) . 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  large 
size  (males  to  100  mm,  females  to  74  mm 
snout- vent  length),  more  often  10  trans- 
verse rows  of  ventrals  (although  the  inci- 
dence of  S  rows  is  almost  equal  to  that  of 
10),  moderate  number  of  fourth  toe  scales, 
very  high  number  of  femoral  pores,  and 
low  number  of  scales  in  the  fifteenth  caudal 
verticil;  dorsal  pattern  consisting  of  a  broad 
median  dorsal  metallic  tan  zone  bordered 
directly  by  black  and  undotted  lateral  fields 
without  an  intervening  dorsolateral  pale 
stripe;  tail  blue-green  and  unpattemed; 
throat  orange. 

Distribution:  He  de  la  Gonave,  and  the 
adjacent  mainland  of  Haiti  in  the  vicinity  of 
Trou  Forban  ( Fig.  1 ) . 

Diseussion:  There  are  now  a\'ailable 
twenty-seven  specimens  of  A.  t.  barbouri 
from  Gonave,  and  another  from  the  main- 
land near  Trou  Forban.  Coloration  and  pat- 
tern of  tv\'o  specimens  from  Etroits  were 
recorded  as:  dorsal  band  metallic  tan 
(Maerz  and  Paul,  1950:  pi.  12D5),  bordered 
directly  by  black  and  unspotted  lateral 
fields.  Sides  grayish,  not  separated  from 
lateral  field  by  an  intervening  pale  lateral 
line.  Head  rich  tan.  Throat  orange  (pi. 
3B12),  ventral  ground  color  entirely  blue 
(pi.  27E1).  Tail  blue-green  (pi.  26J2),  and 
without  any  pale  or  dark  lines.  Hands  and 
feet  pale  blue,  limbs  clear  gray,  unspotted. 
The  specimen  from  Trou  Forban  had  a 
metallic  tan  back  (pi.  14G8),  grading  to 
green  on  the  base  of  the  blue-green  tail. 
The  lateral  field  was  completely  black,  the 
throat  vivid  orange  (pi.  4B12),  with  the 
chest  paler,  fading  to  dull  blue  on  the  rest 
of  the  venter.    The  color  description  of  the 


358        Biillitiii  Mti.scuiu  of  Comparative  Zoology,  Vol.  135,  No.  6 


Gonave  lizards  and  of  that  from  Troii  For- 
ban  are  remarkably  similar.  It  is  possible 
that  the  mainland  population  may  later  be 
separated  from  that  from  Gona\e,  but  at 
present  there  is  no  reason  for  so  doing. 

There  are  only  two  females  kno\\m,  the 
larger  with  a  snout-vent  length  of  74  mm; 
the  largest  male  harhouri  has  a  snout-vent 
length  of  100  mm.  The  longitudinal  ven- 
trals  vary  between  30  and  34  (mean  32.7) 
and  these  scales  are  more  often  arranged  in 
10  transverse  rows  (51.9  per  cent),  with 
48.1  per  cent  having  8  transverse  rows  of 
ventrals.  The  fourth  toe  subdigital  scales 
range  from  70  to  80  (mean  74.4),  and  the 
femoral  pores  range  from  33  to  41  (mean 
36.2).  The  scales  in  the  fifteenth  caudal 
verticil  vary  between  18  and  25  (mean 
22.0).  The  Trou  Forban  male  (not  included 
in  the  above  series)  has  a  snout-vent  length 
of  69  mm,  32  longitudinal  and  8  transverse 
rows  of  \'entrals,  76  fourth  toe  scales,  22 
scales  in  the  fifteenth  verticil,  and  18  fem- 
oral pores  on  the  one  uninjured  leg. 

Comparisons:  The  Gonave  race  of  A. 
faeniura  requires  no  comparison  with  the 
mainland  races  to  the  south.  The  absence 
of  any  longitudinal  lines  on  the  back,  the 
juxtaposed  dorsal  band  and  unspotted  black 
lateral  fields,  and  the  unicolor  and  pattern- 
less  tail  will  distinguish  harhouri  from  the 
described  races.  The  very  high  number  of 
femoral  pores  (36.2  versus  28.3  to  30.8) 
is  distinctive;  there  is  no  overlap  in  this 
count  between  harhouri  and  aequorea,  and 
an  overlap  of  only  two  or  three  scales  be- 
tween harhouri  and  taeniura,  regnatrix  and 
varica. 

Remarks:  The  occurrence  of  A.  t.  har- 
houri on  the  mainland  is  suggestive  of  the 
relationships  of  the  xeric  littoral  along  the 
north  shore  of  the  Golfe  de  la  Gonave  and 
the  He  de  la  Gonave.  The  occurrence  on 
this  strip  of  such  species  as  Anolis  hrevi- 
rostris  and  Dip]o<iJoss%ifi  cwtissi  confirms 
the  relationship  of  these  two  regions.  Since 
both  the  anole  and  the  galliwasp  occur  as 
well  in  the  Cul  de  Sac  plain,  it  is  not  com- 
pletely unlikely  that  parts  of  this  plain  are 


(were)  occupied  by  A.  t.  Ijarhouri,  al- 
though there  is  no  evidence  at  present  of 
such   occurrence    (see   Discussion). 

The  record  of  harhouri  from  Trou  Forban 
is  one  of  three  records  of  A.  taeniura  from 
north  of  the  Cul  de  Sac  plain  in  Haiti.  To 
the  south  occurs  the  race  varica,  separated 
by  some  68  kilometers  airline.  The  other 
northern  records  for  A.  taeniura  are  Plai- 
sance  and  St.  Michel  de  L'Atalaye;  the  near- 
est of  these  localities  is  about  60  kilometers 
airline.  None  of  the  northern  lizards  is 
close  to  harhouri.  The  mainland  distribu- 
tion of  A.  t.  harhouri  is  at  present  unknown. 

On  Gonave,  A.  t.  harhouri  is  widespread, 
being  known  from  one  northern  and  three 
more  southern  localities.  The  \\vo  speci- 
mens collected  by  us  were  taken  on  a 
rocky  hillside  at  the  foot  of  the  central  hills; 
the  immediate  area  was  xeric  scrub  with 
some  large  shade  trees.  At  Trou  Forban, 
the  single  lizard  was  taken  on  a  rocky  path 
into  a  moist  depression  with  a  dense  stand 
of  Acacia  trees  in  an  otherwise  very  arid 
area.  The  elevation  at  Nan  Cafe  on  Gonave 
is  about  1260  feet  (384  meters)  an  upper 
limit  for  the  occurrence  of  harhouri  on 
Gonave. 

Specimens  examined:  Haiti,  He  de  la 
Gomive,  La  Source,  2  ( MCZ  25537-38, 
type  and  paratype);  1.5  mi.  (2.4  km)  SW 
Etroits,  2  (ASFS  X2506-07);  Nan  Cafe,  20 
(MCZ  61064-66,  UF  12242  (2  specimens), 
UF  12243  (4  specimens),  UF  12244,  12245, 
12246  (4  specimens),  YPM  3308-09,  YPM 
3311,  YPM  3313-14);  Pointe  a  Raquettes,  2 
(YPM  3315-16);  no  other  locality,  1  (USNM 
S0827);  Depf.  de  TOuest,  2.2  mi.  (3.5  km) 
SW  Trou  Forban,  1  (ASFS  X1926). 

Ameiva  taeniura  vulcanalis^  new  subspecies 

Holotijpe:  MCZ  81077,  an  adult  male, 
from  5  mi.  (8  km)  NE  Oviedo,  Pedemales 
Province,  Republica  Dominicana,  one  of  a 
series  taken  4  August  1963  by  David  C. 

^  Latin,  belonging  to  Vulcan,  in  allusion  to  the 
\'ivid  orange  throat. 


Ameiva    taeniura   in  Hispaniola  •  Schwartz        359 


Leber  and  Richard  Thomas.   Original  num- 
ber V281. 

Parah/pes:  ASFS  V282-84,  same  data  as 
type;  ASFS  X9959-60,  RT  753,  same  locaHty 
as  type,  30  July  1963,  A.  Schwartz,  R. 
Thomas;  RT  777,  same  locality  as  type,  7 
August  1963;  ASFS  X9954,  13.1  mi.  (21  km) 
SW  Enriquillo,  Pedemales  Province,  Re- 
publica  Dominicana,  30  July  1963,  A. 
Schwartz;  ASFS  X9410-15,  13.1  mi.  (21 
km)  SW  Enriquillo,  Pedemales  Province, 
Republica  Dominicana,  22  July  1963,  A. 
Schwartz,  R.  Thomas;  DWB  296,  13.1  mi. 
(21  km)  SW  Enriquillo,  Pedemales  Prov- 
ince, Republica  Dominicana,  7  December 
1964,  R.  Thomas;  UF  21324,  1.3  mi.  (2.1 
km)  NW  Oviedo,  Pedemales  Province,  Re- 
publica Dominicana,  30  July  1963,  R. 
Thomas;  KU  93312,  3  km  SW  Enriquillo, 
Barahona  Province,  Republica  Dominicana, 
7  August  1963,  R.  Thomas. 

Associated  specimens:  Haiti,  Dept.  de 
rOucsf,  Tean,  nr.  Saltrou,  4  (MCZ  68576-79). 
Republica  Dominicana,  Pcdemoles  Prov- 
ince, Pedemales,  9  (ASFS  V2667,  V2787- 
94);  8  km  N  Pedemales,  1  (ASFS  V2602); 
6  km  NE  Las  Mercedes,  2600  feet  (793 
meters),  1  (ASFS  V2648);  30  km  NW 
Oviedo,  1  (MCZ  57731);  Barahona  Prov- 
ince, Barahona,  10  (AMNH  37203-06, 
37208,  MCZ  63192,  58020-22,  ASFS  X9749); 
1  mi.  (1.6  bn)  N  Barahona,  1  (MCZ  43812); 
4  km  NW,  1  km  SW  Barahona,  1  (ASFS 
V201);  4  km  NW,  2  km  SW  Barahona,  500 
feet,  2  (ASFS  V203-04);  west  side,  Punta 
Martin  Garcia,  11  (ASFS  V89-99);  Inde- 
pendencia  Province,  3  km  WNW  El  Na- 
ranjo,  1000  feet  (305  meters),  3  (ASFS 
X9946-48). 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  large  size 
(males  to  95  mm,  females  to  82  mm  snout- 
vent  length),  usually  10  transverse  rows  of 
ventrals,  low  number  of  fourth  toe  sub- 
digital  scales,  high  number  of  femoral 
pores,  and  moderate  number  of  scales  in 
the  fifteenth  caudal  verticil;  dorsal  pattem 
consisting  of  a  broad  pale  (tan,  gray,  or 
greenish  tan )  zone,  at  times  with  remnants 


of  the  dorsolateral  pale  (lemon  yellow  to 
buffy)  longitudinal  lines,  lateral  fields  black 
with  many  small  orange  flecks;  throat  fire 
orange. 

Distribution:  From  the  vicinity  of  Sal- 
trou in  extreme  southeastern  Haiti,  east 
across  the  Peninsula  de  Barahona  ( south  of 
the  Sierra  de  Baoruco)  in  the  Republica 
Dominicana  to  Oviedo;  thence  northward 
along  the  coast  to  Barahona  and  west  along 
the  north  flank  of  the  Sierra  de  Baoruco  to 
El  Naranjo,  and  east  around  the  Bahia  de 
Neiba  to  Punta  Martin  Garcia  ( Fig.  1 ) . 

Description  of  type:  An  adult  male  with 
the  following  counts  and  measurements: 
snout-vent  length  90  mm,  tail  224  mm;  ven- 
trals in  32  longitudinal  and  8  transverse 
rows;  fourth  toe  subdigital  scales  37  and  37 
(total  74);  femoral  pores  17  and  20  (total 
37);  27  scales  in  the  fifteenth  caudal  ver- 
ticil. A  broad  tan  middorsal  zone  from  the 
occiput  onto  the  basal  quarter  of  the  tail, 
this  zone  only  very  vaguely  outlined  with 
buffy,  the  lines  not  extending  onto  the  head 
shields  nor  prominently  onto  the  tail.  Lat- 
eral fields  obsolescent  and  gray  on  the  tem- 
poral region,  black  between  the  limbs, 
heavily  flecked  with  orange  dots  through- 
out their  length,  and  becoming  faint  on  the 
sides  of  the  tail  near  the  base.  Lateral  fields 
bordered  below  by  a  faint  and  obsolescent 
buffy  line,  which  behind  the  hindlimbs 
forms  a  fairly  prominent  pale  ventrolateral 
tail  stripe.  Pale  cheek  spot  absent,  preauric- 
ular spot  present,  buffy,  and  not  conspicu- 
ous. Lower  sides  gray  with  indistinct  dark 
mottling.  Limbs  tan,  mottled  with  darker 
on  the  thighs.  Throat  and  chest  vivid  fire 
orange,  bellv  dull  gravish  blue.  Tail  green- 
ish  tan  above,  bluish  green  belo\\'. 

Variation:  The  series  of  57  A.  t.  vulcan- 
alis  has  the  following  counts:  longitudinal 
ventrals  31-35  (mean  32.3);  rows  of  trans- 
verse ventrals  10  (80.3  per  cent)  or  8  (19.7 
per  cent);  fourth  toe  scales  66-82  (mean 
73.8);  femoral  pores  30— tO  (mean  .33.9); 
fifteenth  verticil  24-30  (mean  26.7). 

Despite  its  extensive  range,  A.  t.  vulcan- 
alis  is  very  constant  in  pattem,  and  most 


360 


IhiUctin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  6 


specimens  resemble  the  description  of  the 
type.    In  precise  shade  of  the  dorsal  zone, 
there  is  some  variation,  however;  the  range 
in  color  includes  tan  to  greenish  tan   ( En- 
riquillo),   gray    (Las   Mercedes),  tan  with 
a  faint  greenish  border  or  tan  anteriorly  and 
greenish  posteriorly  (Pedernales),  and  red- 
dish tan  anteriorly  and  grayish  tan  poster- 
iorly  (Punta  Martin  Garcia).    The  lateral 
fields  are  black  and  almost  alwa\'s  are  heav- 
ily flecked  with  orange  flecks,  although  oc- 
casional specimens  (ASFS  V93,  for  exam- 
ple)  lack  flecks  completely.    If  there  is  a 
dorsolateral  line  separating  the  dorsal  zone 
from  the  lateral  field  ( this  is  not  the  usual 
condition),  it  is  yellow.   If  there  is  a  lateral 
stripe  ( and  there  often  is  not ) ,  it  is  cream. 
The   throats   and   chests   are   always   vivid 
flame  orange,   and  the  venters  vary  from 
whitish  (Pedernales)  to  grayish  blue  (En- 
riciuillo)    or    dull    grayish    orange    (Punta 
Martin   Garcia).    The   head  markings   are 
always  obsolete,  the  preauricular  spot  being 
the  more  persistent  of  the  two.    The  tails 
are  not  prominently  striped  dorsally,   and 
are  greenish  or  tannish  green  dorsally,  dark 
gray  to  black  lateralh',  \\'ith  a  broad  cream 
stripe    \entrolaterall\-.     The   hindlimbs   are 
dark,    almost  black,    in    many   individuals, 
and  the  thighs  are  flecked  with  greenish  or 
tan. 

Ccnnparisonfi:  No  other  race  thus  far 
described  has  heavily  flecked  lateral  fields 
and  lacks  dorsal  stripes.  A.  t.  harhouri 
superficially  resembles  viilcanalis,  but  in 
details  the  two  subspecies  are  very  differ- 
ent; the  solid  black  lateral  fields  of  the 
former  blending  into  the  dorsal  metallic  tan 
zone  are  distinct  from  the  sharp-edged 
dorsal  zone  of  vulcanalis.  The  unicolor  tail 
of  ])arhouri  likewise  distinguishes  it  from 
vulcanalis.  The  tan  zonate  dorsum  of  vul- 
canalis will  distinguish  it  from  the  striped 
or  brown  or  reddish  brown  dorsa  of  the 
western  races. 

Remarks:  A.  t.  vulcanalis  is  an  inhabi- 
tant of  some  of  the  more  arid  areas  in  His- 
paniola,  but  in  this  region  it  occupies  shady 
situations  such  as  stands  of  deciduous  trees, 


wooded    mountain    foothills,    and    shaded 
Acacia  stands.  Its  interaction  with  A.  chnjs- 
olacma  at  Oviedo  has  been  described  by 
Schwartz  and  Klinikowski  (1966).  Although 
several  species  of  reptiles  are  restricted  to 
the  tip  of  the  Peninsula  de  Barahona  by  the 
Sierra  de  Baoruco  and  the  virtually  non-ex- 
istent eastern  coastal  plain   (and  these  re- 
stricted species  include  A.  chrysolaema  and 
A.  Uncolafa,  each  of  which  has  developed 
races  both  to  the  north  and  south  of  the 
Sierra  de  Baoruco  in  xeric  habitats),  such 
is  not  the  case  with  A.  taeniura,  where  vul- 
canalis occurs  both  to  the  north  and  south 
of  the  mountains.  Undoubtedly,  the  eastern 
edge  of  the  Sierra  de  Baoruco  provides  ex- 
cellent mesic  habitat  for  this  lizard  and  this 
accounts  for  the  continuity  of  the  popula- 
tions between  Barahona  and  Oviedo.   Also, 
vulcanalis  has  crossed  the  lower  reaches  of 
the  Rio  Yaque  del  Sur.    Near  the  mouth  of 
this  river,  the  Valle  de  Neiba  is  distinctly 
mesic,  and  this  feature  has  presumably  al- 
lowed vulcanalis  to  cross  the  otherwise  xeric 
valley    into    the    region    of    Punta    Martin 
Garcia.     Although   A.   t.    vulcanalis  is   not 
known  to  occur  in  the  xeric  regions  of  the 
\^alle   de   Neiba,   it   does   occur   along   the 
northern   lower  foothills   of  the   Sierra   de 
Baoruco  as  far  west  as  El  Naranjo.    Since 
this  locality  is  very  close  to  the  Dominico- 
Haitian  border,  vulcanalis  is  to  be  expected 
along  the  northern  slopes  of  the  Mome  des 
Enfants  Perdus  in  Haiti. 

The  highest  elevation  for  vulcanalis  is 
2600  feet  (793  meters)  above  Las  Mercedes. 
The  species  is  presumably  absent  from  high 
elevations  in  the  Sierra  de  Baoruco,  since 
there  has  been  much  collecting  in  this 
range,  especially  in  the  Valle  de  Polo  region. 
A.  t.  vulcanalis  might  be  expected  to  occur 
at  Foret  des  Pins  in  Haiti,  at  5800  feet 
( 1768  meters )  near  the  Dominico-Haitian 
border;  it  has  not  been  taken  there  nor  at 
intermediate  or  high  elevations  on  the  Do- 
minican side  of  the  boundary.  It  is  inter- 
esting that  of  the  four  southern  subspecies 
(taeniura,  reii,natrix,  varica,  vulcanalis)  as- 
sociated  directlv  with  mountainous   areas, 


Ameiva    taeniura    in  Hispaniola  •  Schwartz        361 


only  varica  occurs  at  very  high  elevations. 
A.  t.  vuJcanalis  and  A.  t.  varica  approach 
one  another  along  the  southern  coast  of 
Haiti;  the  easternmost  record  of  varica 
(halfway  between  Cayes  Jacmel  and  Mari- 
got)  and  the  westernmost  record  for  vul- 
canalis  (Trou  Roche  near  Saltrou)  are  sep- 
arated by  about  38  kilometers  ( Trou  Roche 
cannot  be  precisely  located).  Between  Mari- 
got  and  Saltrou,  the  Morne  Fortune  fonns  a 
steep  scarjD  adjacent  to  the  ocean,  and  this 
may  effectively  separate  varica  and  vulcan- 
alis. 

Ameiva  taeniura  azuae   new  subspecies 

HoJotijpe:  MCZ  81078,  a  subadult  male, 
from  22  km  NW  Azua,  Azua  Province,  Re- 
publica  Dominicana,  taken  14  August  1963 
by  David  C.  Leber.   Original  number  V459. 

Paratype:   ASFS  V458,  same  data  as  type. 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  small  ( ? ) 
size  ( male  65  mm,  female  70  mm  snout- vent 
length  ),  10  transverse  rows  of  ventrals,  very 
high  number  of  fourth  toe  subdigital  scales, 
moderate  number  of  femoral  pores,  and 
high  number  of  scales  in  the  fifteenth 
caudal  verticil;  dorsal  pattern  consisting  of 
a  broad  bro\\'n  dorsal  zone,  bordered  by 
lemon  yello\\'  dorsolateral  lines,  black  lat- 
eral fields  \\'ith  many  large  brick  dots; 
throat  black. 

Distribution:  Kno\Mi  only  from  the  type 
locality  in  the  Llanos  de  Azua,  Republica 
Dominicana   (Fig.  1). 

Description  of  type:  A  subadult  male 
with  the  following  counts  and  measure- 
ments: snout-vent  length  65  mm,  tail  127 
mm,  distal  half  regenerated;  ventrals  in  34 
longitudinal  and  10  transverse  rows;  fourth 
toe  subdigital  scales  44  and  41  (total  85); 
femoral  pores  16  and  15  (total  31);  31 
scales  in  the  fifteenth  caudal  verticil.  A 
broad  brown  dorsal  zone,  bordered  anter- 
iorly by  a  pair  of  narrow  dorsolateral  lemon 
yellow  lines,  extending  onto  the  unregener- 
ated  portion  of  the  tail.  Lateral  fields  black, 
contiimous  from  temporal  region  onto  sides 


of  tail  basally,  and  heavily  spotted  with 
large  brick  dots.  Lateral  line  below  lateral 
fields  grayish  yellow,  fairly  prominent. 
Cheek  and  auricular  spots  yellow  and  mod- 
erately prominent.  Throat  and  chest  black, 
infralabials  and  chin  shields  dull  gray.  Ven- 
ter and  underside  of  hindlimbs  bronzy,  un- 
derside of  tail  grayisli  blue.  Limbs  brown, 
somewhat  marbled  with  darker  gray  or 
brown.  Tail  not  striped  dorsally,  but  with 
a  pale  ventrolateral  stripe,  the  continuation 
of  the  lateral  body  stripe. 

Variation:  The  only  other  specimen  is  a 
female  with  a  snout- vent  length  of  70  mm, 
ventrals  in  31  longitudinal  and  10  transverse 
rows,  79  fourth  toe  scales,  31  femoral  pores, 
and  30  scales  in  the  fifteenth  verticil.  In 
coloration  and  pattern,  the  female  is  identi- 
cal to  the  type,  except  that  there  are  more 
brick  dots  in  the  lateral  fields,  and  these 
dots  are  arranged  into  a  series  of  about  nine 
vertical  bars  in  the  posterior  half  of  the 
fields.  The  throat  of  the  female  was  dull 
gray  rather  than  black;  the  ventral  colora- 
tion was  bronzy  like  that  of  the  male. 

Comparisons:  No  previously  described 
race  has  a  black  throat,  and  azuae  can  be 
thus  easily  distinguished  from  all  other 
subspecies.  The  liigh  counts  of  fourth  toe 
scales  separate  azuae  from  all  other  races; 
the  only  exception  to  this  are  those  lizards 
from  the  Cayemites  which  have  counts  from 
81  to  91.  The  high  fifteenth  verticil  counts 
of  azuae  distinguish  it  from  taeniura,  reg- 
natrix,  aec/uorea,  varica  and  barbouri.  Ad- 
ditional specimens  of  azuue  will  doubtless 
bring  about  some  overlap  in  these  counts. 

Remarks:  Although  A.  t.  azuae  is  known 
only  from  two  specimens,  it  is  eminently 
distinct.  Of  all  the  specimens  of  A.  t.  vul- 
eanalis,  its  neighbor  to  the  south,  none  has 
a  black  or  gray  throat — in  fact  the  vivid 
flame  orange  throats  of  vulcanalis  offer 
strong  contrast  to  the  black  (and  gray) 
throats  of  azuue.  The  closest  approxima- 
tion of  vulcanalis  ( Punta  Martin  Garcia)  to 
the  type  locality  of  azuae  is  only  about  23 
kilometers  airline.  The  specimens  of  azuae 
were  taken  in  a  moderatelv  mesic  ravine  in 


362         Biillcliii  Museum  of  Cotiijxinitive  Zoology,  Vol.  135,  No.  6 


Acacia  scrub.  Although  \\'e  collected  ex- 
tensively in  the  Llanos  de  Azua,  often  in 
mesic  and  shady  areas,  we  did  not  encoun- 
ter A.  tacniiira  elsewhere.  Surely  the  dis- 
tribution of  A.  f.  aztiac  is  more  extensive 
than  the  present  record  indicates.  Schmidt 
(1921a:  17)  reported  Beck's  taking  of  A. 
faeniiira  in  "the  interior  of  Azua  Province"; 
whether  this  specimen  is  from  the  Llanos 
(and  thus  ]-)robably  azuac)  or  is  from  the 
interior  uplands  ( where  much  of  Beck's  col- 
lecting in  this  area  was  carried  on;  see  Wet- 
more  and  Swales,  1931)  is  unknown.  A.  t. 
azuac,  in  addition  to  the  Llanos  de  Azua, 
may  occur  as  well  in  the  Valle  de  San  Juan. 

Ameiva  foeniura  tofacea^  new  subspecies 

Holotij))c:  MCZ  81079,  an  adult  male, 
from  mouth  of  the  Rio  Chavon,  west  side. 
La  Romana  Province,  Republica  Domini- 
cana,  one  of  a  series  taken  4  September 
1963  by  Ronald  F.  Klinikowski,  Albert 
Schwartz  and  Richard  Thomas.  Original 
number  \T065. 

Parahjpes:  ASFS  V1064,  V1066-68,  UI- 
MNH  61611-13,  same  data  as  type;  CM 
40562-63,  Rio  Cumayasa,  17  km  W  La 
Romana,  La  Romana  Province,  Republica 
Dominicana,  28  June  1963,  D.  C.  Leber  and 
R.  Tliomas;  ASFS  X9293,  8  km  E  La  Ro- 
mana, La  Romana  Province,  Republica  Do- 
minicana, 19  July  1963,  R.  Thomas;  AMNH 
7567,  San  Pedro  de  Macoris,  San  Pedro  de 
Macoris  Province,  Republica  Dominicana 
(no  date),  0.  K.  Noble;  SMF  25700,  San 
Pedro  de  Macoris,  San  Pedro  de  Macoris 
Province,  Republica  Dominicana,  6  March 
1939,  R.  Mertens;  SMF  25553,  Tres  Ojos, 
Distrito  Nacional,  Republica  Dominicana, 
16  March  1939,  R.  Mertens. 

Associated  specimen:  Republica  Domini- 
cana, El  Seiho  Province,  "San  I'^rancisco 
Mountains,  ±  2500  feet,"  1  (USNM  35982). 

Diagnosis:  A  subspecies  of  A.  faeniura 
characterized  by  a  combination  of  large 
size  (males  to  96  mm,  females  to  83  mm 

^  Latin,  like  sandstone,  in  allusion  to  the  pale 
sandy  dorsum. 


snout-vent  length),  more  often  10  than  8 
transverse  rows  of  ventrals  (although  the 
difference  in  incidence  between  the  two 
categories  is  slight),  and  moderate  number 
of  fourth  toe  subdigital  scales,  femoral 
pores,  and  scales  in  the  fifteenth  caudal 
verticil;  dorsal  pattern  consisting  of  a  very 
pale  greenish  tan  to  sandy  dorsal  zone,  bor- 
dered by  yellow-green  dorsolateral  lines, 
lateral  fields  black  with  some  scattered 
small  brick  dots;  throat  pale  orange. 

Distrihiition:  Known  from  Tres  Ojos  east 
to  the  mouth  of  the  Rio  Chavon;  specimens 
reported  by  Cochran  (1941:  274)  from  the 
city  of  Santo  Domingo  in  the  Distrito  Na- 
cional may  be  assignable  to  this  subspecies. 
The  single  specimen  from  the  "San  Fran- 
cisco Mountains,  about  2500  feet"  likewise 
seems  close  to  tofacea,  and  extends  the 
range  of  this  race  into  the  interior  of  eastern 
Hispaniola   ( Fig.  1 ) . 

Description  of  type:  An  adult  male  with 
the  following  counts  and  measurements: 
snout-vent  length  77  mm,  tail  143  mm;  ven- 
trals in  32  longitudinal  and  8  transverse 
rows;  fourth  toe  scales  36  on  one  leg,  other 
leg  damaged;  femoral  pores  14  and  13  (to- 
tal 27);  23  scales  in  the  fifteenth  caudal 
verticil.  A  broad  pale  greenish  tan  dorsal 
zone,  bounded  by  conspicuous  yellow-green 
dorsolateral  lines,  and  grading  to  greenish 
on  sacrum  and  base  of  tail.  Dorsolateral 
lines  continue  onto  tail  and  are  discernible 
to  near  tip.  Head  pale  tannish  brown, 
slightly  darker  than  dorsal  zone.  Lateral 
fields  black,  bordered  below  by  a  yellow- 
green  line,  and  with  some  scattered  rusty 
flecks.  Cheek  and  preauricular  spots  yel- 
low, fairly  prominent.  Lower  sides  dark 
gray  flecked  with  cream.  Lateral  pale  stripe 
continues  onto  anterior  face  of  thigh,  and 
also  resumes  on  tail  as  a  broad  ventrolateral 
pale  stripe;  sides  of  tail  black,  continuous 
with  lateral  fields.  Both  fore-  and  hind- 
limbs  tan,  much  spotted  with  black.  Throat 
and  chest  pale  orange,  remainder  of  venter 
pale  bluish.    Underside  of  tail  pale  blue. 

Variation:  The  series  of  15  specimens  of 
A.  t.  tcrfacea  has  the  following  counts;  Ion- 


AmEIVA    TAENWRA    IX   HiSPANIOLA   •    Scliicaitz- 


363 


gitudinal  ventrals  31-33  (mean  32.1);  rows 
of  transverse  ventrals  10  (53.3  per  cent)  or 
8  (46.7  per  cent);  fourth  toe  scales  66-88 
(mean  77.0);  femoral  pores  27-35  (mean 
30.9);  fifteenth  verticil  21-29  (mean  24.2). 

The  paratypes  agree  closely  with  the  type 
in  coloration  and  pattern.  The  specimens 
from  the  Rio  Cumayasa  were  brown  to 
greenish  brown  dorsally  in  life,  with  yellow- 
green  dorsolateral  lines  and  orange  cheek 
and  preauricular  markings.  The  specimen 
from  La  Romana  was  bronzy  tan  above, 
with  yellow  dorsolateral  lines.  Tlie  lateral 
fields  usually  have  some  rusty  flecks,  al- 
though three  juveniles  lack  this  feature. 
The  throats  and  chests  are  regularly  pale 
orange;  the  ventral  ground  color  varies  from 
pale  bluish  to  pale  orange,  and  in  the  latter 
case,  the  lower  sides  are  also  dotted  \\ith 
orange.  The  underside  of  the  tails  varies 
from  clayey  gray  to  blue  or  blue-green. 

Comparisons:  From  the  four  extreme 
western  subspecies,  A.  t.  tofocea  differs  in 
having  a  dorsal  band  bordered  by  dorsolat- 
eral stripes,  rather  than  having  a  striped 
dorsum.  Tofacca  most  closely  resembles 
both  video nol is  and  azuae,  but  may  be  dis- 
tinguished from  the  latter  in  having  a  pale 
orange  rather  than  a  black  or  gray  throat. 
The  throat  color  of  viilcanalis  is  brilliant 
orange,  rather  than  pale  orange.  The  heav- 
ily flecked  lateral  fields  of  vulcanalis  differ 
from  the  more  sparsely  flecked  fields  of 
tofacca.  The  obsolete  head  markings  of 
vulcanalis  also  \\i\\  differentiate  the  two 
races.  From  harhouri,  tofacca  differs  in 
having  the  dorsal  zone  bordered  by  the 
dorsolateral  light  lines,  by  having  flecks  in 
the  lateral  fields,  and  by  having  a  patterned 
tail. 

Remarks:  A.  t.  tofacea  occurs  from  the 
coast  up  to  elevations  of  about  2500  feet 
( 762  meters ) ,  if  the  elevation  noted  for  the 
specimen  from  the  "San  Francisco  Moun- 
tains" is  correct.  Since  these  mountains  are 
presently  not  locatable  on  an\-  map,  I  am 
not  completely  sure  where  they  are;  the 
major  mountain  range  in  extreme  eastern 
Hispaniola  is  the  Cordillera  Oriental,  whose 


maximum  elevation  is  about  2300  feet  (701 
meters ) . 

The  specimen  from  La  Romana  was 
taken  in  xeric  scmb  but  adjacent  to  a  fence 
row  of  shade  trees;  the  series  from  the  Rio 
Chavon  was  taken  along  the  coast  in  shaded 
thom-scrub  and  sea-grape.  At  the  Rio  Cu- 
mayasa, on  28  July  1963,  Richard  Thomas 
collected  five  eggs  under  a  large  flat  rock 
in  the  river  valley;  two  of  these  eggs  mea- 
sured 18.9  X  13.6  mm  and  18.8  X  11.5  mm. 
One  was  opened  on  the  following  day  and 
contained  a  young  Ameiva.  This  foetus 
(ASFS  X9927)  clearly  shows  the  pattern 
characteristics  of  A.  t.  tofacca.  As  far  as 
I  am  aware,  the  eggs  of  A.  taeniura  have 
never  before  been  found  in  the  field. 

Ame'iva  taeniuro   vafra'    new  subspecies 

Holotype:  MCZ  81080,  an  adult  female, 
from  0.5  mi.  ( 0.8  km )  NW  Boca  de  Yuma, 
La  Romana  Province,  Republica  Domini- 
cana,  one  of  a  series  taken  30  August  1963 
by  Albert  Schwartz  and  Richard  Tliomas. 
Original  number  V862. 

Parafi/pcs:  ASFS  V863-66,  AMNH  942,38- 
40,  KU  93313-15,  RT  798,  same  data  as 
type. 

Associated  specimen:  Republica  Dotnini- 
cana,  La  Romana  Province,  0.7  mi.  SE  El 
Macao,  1  (ASFS  X7878). 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  small 
size  (males  to  74  mm,  females  to  71  mm 
snout-vent  length),  usually  10  transverse 
rows  of  \'entrals,  moderate  number  of  fourth 
toe  subdigital  scales,  high  number  of  fem- 
oral pores,  and  low  number  of  scales  in 
the  fifteenth  caudal  verticil;  dorsal  pattern 
consisting  of  a  rather  narrow  yello\\'ish  tan 
dorsal  zone  bordered  by  two  indistinct 
yellow  dorsolateral  lines,  lateral  fields  solid 
black;  throat  bright  fire  orange. 

Distribution:  Known  only  from  two  lo- 
calities in  extreme  eastern  Hispaniola,  to 
the  north  and  south  of  Cabo  Engaiio  (Fig. 

!)• 

^  Latin,  cunning,  in  allusion  to  their  wariness. 


364        Bulletin  Mtisciini  of  Cumpdiatwc  Zoology.  Vol.  135,  No.  6 


Description  of  type:  An  adult  female 
with  the  following  counts  and  measure- 
ments: snout-vent  length  71  mm,  tail  158 
mm;  ventrals  in  31  longitudinal  and  10 
transverse  rows;  fourth  toe  suhdigital  scales 
41  and  41  (total  82);  femoral  pores  17  and 
16  (total  33);  26  scales  in  the  fifteenth 
caudal  \ertieil.  A  broad  yellowish  tan  dor- 
sal zone  bordered  by  a  pair  of  dark  yellow 
dorsolateral  lines.  Both  dorsal  zone  and 
dorsolateral  lines  extend  onto  the  tail,  where 
the  lines  become  pale  blue  and  very  wide, 
and  continue  down  the  length  of  the  tail 
but  are  separated  proximally  by  a  black 
attenuated  triangular  figure.  Lateral  fields 
black,  extending  from  the  loreal  region 
along  the  sides  onto  the  lateral  surface  of 
the  tail,  bordered  below  by  a  yellow-orange 
line,  and  without  any  included  flecking. 
Lateral  line  resumed  behind  hindlimbs  to 
form  a  pale  blue  ventrolateral  caudal  line. 
Cheek  and  preauricular  markings  bold, 
yellow-orange.  Lower  sides  gray,  flecked 
with  darker  gray.  Throat  and  most  of  ven- 
ter vivid  fire  orange;  underside  of  tail  pale 
blue,  upperside  of  tail  greenish  blue.  Limbs 
marbled  tan  and  dark  brownish  gray. 

Variation:  The  series  of  13  A.  t.  vafra  has 
the  following  counts:  longitudinal  ventrals 
31-33  (mean  31.7);  rows  of  transverse  ven- 
trals 10  ( 84.6  per  cent )  or  8  ( 15.4  per  cent ) ; 
fourth  toe  scales  70-82  (mean  77.8);  fem- 
oral pores  31-37  (mean  33.7);  fifteenth 
verticil  20-26  (mean  22.5). 

The  series  of  paratopotypes  requires  no 
comment;  they  agree  in  detail  with  the  type 
in  coloration  and  pattern.  The  specimen 
from  El  Macao  was  described  in  life  as  hav- 
ing a  tan  dorsal  zone  with  the  edges  a  bit 
paler,  but  without  definitive  dorsolateral 
lines;  the  head  was  slightly  orange.  The 
lateral  fields  were  solid  black  with  a  pale 
yellow^  lateral  line.  The  lower  sides  were 
tan,  flecked  with  pale  yellow.  The  chin, 
throat,  and  subocular  area  were  bright 
orange  (Maerz  and  Paul,  1950:  pi.  4D12). 
These  notes  agree  fairly  well  with  topo- 
typical    vafra,   and   I   have  little  hesitancy 


in   assigning  the   El   Macao   lizard   to   this 
taxon. 

Comparisons:  From  its  neighbor  to  the 
west,  A.  t.  tofacea,  vafra  differs  in  smaller 
size  (96  mm  versus  74  mm  in  males)  and 
apparently  in  having  a  higher  number  of 
femoral  pores.  The  most  diagnostic  features 
are  the  vivid  (versus  pale)  orange  throats 
and  solid  black  lateral  fields  of  vafra.  From 
the  balance  of  the  races,  vafra  differs  in 
lacking  a  lined  dorsum  (as  have  the  four 
Tiburon  and  Ile-a-Vache  races),  in  having 
an  orange  throat  ( in  contrast  to  black  in 
azuae),  and  in  smaller  size  and  details  of 
pattern  and  coloration  from  vidcanalis.  Vul- 
canalis  and  vafra  are  virtually  separable  on 
the  basis  of  number  of  scales  in  the  fifteenth 
verticil;  these  scales  are  24  to  30  in  viiJcan- 
alis  and  20  to  26  in  vafra.  From  harbouri, 
vafra  differs  in  smaller  size,  in  having  a 
patterned  tail,  and  in  having  a  dorsolateral 
line  between  the  dorsal  zone  and  the  lat- 
eral fields. 

Remarks:  The  distribution  of  A.  t.  vafra 
is  apparently  restricted  to  the  more  xeric 
coastal  region  of  the  Cabo  Engano  area. 
The  type  series  was  collected  along  a  road- 
side and  in  forest  clearings  on  the  limestone 
ridge  which  parallels  the  coast  behind  Boca 
de  Yuma.  The  forest  is  rather  mesic  and 
extensive.  At  El  Macao,  the  single  individ- 
ual was  taken  in  a  very  mesic  hammock 
woods  adjacent  to  the  ocean;  several  others 
were  seen  in  a  coastal  Cocos  grove  nearby. 
The  predilection  of  A.  taeniura  for  shady 
and  moist  situations  in  otherwise  arid  re- 
gions is  once  more  demonstrated. 

A.  t.  vafra  approaches  A.  t.  tofacca  by  a 
distance  of  32  kilometers  airline  (Rio  Cha- 
\  on  and  Boca  de  Yuma ) .  The  area  between 
these  two  points  is  presently  virtually  inac- 
cessible. 

Ame/va     taeniura     rosamondae     Cochran, 

1934 

Ameiva  rosantunduc  Cochran,  1934,  Occ.  Papers 
Boston  Soc.  Nat.  Hist.,  8:179  (type  locality— Isla 
Saona ) . 

Diagnosis:    A  subspecies  of  A.  taeniura 


Ameiva    taeniura   in  Hispaniola  •  Schwartz 


365 


characterized  by  a  combination  of  large  size 
(male  to  101  mm  snout-vent  length,  no 
adult  females  known),  10  transverse  rows 
of  ventrals,  moderate  number  of  fourth  toe 
subdigital  scales,  high  number  of  femoral 
pores,  low  number  of  scales  in  the  fifteenth 
caudal  verticil;  dorsal  pattern  consisting  of 
a  narrow  greenish  gray  dorsal  zone,  faint 
yellow-green  dorsolateral  lines,  solid  black 
lateral  field,  and  a  patterned  tail;  throat 
orange. 

Distribution:  Isla  Saona,  Republica  Do- 
minicana  (Fig.  1). 

Discussion:  A.  t.  rosamondae  is  knowai 
from  only  two  specimens,  the  type  which 
is  an  adult  male,  and  a  subadult  female  col- 
lected b\'  Richard  Thomas.  Color  notes  on 
the  latter  describe  the  details  of  pattern: 
top  of  head  and  anterior  portion  of  dorsal 
zone  tan,  fading  to  greenish  gray,  and  be- 
coming green  and  then  blue  on  tail;  the  tail 
has  a  charcoal  wash  down  its  median  basal 
portion.  The  dorsolateral  stripes  are  yellow- 
green,  the  lateral  fields  solid  black  bordered 
below  by  a  pale  green  lateral  stripe.  The 
snout  is  orange;  the  mental  region  is  pink, 
becoming  orange  on  the  throat  and  chest. 
The  venter  is  grayish  green  (the  anterior 
scales  are  edged  with  orange).  The  tail  is 
greenish  dorsally  at  the  base  and  deep  blue 
(pi.  36L6)  for  its  distal  three-quarters.  The 
upper  surface  of  the  limbs  is  charcoal  col- 
ored. 

The  new  specimen  agrees  with  the  type 
in  pattern.  To  the  above  description  may 
be  added  that  the  dorsal  zone  is  narrow  and 
the  lateral  fields  especially  wide.  The  lat- 
eral fields  continue  boldly  onto  the  sides  of 
the  tail  as  broad  black  bands;  the  lateral 
stripe  forms  a  broad  pale  ventrolateral 
caudal  stripe.  The  cheek  and  preauricular 
spots  are  present  but  obsolete.  The  sides 
below  the  lateral  stripe  are  gray  with  black 
flecking. 

Comparisons:  Remarkably,  in  pattern 
A.  t.  rosamondae  most  closely  resembles  A. 
t.  harhouri  from  Gonave.  The  tan  dorsum 
of  the  latter  contrasts  with  the  greenish  gray 


dorsum  of  the  former.  The  dorsal  zone  is 
much  narrower  in  rosamondae  than  in  har- 
houri, and  the  former  has  a  patterned  versus 
an  unpatterned  tail.  From  A.  t.  vafra  on 
the  adjacent  mainland,  rosamondae  differs 
in  being  much  larger  ( 101  versus  74  mm  in 
males),  and  in  lacking  obvious  dorsolateral 
longitudinal  lines.  The  tails  of  these  two 
races  are  very  similar  in  pattern  and  pig- 
mentation. Comparison  with  the  other  sub- 
species is  not  necessary,  since  rosamondae 
is  quickly  distinguishable  both  from  those 
subspecies  with  lined  dorsa  and  those  with 
zonate  dorsa  by  its  coloration  and  pattern. 

Remarks:  A.  t.  rosamondae  is  apparently 
uncommon  on  Isla  Saona;  Thomas  saw  no 
other  individuals  in  his  eight  hours  ashore 
there.  It  is  really  remarkable  that  the  type 
of  rosamondae  (until  now  the  only  known 
specimen)  is  such  a  large  individual;  it  ranks 
third  among  all  specimens  of  A.  taeniura  I 
have  examined. 

Specimens  examined:  Republica  Domini- 
cana,  Isla  Saona,  environs  of  Mano  Juan, 
1  (ASFS  V3003);  no  precise  locality,  1 
(MCZ  37567— type). 

Ameiva  taeniura  ignobilis^   new  subspecies 

Holotype:  MCZ  81081,  an  adult  male, 
from  14.4  km  E  La  Vega,  La  Vega  Province, 
Republica  Dominicana,  one  of  a  series  taken 
27  November  1964  b\'  Richard  Thomas. 
Original  number  V4204. 

Paratypes:  ASFS  V4205-07,  same  data  as 
type;  ASFS  V4270,  12  km  NE  Jarabacoa, 
1400  feet  (427  meters).  La  Vega  Province, 
Republica  Dominicana,  30  November  1964, 
native  collector;  ASFS  V2925-27,  7  km  W 
Santiago,  Santiago  Province,  Republica  Do- 
minicana, 13  July  1964,  R.  Thomas;  MCZ 
58664,  Santiago,  Santiago  Proxince,  Repub- 
lica Dominicana,  (no  date).  Dr.  Jiminez; 
SMF  26124,  SMF  26251,  SMF  26289,  SMF 
26317,  Moca,  Espaillat  Province,  Republica 
Dominicana,  10-16  April  1939,  R.  Mertens; 
MCZ  58667,  Santiago  and  vicinity,  Santiago 

^  Latin,  obscure,  ignoble,  in  allusion  to  the  dark 
throat. 


366        Bulletin  Mtisi'init  of  Comparative  Zoology,  Vol.  135,  No.  6 


Province,  Republica  Dominicana,  (no  date). 
Dr.  Jiminez;  MCZ  57730,  3  km  S  Pena,  San- 
tiago Province,  Republica  Dominicana,  4 
Angnst  1958,  C.  E.  Ray  and  A.  S.  Rand. 

Associated  specimens:  Republica  Domini- 
cana, Saniiaiio  Rodrii^uez  Province,  19  km 
SE  Martin  Garcia,  600  feet  (183  meters), 

5  (ASPS  V125;3-57);  Puerto  Plata  Province, 
Puerto  Plata,  2  ( MCZ  5441,  AMNH  44845); 
Samana  Province.  Samana,  3  (AMNH 
40984-85,  MCZ  43700);  2  mi.  from  Samana 
(not  mapped),  1  (AMNH  42296);  1.5  mi. 
(2.4  km)  from  Samana  (not  mapped),  4 
(AMNM  42304-07);  Rojo  Cabo,  11  (AMNH 
39346-53,  40254-56);  Chico  Puerto  Fran- 
ces (not  mapped),  4  (AMNH  42300-03); 
0.5  mi.  (0.8  km)  inland  at  Puerto  Frances 
(not  mapped),  3  (AMNH  42310-12);  be- 
tween Las  Flechas  and  Clara  (not  mapped), 
1  (AMNH  42297);  Bahia  del  Rincon,  2 
(AMNH  42298-99);  Laguna,  1  (USNM 
65018);  Sanchez,  1  (CM  8137);  •'Samana 
Peninsula,"  1  (USNM  66765);  Isla  Carenero, 

6  (AMNH  42274-79). 

Diagnosis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  large  size 
(males  to  102  mm,  females  to  103  mm 
snout- vent  length),  usually  10  transverse 
rows  of  ventrals,  moderate  number  of  fourth 
toe  subdigital  scales,  femoral  pores,  and 
scales  in  the  fifteenth  caudal  verticil;  dor- 
sal pattern  consisting  of  a  broad  brown 
dorsal  zone,  bordered  by  bright  green  or 
\ fllow-green  dorsolateral  lines,  lateral  fields 
i)laek  with  large  dull  red  flecks;  throat 
black  or  gray. 

Dlstrihnlion:  In  the  west,  from  south  of 
Martin  Ciareia  and  near  La  Vega,  east  to 
the  tip  of  the  Peninsula  de  Samana;  appar- 
ently also  on  the  north  coast  near  Puerto 
Plata  (Fig.  1). 

Description  of  type:  An  adult  male  with 
the  following  counts  and  measurements: 
snout-vent  kiigth  87  mm,  tail  201  mm;  ven- 
trals in  34  longitudinal  and  10  transverse 
rows;  fourth  toe  sul)digital  scales  35  and  33 
(total  68);  femoral  pores  15  and  16  (total 
31);  29  scales  in  the  fifteenth  caudal  verti- 
cil.   A  broad  deep  biown  dorsal  zone,  bor- 


dered by  bright  green  dorsolateral  lines, 
the  dorsal  zone  continuing  onto  the  tail 
where  it  gradually  becomes  checkerboarded 
and  then  inconspicuous;  dorsolateral  lines 
on  tail  faint.  Lateral  fields  with  faint  or- 
ange, large,  scattered  spots.  Lateral  field 
begins  on  temporal  region  and  continues 
onto  basal  portion  of  tail,  where  it  is  in- 
vaded by  brown  scales.  Lateral  field  bor- 
dered below  by  a  greenish  yellow  lateral 
line,  which  stops  at  the  hindlimbs,  and  then 
continues  onto  the  tail  as  a  ventrolateral 
pale  greenish  gray  line.  Lower  sides  black 
mottled  wdth  reddish.  Low^er  labials  and 
tip  of  chin  orange,  throat  black;  chest  dull 
gray,  venter  light  gray.  Underside  of  tail 
blue-black.  Both  fore-  and  hindlimbs  heav- 
ily blotched  brown  and  black.  Cheek  and 
preauricular  spots  orange,  fairly  prominent. 

Variation:  Tlie  series  of  53  A.  t.  igno- 
bilis  has  the  following  counts:  longitudinal 
ventrals  30-34  (mean  32.3);  rows  of  trans- 
verse ventrals  10  (82.7  per  cent)  or  8  (17.3 
per  cent);  fourth  toe  scales  68-86  (mean 
76.0);  femoral  pores  26-36  (mean  30.4); 
fifteenth  verticil  23-30  (mean  26.0). 

I  am  not  certain  that  the  large  series  of 
specimens  from  the  Peninsula  de  Samana 
( 38  lizards )  is  correctly  associated  with  the 
lizards  from  the  interior.  This  is  partly  due 
to  the  fact  that  I  have  never  seen  the  Sa- 
mana lizards  in  life,  despite  three  trips  to 
the  peninsula  by  myself  and  Richard 
Thomas.  On  the  other  hand,  there  are  no 
scale  differences  between  the  two  major 
samples,  and  I  cannot  at  present  detemiine 
any  coloration  or  pattern  differences.  The 
Samana  lizards  reach  a  larger  size  than  do 
those  from  the  interior,  and  in  fact  the 
largest  female  of  any  subspecies  of  A.  tae- 
niura is  a  Samana  lizard  (USNM  65018). 
This  female  exceeds  the  largest  female  of 
any  other  races  (A.  t.  tofacea,  83  mm)  by 
20  mm. 

The  dorsal  band  in  western  (interior) 
specimens  of  A.  t.  ignobilis  is  dark  brown; 
it  was  noted  in  the  specimen  from  Jara- 
bacoa  that  the  dorsal  zone  granules  are 
Ureen  basallv,   so   that  when  viewed  from 


Ameiva   TAENiuRA   IN  HisPANiOLA  •  Schwartz        367 


behind  or  above,  the  zone  appears  to  be 
stippled  with  bright  green.  The  dorsolateral 
lines  are  fairly  conspicuous  and  vary  from 
bright  green  to  greenish  yellow.  The  lateral 
fields  are  black,  with  rather  large  and  scat- 
tered faint  reddish  to  orange  spots  (al- 
though five  young  individuals  with  snout- 
vent  lengths  to  47  mm  lacks  dots).  The 
cheek  and  preauricular  markings  may  be 
orange  or  grayish  yellow.  The  lateral  stripes 
vary  between  greenish  yellow  and  cream. 
The  throat  is  always  black  or  gray  (in  fe- 
males or  subadult  males),  although  the  la- 
bials may  be  bright  orange.  The  venter  is 
\'ariable,  having  been  recorded  as  yellow- 
ish gray  or  gray  with  an  orange  wash  ( San- 
tiago), light  gray  (adult  male)  or  orange 
(juveniles  and  females)  (La  Vega),  pink 
(Jarabacoa),  and  bluish  gray  with  faint 
orange  posteriorly  (Martin  Garcia).  The 
upper  surface  of  the  tail  is  tan  or  brown 
proximally,  usually  with  some  checker- 
boarding, and  black  distally.  One  speci- 
men (MCZ  57730)  has  the  checkerboarding 
continued  onto  the  posterior  third  of  the 
dorsal  zone. 

A  conspicuous  pattern  difference  be- 
tween these  interior  specimens  and  those 
from  the  Samana  is  that  the  tails  of  Samana 
lizards  are  prominently  lined  longitudinally, 
and  lack  the  uniform  coloration  of  the  tails 
of  interior  individuals.  I  have  little  doubt 
that  fresh  Samana  specimens  will  be  distinct 
from  lizards  from  the  interior  region. 

Comparisons:  A.  t.  ignobilis  requires 
comparison  only  with  A.  t.  azuae;  all  other 
described  races  have  orange  rather  than 
black  throats.  From  aziiae,  ig,nohiIis  differs 
in  much  larger  size  ( 102  mm  versus  65 
mm),  in  having  a  dark  brown  rather  than 
brown  dorsal  zone,  and  in  not  having  so 
many  dots  in  the  lateral  fields.  The  means 
of  fourth  toe  scales  are  quite  different  ( 82.0 
in  azuoe,  76.0  in  ignobilis),  but  the  counts 
on  the  two  specimens  of  azuae  are  em- 
braced by  the  counts  of  ignobilis. 

Remarks:  Additional  specimens,  presum- 
ably ignobilis,  have  been  reported  by  Mer- 
tens    (1939:    73)    from  the  Rio  Mao  near 


Moncion,  Santiago  Rodriguez  Province,  and 
by  Schmidt  (1921a:  17)  from  the  Rio  Gu- 
rabo  and  the  Rio  Cana,  probably  also  in 
Santiago  Rodriguez  Province,  and  from 
Villa  Riva,  Duarte  Province.  The  latter 
record  bridges  the  gap  between  the  interior 
and  Samana  localities. 

The  strange  rarity  of  A.  tacniura  on  the 
north  coast  of  the  Republica  Dominicana 
deserves  comment;  there  are  but  two  speci- 
mens from  this  region,  from  Puerto  Plata. 
During  a  lengthy  stay  at  Sosiia  in  this  area, 
and  extensive  travel  along  the  north  coast 
from  Imbert  to  Gaspar  Hernandez,  we  en- 
countered no  A.  taeniura.  It  is  possible  that 
specimens  from  along  this  north  coast  will 
differ  considerably  from  material  to  the 
south  and  east. 

In  the  interior,  A.  t.  ignobilis  occupies  the 
foothills  and  northern  slopes  of  the  Cordil- 
lera Central  to  elevations  of  1400  feet  (427 
meters).  It  occurs  as  well  in  the  eastern 
(and  more  mesic)  extremity  of  the  Valle 
de  Cibao  near  Santiago.  The  type  and  para- 
topotypes  were  collected  in  a  plantain  plan- 
tation near  the  Rio  Camu,  and  the  speci- 
mens from  Martin  Garcia  were  taken  in 
woods  along  the  edge  of  a  stream.  The 
specimen  from  Jarabacoa  apparently  came 
from  pine  forest. 

Ameiva   taeniura   olgida^    new  subspecies 

Holofijpe:  MCZ  81082,  an  adult  male, 
from  1  mi.  (1.6  km)  WSW  Constanza,  4000 
feet  (1220  meters),  La  Vega  Province,  Re- 
publica Dominicana,  one  of  a  series  taken 
2  July  1963  by  native  collector.  Original 
number  X8503. 

Paratijpes:  ASFS  X8502,  X8504-06,  same 
data  as  type;  AMNH  94241-44,  MCZ  81083- 
85,  RT  683,  same  locality  as  type,  3  July 
1963,  nati\'e  collector;  ASFS  X8653-54, 
UIMNH  61614-15,  same  locality  as  type,  4 
July  1963,  native  collector;  ASFS  X8825,  6 
km  W  Constanza,  4250  feet  (1296  meters), 

^  Latin  algida,  cold,  referring  to  the  high  ele\a- 
tion  of  this  subspecies. 


368 


Bulletin  Miiscinn  of  Comparative  Zoology,  Vol.  135,  No.  6 


La  Vega  Province,  Republica  Dominicana, 
9  July  1963,  R.  Thomas. 

Diagno.sis:  A  subspecies  of  A.  taeniura 
characterized  by  a  combination  of  moderate 
size  (males  to  92  mm,  females  to  76  mm 
snout-vent  length),  always  10  transverse 
rows  of  ventrals,  moderate  number  of  fourth 
toe  subdigital  scales  and  femoral  pores,  and 
high  number  of  scales  in  the  fifteenth 
caudal  verticil;  dorsal  pattern  consisting  of 
a  brown  to  reddish  brown  dorsal  zone  with- 
out dorsolateral  light  lines  in  adults  (but 
present  and  yellow  in  juveniles ) ,  the  dorsal 
zone  heavily  dotted  with  conspicuous  yellow 
dots  in  males  but  not  in  females,  lateral 
fields  l:)lack  to  dark  reddish  brown  flecked 
with  brick  or  golden,  lower  sides  heavily 
and  boldly  dotted  with  cream;  throat  and 
chest  black. 

Distribution:  Known  only  from  the  vi- 
cinity of  Constanza  in  the  Cordillera  Cen- 
tral, Republica  Dominicana  (Fig.  1). 

Description  of  type:    An  adult  male  with 
the   following    counts    and    measurements: 
snout-\'ent    length    91    mm,    tail    79    mm, 
broken;  ventrals  in  32  longitudinal  and  10 
transverse  rows;  fourth  toe  subdigital  scales 
41  and  40  (total  81);  femoral  pores  16  and 
16  ( total  32 ) ;  26  scales  in  the  fifteenth  cau- 
dal verticil.    A  broad  reddish  brown  dorsal 
zone,    without    indications    of    dorsolateral 
light  lines,  heavily  dotted  with  yellow  dots 
from  the  neck  to  the  base  of  the  tail,  but 
more  abundant  and  clear  posteriorly;  lateral 
field  black,  flecked  with  golden  dots.    Lat- 
eral  line   absent,   the   region   between   the 
lateral  edges  of  the  ventral  plates  and  the 
lateral  fields  heavily  sprinkled  with  creamy 
to  golden  dots.   Temporal  and  preauricular 
markings    absent.     Upper    surface    of    tail 
brown  with   some   darker  brown   checker- 
boarding basally,   and  with  no  prominent 
longitudinal  lines  or  dark  lateral  band,  the 
underside  of  the  tail  dull  grayish  tan.   Chin 
and  snout  bright  orange,  infralabials  green- 
ish yellow,  throat  and  chest  ( including  the 
first    seven    transverse    rows    of    ventrals ) 
black.     Venter   black,    dotted    with    bright 
blue.    Forelimbs  marbled  black  and  brown. 


hindlimbs  blotched  with  reddish  brown  dor- 
sally  and  spotted  bright  blue  on  their  an- 
terior faces. 

Variation:  The  series  of  18  A.  t.  algida 
has  the  following  counts:  longitudinal  ven- 
trals 31-33  ( mean  32.0 ) ;  rows  of  transverse 
ventrals  always  10;  fourth  toe  scales  67-86 
(mean  74.6);  femoral  pores  27-38  (mean 
31.2);  fifteenth  verticil  24-31  (mean  27.8). 

Male  A.  t.  ali^ida  agree  with  the  descrip- 
tion of  the  type;  the  dorsal  zone  may  be 
brown  or  reddish  brown,  and  the  lateral 
fields  vaiy  between  black  and  dark  reddish 
brown,  flecked  in  smaller  males  with  brick 
and  with  golden  in  adults.  The  dorsal  sur- 
face of  the  hindlimbs  may  be  dotted  with 
golden  flecks.  The  four  smallest  males 
(snout- vent  lengths  to  73  mm)  lack  dorsal 
dotting.  The  females  lack  dotting,  but  have 
the  dorsal  zone,  especially  posteriorly,  mar- 
bled with  darker  brown;  the  snout  in  fe- 
males is  pinkish,  not  orange.  In  females 
the  lateral  fields  are  black  with  many  brick 
dots,  and  the  lateral  and  dorsolateral  lines 
vary  between  yellow  and  pale  yellow-green. 
The  throats  are  gray  and  the  venter  dull 
reddish  orange.  All  adult  males  have 
black  throats  and  chests,  and  in  some  speci- 
mens the  black  continues  posteriorly  to  the 
center  of  the  abdomen.  The  smallest  male 
with  a  black  throat  has  a  snout-vent  length 
of  65  mm,  although  two  slightly  larger 
males  (66  and  73  mm)  have  only  gray 
throats.  In  females,  the  cheek  and  preauric- 
ular spots  are  more  clearly  defined  than 
in  males. 

Comparisons:  A.  t.  ali^ida  requires  com- 
parison only  with  the  two  other  black- 
throated  races,  azuae  and  ignobili.s:  All 
other  subspecies  have  orange  throats.  From 
both  azuae  and  ignobilis,  aliiida  differs  in 
having  a  black  chest  (and  at  times  part  of 
the  abdomen)  and  in  having  the  dorsum 
in  males  dotted  with  bright  yellow.  The 
heavily  dotted  sides  and  obsolescent  lateral 
line  in  adult  males  will  also  distinguish 
(d<iida  from  the  two  other  subspecies. 

Remarks:  A.  t.  algida  is  known  only  from 
a  rather  circumscribed  area  in  the  Cordil- 


Ameiva    taeniura    in-  Hispaxiola  •  Schwartz 


369 


lera  Central  at  ele\'ations  of  4000  and  4250 
feet  (1220  and  1296  meters);  undoubtedly 
it  is  more  widespread  than  these  data  indi- 
cate. We  spent  two  weeks  at  Constanza 
and  saw  only  one  lizard,  which  was  col- 
lected by  Richard  Thomas  late  in  a  wann 
morning  in  a  wooded  but  cut-over  ravine. 
The  natives  who  collected  most  of  the  speci- 
mens indicated  that  they  had  been  taken 
in  open  areas  near  Constanza.  Much  of  the 
slopes  above  the  Valle  de  Constanza  today 
is  covered  with  mixed  pine  and  deciduous 
shrubs  and  low  trees;  such  a  shaded  habitat 
seems  a  very  suitable  situation  for  A.  taeni- 
ura. 

The  only  subspecies  of  A.  taeniura  ad- 
jacent to  algicla  is  ignobili.'i.  The  closest 
these  two  races  are  known  to  approach  one 
another  is  about  37  kilometers,  airline.  The 
intei-vening  area,  however,  is  extremely 
rugged  and  dissected,  and  the  two  races 
may  not  be  in  direct  contact. 

NORTHWESTERN  REPUBUCA 
DOMINICANA  AND 
NORTHERN  HAITI 

There  remain  six  other  specimens,  two 
from  extreme  w^estern  Republica  Domini- 
cana,  and  four  from  Haiti  north  of  the  Cul 
de  Sac  Plain,  which  require  special  com- 
ment. These  are  the  only  specimens  of  A. 
taeniura  available  from  this  region,  and 
although  they  are  suggestive,  they  are  in- 
adequate for  systematic  treatment. 

1)  ASFS  Vi  168-69,  1  km  S  Loma  de 
Cabrera,  900  feet  (274  meters),  Dajabon 
Province,  Republica  Dominicana.  These  are 
two  males  with  the  largest  having  a  snout- 
vent  length  of  50  mm;  they  are  both  obvi- 
ously young.  In  addition,  the  smaller  is 
badly  damaged.  There  is  nothing  distinc- 
tive about  the  scale  counts.  The  throats 
were  grayish  orange  in  life  (and  thus  not 
like  either  the  adjacent  ignobilis  or  algida), 
and  the  dorsal  zones  were  olive,  almost 
black,  \\'ith  the  smaller  having  bright  yellow 
dorsolateral  lines.  The  lateral  fields  are 
black  with  red  dots.   The  entire  tail  is  very 


dark  blue-black,  with  the  dorsolateral  pale 
lines  very  much  reduced  and  almost  absent. 
Quite  obviously,  these  two  lizards  are  not 
assignable  to  either  ignobilis  or  algida. 
Their  correct  designation  must  await  fur- 
ther material. 

2)  AMNH  49848,  near  Plaisance,  Dept. 
du  Nord,  Haiti.  This  is  a  large  male  with  a 
snout- vent  length  of  87  mm.  It  is  presently 
very  discolored,  but  a  dark  median  zone 
can  be  ascertained,  and  there  appear  to  be 
dorsolateral  lines.  The  lateral  fields  are 
flecked  with  pale.  The  entire  venter  is  pres- 
ently black,  and  presumably  in  life  at  least 
the  throat  (and  chest)  ma\'  have  been  black. 
The  tail  is  patternless.  This  individual,  sep- 
arated from  the  nearest  record  of  ignoJ)iJis 
by  about  120  kilometers,  airline,  and  from 
the  Dajabon  specimens  noted  above  by 
about  85  kilometers,  might  be  considered 
to  be  ignobilis.  I  prefer  to  consider  it  pres- 
ently unidentifiable  to  subspecies. 

3)  USNM  74133-34.  St.  Michel  de  I'At- 
alaye,  Dept.  de  TArtibonite,  Haiti.  These 
are  two  males  with  snout-\'ent  length  of  70 
and  64  mm.  The  smaller  has  a  black  chest 
and  throat,  the  larger  has  these  regions  gray. 
There  is  a  broad  dorsal  zone  with  prominent 
pale  dorsolateral  lines.  The  lateral  fields 
are  hea\ily  dotted  with  pale,  and  the  lateral 
line  is  especially  prominent.  These  two  liz- 
ards might  also  be  regarded  as  ignobilis, 
but  they  may  well  belong  to  the  same  taxon 
as  the  specimen  from  Plaisance  (from 
which  St.  Michel  is  separated  by  only  28 
kilometers ).  More  material  is  badly  needed 
from  northern  Haiti  before  any  of  these 
lizards   can  be   evaluated  properly. 

4)  USNM  75922,  "Artibonite  Valley,"  Hai- 
ti. I  have  commented  elsewhere  (Schwartz, 
1966b)  on  the  status  of  a  specimen  of 
Leiocephahis  melanochlorus  supposedly 
collected  by  J.  S.  C.  Bos  well  in  the  Arti- 
bonite Valley;  that  lizard  clearly  came  from 
the  southwestern  portion  of  the  Tiburon 
Peninsula,  probably  in  the  \icinity  of  Les 
Cayes  where  Boswell  is  known  to  have 
collected.  The  Ameiva,  although  much  dis- 
colored, is  dorsally  lined,  and  resembles  ( in 


370         Bulletin  Miiscuin  of  Conipanilii-v  Zoology,  Vol.  135,  No.  6 


-? 


b 

_1 


CD 
I 


CL 


D      ^ 


3 

n 

r 

Q. 

(D 

H 

(1 

>v 

l/l 

n 

a) 

> 

n 

ID 

E 

-D 

CT 

o 

o 

^ 

o 

C 

o 

E 

0) 
D 

g  5 

-o  g 

E  o 

o  . 


o 


o 


CD     O 

u:  z 


Ameiva    taeniura   in  Hispaniola  •  Schwartz        371 


what  details  are  discernible)  specimens  of 
A.  t.  rcgnatrix.  Since  there  are  no  lined 
populations  known  from  north  of  the  Cul 
de  Sac-Valle  de  Neiba  plain,  and  since  Bos- 
well  is  known  to  have  collected  in  the  Les 
Cayes  region,  it  seems  likely  that  this 
Ameiva  originated  in  that  region  and  not  in 
the  Artibonite  Valley.  The  closest  record 
of  A.  taeniura  to  the  Artibonite  Valley  itself 
is  that  of  A.  t.  harhouri,  a  very  distinctive 
subspecies.  It  is  possible  that  there  is  a 
population  of  A.  taeniura  in  the  Artibonite 
Valley,  but  I  consider  it  unlikely  that  this 
specimen  originated  there. 

DISCUSSION 

Ameiva  taeniura,  along  with  Ameiva  lin- 
eolata  and  Ameiva  ehrysolaema,  completes 
the  roster  of  Hispaniolan  teiids.  The  latter 
two  species  have  been  recently  discussed 
(Schwartz,  1966a,  Schwartz  and  Klini- 
kowski,  1966)  and  the  conclusion  reached 
that  both  are  north  island  ( sensu  Williams ) 
species.  Both  lineolata  and  ehrysolaema  are 
confirmed  inhabitants  of  xeric  regions,  the 
former  somewhat  more  so  than  the  latter. 
A.  lineolata  occurs  on  the  south  island  only 
on  the  Peninsula  de  Barahona  (and  Isla 
Beata),  and  A.  ehrysolaema  occurs  only  east 
of  a  line  drawn  between  Leogane  and  Sal- 
trou  in  Haiti  (except  for  an  isolated  record 
at  Aquin).  Thus  neither  lineolata  nor  ehrys- 
olaema have  extensive  south  island  distri- 
butions. Reasons  for  suggesting  that  these 
two  species  are  immigrants  onto  the  south 
island  have  been  discussed  in  the  two 
papers  mentioned  above. 

A.  taeniura,  on  the  other  hand,  occurs  (in 
a  general  fashion)  throughout  much  of  the 
south  island,  and  on  Ile-a-Vache  and  the 
Cayemites.  The  peninsular  races  {taeniura, 
regnatrix,  variea,  and  aequorea)  share  in  a 
community  of  characters  which  include  or- 
ange throats  and  lined  dorsa.  The  eastern 
south  island  race  vulcanalis,  as  well  as  all 
the  north  island  races,  have  dorsa  which 
show  a  dorsal  zone.  The  races  on  Gonave 
and  Saona  have  this  type  of  pattern  also. 


It  seems  likely  that  A.  taeniura  is  the  south 
island  Ameiva  and  that  it  was,  prior  to  the 
invasion  of  A.  ehrysolaema  and  A.  lineo- 
lata, the  dominant  and  only  ground  lizard 
on  this  southeni  land  mass. 

Of  the  north  island  Dominican  races 
(azuae,  tofacea,  vafra,  ignohilis,  and  al- 
gida),  two  have  orange  throats  (tofacea, 
vafra)  and  the  balance  have  black  throats. 
The  orange-throated  races  are,  I  believe, 
derivatives  of  the  eastern  south  island  vul- 
canalis, which  has  successfully  been  able  to 
cross  the  eastern  (mesic)  end  of  the  arid 
Valle  de  Neiba  (vulcanalis  occurs  today  on 
the  north  island  at  Punta  Martin  Garcia) 
and  from  this  region  has  expanded  to  the 
east  along  the  southern  Dominican  littoral 
as  far  as  Cabo  Engano  (and  has  reached 
Isla  Saona  as  well ) .  The  current  absence  of 
records  for  A.  taeniura  between  Punta  Mar- 
tin Garcia  and  Santo  Domingo  may  reflect 
only  that  the  proper  microhabitats  in  the 
intervening  region  have  not  been  sampled; 
the  fortuitous  taking  of  A.  t.  azuae  in  the 
Llanos  de  Azua  in  a  particularly  favorable 
niche  in  an  othenvise  inhospitable  (for  A. 
taeniura)  environment  shows  how  isolated 
populations  of  this  species  might  easily  be 
overlooked.  On  the  other  hand,  the  gap 
between  vulcanalis  and  tofacea  may  be 
real;  since  much  of  the  intermediate  area  is 
today  the  hot  and  dry  Llanos  de  Azua,  it  is 
possible  that  there  may  not  be  populations 
of  A.  taeniura  throughout  the  entire  region. 

The  origin  of  the  northern  black-throated 
races  (algida,  ignohilis)  is  difficult  to  de- 
termine. The  situation  in  northern  Haiti 
is  presently  completely  unknown.  As  far 
as  we  now  know,  tofacea  ( orange-throated ) 
and  ignohilis  (black-throated)  approach 
one  another  most  closely  in  the  vicinity  of 
the  Bahia  de  Samana  ( for  the  moment  I  am 
disregarding  the  approximation  of  black- 
throated  azuae  and  orange-throated  vulcan- 
alis to  the  south).  The  possibility  suggests 
itself  that  the  black-throated  forms  repre- 
sent a  long  isolated  off-shoot  from  the  south 
island  stock  which  has  become  restricted  to 
the  more  northern  and  (generally)  interior 


372         Bulletin  Miis(  uiii  of  Cojiiixinitive  Zoology.  Vol.  135,  No.  6 


regions  of  the  north  island.  The  scattered 
natiu-e  of  the  records  for  the  l:)lack-throated 
subspecies  suggests  as  well  that  these  pop- 
ulations are  in  the  process  of  becoming  re- 
stricted in  distribution,  and  what  we  see 
today  are  mere  remnants  of  a  formerly 
much  more  widespread  range.  In  confirma- 
tion of  this  supposition  is  the  finding  of 
fossil  A.  taeniura  (Etheridge,  1965:99)  at 
Pedro  Santana,  San  Rafael  Province,  Re- 
piiblica  Dominicana,  in  an  area  where  to- 
da)'  the  species  is  not  known  to  occur.  If 
we  consider  the  black-throated  races  as 
being  an  old  north  island  element,  then 
azuac  must  be  included,  despite  its  prox- 
imity to  orange-throated  vulcanalis.  The 
precise  geographical  relationships  between 
the  races  vulcanalis,  azuac,  and  tofacca  in 
the  Llanos  de  Azua  and  along  the  southern 
Dominican  coast  would  be  of  extreme  in- 
terest in  clarifying  the  patterns  of  distribu- 
tion of  orange-  and  black-throated  races  in 
this  area,  but  material  is  presently  not  avail- 
able. 

To  sum  up  the  above  interpretations,  I 
visualize  A.  taeniura  as  the  south  island 
Hispaniolan  Ameiva;  at  some  distant  time,  a 
stock  of  A.  taeniura  invaded  the  north  island 
( either  when  the  interisland  strait  was  tem- 
porarily closed,  or  across  the  water  gap ) 
and  evolved  into  the  black-throated  fomn  of 
which  ii!,no])iJi,s\  azuac,  and  al'^ida  are  now 
remnants.  Secondly,  vulcanali.s  from  the 
south  island  later  invaded  the  southern 
shore  of  the  north  island,  and  has  since 
spread  to  the  east  and  onto  Isla  Saona,  and 
has  developed  two  subspecies  in  the  eastern 
portion  of  its  range. 

I  have  made  no  mention  of  A.  /.  harhouri 
in  the  above  discussion.  Its  occurrence  on 
the  Hispaniolan  mainland  and  on  He  de  la 
Gonave  suggest  that,  rather  than  having 
evolved  on  Gonave,  this  race  has  invaded 
Gonave  from  the  mainland.  The  Gonave 
faima  includes  such  elements  as  Diplo- 
^j,lossus  curtissi,  Anolis  ])rcviro.stri.s-  and  Dro- 
micufi  parvifrons  alleni;  of  these  three 
forms,  the  galliwasp  and  the  anole  occur 
along  the  northern  shore  of  the  Golfe  de  la 


Gonave  and  in  the  Haitian  Cul  de  Sac  Plain, 
and    the    snake    has    been    shown    to    be 
strangely  like  some  specimens  of  D.  p.  pro- 
tenus  from  the  Cul  de  Sac   (Thomas  and 
Schwartz,  1965 ) .   The  Cul  de  Sacian  affini- 
ties with  Gonave,  and  additionally  with  the 
adjacent  mainland  coast  to  the  northwest, 
are  rather  striking.   Tliere  would  thus  seem 
the  possibility  that  harhouri  represents  a  de- 
rivative from  the   (proto)   vulcanaJis  stock 
which  early  crossed  the  Cul  de  Sac  strait, 
and  developed  along  the  southern  littoral  of 
the  north  island.    With  the  closure  of  the 
strait,  the  resulting  arid  plain  was  too  xeric 
for  harhouri  (and  this  plain  likely  was  rap- 
idly colonized  by  A.  Uncolata  and  A.  chrijso- 
lacma,  as  well  as  by  Lcioccphahis  semilin- 
catu.s  and   L.   schrcihersi — four  species   of 
ground  dwelling  lizards  which  bracket  in 
size  the  intermediately-sized  harhouri),  and 
the  race  has  become  increasingly  restricted 
in   distribution  to   the  shore  of  the  Golfe 
de  la  Gonave.    At  some  time,  harhouri  has 
reached  Gonave,  as  have  the  other  species 
noted  above.  Such  a  proposed  history  would 
be  confirmed  if  harhouri  were  to  be  taken 
along  the  northern  side  of  the  Cul  de  Sac- 
Valle  de  Neiba  plain  in  the  foothills  of  the 
Montagues  du  Trou  d'Eau  or  the  Sierra  de 
Neiba;  these  particular  areas  have  not  been 
well  collected.   Casual  observation  of  much 
of  these  foothill  areas  indicates  that,  com- 
pared with  the  ecological  situations  where 
Ixirhouri  was  taken  on  Gonave  and  at  Trou 
Forban,  they  might  well  be  very  suitable 
for  this   subspecies. 

As  has  been  stated  previously,  A.  chrys- 
olaema  and  A.  Uncolata  are  both  inhabitants 
of  xeric  environments,  whereas  A.  taeniura 
prefers  cool  and  shady  habitats.  Inspection 
of  the  map  ( Fig.  2 )  showing  the  known  dis- 
tributions of  the  three  species  on  Hispaniola 
and  its  satellite  islands,  shows  that  the 
ranges  of  A.  chn/.solaema  and  A.  Uncolata 
correspond  very  closely.  Only  in  occasional 
areas  does  Uncolata  occur  without  chrijso- 
laema.  A.  taeniura  overlaps  A.  Uncolata  in 
three  major  areas:  the  Peninsula  de  Bara- 
hona,  Trou  Forban,  and  the  Llanos  de  Azua. 


Ameiva    taeniura    in  Hispamola  •  Scliwartz        373 


In  the  latter  two  areas,  the  known  distribu- 
tion of  A.  taeniura  is  confined  to  but  a  single 
loeality.  A.  taeniura  and  A.  chrijsolaema  are 
somewhat  more  widely  sympatric;  known 
areas  include  the  northeastern  shore  of  the 
Tiburon  Peninsula,  portions  of  the  Penin- 
sula de  Barahona  and  the  southern  shore 
near  Saltrou,  the  southeastern  coast  near 
Santo  Domingo,  the  extreme  eastern  end  of 
the  island,  the  eastern  end  of  the  Valle  de 
Cibao,  and  the  islands  of  Gonave  and 
Saona.  The  three  species  are  sympatric  in 
only  four  areas:  Trou  Forban,  Peninsula  de 
Barahona,  Llanos  de  Azua,  and  Punta  Mar- 
tin Garcia.  In  three  of  these  regions  of  triple 
overlap,  A.  taeniura  is  distinctly  the  less 
common  of  the  three  species,  and  is  re- 
stricted to  the  more  mesic  microsituations 
within  the  widespread  arid  macrosituation. 
In  the  fourth  region  (Peninsula  de  Bara- 
hona), the  same  ecological  arrangement  of 
species  occurs,  but  all  three  are  widespread 


throughout  the  Peninsula,  with  Ii)u'ohita 
having  the  most  restricted  distribution.  The 
more  stringent  ecological  requirements  of 
lineolata  have  doubtless  brought  this  about; 
the  eastern  coast  of  the  Peninsula  is  more 
mesic  and  unsuitable  for  lineolata. 

There  are  still  wide  areas  in  Hispaniola 
where  Ameiva  is  unknown.  Much  of  central 
Haiti  is  still  terra  incognita  as  far  as  the 
genus  is  concerned.  At  least  A.  taeniura 
(and  possibly  A.  chrijsolaema)  should  have 
wider  distributions  in  this  section.  The 
same  statement  may  be  made  concerning 
the  extreme  eastern  and  central  Republica 
Dominicana.  The  status  of  A.  taeniura  along 
the  north  Dominican  coast  in  the  region  of 
Puerto  Plata  and  the  interrelationships  of 
the  races  ignohilis,  tofacea,  and  vafra,  on 
one  hand,  and  of  vulcanulis,  azuae  and  to- 
facea, on  the  other,  all  require  additional 
study. 


Key  to  the  subspecies  of  Hispaniolan  Ameiva^ 


1.  Size  small  (to  59  mm  snout- vent  length); 
8  (occasionally  10)  transverse  and  26-33 
longitudinal  rows  of  ventrals;  14-21  scales 
in  fifteenth  caudal  verticil;  caudal  scales 
smooth  and  oblique;  dorsal  pattern  a  se- 
ries   of   lioldly   contrasting    narrow    black 

and  white  lines   (A.  lineolata)   2 

Size  larger;  8-12  transverse  and  28-41 
longitudinal  rows  of  ventrals;  18-52  scales 

in  fifteenth  caudal  \'erticil;  caudal  scales 
keeled,  and  straight  or  oblique;  dorsal 
pattern  never  as  described  above  7 

2.  Usually  9  dorsal  black  lines  at  midbody  ..       3 
Usually    10   or   11    dorsal   black   lines   at 
midbody     5 

3.  Snout  and  top  of  head  black  4 

Snout  and  top  of  head  pale  A.  I.  semota 

4.  Modal  black  stripe  formula  7-9-7;  throat 
creamy,    not    pale   blue    like    balance    of 

venter  A.  I.  privigna 

Modal  black  stripe  fonnula  7-9-8;  throat 
pale  blue  A.  I.  heatensis 

''■  This  key  depends  in  usefulness  on  having 
freshly  taken  specimens  which  still  retain  their 
original  colors  and  patterns.  Attempts  to  deter- 
mine old  badly  faded  or  discolored  specimens  will 
meet  with  limited  success,  except  in  cases  where 
scale  counts  or  gross  pattern  are  definitive. 


5.  Usually  1 1  dorsal  black  lines  at  midbody; 

snout  clear  pale  sandy  A.  /.  pcrplicata 

Usually  10  dorsal  black  lines  at  midbody       6 

6.  Usually  7  black  lines  at  level  of  sacrum; 
size  smaller  ( to  55  mm  snout-vent  length ) ; 
median    black   line    broken   on    head    or 

neck  A.  I.  meravula 

Usually  8  black  lines  at  level  of  sacrum; 
size  larger  ( to  59  mm  snout-vent  length ) ; 

median  black  line  entire  or  broken  

A.   /.   lineolata 

7.  Size  large  (to  160  mm  snout-vent  length); 
10-12  transverse  and  33—41  longitudinal 
rows  of  ventrals;  30-52  scales  in  fifteenth 
caudal  verticil;  caudal  scales  keeled  and 

straight   (A.  chrijsolaema)  8 

Size  moderate  (to  103  mm  snout-vent 
length);  8-10  (usually  10)  transverse 
and  28-35  longitudinal  rows  of  ventrals; 
18-31  scales  in  fifteenth  caudal  verticil; 
caudal  scales  keeled  and  oblique  (A. 
taeniura )    24 

8.  Ventrals  modally  in  10  transverse  rows  ____       9 
Ventrals  modally  in  12  transverse  rows  ___.      19 

9.  Dorsmn  patternless  10 

Dorsum  with  jjattern  12 

10.  Size  large  (to  137  mm  snout-vent  length); 
dorsum  gray-green  with  indistinct  gray- 
brown  mottling  in  lateral  field   area;   no 


374        Bulletin  Mu.scitni  of  Comparative  Zoology,  Vol.  135,  No.  6 


black  gular  band  A.  c.  lichunlttHnna.si 

Size  smaller  ( to  126  mm  snoiit-\  cut 
length ) ;  dorsum  rusty  brown,  yellowish 
tan,  grayish  brown  or  olive;  lateral  fields 
absent  or  only  vaguely  indicated;  l^lack 
gular  band  present  11 

11.  Size  small  (to  111  mm  snout-vent  length); 
venter  deep  orange-red;  lateral  fields  ab- 
sent    A.  c.  lehcri 

Size  larger  (to  126  mm  snout-vent 
length ) ;  venter  some  shade  of  blue  to 
orange-gray;  lateral  fields  absent  or  only 
indicated  by  a  gray  lateral  stripe  A.  c.  hoekeri 

12.  Dorsal    pattern    consisting   of    pale    blue 
spots  on  a  tan  to  brown  ground  .—  A.  c.  ficta 
Dorsal  pattern  not  consisting  of  spots  _—     13 

13.  Dorsum  gray-green,  hea\'ily  mottled  with 
black;    sides    with    black    tigroid    lateral 

markings  A.  c.  richardthomasi 

Dorsum   lined   14 

14.  Dorsal  pattern  of  5  pale  lines  on  a  very 
dark  brown  ground;  pattern  often  highly 
modified  to  give  complex  longitudinal 
dorsal  figures;   sides  with  vertical  tigroid 

markings  A.  c.  uoodi 

Dorsal  pattern  of  5  or  more  lines,  each 
line  often  consisting  of  longitudinal  series 

of  pale  dots;  sides  without  tigroid  mark- 


15. 


16. 


17. 


IS. 

19. 

20. 


21. 


mgs 


15 


Dorsum  tan  to  brown,  with  6  or  7  dull 
pale  lines;  lateral  fields  brown  and  often 
without  included  pale  dots;  no  black  gular 

band   A.    c.    dcjemor 

Dorsum  seldom  brown,  with  .5  to  10  lines; 
lateral  fields  ])lack;  black  gular  band  pres- 
ent     __.     16 

5-7  dorsal  lines  entire;  dorsum  brown  . 

A.  c.  alacris 

6-10  dorsal  lines  fragmented  17 

Dorsum    reddish    brown;    venter   blue   to 

solid  black  A.  c.  proca.x 

Dorsum  gray,  olive,  to  black;  ventral  col- 
oration varying  from  gray  to  grayish  or- 
ange,   never   black    18 

Lateral  fields  prominent,  outlined  by  pale 

longitudinal    lines    A.    c.    hoekeri 

Lateral    fields    obscure,    not    outlined    by 

pale  longitudinal  lines  A.  c.  mnhratilis 

Dorsum  spotted  or  reticulate  20 

Dorsum   lined   21 

Dorsum  with  discrete  sky-blue  spots  on  a 

black   ground   A.    c.   (d)hotti 

Dorsuiu  with  discrete  or  confluent  ( yield- 
ing  a    rctieuhun  )    yellow   spots    on    a    tan 

to  blackish  Inown  ground  A.  c.  parvoris 

Dorsum  with  5  wide  black  and  confused 
longitudinal  lines  on  a  tannish  gray  to 
dark  brown  groimd;  lateral  tigroid  mark- 
ings present  and  joined  to  the  dorsal  pat- 
tern;  lateral  fields  absent  A.   c.   iacta 


Dorsum  with  5  to  7  pale  lines;  lateral 
fields  present,  no  tigroid  markings  22 

22.  Aspect  faded;  6-7  dull  huffy  lines  on  red- 
dish brown  ground;  lateral  fields  grayish 
brown;  usually  with  black  gular  band  re- 
duced or  absent A.  c.  secessa 

Aspect  not  faded;  lateral  fields  black; 
black  gular  band  present  or  absent  23 

23.  Size  large  (to  160  mm  snout-vent  length); 
dorsal  pattern  a  series  of  6  lemon  yellow 
lines  and/or  linear  series  of  dots  on  a 
dark  brown  to  reddish  brown  ground; 
black  gular  band  present  —-  A.  c.  chnjsolaema 
Size  smaller  (to  132  mm  snout-vent 
length);  dorsal  pattern  a  series  of  5-7  pale 
yellow  lines  ( at  times  modified  into  a 
clear  tan  middorsal  zone)  on  a  tan  to 
brown  ground;  black  gular  band  absent, 

or  present  but  not  e.xtensive  ___.  A.  c.  regidari.s 

24.  Dorsum  with  a  series  of  longitudinal  lines 
and/or  a  middorsal  zone  accompanied  by 

longitudinal   lines;   throat  orange   25 

Dorsum  without  a  series  of  longitudinal 
lines  but  with  a  middorsal  zone;  throat 
orange  or  black  (  including  gray)  29 

25.  Usually  10  transverse  rows  of  ventrals  _—     26 

Eight  transverse  rows  of  ventrals  

A.  t.  regnatrix 

26.  Total  number  of  fourth  toe  scales  85  _.  . 

A.  t.  navassae 

Total  number  of  fourth  toe  scales  83  or 
less   27 

27.  Dorsum  with  5  longitudinal  lines;  lateral 

fields  without  dots  _ A.  t.  aequorea 

Dorsum  with  a  dorsal  zone  and  associated 
longitudinal  lines  28 

28.  Dorsum  with  a  middorsal  zone  and  one 
or  two  pairs  of  dorsolateral  lines;  lateral 
fields  with  only  a  few  scattered  pale  dots, 

often   only   posteriorly   A.  t.   taeniura 

Dorsum  with  a  middorsal  zone  and  one 
pair  of  dorsolateral  lines;  lateral  fields 
with  prominent  and  scattered  pale   ( red 

to   buffy)    dots   A.   t.   varico 

29.  Throat  orange  30 

Throat  (or  throat  and  chest)  black  (in- 
cluding  gray)    34 

30.  Dorsolateral  lines  completely  absent;  lat- 
eral fields  immediately  adjacent  to  dorsal 

zone;  tail  unlined  A.  t.   harhouri 

Dorsolateral  lines  present  or  at  least  in- 
dicated         31 

31.  'I'luoat  pale  orange;  lateral  fields  sparsely 

flecked  with  rusty  A.  t.  tofacea 

Tliroat  orange,  but  not  pale  32 

32.  Throat    vivid    fire    orange;    lateral    fields 
heavily  flecked  with  orange  .—  A.  t.  iiilraiudi.s 
Throat     orange;     lateral     fields     witliout 
flecks    33 

33.  Dorsolateral   lines   distinct,  yellow-green; 


Ameiva    taeniura    i.\  Hispaniola  •  Schwartz        375 


dorsal  zone  gret-nish  gray   _.   A.  t.  rosamondae 
Dorsolateral  lines  indistinct,  yellow;  dor- 
sal zone  yellowish  tan  A.  t.  vafra 

34.    Only   throat   black    35 

Throat  and   chest   black;   dorsum   in    ^ 's 
heavily  dotted  with  yellow,  in  9  's  marbled 


with  dark  brown  A.  t.  algida 

35.    Size  small  (to  70  mm  snout-vent  length); 

dorsolateral  lines  lemon  yellow  _„.  A.  t.  azuae 
Size  large  (to  103  mm  snout-vent  length); 
dorsolateral  lines  green  or  yellow-green  __ 
y\,   f^   ianohilifi 


LITERATURE  CITED 

Bakbour,  Thomas  and  Arthur  Loveridge.  1929. 
Typical  reptiles  and  amphibians.  Bull.  Mus. 
Comp.   Zool.,   69    (10):     206-360. 

Barbour,  Thomas  axd  G.  K.  Noble.  1915.  A 
revision  of  the  lizards  of  the  genus  Ameiva. 
Bull.  Mus.  Comp.  Zool,  59  (6):  417-479. 

Cochrax,  Doris  M.  1928.  The  herpetological 
collections  made  in  Haiti  and  its  adjoining 
islands  by  Walter  J.  Eyerdam.  Proc.  Biol. 
Soc.  Washington,  41:  53-59. 

.      1934.     Herpetological    collections    made 

in  Hispaniola  by  the  Ufowana  expedition, 
1934.  Occ.  Papers  Boston  Soc.  Nat.  Hist., 
8:   163-188. 

.     1941.     The    herpetologv    of    Hispaniola. 

Bull.  U.S.  Nat.  Mus.,  177:  i-vii,  1-398,  120 
figs.,  12  pis. 

Etheridge,  Richard.  1965.  Fossil  lizards  from 
the  Dominican  Republic.  Quart.  Jour.  Flor- 
ida Acad.  Sci.,  28  ( 1 ) :  83-105,  3  figs. 

Maerz,  a.,  and  M.  Rea  Paul.  1950.  A  diction- 
ary of  color.  New  York,  McGraw-Hill  Book 
Co.,  pp.  i-vii,  1-23,  137-208,  56  pis. 

Mertexs,     Robert.      1939.     Herpetologische     Er- 
gebnisse  einer  Reise  nach  der  Insel  Hispaniola, 
Westindien.      Abh.  Senckenberg.  Naturf.  Ges. 
449:    1-84,   10  pis. 

Schmidt,  Karl  P.  1919.  Descriptions  of  new 
amphibians  and  reptiles  from  Santo  Domingo 
and  Navassa.  Bull.  Amer.  Mus.  Nat.  Hist., 
41  (12):  519-525. 

.      1921a.     Notes    on    the    herpetology    of 


Santo  Domingo.     Bull.  Amer.  Mus.  Nat.  Hist. 
44  ( 2 ) :  7-20,  12  figs. 

.  1921b.  The  herpetology  of  Navassa  Is- 
land. Bull.  Amer.  Mus.  Nat.  Hist.,  44  (18): 
555-559,  2  pis.,  5  figs. 

Schwartz,  Albert.  1966a.  The  Ameiva  ( Rep- 
tilia,  Teiidae)  of  Hispaniola.  I.  Ameiva  line- 
olata  Dumeril  and  Bibron.  Carib.  Jour.  Sci., 
5  (1,  2):  45-57,  4  figs. 

.      1966b.     The     Leioeephalus     ( Lacertiha, 

Iguanidae)    of    Hispaniola.     I.     Leioeephalus 
melanoehloms  Cope.     Jour.  Ohio  Herp.  Soc 
5  (2):    ,39-48,  1  fig. 

Schwartz,  A.  axd  Roxald  F.  Klixikowski. 
1966.  The  Ameiva  (Lacertiha,  Teiidae)  of 
Hispaniola.  II.  Geographic  variation  in  Ameiva 
ehnjsolaema  Cope.  Bull.  Mus.  Comp.  Zool, 
133   (10):  425-487,   11  figs. 

Thomas,  Richard.  1966.  A  reassessment  of  the 
herpetofauna  of  Navassa  Island.  Jour.  Ohio 
Herp.  Soc,  5  ( 3 ) :  73-89,  5  figs. 

Thomas,  R.  axd  Albert  Schwartz.  1965.  His- 
paniolan  snakes  of  the  genus  Diomicus  ( Colu- 
bridae).  Rev.  Trop.  Biol.,  13  (1):  59-83, 
10  figs. 

Wetmohe,  Ale.xaxder,  axd  Bradshaw  H.  Swales. 
1931.  The  birds  of  Haiti  and  the  Dominican 
Republic.  Bull.  U.S.  Nat.  Mus.,  155:  i-iv, 
1-483.  26  pis.,  2  figs. 

Williams,  Erxest  E.  1961.  Notes  on  Hispani- 
olan  herpetology.  3.  The  evolution  and  rela- 
tionships of  the  Anolis  semilineatus  group. 
Breviora,  Mus.  Comp.  Zool.,  No.  136:  1-8, 
1   map. 


-mtmm^^mm^ 


in     O  F     T  H  E 


seum   o 


oology 


iigjJiSJiSMiMmmmdimmM 


New    Cyclopoid    Copepods    Associated 
with    Polychaete    Annelids    in    Madagascar 


ARTHUR  G.  HUMES  and  JU-SHEY  HO 


HARVARD   UNIVERSITY  VOLUME    135,   NUMBER  7 

CAMBRIDGE,  MASSACHUSETTS,   U.S.A.  APRIL  24,    1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

Memoris  1864-1938 

JoHNSONiA,  Department  of  Mollusks,  1941- 

OccASioNAL  Papers  on  Mollusks,  1945- 


Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).    $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.   $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.    ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©  The   President   and    Fellows   of   Horvord   College    1967. 


NEW  CYCLOPOID  COPEPODS  ASSOCIATED  WITH  POLYCHAETE 
ANNELIDS  IN  MADAGASCAR 


ARTHUR  G.   HUMES'  -  AND  JU-SHEY   HO' 


INTRODUCTION 

At  Nosy  Be,  in  northwestern  Madagascar, 
copepods  are  known  to  be  associated  ^^'ith 
many  different  marine  invertebrates,  but 
as  yet  none  has  been  described  from  poly- 
chaete  annehds.  This  paper  deals  with  six 
new  cyclopoid  copepods  collected  from 
polychaetes  at  Nosy  Be  in  1960  and  1963- 
64. 

All  collections  were  made  by  A.  G. 
Humes,  those  in  1960  during  an  expedition 
of  the  Academy  of  Natural  Sciences  of 
Philadelphia,  and  those  in  1963-64  as  part 
of  the  U.S.  Program  in  Biology  of  the  Inter- 
national Indian  Ocean  Expedition.  Type 
material  has  been  deposited  in  the  large 
copepod  collection  of  the  United  States 
National  Museum.  Other  specimens  of  the 
new  species  (with  the  exception  of  Naso- 
molgus  leptus)  have  been  placed  in  the 
Museum  of  Comparative  Zoology. 

The  study  of  the  specimens  has  been 
aided  by  a  grant  (GB-1809)  from  the  Na- 
tional Science  Foundation  of  the  United 
States. 

All  figures  were  dra\Mi  with  the  aid  of 
a  camera  lucida.  The  letter  after  the  ex- 
planation of  each  figure  refers  to  the  scale 
at  which  it  was  drawn. 

The  abbreviations  used  are:  ai  =  first 
antenna,  ao  =  second  antenna,  md  =  man- 
dible,   p  =  paragnath,    mxi  =  first    maxilla, 

^  Boston  University,  Boston,  Massachusetts 
-  Associate  in  Marine  In\  ertebrates,  Museum  of 
Comparati\e   Zoology 


mx2  =  second  maxilla,  nixpd  =  maxilHped, 
p,  =  leg  1. 

We  \\ish  to  thank  Dr.  Marian  H.  Pet- 
tibone  of  the  United  States  National  Mu- 
seum for  the  identification  of  the  polychaete 
hosts  and  to  acknowledge  with  apprecia- 
tion the  assistance  to  the  field  work  given 
by  the  staff  of  the  Centre  d'Oceanographie 
et  des  Peches  at  Nosy  Be.  We  are  indebted 
to  Dr.  J.  P.  Harding  and  Miss  P.  D.  Loft- 
house  of  the  British  Museum  (  Natural  His- 
tory) who  have  examined  for  us  the  single 
type  specimen  of  Nasomolgus  ciistatiis. 

The  copepods  described  in  this  paper 
comprise  the  following: 

1)  from  Lepidonotus  cristatus   (Grube) 

Cotylomolgiis  Icpidonoti  n.  gen.,  n.  sp. 

2)  from  Sabelhi  fusca  (Grube) 

Pseudanthcssius  ferox  n.  sp. 

3)  from  SabeUastaife  magnifica  (Shaw) 

Nasomolgus  firmus  n.  sp. 
Nasomolgus  leptus  n.  sp. 
Nasomolgus  rudis  n.  sp. 
Nasomolgus  pai  cuius  n.  sp. 

SYSTEMATIC   DESCRIPTION 

CLAUSIDIIDAE  Embleton,  1901 

COTYLOMOLGUS'  r.  gen. 

Body  cyclopoid,  elongated,  moderateh 
widened  and  rather  flattened  in  the  pro- 


^  The  generic  name  is  a  combination  of  KorvXri 
—  a  small  cup,  alluding  to  the  sucker  on  the  second 
antenna,    and    /jlo\;6s  —  a    sack    made    of    leather. 

Bull.  Mus.  Comp.  Zool.,  135(7):  .377-414,  April,  1967        377 


378        Bulletin  Mu.srinii  of  Comparative  Zoology,  Vol.  135,  No.  7 


some.  Segment  of  leg  1  completely  fused 
with  the  head.  Urosome  5-segmented  in 
the  female,  6-segmented  in  the  male.  Cau- 
dal ramus  with  6  setae.  First  antenna  6- 
segmented.  Second  antenna  4-segmented, 
with  a  large  pedunculate  sucker  on  the 
third  segment.  Mandible  with  4  terminal 
elements  ( 1  flattened  recurved  attenuated 
spine  and  3  setae).  Paragnath  a  spinulose 
lobe.  First  maxilla  a  small  lobe  with  4 
elements.  Second  maxilla  probably  2-seg- 
mented,  with  terminally  a  recurved  spine 
and  a  long  spinifonn  element.  Maxilliped 
absent  in  the  female,  Init  well  formed  in 
the  male,  where  it  is  4-segmented  (assum- 
ing that  part  of  the  tenninal  claw  represents 
the  fourth  segment ) . 

Legs  1  and  2  with  3-segmented  rami. 
Leg  3  reduced  to  a  single  free  segment. 
Leg  4  absent  in  both  sexes.  Leg  5  two- 
segmented  in  the  female,  the  first  segment 
with  a  single  seta,  the  second  segment 
amied  with  4  setae,  3  of  them  long  and 
spiniform,  the  other  short  and  slender.  Leg 
5  in  the  male  with  a  single  free  segment. 

Other  features  as  in  the  species  described 
below. 

Living  on  polychaete  annelids. 

Type  and  only  known  species:  CotijJo- 
molgus  lepiclonoti  n.  sp. 

Gender  masculine. 

Cofylomolgus  /ep/c/onof/'  n.  sp. 
Figs.  1-29 

Type  material. — 10  females,  8  males,  and 
3  immature  specimens  from  5  Lepidonotus 
eristatus  (Grube)  under  intertidal  rocks  at 
Antsakoabe,  on  the  northern  shore  of  Nosy 
Be,  Madagascar.  Collected  December  1, 
1963.  Holotype  female,  allotype,  and  11 
paratypes  (6  females  and  5  males)  deposited 
in  the  United  States  National  Museum, 
Washington;  the  remaining  paratypes  in 
the  collection  of  A.  G.  Humes. 

Other  specimen.'^  (all  from  Lepidonotus 
eristatus  collected  intertidally  at  various 
localities  on  Nosy  Be). — 1   female  from   1 

'^  The  specific  name  lepidnuofi  is  (lerived  from 
the  generic  name  of  the  liost. 


host,  Ambatoloaka,  September  2,  1960;  7 
females  and  6  males  from  2  hosts, 
Antsakoabe,  November  1,  1963;  6  females, 
4  males,  and  2  immature  from  2  hosts, 
Navetsy,  November  3,  1963;  2  females  and 
3  males  from  4  hosts,  Antsakoabe,  February 
16,  1964;  4  females  and  4  males  from  2 
hosts,  Antsakoabe,  September  7,  1964;  and 

1  female  and  1  male  from  1  host,  Befifika, 
October  7,  1964. 

Female. — The  body  (Figs.  1  and  2),  with 
a  moderately  broadened  and  somewhat 
flattened  prosome,  has  a  length  (excluding 
the  setae  on  the  caudal  rami)  of  1.41  mm 
(1.29-1.50  mm)  and  a  greatest  width  of 
0.63  mm  (0.59-0.72  mm),  based  on  8  speci- 
mens. The  ratio  of  length  to  width  of  the 
prosome  is  1.35  :  1.  The  segment  of  leg  1 
is  completely  fused  with  the  head.  The 
epimeral  areas  of  the  metasomal  segments 
are  rounded  posteriorly.  These  segments 
are  separated  by  faintly  striated  interseg- 
mental membranes. 

The  segment  of  leg  5  (Figs.  3  and  4)  is 
wider  than  long,  101  X  257  fx,  and  bears 
the  fifth  legs  ventrally  on  the  posterolateral 
areas.  Dorsally  the  segment  bears  pos- 
teriorly a  transverse  striated  membrane,  and 
ventrally  between  the  insertions  of  the  legs 
there  are  2  patches  of  spinules  arranged 
in  somewhat  irregular  rows.  The  genital 
segment  is  wider  than  long,  141  X  177  fx. 
in  greatest  dimensions,  with  the  broadened 
anterior  two-thirds  separated  abruptly  from 
the  narrowed  posterior  third  (where  the 
width  is  130  /x).  The  posterior  margin  of 
the  segment  bears  a  striated  membrane 
dorsally  and  ventrally  and  has  delicate 
spinules  laterally.  The  areas  of  attachment 
of  the  egg  sacs  are  situated  dorsolaterally 
in  the  middle  of  the  segment.  Each  area 
(Fig.  5)  bears  2  small  setae  about  7  /x  in 
length  and  a  small  setiform  projection. 
There  are  3  postgenital  segments,  the  first 

2  bearing  a  posterior  membrane  dorsally 
and  ventrally  and  lateral  spinules  as  on  the 
genital  segment  (though  the  membrane  on 
the  second  postgenital  segment  is  incom- 
plete  midventrally    and   is   indented   mid- 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ilo        379 


dorsally).    The  first  postgenital  segment  is  hairs);  the  second  bears  11  proximal  setae 

75  X  120  ^,  the  second  49  X  114  /.,  and  the  and  3  anterodistal  setae;  the  third  6  medio- 

third  60  fM  (greatest  length)   X  101  ^.    The  anterior  setae,  a  single  mediopostcrior  seta, 

anal  segment,   on   which  the   caudal  rami  and  2  distal  setae;  the  fourth  2  setae  and 

are  inserted  dorsally,  bears  ventrally  near  one  aesthete  proximallv  and  2  setae  distally; 

the  base  of  each  ramus   a  patch  of  very  the  fifth  2  setae  and  one  aesthete  distally; 

small  spinules.  and  the  sixth  7  setae  and  one  aesthete.  The 

The  caudal  ramus  (Fig.  6)  is  elongated,  formula  thus  is:    5,  14   (11  +.3),  9   (6  +  1 

78  IX  in  length  along  its  outer  edge   (70  /x  +  2),  4  +  1  aesthete  (2  and  1  aesthete  +  2), 

along  the  inner  edge)   and  33  /x  in  width,  2  +  1  aesthete,  and  7+1  aesthete.    All  the 

or  about  2.4  times  longer  than  wide.    The  setae  are  delicately  annulated  and  naked 

outer  lateral  seta  is  45  /x  long  and  naked,  except  for  the  single  haired  seta  on   seg- 

The   pedicellate    dorsal    seta   is   36   ij.   and  ment  1. 

naked.    The   outermost  terminal   seta    (86         The  second  antenna   (Figs.  10  and  11) 

ix)  and  the  innermost  teiTninal  seta  (66  /x)  is  4-segmented  and  distinctly  flexed,  with 

are  minutely  spinulose.    There  is  a  single  the  3  short  distal  segments  directed  back 

well  developed  long  median  terminal  seta  toward   the   relatively   elongated   proximal 

314  fji  in  length  and  naked.    From  its  outer  segment.    The  first  segment  bears  antero- 

basal  area  there   arises  a  finely  spinulose  distally  a  cluster  of  hairs.    The  second  seg- 

slender  seta  53  ^  in  length.    This  seta  has  ment  bears  a  row  of  spinules.    The  third 

no    apparent    articulation;    presumably    it  segment  bears  a  hyaline  seta,  a  spine  ^^'ith 

represents  the  outer  of  the  2  long  terminal  its  distal  third  formed  like  a  crooked  thumb 

setae    commonly    found    in    poecilostomes  and    spinulose,    and    a    large    pedunculate 

which  here  has  fused  with  the  base  of  the  sucker  56  /x  in  diameter,  its  cup  having  well 

inner  long  seta.    On  the  proximal  half  of  sclerotized    supporting    rays    and    with   its 

the  dorsal  surface  of  the  ramus  there  are  rim  formed  by  a  hyaline  lamella  ornamented 

a  few  small  spinules  and  on  the  distal  ven-  with  minute  hairlike  processes  (spinules?). 

tral  surface  there  is  a  patch  of  very  small  The   fourth   segment   is    longer   and   more 

spinules.  A  small  hair  arises  on  the  proximal  slender  than  the  preceding  2  and  bears  2 

outer  margin  of  the  ramus.  setae   on   a   subtemiinal   expansion    and   4 

The  dorsal  surface  of  the  prosome  and  tenninal  setae,  all  of  them  annulated  and 

the    dorsal    and    ventral    surfaces    of    the  naked. 

urosome  bear  scattered  minute  setules.  The         The    labrum    (Fig.    12),    held    erect    in 

ratio  of  the  length  of  the  prosome  to  that  alcoholized  specimens,  is  linguiform  in  out- 

of  the  urosome  is  2  :  1.  line,  with  a  small  terminal  indentation  and 

The  egg  sacs  are  moderately  elongated  2  lateral  hyaline  lobes. 
(Fig.  7),  593  X  246  /x,  and  contain  numerous  The  mandible  (Fig.  13)  is  a  single  elon- 

mostly   hexagonal   eggs   about   55-60  /x  in  gated  segment  bearing  4  terminal  elements: 

diameter.  a  recurved  attenuated  flattened  spine  with 

The  rostral  area  (Fig.  8)  is  not  well  de-  a  few  short  lateral  spinules,   and  3   setae 

\eloped  and  consists  of  a  small  lobe  lying  with  lateral  spinules.    The  paragnath  (Fig. 

behind  the  prominent  crescentic  ridge  be-  14)  is  a  rounded  lobe  bearing  slender  spi- 

tween  the  bases  of  the  first  antennae.  nules  and  showing  small  circular  markings 

The  first  antenna  (Fig.  9)  is  6-segmented.  in  its  cuticle.    The  first  maxilla   (Fig.   15) 

The  lengths    of   the    segments    (measured  is  a  small  segment  armed  with  4  elements: 

along  their  posterior  margins)  are:    36  (30  1  seta  on  the  anterior  surface,  and  3  ter- 

/x  along  the  anterior  margin),  73,  48,  50,  minal  setae   (2  large,   the  longer  33  /x  in 

42,  and  32  ix  respectively.  The  first  segment  length,  and  1  shorter  and  slender ) .    Near 

liears  5  setae  (the  posteriormost  with  lateral  the  base  of  the  seta  on  the  anterior  surface 


380 


Bulletin  Museum  of  Cuniparativc  Zoology,  Vol.  135,  No.  7 


there  is  a  group  of  surficial  markings.  The 
second  maxilla  ( Figs.  16  and  17 )  probably 
consists  of  2  segments,  though  the  seg- 
mentation is  obscure.  The  proximal  portion 
( first  segment? )  bears  a  large  ventral  patch 
of  small  spinules  arranged  in  irregular  rows. 
The  distal  portion  (second  segment?)  has 
surficial  creases  and  folds,  and  bears  2 
tenninal  elements:  a  greatly  recurved  spine 
having  on  its  concave  surface  2  relatively 
long  hyaline  spinules  followed  by  2  rows 
of  minute  spinules,  and  a  long  attenuated 
spiniform  element  not  clearly  articulated 
\\'ith  the  segment  and  bearing  surficial 
punctations  and  a  minute  subterminal 
process.    The  maxilliped  is  absent. 

The  postoral  area  (Fig.  IS)  shows  a 
number  of  sclerotized  regions,  with  a  trans- 
verse lobed  area  just  posterior  to  the  level 
of  the  second  maxillae  having  a  pair  of 
irregular  somewhat  spherical  sclerotizations. 
These  2  sclerotized  pieces  might  be  con- 
sidered as  remnants  of  maxillipeds,  but 
since  they  are  so  far  removed  (see  Fig. 
8)  from  the  usual  position  of  maxillipeds 
in  other  poecilostomes  they  probably  do 
not  represent  appendages.  Posterior  to  the 
region  shown  in  Figure  IS  there  is  a  bal- 
loonlike expansion  (see  Fig.  2)  bearing  2 
patches  of  small  spinules  arranged  in  ir- 
regular rows   (Fig.  19). 

As  shown  in  a  ventral  view  of  the  cephalo- 
some  (Fig.  8),  the  area  between  the  bases 
of  the  first  antennae  is  raised  to  fonn  a 
crescentic  ridge,  and  bears  groups  of  small 
spinules.  On  either  side,  posterior  to  the 
bases  of  the  antennae,  the  ventral  wall  of 
the  cephalosome  is  raised  to  form  a  spinu- 
lose  ridge  that  extends  nearly  to  the  postero- 
lateral comers  of  the  cephalosome.  The 
rostrum  and  head  appendages  are  thus  svn- 
rounded  by  these  ridges  (except  posteriorly). 
The  ridges  may  aid  (together  with  the  2 
suckers  on  the  second  antennae)  in  ad- 
hesion to  the  host. 

Legs  1  and  2  (Figs.  20  and  21)  have  3- 
segmented  rami.  Leg  3  is  reduced  to  a 
single  segment.  Leg  4  is  absent.  The  spine 
and  setal  formula  is  as  follows  ( the  Arabic 


numerals  indicating  setae,  there  being  no 
spines  on  the  legs ) : 

PI   protopod  0-0;    1-0  exp   1-0;    1-1;  1,1,4 

end  0-1;   0-1;  4 

P2  protopod  0-0;   1-0  exp  1-0;   1-1;  1,4 

end  0-1;   0-1;  4 

P3  reduced,    1-0;       2 

P  4  absent 

Leg  1  (Fig.  20)  shows  a  row  of  spinules 
on  the  outer  distal  area  of  the  coxa,  but 
there  is  no  inner  spine  or  seta  on  this  seg- 
ment. The  intercoxal  plate  is  broad  and 
its  distal  edge  is  ornamented  anteriorly  and 
posteriorly  with  2  groups  of  small  spinules 
arranged  in  irregular  rows.  The  basis  has 
an  outer  haired  seta  and  is  ornamented  on 
its  anterior  surface  with  a  row  of  spinules 
between  the  insertions  of  the  rami  and 
with  3  or  4  rows  of  spinules  medial  to  the 
insertion  of  the  endopod;  on  the  margin  of 
the  inner  expansion  of  the  basis  there  is  a 
row  of  long  hairs.  The  outer  margins  of 
the  segments  of  both  rami  are  well  orna- 
mented with  spinules.  Hairs  occur  along 
the  inner  margin  of  the  first  segment  of  the 
exopod  and  along  the  outer  margins  of  the 
first  and  third  segments  of  the  endopod. 
The  segments  of  the  exopod  bear  slender 
annulated  outer  setae  instead  of  the  spines 
often  seen  in  other  poecilostomes.  The  in- 
ner margin  of  the  third  segment  of  the 
endopod  shows  an  interruption  of  the 
sclerotization. 

Leg  2  ( Fig.  21 )  is  in  general  similar  to 
leg  1,  but  the  several  rows  of  spinules  on 
the  anterior  surface  of  the  basis  medial  to 
the  insertion  of  the  endopod  are  absent, 
the  outer  seta  on  the  basis  is  smaller  and 
naked,  there  is  only  1  small  annulated  seta 
on  the  third  segment  of  the  exopod,  and 
there  is  no  interruption  in  the  sclerotization 
of  the  inner  margin  of  the  third  segment 
of  the  endopod. 

Leg  3  (Fig.  22)  consists  of  only  a  single 
elongated  free  segment,  58  X  22  jjl,  anned 
with  2  very  unequal  tenninal  setae,  the 
inner  94  /x  in  length  and  bearing  rows  of 
very  short  spinules  on  its  distal  two-thirds, 
the  outer  35  /x  long,   slender,   and  naked. 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        381 


Near  the  insertions  of  the  2  setae  there  is 
a  ventral  patch  of  minute  spinules.  Close 
to  the  insertion  of  the  free  segment  there 
is  an  outer  naked  seta  52  /x  long  arising  from 
the  body. 

No  trace  could  be  found  of  leg  4. 

Leg  5  (Fig.  23)  is  2-segmented.  The 
first  segment  is  88  x  73  /x,  and  bears  an 
outer  naked  seta  52  /jl  long.  On  the  dorsal 
sm-face  of  the  segment  there  is  a  diagonal 
line  from  the  base  of  the  seta  to  the  inner 
distal  angle;  on  the  ventral  surface  there 
are  transverse  rows  of  minute  spinules  near 
the  distal  margin.  The  second  segment  is 
117  /i.  along  the  inner  margin  and  68  ji  along 
the  outer  margin  to  the  base  of  the  first 
seta;  its  greatest  diagonal  length  in  135  [x 
and  its  greatest  width  is  70  /x.  The  segment 
is  armed  \\'ith  2  outer  lateral  spinifonn 
setae  (99  and  112  /x  long  respectively)  and 
2  very  unequal  terminal  setae,  1  slender 
and  55  fx,  the  other  spinifonn  and  140  n 
in  length.  All  4  setae  bear  extremely  minute 
lateral  spinules  and  are  delicately  annu- 
lated.  Diagonal  rows  of  minute  spinules 
occur  along  the  distal  dorso-inner  margin 
and  extend  around  on  the  ventral  surface 
to  form  a  patch  near  the  insertion  of  the 
innermost  long  seta.  Another  patch  of 
minute  spinules  arranged  in  diagonal  rows 
may  occur  (not  in  all  specimens)  on  the 
ventral  margin  just  proximal  to  the  first 
outer  seta. 

Leg  6  is  probably  represented  by  the  2 
small  setae  on  the  areas  of  attachment  of 
the  egg  sacs  ( see  Fig.  5 ) . 

The  color  in  life  in  transmitted  light  is 
translucid  to  slightly  opaque,  the  eye  red, 
the  egg  sacs  opaque  gray. 

Male.— The  fomi  of  the  body  (Fig.  24) 
resembles  that  of  the  female.  The  length 
(without  the  setae  on  the  caudal  rami)  is 
1.06  mm  (0.90-1.17  mm)  and  the  greatest 
width  is  0.46  mm  (0.39-0.50  mm),  based 
on  10  specimens.  The  ratio  of  length  to 
width  of  the  prosome  is  1.4  :  1. 

The  segment  of  leg  5  (Figs.  25  and  26), 
measuring  81  x  172  /x,  resembles  tliat  of 
the  female.    The  genital  segment  is  wider 


than  long,  91  X  143  /x.  The  4  postgenital 
segments  are  55  X  107,  47  X  97,  33  X  92, 
and  40  X  86  /x  from  anterior  to  posterior. 

The  caudal  ramus  is  like  that  of  the  fe- 
male. 

The  surfaces  of  the  prosome  and  urosome 
bear  minute  setules  as  in  the  opposite  sex. 
The  ratio  of  the  length  of  the  prosome  to 
that  of  the  urosome  is  about  1.8  :  1. 

The  rostral  area,  first  antenna,  second 
antenna,  labrum,  mandible,  paragnath,  first 
maxilla,  and  second  maxilla  resemble  those 
of  the  female.  The  maxilliped  (Fig.  27) 
has  a  rather  poorly  defined  basal  segment. 
The  large  second  segment  bears  2  hyaline 
naked  setae  on  its  inner  surface  and  a  large 
patch  of  small  blunt  spinules  arranged  in 
longitudinal  rows  on  its  posterior  surface. 
The  third  segment  is  small  and  unanned. 
The  fourth  segment  probably  fomis  part 
of  the  short  and  rather  stout  tenninal  claw, 
which  is  54  /j.  in  length  (measured  along 
its  axis )  and  bears  near  its  base  a  slender 
naked  annulated  seta  and  a  very  small 
setule.  On  the  concave  surface  of  the  claw 
there  is  proximally  a  minute  hyaline  process 
and  more  distally  2  parallel  rows  of  minute 
denticles. 

The  postoral  area  and  the  cephalosome 
resemble  generally  those  areas  in  the  fe- 
male. 

Legs  1,  2,  and  3  are  like  those  in  the 
female.    Leg  4  is  absent. 

Leg  5  ( Fig.  28 )  has  a  single  free  segment 
and  is  held  against  the  sides  of  the  genital 
segment  in  alcholic  specimens  (see  Fig. 
25)  rather  than  diverging  as  in  the  female. 
The  free  segment  is  more  elongated  than  in 
the  female,  59  X  25  fx  in  greatest  dimensions. 
The  4  setae  measure  from  outer  to  inner 
79,  95,  39,  and  127  /x  in  length.  There  is 
a  patch  of  small  spinules  on  the  ventral 
surface  near  the  insertion  of  the  longest 
seta.  The  group  of  spinules  seen  in  the 
female  on  the  ventral  margin  proximal  to 
the  first  outer  seta  is  absent  here.  The 
seta  arising  from  the  body  near  the  free 
segment  is  55  /x  long  and  naked. 

Leg  6   (Fig.   26)    consists   of  a   postero- 


382         Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  7 


lateral  flap  on  the  ventral  surfaci'  ol  the 
genital  segment,  ornamented  with  minute 
spiniiles  but  apparently  lacking  spines  or 
setae. 

The  spennatophore  (Fig.  29),  attached 
to  the  female,  is  elongated,  234  X  68  /x,  in- 
cluding the  neck. 

The  color  in  life  in  transmitted  light 
resembles  that  of  the  female. 

Taxonomic  ))(>sition  of  the  iienu.'i. — The 
new  genus  Cutylouiol^u.s  appears  to  be  re- 
lated to  Myzomoliiii.'i  Bocquet  and  Stock, 
1957  (Clausidiidae),  and  to  Catinia  Bocquet 
and  Stock,  1957  (Catiniidae),  genera  which 
live  on  sipunculid  wonns.  As  in  these  two 
genera,  Cotijlomolfiu.s  has  the  segment  of 
leg  1  fused  with  the  head,  the  urosome  in 
the  female  is  5-segmented  and  in  the  male 
6-segmented,  the  first  antenna  is  6-seg- 
mented,  the  second  antenna  is  4-segmented 
with  a  large  sucker  on  the  third  segment, 
the  first  maxilla  bears  four  elements,  leg  5 
in  the  female  is  2-segmented  and  in  the 
male  has  only  a  single  free  segment,  and 
the  maxilliped  in  the  male  is  4-segmented 
(assuming  that  the  fourth  segment  forms 
part  of  the  claw). 

CotijlomoJiius  differs  from  Mijzomolgus 
chiefly  in  lacking  the  maxilliped  in  the 
female,  in  the  reduction  of  leg  3  and  the 
absence  of  leg  4,  and  in  the  absence  of  an 
inner  spine  on  the  basis  of  leg  1. 

The  new  genus  differs  from  Catinia  in 
having  a  well  developed  mandible  with 
four  tenninal  elements,  in  the  reduction 
of  leg  3  and  the  absence  of  leg  4,  and  in 
the  absence  of  an  inner  spine  on  the  basis 
of  legs  1-3. 

Cotijlomolii^us  differs  from  both  Mijzo- 
mulgus  and  Catinia  in  certain  features  of 
legs  1-4,  principally,  the  absence  of  an 
inner  spine  on  the  basis  of  legs  1  and  2, 
the  reduction  of  the  outer  spines  on  the 
exopods  of  legs  1  and  2  to  simple  setae, 
the  reduction  of  leg  3  to  a  single  free  seg- 
ment, and  the  absence  of  leg  4. 

The  nature  of  the  mandible  in  Cotylomol- 
gf/.y,  with  its  four  tenuinal  elements,  is 
more  like  that  of  Myzomoli^us  than  that  of 


Catinia.  (Gooding,  1963,  in  an  unpublished 
thesis,  after  examination  of  specimens  of 
Catinia  plana  Bocquet  and  Stock,  1957,  has 
shown  that  a  pair  of  small,  weakly  cuticu- 
larized  mandibles,  bent  midway  almost  at 
a  right  angle  and  apparently  without  major 
spines  or  setae,  exists  in  that  species.) 
Since  the  stiTicture  of  the  mandible  in 
poecilostomes  is  characteristic  within  supra- 
specific  groups  and  thus  may  be  regarded 
as  indicative  of  phylogenetic  relationship, 
CotylomoJgus  appears  to  be  closer  to 
Myzo7noJgus  than  to  Catinia. 

Bocquet  and  Stock  ( 1957,  p.  430 )  placed 
their  new  genus  Myzomolgus  in  the  Clau- 
sidiidae largely  on  the  basis  of  the  structure 
of  the  mouthparts,  notably  the  mandible, 
the  antennae,  and  the  well  developed 
thoracic  legs.  While  Cotylomolgu.s  shows  a 
reduction  of  the  legs,  a  feature  which  is 
characteristic  of  many  of  the  genera  in  the 
Clausiidae,  the  fonn  of  its  mandible  is 
clausidiid  rather  than  clausiid.  We  are  led, 
therefore,  to  include  provisionally  the  genus 
Cotylomolgu.s  in  the  Clausidiidae,  recog- 
nizing at  the  same  time  that  there  is  a 
close  relationship  between  the  Clausidiidae 
and  the  Clausiidae  and  that  the  two  fami- 
lies may  actually  represent  a  single  cate- 
gory (see  Wilson  and  Illg,  1955,  p.  137). 

LICHOMOLGIDAE  Kossmann,  1877 
PSEUDANTHESSIUS  Claus,  1889 

Pseudanthessius  ferox'  n.  sp. 
Figs.  30-59 

Type  material. — 14  females,  5  males,  and 
5  immature  specimens  from  3  SahelJa  fusca 
(Grube),  in  1  m,  at  Ambariobe,  a  small 
island  between  Nosy  Be  and  Nosy  Komba, 
Madagascar.  Collected  October  4,  1964. 
llolotype  female,  allotype,  and  14  para- 
types  (11  females  and  3  males)  deposited 
in  the  United  States  National  Museum, 
Washington;  the  remaining  paratypes  (dis- 
sected )  in  the  collection  of  A.  G.  Humes. 


^  The  specific  name  fcrox  (from  Latin  =:  war- 
like, savage)  alludes  to  the  rather  formidable 
appearance   of   the   labrnin. 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho 


383 


OtJicr  specimens  (all  from  SahcJIa  ftisca). 
— 1  female  and  2  males  from  1  host,  in 
6-8  m,  Ambariobe,  December  27,  1963;  4 
females,  10  males,  and  12  immature  speci- 
mens from  1  host,  in  1  m,  west  of  Pte. 
Mahatsinjo,  Nosy  Be,  March  27,  1964;  and 
4  females  and  5  males  from  1  host,  in  2  m, 
Andraikarebe,  Nosy  Komba,  October  9, 
1964  (these  specimens  placed  in  the  Mu- 
seum   of   Comparative   Zoology). 

Female. — The  body  (Fig.  30)  is  rather 
elongated,  with  the  prosome  moderately 
broadened.  The  length  (not  including  the 
setae  on  the  caudal  rami )  is  1.73  mm  ( 1.63- 
1.97  mm)  and  the  greatest  width  is  0.71 
mm  (0.67-0.75  mm),  based  on  10  speci- 
mens. The  ratio  of  length  to  width  of  the 
prosome  is  1.67  :  1.  The  segment  of  leg  1 
is  separated  from  the  head  by  a  dorsal 
furrow.  Near  the  level  of  the  maxillipeds 
on  each  side  of  the  cephalosome  there  is  a 
slight  notch,  perhaps  indicating  the  bound- 
ary of  the  maxillipedal  segment.  The 
epimeral  areas  of  the  metasomal  segments 
are  rather  pointed  posteriorly. 

The  segment  of  leg  5  (Fig.  31)  is  104  /x 
long  and  bears  a  transverse  sclerotized  area 
153  IX  in  width.  (The  exact  width  of  the 
segment  is  difficult  to  detennine,  since  the 
fifth  legs  are  fused  with  it.)  The  genital 
segment  (Figs.  31,  32,  and  33)  is  224  /x  in 
length.  In  dorsal  view  the  segment  is 
widened  in  its  anterior  two-thirds  (180  /x), 
then  narrowed  in  its  posterior  third  ( 130  fi). 
On  the  ventral  surface  of  the  anterior  half 
there  are  2  swollen  areas,  each  about  112  X 
68  ju.  In  lateral  view  these  areas  protrude 
conspicuously.  These  swellings  were  seen 
on  all  females  examined.  Their  size,  shape, 
position,  and  lack  of  a  neck  make  it  un- 
likely that  they  represent  spermatophores. 
The  areas  of  attachment  of  the  egg  sacs 
are  located  dorsolaterally  at  the  level  of 
the  segmental  constriction.  Each  area  in 
dorsal  view  (Fig.  34)  shows  a  pedicellate 
naked  seta  37  /x  long  set  upon  a  sclerotized 
base  and  2  spinelike  elements,  one  22  [x 
long  and  naked,  the  other  14  jx  and  bearing 
a  subterminal  setuliform  process.    In  lateral 


view  ( Fig.  35 )  the  armature  is  more  clearly 
visible.  The  3  postgenital  segments  are 
78  X  112,  62  X  95,  and  90  X  91  yx  from  an- 
terior to  posterior. 

The  caudal  ramus  ( Fig.  36 )  is  elongated, 
109  X  38  fx,  or  about  2.9  times  longer  than 
wide.  The  outer  lateral  seta  is  92  fx  long, 
the  pedicellate  dorsal  seta  31  /x,  the  outer- 
most terminal  seta  81  /x,  the  innennost 
tenninal  seta  138  /x,  and  the  2  median 
terminal  setae  360  /x  (outer)  and  .560  /x 
(inner).  All  the  setae  except  the  2  median 
terminal  ones  are  finely  annulated  and  all 
are  naked  except  the  innennost  terminal  1 
which  bears  an  inner  row  of  hairs  near  its 
base.  The  dorsal  and  ventral  surfaces  of 
the  ramus  bear  scattered  minute  setules. 

The  dorsal  surface  of  the  prosome  and 
the  dorsal  and  ventral  surfaces  of  the  uro- 
some  bear  minute  setules.  The  ratio  of  the 
length  of  the  prosome  to  that  of  the  uro- 
some  is  1.45  :  1. 

The  egg  sacs  are  elongated  (Fig.  30),  in 
one  female  measuring  952  X  190  /x,  though 
there  is  some  variation  in  length  in  differ- 
ent individuals,  and  contain  numerous  eggs, 
each  about  60  ^  in  diameter. 

The  rostral  area  ( Fig.  37)  is  undeveloped 
and  represented  by  a  crescentic  line  be- 
tween the  bases  of  the  first  antennae. 

The  first  antenna  (Fig.  38)  is  7-seg- 
merited,  but  the  third  segment  has  on  its 
ventral  surface  a  small  proximal  sclerotized 
area  suggesting  an  intercalary  segment. 
The  lengths  of  the  segments  (measured 
along  their  posterior  non-setiferous  mar- 
gins )  are:  15  (70  /x  along  the  anterior  mar- 
gin), 75,  33,  51,  47,  33,  and  24  ,x,  respec- 
tiveh'.  The  formula  for  the  annature  is  4, 
13,  6,  3,  4  -f  1  aesthete,  2  +  1  aesthete,  and 
7  +  1  aesthete.  All  the  setae  are  naked  ex- 
cept 1  on  segment  5  and  4  on  segment  7 
which  are  haired.  One  of  the  tenninal 
setae  on  the  last  segment  is  longer  (150  /x) 
and  stronger  than  any  of  the  others. 

The  second  antenna  (Fig.  39)  is  4-seg- 
mented,  with  the  formula  1,  1,  3,  and  5  +  II. 
The  last  segment,  about  55  X  39  fx,  bears 
terminally  2  strong  unequal  claws  72  and 


384        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  7 


36  /x  long,  a  slender  transversely  divided 
seta,  and  a  small  seta  near  the  insertion  of 
the  smaller  claw;  subterminally  there  are 
3  setae,  1  short,  the  other  2  long.  All  the 
setae  arc  naked,  and  those  on  the  last  2 
segments  are  slightly  annulated. 

The  labrum  (Fig.  40)  has  2  medial 
rounded  and  rather  hyaline  lobes,  external 
to  w  hich  there  are  2  large  well  sclerotized 
spikelike  processes  about  50  /x  in  length 
which  extend  ventrally  to  the  mouthparts. 
The  labrum  lacks  fine  ornamentation. 

The  mandible  ( Fig.  41 )  has  a  small  naked 
spinelike  element  on  the  convex  side  at 
the  base  of  the  blade.  The  blade  is  at- 
tenuated distally  and  bears  a  striated  flange 
on  the  convex  side  and  a  row  of  spinules 
on  the  concave  side.  The  paragnath  (Fig. 
40)  is  probably  represented  by  a  small 
unoramented  rather  hyaline  lobe  about  17 
IX  long  located  between  the  base  of  the 
mandible  and  the  outer  corner  of  the 
labrum.  The  first  maxilla  (Fig.  42)  con- 
sists of  a  single  segment  armed  with  4  setae, 
all  of  which  are  indistinctly  articulated. 
The  second  maxilla  (Fig.  43)  is  2-seg- 
mented.  The  first  segment  is  unanned. 
The  second  segment  bears  the  usual  arma- 
ture consisting  of  a  small  hyaline  basal 
outer  seta,  a  larger  seta  on  the  posterodor- 
sal  surface,  and  a  long  subterminal  spinu- 
lose  seta,  and  terminates  in  a  moderately 
short  lash  with  prominent  dentiform  spines 
proximally  becoming  slender  spinules  dis- 
tally. The  maxilliped  (Fig.  44)  has  3 
segments:  an  elongated  unarmed  basal 
segment,  a  rather  swollen  second  segment 
partially  divided  by  surficial  creases  and 
bearing  on  the  inner  surface  proximally  a 
patch  of  spinules  and  distally  2  naked  ele- 
ments ( 1  spiniform  and  1  setifomi ) ,  and  a 
small  slender  terminal  segment  bearing  on 
its  inner  surface  2  naked  elements  ( 1 
spiniform  and  1  setifonn)  and  terminating 
ill  a  spiniform  process  (not  articulated). 

The  postoral  area  (see  Fig.  37)  does  not 
protrude  ventrally.  A  sclerotized  line  con- 
nects the  bases  of  the  maxillipeds. 

Legs  1-4  ( Figs.  45,  47,  48,  and  49)  have 


3-segmented  rami,  with  the  exception  of 
the  endopod  of  leg  4  which  consists  of  a 
single  segment.  The  spine  and  setal 
fomiula  is  as  follows  ( the  Arabic  numerals 
representing  setae,  the  Roman  numerals 
spines ) : 

PI   protopod  0-1;    1-0  exp  I-O;  I-l;   111,1,4 

end  0-1;  0-1;   1,5 

P2  protopod   0-1;    1-0  exp  I-O;  I-l;   111,1,5 

end  0-1;  0-2;   11,1,3 

P3   protopod   0-1;    1-0  exp  I-O;  I-l;   111,1,5 

end  0-1;  0-2;   11,1,2 

P4  protopod  0-1;   1-0  exp  I-O;  I-l;  11,1,5 

end  II 

The  inner  seta  on  the  coxa  of  legs  1-3  is 
long  and  feathered,  but  in  leg  4  this  seta 
is  minute  (12  [x  long)  and  naked.  In  the 
first  3  legs  the  inner  margin  of  the  basis 
bears  a  row  of  hairs,  but  these  hairs  are 
absent  in  leg  4.  In  leg  1  (Fig.  45)  the 
outer  spines  of  the  exopod  have  short  blunt 
spinules  along  one  side.  Between  the  rami 
the  basis  (as  in  legs  2  and  3  also)  forms  a 
shaiply  pointed  process  (Fig.  46).  In  leg 
2  ( Fig.  47 )  the  outer  spines  of  the  exopod 
have  delicate  spinulose  lamellae.  The 
terminal  spine  on  the  last  segment  of  the 
endopod  is  66  /x  long,  naked,  somewhat 
irregular,  and  has  a  minutely  pointed  tip. 
Leg  3  (Fig.  48)  is  similar  to  leg  2,  except 
for  differences  in  the  spine  and  setal 
formula.  Leg  4  (Fig.  49)  has  a  less  promi- 
nent and  more  rounded  process  on  the 
basis  be'tween  the  rami.  The  endopod  is 
1 -segmented,  though  the  restriction  of  the 
inner  lateral  hairs  to  the  proximal  portion 
and  the  interruption  in  the  sclerotization 
suggest  a  division  of  the  segment.  The 
segment  measures  115  X  57  /x  in  greatest 
dimensions.  The  outer  margin  of  the  endo- 
pod is  somewhat  irregular  and  on  its  distal 
half  there  are  3  groups  of  minute  denticles 
borne  on  3  marginal  lobes.  The  2  terminal 
spines  are  44  fx  (inner)  and  29  /x  (outer) 
in  length.  There  is  an  anterior  row  of 
minute  spinules  near  the  insertions  of  these 
spines. 

Leg  5  (Fig.  50)  does  not  have  a  free 
segment  and  bears  terminally  a  naked 
spine  37  //,  and  a  naked  seta  56  fx  in  length. 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho 


385 


and  dorsally  a  seta  55  jx  long  with  lateral 
hairs. 

Leg  6  is  probably  represented  by  the 
armature  on  the  area  of  attachment  of  the 
egg  sacs  (see  Figs.  34  and  35). 

The  color  in  life  in  transmitted  light  is 
slightly  opaque,  the  eye  red,  the  egg  sacs 
gray. 

Male— The  form  of  the  body  (Fig.  51) 
resembles  that  of  the  female.  The  length 
(not  including  the  setae  on  the  caudal  rami) 
is  1.41  mm  (1.28-1.52  mm),  and  the  great- 
est width  is  0.46  mm  (0.42-0.50  mm), 
based  on  10  specimens  (the  allotype,  4 
parat\'pes,  and  5  specimens  from  Andrai- 
karebe  collected  October  9,  1964).  The 
epimeral  areas  of  the  segments  of  legs  2-A 
are  more  rounded  posteriorly  than  in  the 
female.  The  ratio  of  length  to  \\'idth  of  the 
prosome  is  1.66  :  1. 

The  segment  of  leg  5  is  smaller  than  in 
the  female,  being  only  52  X  96  ^,  but  other- 
wise similar.  The  genital  segment  (Fig. 
52)  is  longer  than  wide,  161  X  146  /x,  with 
gently  arcuate  lateral  margins  in  dorsal 
\'iew.  The  4  postgenital  segments  are  77  X 
90,  73  X  77,  50  x  64,  and  64  x  64  /.  from 
anterior  to  posterior. 

The  caudal  ramus  is  like  that  of  the  fe- 
male. 

The  surfaces  of  the  prosome  and  uro- 
some  bear  minute  setules  as  in  the  female. 
The  ratio  of  the  length  of  the  prosome  to 
that  of  the  urosome  is  about  1.15  :  1. 

The  rostral  area  and  first  antenna  are 
like  those  in  the  female.  The  second  an- 
tenna also  resembles  that  of  the  female,  but 
the  slender  transversely  divided  seta  on  the 
last  segment  next  to  the  2  claws  is  longer 
and  more  clawlike  (Fig.  53).  The  labrum, 
mandible,  paragnath,  and  first  maxilla  re- 
semble those  of  the  female.  The  second 
maxilla  (Fig.  54)  is  much  like  that  of  the 
opposite  sex,  but  the  first  segment  appears 
to  be  a  little  more  swollen  and  the  terminal 
lash  is  relatively  shorter.  The  maxilliped 
(Fig.  55)  is  4-segmented  (assuming  that 
the  fourth  segment  forms  part  of  the  claw ) . 
The  first  segment  bears  a  prominent  weakly 


sclerotized  digitiform  process  on  its  distal 
inner  corner.  The  second  segment  bears  on 
its  inner  surface  2  setae  ( 31  and  19  /a  long ) 
and  a  row  of  small  spinules.  The  short 
third  segment  is  unanned.  The  slender 
claw  (Fig.  56),  130  /jl  in  length  (measured 
along  its  axis),  bears  near  its  base  a  poste- 
rior seta  28  fx.  and  an  anterior  seta  11  /x  in 
length.  The  marginal  lamella  along  the 
concave  surface  of  the  claw  shows  a  minute 
interruption  about  midway,  perhaps  repre- 
senting the  distal  boundary  of  the  fourth 
segment.  The  tip  of  the  claw  has  a  very 
narrow  lamella. 

The  postoral  area  resembles  that  of  the 
female. 

Leg  1  (Fig.  57)  shows  several  differ- 
ences from  that  of  the  female,  especially 
in  the  endopod.  The  outer  spines  on  the 
exopod  are  more  slender  and  acutely 
pointed.  The  last  2  segments  of  the  endo- 
pod are  almost  completely  fused,  the  only 
evidence  of  the  former  articulation  between 
them  seen  on  the  anterior  surface  where  a 
line  extends  halfway  across  the  ramus  at 
the  level  of  the  spinous  process.  The  distal 
part  of  the  endopod  is  much  modified, 
terminating  in  a  sclerotized  clawlike  struc- 
ture. There  is  a  single  short  ( 10  ^ )  naked 
outer  spine  and  4  setae  ( 1  naked,  20  jx  long, 
with  a  narrow  lamella,  arising  from  the 
convex  edge;  1  feathered,  32  jn  long,  arising 
on  the  posterior  surface;  and  2  feathered, 
44  and  48  /j.  long,  arising  on  the  inner  side). 
On  the  anterior  surface  between  the  spine 
and  the  first  seta  there  is  a  sclerotized  claw- 
like projection  (for  \\'hich  no  articulation 
could  be  seen ) .  This  projection  may  per- 
haps be  derived  from  the  outermost  seta 
in  the  female.  Two  other  small  projections 
are  located  subterminally  on  the  anterior 
surface.  The  presence  of  the  clavvlike  struc- 
tures and  the  curvature  of  the  distal  part  of 
the  endopod  suggest  a  prehensile  function. 
The  formula  for  the  endopod  is  0-1;  0-1 
.  .  .  1,4. 

Legs  2-5  resemble  those  of  the  female. 

Leg  6  (Fig.  58)  consists  of  a  postero- 
lateral flap  on  the  ventral  side  of  the  genital 


386 


Bulletin  Mtisctntt  of  Coinixirdtivc  Zoology,  Vol.  135,  No.  7 


segment,  bearing  2  naked  setae  27  and  44 
jjL  in  length  and  having  a  surficial  spiniform 
projection  4.5  /x  long. 

The  spermatophore  (Fig.  59),  as  seen 
within  the  bod\-  of  the  male,  is  elongated, 
151  X  62  fjL,  not  inclnding  the  short  neck  of 
10  ^. 

The  color  in  life  in  transmitted  light  re- 
sembles that  of  the  female. 

Comparison  uith  other  species  in  the 
<i,enus. — Pseudcnitlicssius  ferox  may  be 
readily  distinguished  from  13  of  the  22 
species  in  the  genus  listed  by  Stock,  Humes, 
and  Gooding  ( 1963 )  on  the  basis  of  three 
recognition  characters:  the  two  spikelike 
processes  on  the  labrum  in  both  sexes,  the 
process  on  the  first  segment  of  the  male 
maxilliped,  and  the  clawlike  modification 
of  the  endopod  of  leg  1  in  the  male.  These 
13  species,  which  lack  one  or  more  of  the 
three  characters  and  thereby  differ  from 
P.  ferox,  are:  aestheticus  Stock,  Humes, 
and  Gooding,  1963;  assimiUs  G.  O.  Sars, 
1917;  deficicns  Stock,  Humes,  and  Gooding, 
1963;  hfus  Illg,  1950;  liher  (Brady,  1880); 
luculentus  Humes  and  Cressey,  1961;  nm- 
cronatus  Guniey,  1927;  nemcrtophiliis  Gal- 
lien,  1935;  notahilis  Humes  and  Cressey, 
1961;  pectinifer  Stock,  Humes,  and  Good- 
ing, 1963;  sauva^ei  Canu,  1892;  thorelli 
(Brady,  ISSO);  and  tortuostis  Stock,  Humes, 
and  Gooding,  1963.  Fseudanthessius  pro- 
currens  Humes,  1966  from  a  cidarid  echi- 
noid  in  Madagascar  lacks  all  three  of  the 
characters  just  mentioned  and  thus  is  easily 
separated  from  P.  ferox. 

Unfortunately,  in  the  remaining  9  species 
no  information  is  available  regarding  these 
three  recognition  characters,  and  other 
features  must  be  used  to  separate  them 
from  P.  ferox.  Of  these  species,  liber,  sensu 
Sewell,  1949,  has  two  long  elements  on 
the  first  segment  of  the  first  antenna; 
duhius  G.  O.  Sars,  1918,  has  a  4-segmented 
urosome  in  the  female;  and  the  remaining 
7  species  (concinnus  Thompson  and  A. 
Scott,  1903,  'gracilis  Glaus,  1889,  graciloides 
Sewell,  1949,  ohscurus  A.  Scott,  1909,  spi- 
nifer  Lindberg,  1945,  tenuis  Nicholls,  1944, 


and  iceheri  A.  Scott,  1909)  have  slender 
setiform  elements  instead  of  strong  claws 
on  the  last  segment  of  the  second  antenna. 
In  addition  to  these  differences,  other 
features,  such  as  the  length  to  width  ratio 
of  the  caudal  rami,  may  be  useful  in  sepa- 
rating many  of  the  species  from  P.  ferox. 
None  of  the  known  species  of  Pseudan- 
thessitis  seems  to  be  closely  related  to  the 
new  species  from   Madagascar. 

NASOtAOLGUS  Sewell,  1949 

This  genus  was  established  by  Sewell  on 
the  basis  of  a  new  species,  Nasomolgiis 
eristatiis,  of  which  he  fomid  only  a  single 
female  in  debris  at  a  depth  of  38  m  off  the 
South  Arabian  coast.  In  describing  the  four 
new  species  which  follow,  we  have  been  un- 
able to  compare  at  firsthand  this  type  speci- 
men (which  is  in  the  British  Museum).  How- 
ever, Dr.  J.  P.  Harding  and  Miss  P.  D. 
Lofthouse  of  the  Museum  staff  have  ex- 
amined the  pennanent  slide  of  N.  cristatus 
and  have  supplied  us  with  infonnation  on 
several   critical   points. 

Nasomolgus  firmus^  n.  sp. 

Figs.  60-86 

Type  material. — 28  females  and  4  males 
from  one  sabellid  polychaete,  Sabellastarte 
magnifica  (Shaw),  in  2  m,  at  Ambariotelo, 
a  small  island  between  Nosy  Be  and  Nosy 
Komba,  Madagascar.  Collected  May  15, 
1964.  Holotype  female,  allotype,  and  20 
paratypes  (females)  deposited  in  the  United 
States  National  Museum,  Washington;  the 
remaining  paratypes  in  the  collection  of 
A.  G.  Humes. 

Other  specimens  (all  from  Sabellasfaiie 
magnifica). — 14  females  and  3  males  from 
2  hosts,  under  intertidal  dead  coral  at 
Antsakoabe,  on  the  northern  shore  of  Nosy 
Be,  November  1,  1963  (of  this  collection  7 
females  and  2  males  placed  in  the  Museum 
of  Comparative  Zoology);  5  females  and  3 
males  from  one  host,  under  intertidal  rock 
at  Antsakoabe,  September  7,  1964. 

^  The  specific  name  finini.s  (from  Latin  =  fimi, 
strong,  robust)  refers  to  the  strong  sclerotization 
of  the  body  wall  and  appendages  in  this  species. 


Ne\\'  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        387 


Female. — The  body  (Fig.  60)  is  mod- 
erately broadened  with  a  flattened  prosome. 
The  length  (without  the  setae  on  the  caudal 
rami)  is  0.81  mm  (0.78-0.84  mm)  and  the 
greatest  width  is  0.36  mm  ( 0.33-0.38  mm ) , 
based  on  10  specimens.  The  ratio  of  length 
to  width  of  the  prosome  is  1.46  :  1.  The 
segment  of  leg  1  is  rather  indistinctly  set 
off  from  the  head,  and  its  lateral  areas  are 
rounded  posteriorly.  The  epimeral  areas 
of  the  segment  of  leg  2  are  truncated,  those 
of  the  segment  of  leg  3  are  broadly  rounded, 
and  those  of  the  small  segment  of  leg  4  are 
rather  truncated. 

The  segment  of  leg  5  (Fig.  61)  is  rather 
narrow,  36  X  68  /x,  and  bears  the  small  fifth 
legs  slightly  ventrally  a  little  posterior  to 
the  midlateral  areas.  There  is  a  narrow 
\\'eakly  sclerotized  intersegmental  sclerite 
ventrally  between  the  segment  of  leg  5  and 
the  genital  segment.  The  genital  segment 
is  longer  than  wide,  and  (as  seen  in  dorsal 
view)  is  expanded  in  its  anterior  part  to 
form  2  broadly  rounded  wings  but  is  nar- 
rowed posteriorly  where  its  sides  are  par- 
allel. The  length  of  the  segment  is  110  jx, 
the  width  at  the  expanded  part  100  /^,  and 
the  width  in  the  narrow  posterior  portion 
58  IX.  The  areas  of  attachment  of  the  egg 
sacs  are  situated  dorsolaterally  on  the  pos- 
terior halves  of  the  expansions.  Each  area 
(Fig.  62)  bears  2  small  naked  setae  20  /x 
and  10  /I  in  length.  There  are  3  postgenital 
segments,  without  posterior  rows  of  spi- 
nules,  the  first  46  X  48  ^,  the  second  41  X 
43  fx,  and  the  third  44  X  42  ^. 

The  caudal  ramus  (Fig.  63)  is  moderately 
elongated,  39  /x  along  its  outer  edge,  35  jx 
along  its  inner  edge,  and  43  /x  in  greatest 
length  (including  the  subconical  terminal 
expansion).  Its  width  proximal  to  the 
lateral  seta  is  18  ^,  and  distal  to  that  seta 
15.5  IX.  Taking  the  greatest  dimensions,  the 
ratio  of  length  to  width  is  2.9  :  1.  The  outer 
lateral  seta,  situated  21  p-  from  the  base  of 
the  ramus,  is  33  /x  long.  The  j)edicellate 
dorsal  seta  is  33  ix,  the  outennost  terminal 
seta  40  ^i,  and  the  innermost  terminal  seta 
39  IX.    The  2  long  median  terminal  setae, 


inserted  between  unomamented  dorsal  and 
ventral  flaps,  are  180  /x  (outer)  and  325  yu, 
(inner),  and  do  not  show  the  basal  "pegs" 
often  seen  in  lichomolgids.  The  surfaces 
of  the  ramus  are  without  ornamentation  ex- 
cept for  a  setule  and  a  refractile  point  dor- 
sally. 

The  dorsal  surface  of  the  prosome  is 
ornamented  \\'ith  minute  setules  and  re- 
fractile points,  and  in  addition  the  posterior 
half  of  the  cephalosome  shows  dorsally  a 
broad  band  of  extremely  fine  transverse  stri- 
ations  (visible  only  under  very  high  magnifi- 
cation and  not  shown  in  Fig.  60).  In  dorsal 
\'iew  there  are  2  internal  longitudinal 
sclerotized  bars  (Fig.  60)  extending  pos- 
teriorly from  the  rostral  area  to  nearly  the 
middle  of  the  cephalosome.  (The  dorsal 
surface  of  the  cephalosome  in  this  region 
is  smooth,  there  being  no  crest  such  as 
Sew^ll  described  in  N.  cristatus. )  The  dor- 
sal and  ventral  surfaces  of  the  urosome 
bear  relatively  few  minute  setules  and  re- 
fractile points.  The  ratio  of  the  length  of 
the  prosome  to  that  of  the  urosome  is 
1.6:  1. 

The  body  wall  (and,  as  will  be  seen 
below,  the  appendages)  is  strongly  sclero- 
tized, in  contrast  to  the  3  species  whose 
descriptions  follow. 

The  egg  sacs  are  slender  (Fig.  64)  and 
reach  \\'ell  beyond  the  ends  of  the  caudal 
rami.  Each  sac  is  about  418  X  99  /x,  and 
contains  numerous  eggs  about  47  /x  in 
diameter. 

The  rostral  area  (Fig.  65)  projects 
slightly  in  front  of  the  head.  Posterior  to 
it,  on  the  ventral  surface  of  the  head  be- 
t\\'een  the  insertions  of  the  second  anten- 
nae, there  is  a  slight  longitudinal  ridge. 

The  first  antenna  (Fig.  66)  is  7-seg- 
mented.  The  lengths  of  the  segments 
(measured  along  their  posterior  non-setif- 
erous  margins)  are:  13  (35  ix  along  the 
anterior  edge),  57,  22,  28,  25,  17,  and  13  ^ 
respectively.  The  entire  first  antenna 
(without  the  terminal  setae)  is  about  185 
/x  long.  The  formula  is:  4,  13,  6,  3,  4  +  1 
aesthete,  2  -f-  1  aesthete,  and  7  -f-  1  aesthete. 


388        BuUctin  Museum  of  Comparative  Zoology,  Vol  135,  No.  7 


All  the  setae  are  naked.  The  third  segment 
shows  ventrally  a  sclerotization  suggesting 
an  intercalary  segment. 

The  second  antenna  (P'ig.  67)  is  4-seg- 
mented  and  rather  robust.  The  first  and 
second  segments  bear  a  single  small  inner 
seta.  The  short  third  segment  bears  3  setae, 
one  much  larger  than  the  other  two.  The 
fourth  segment  bears  7  elements:  3  proximal 
slender  setae,  another  more  distal  slender 
seta,  a  slender  clawlike  seta,  and  2  stout 
recurved  claws.  All  the  setae  are  naked. 
The  entire  second  antenna  is  about  108  /j, 
in  length. 

The  labrum  (Fig.  65)  bears  anterolater- 
ally  a  pair  of  naked  setae,  each  44  /a  in 
length,  directed  ventrally  and  somewhat 
anteriorly,  and  rather  indistinctly  articu- 
lated with  the  surface  of  the  labrum.  Near 
the  extreme  anterior  end  of  the  labrum, 
between  these  2  setae,  there  is  a  somewhat 
triangular  raised  sclerotized  area  (rather 
similar  to  that  described  by  Sewell  in  N. 
cristatus).  Posteriorly  the  labrimi  is  bifur- 
cated to  form  2  rounded  lobes  without 
ornamentation. 

The  mandible  (Fig.  68)  has  a  moderately 
elongated  blade,  with  a  row  of  small 
spinules  along  the  concave  outer  margin 
and  a  fringe  of  large  hyaline  spinifomi 
elements  along  the  convex  inner  margin 
(the  proximal  spinelike  element  in  this  row 
being  slightly  more  sclerotized  and  promi- 
nent that  the  succeeding  ones).  There  are 
2  sclerotized  spinifonn  elements  on  the 
dorsal  surface  of  the  blade.  The  paragnath 
(Fig.  69)  is  a  small  lobe  with  hairs  along 
its  medial  side.  The  first  maxilla  (Fig.  70) 
is  a  small  lobe  bearing  4  elements:  termi- 
nally 2  long  unilaterally  barbed  spines  and 
a  minute  naked  setule,  subterminally  a 
shorter  naked  spine.  The  second  maxilla 
(Fig.  71)  is  2-segmented,  the  first  segment 
bearing  a  prominent  sclerotized  conical 
projection  on  its  outer  margin.  The  second 
segment  bears  proximally  2  slender  naked 
inner  setae  and  a  minute  outer  setule;  the 
segment  is  i:)rolonged  to  form  a  short 
process    bearing   a   row   of   about   8   outer 


spines  graduated  in  length  and  becoming 
smaller  distally.  The  maxilliped  (Fig.  72)  is 
3-segmented  and  slender.  The  first  segment 
is  unarmed.  The  second  segment  bears  2 
small  inner  naked  setae.  The  clawlike  third 
segment  is  very  long  (110  /x)  and  slender; 
proximally  on  the  inner  side  there  is  a  small 
naked  seta  and  a  minute  setule.  The  distal 
half  of  the  segment  is  more  slender  and  less 
sclerotized  than  proximally  and  bears  a 
row  of  short  hairs  along  each  side. 

The  postoral  area  ( Fig.  65 )  shows  a  pair 
of  sclerotizations  extending  medially  from 
the  bases  of  the  maxillipeds,  but  not  joining 
each  other.  There  is  a  slightly  projecting 
cordiform  area  just  posterior  to  the  mouth 
region.  The  area  between  the  maxillipeds 
and  the  first  pair  of  legs  projects  only 
slightly. 

Legs  1^  (Figs.  73,  74,  75,  and  76)  have 
3-segmented  rami  except  for  the  endopod 
of  leg  4  which  is  2-segmented.  The  spine 
and  setal  formula  is  as  follows  (the  Roman 
numerals  indicating  spines,  the  Arabic 
numerals  setae ) : 

PI   piotopod  0-1;    1-0   exp  I-O;  I-l;  111,1,4 

end  0-1;  0-1;  1,5 

P2  piotopod  0-1;   1-0   exp  I-O;  I-l;  111,1,5 

end  0-1;  0-2;  1,11,3 

P3  protopod  0-1;   1-0   exp  I-O;  I-l;  11,1,5 

end  0-1;  0-2;  11,2 

P4  protopod  0-1;    1-0    exp  I-O;  I-l;  11,1,5 

end  0-1;  2 

The  inner  seta  on  the  coxa  of  legs  1-3 
is  long  and  feathered,  but  in  leg  4  tliis  seta 
is  shorter  ( 20  /u. )  and  naked.  A  row  of  hairs 
occurs  on  the  inner  margin  of  the  basis  in 
all  4  legs.  The  spines  on  the  exopods  are 
recurved  posteriorly,  with  the  lateral  fringe 
along  their  proximal  sides  more  prominent 
than  that  on  their  distal  margins.  The  outer 
spines  on  the  exopod  of  leg  1  show  small 
subterminal  flagella;  the  tenninal  spine  on 
this  ramus  has  a  straight  tip,  while  in  legs 
2-4  the  tip  is  reflexed.  The  outer  spine  on 
the  last  segment  of  the  endopod  of  leg  1  is 
7  IX  in  length.  The  endopod  of  leg  4  is 
elongated  (Fig.  76).  The  first  segment  is 
18x11  IX,  having  a  row  of  hairs  on  the 
outer  margin  and  bearing  a  long  feathered 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho 


389 


inner  distal  seta  68  ^  in  length.  The  second 
segment  measures  39  X  8  /x,  having  a  row 
of  hairs  on  both  outer  and  inner  margins 
and  bearing  2  unequal  terminal  naked 
setae,  the  outer  one  31  /x,  the  inner  one 
50  [x  in  length.  In  addition,  there  is  an 
anterior  row  of  minute  spinules  on  the 
end  of  the  segment  near  the  insertions  of 
the  2  setae. 

Leg  5  (Fig.  77)  has  a  very  small  quad- 
rate free  segment,  6  x  6  /x,  bearing  2  termi- 
nal naked  setae,  the  anterior  1  slender  and 
28  ix  long,  the  posterior  1  much  stouter  and 
44  fx  long.  The  seta  arising  from  the  body 
wall  adjacent  to  the  segment  is  naked  and 
26  IX  long. 

Leg  6  is  probably  represented  by  the  2 
setae  near  the  attachment  of  the  egg  sacs 
(see  Fig.  62). 

The  color  in  life  in  transmitted  light  is 
slightly  opaque,  the  eye  red,  the  egg  sacs 
gray. 

Male. — The  body  (Fig.  78)  is  more 
slender  than  in  the  female.  The  length 
(excluding  the  setae  on  the  caudal  rami) 
is  0.58  mm  (0.52-0.63  mm)  and  the  great- 
est width  is  0.16  mm  (0.14-0.18  mm),  based 
on  10  specimens  ( the  allotype,  3  paratypes, 
3  specimens  collected  at  Antsakoabe  on 
November  1,  1963,  and  3  collected  at  the 
same  locality  on  September  7,  1964).  The 
ratio  of  length  to  width  of  the  prosome  is 
1.8  :  1.  The  segment  of  leg  1  is  more  dis- 
tinctly separated  from  the  head  than  in  the 
female. 

The  segment  of  leg  5  (Fig.  79),  37  X  29 
IX,  resembles  that  of  the  female.  The  ven- 
tral intersegmental  sclerite  between  the 
segment  of  leg  5  and  the  genital  segment 
is  not  evident.  The  genital  segment  is 
longer  than  wide,  97  X  75  /x,  with  its  lateral 
borders  in  dorsal  view  only  slightly 
rounded.  The  4  postgenital  segments  are 
24  X  37,  29  X  34,  23  x  31,  and  25  X  30  ,x 
from  anterior  to  posterior. 

The  caudal  ramus  (Fig.  80)  resembles 
that  of  the  female,  but  is  relatively  shorter, 
the    greatest   dimensions   being   27  X  13  /x, 


or  2  times  longer  than  wide.    The  2  long 
tenninal  setae  show  basal  "pegs." 

The  surfaces  of  the  prosome  and  urosome 
bear  scattered  hairs  and  refractile  points. 
The  fine  transverse  striations  seen  on  the 
dorsal  posterior  half  of  the  cephalosome  in 
the  female  are  absent  here.  The  ratio  of 
the  length  of  the  prosome  to  that  of  the 
urosome  is  1.25  :  1. 

The  rostral  area  and  first  antenna  re- 
semble those  of  the  female.  The  second 
antenna  (Fig.  81)  is  a  little  more  slender 
than  in  the  female.  The  first  and  second 
segments  bear  a  small  seta  as  in  the  female, 
but,  in  addition,  the  second  segment  is 
ornamented  on  its  inner  surface  with  2 
rows  of  minute  spinules.  The  third  seg- 
ment bears  4  setae,  3  of  them  slender,  the 
other  larger  and  placed  more  distally,  veiy 
near  the  base  of  the  fourth  segment.  The 
fourth  segment  bears  7  elements:  3  slender 
outer  setae,  2  setae  near  the  bases  of  the 
claws,  and  2  recurved  claws,  1  slender,  the 
other  stout.  The  labrum,  mandible,  parag- 
nath,  and  first  maxilla  resemble  those  of 
the  female.  The  second  maxilla  (Fig.  82) 
has  on  the  outer  margin  of  its  first  segment 
groups  of  small  spinules  and  a  small  proc- 
ess which  is  perhaps  homologous  to  the 
prominent  conical  projection  seen  in  the 
female.  The  maxilliped  (Fig.  83)  is  slender 
and  4-segmented  ( assuming  that  the  fourth 
segment  forms  part  of  the  claw).  The  first 
segment  shows  on  its  inner  edge  a  small 
knob  and  a  rather  pointed  sclerotized 
process.  The  second  segment  has  2  small 
inner  naked  setae  and  2  inner  rows  of  small 
spinules.  The  small  third  segment  is  un- 
armed. The  recurved  claw,  58  /x  in  length 
(measured  along  its  axis),  bears  proximally 
a  posteromedial,  minutely  barbed  seta  20 
jx  long  and  on  its  inner  edge  a  slender  naked 
setule  6  jx  long  and  a  small  process.  The 
posteroinner  surface  of  the  claw  is  covered 
with  minute  blunt  spinules.  At  the  tip  of 
the  claw  there  is  a  conspicuous  lamella. 

The  postoral  area  resembles  that  of  the 
female. 

Legs    1-4   in   general   resemble  those   of 


390 


BuUciin  Muscnni  of  Coitiparative  Zoology,  Vol.  135,  No.  7 


the  female,  having  the  same  spine  and  setal 
fonnula.  Tlie  outer  distal  comers  of  the 
first  2  segments  of  the  endopods  in  legs 
1-3  bear  small  spinelike  processes  not  pres- 
ent in  the  female.  The  outer  spine  on  the 
last  segment  of  the  endopod  of  leg  1  ( Fig. 
84)  is  11  ix  in  length,  a  little  longer  than  in 
the  female. 

Leg  5  is  hke  that  of  the  female. 

Leg  6  (Fig.  85)  consists  of  a  postero- 
lateral flap  on  the  ventral  surface  of  the 
genital  segment,  bearing  2  naked  setae  17 
and  28  /a  long. 

In  2  females  spermatophores  ( Fig.  86 ) 
were  attached  in  pairs.  In  both  cases  the  2 
elongated  spermatophores  ( 86  X  31  /x  with- 
out the  neck)  were  joined  in  a  common 
tube  which  led  into  the  genital  segment. 

The  color  in  life  in  transmitted  light  re- 
sembles that  of  the  female. 

Comparison  icith  NasomoJgus  cristatus. 
— Na,somolg,u.s  jinmis  seems  to  be  close  to 
N.  cristatus,  but  shows  several  differences. 
In  N.  firmus  the  caudal  ramus  is  a  little 
longer  (with  a  ratio  of  length  to  width  of 
2.9  :  1  instead  of  about  2.5  :  1),  the  lateral 
seta  on  the  ramus  is  inserted  halfway  along 
the  margin  instead  of  at  about  the  junction 
of  the  middle  and  distal  thirds  as  in  Sewell's 
species,  the  formula  for  the  last  segment 
of  the  endopod  of  leg  3  is  11,2  instead  of 
1,11,2,  the  two  setae  on  the  end  of  the  last 
segment  of  the  endopod  of  leg  4  have  a 
ratio  of  1  :  1.6  instead  of  about  1  :  2.2  as 
in  Sewell's  fig.  35E,  and  the  outline  of  the 
genital  segment  in  dorsal  view  is  slightly 
different. 

In  both  N.  cristatus  and  N.  firmus  the 
labrum  bears  anteriorly  a  pair  of  prominent 
ventrally  directed  setae,  there  is  a  some- 
what triangular  sclerotized  raised  area  near 
the  front  of  the  labrum  at  the  posterior  end 
of  the  rostral  area,  the  maxilliped  is  3- 
segmented  \\'ith  the  last  segment  very  long 
and  slender,  and  the  arrangement  of  the 
spines  on  the  last  segment  of  the  exopod  of 
leg  1  is  III, I  instead  of  II, I  as  in  Sewell's 
formula  (p.  125).  Since  it  has  been  im- 
possible to  dissect  the  single  type  specimen 


of  N.  cristatus,  the  mouthparts  cannot  be 
compared  in  detail. 

Nasomolgus  leptus^  n.  sp. 
Figs.  87-109 

Type  materia]. — 4  females  and  4  males 
from  2  Sahelhstarte  magnijica  (Shaw), 
under  intertidal  dead  coral  at  Antsakoabe, 
on  the  northern  shore  of  Nosy  Be,  Mada- 
gascar. Collected  November  1,  1963.  Holo- 
type  female,  allotype,  and  4  paratypes  (2 
females  and  2  males)  deposited  in  the 
United  States  National  Museum,  Washing- 
ton; the  remaining  paratypes  (both  dis- 
sected) in  the  collection  of  A.  G.  Humes. 
(This  species  was  collected  in  company 
with  N.  firmus,  from  the  same  2  poly- 
chaetes. ) 

Female. — The  body  (Fig.  87)  is  elon- 
gated and  rather  slender,  with  the  prosome 
not  broadened  and  not  as  flattened  as  in 
the  previous  species.  The  length  (not  in- 
cluding the  setae  on  the  caudal  rami)  is 
1.40  mm  (1.37-1.44  mm)  and  the  greatest 
width  is  0.35  mm  (0.31-0.39  mm),  based 
on  4  specimens.  The  ratio  of  length  to 
width  of  the  prosome  is  1.8  :  1.  The  seg- 
ment of  leg  1  is  distinctly  separated  from 
the  head.  The  epimeral  areas  of  the  meta- 
somal  segments  resemble  fairly  closely 
those  in  the  previous  species. 

The  segment  of  leg  5  ( Fig.  88 )  is  60  X 
94  IX,  with  the  small  fifth  legs  borne  as  in 
N.  firmus.  A  weakly  developed  interseg- 
mental sclerite  occurs  ventrally  between 
the  segment  of  leg  5  and  the  genital  seg- 
ment. The  genital  segment  is  longer  than 
wide,  in  dorsal  view  being  broadly  ex- 
panded laterally  in  its  anterior  half  and 
constricted  in  its  posterior  half.  The  length 
of  the  segment  is  174  /x,  the  width  at  the 
expansions  140  /x,  and  the  width  in  the 
posterior  part  of  the  constricted  area  73  /x 
(at  this  level  the  segment  being  slightly 
wider  than  more  anteriorly).  The  areas  of 
attachment    of    the    egg    sacs    are    placed 

^  The  specific  name  leptus  ( from  XewTO's  =  thin, 
slender)  refers  to  the  elongated  slender  form  of 
the  b()d>'  in  this  species. 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        391 


dorsolaterally  on  the  posterior  halves  of  the 
expansions.     Each   area    (Fig.    89)    carries 

2  naked  setae,  11  and  24  fj.  in  length.    The 

3  postgenital  segments,  without  posterior 
rows  of  spinules,  are  96  X  60,  82  X  52,  and 
96  X  55  /x  (the  last  segment  being  slightly 
expanded  laterally  in  its  posterior  half 
where  the  width  was  measured). 

The  caudal  ramus  (Fig.  90)  is  very 
elongated  and  slender,  278  /x  in  length,  25 
(U,  wide  in  its  basal  part  and  19  ^  wide  at 
the  level  of  the  outer  lateral  seta.  Taking 
the  latter  width,  the  ratio  of  length  to 
width  is  14.6  :  1.  The  outer  lateral  seta, 
located  177  /x  from  the  base  of  the  ramus, 
is  36  /JL.  long.  The  pedicellate  dorsal  seta  is 
26  /x,  the  outemiost  temiinal  seta  40  /x,  the 
innermost  terminal  seta  36  /x,  and  the  2 
long  median  terminal  setae  are  143  /x 
(outer)  and  260  /x  (inner)  and  show  weak 
basal  "pegs."  All  the  setae  are  naked.  A 
minute  setule  3  p.  long  occurs  on  the  outer 
proximal  margin  of  the  ramus.  The  ramus 
is  ornamented  with  a  few  minute  hairs. 

The  dorsal  surface  of  the  prosome  bears 
very  few  hairs  and  no  retractile  points.  The 
dorsal  and  ventral  surfaces  of  the  urosome 
are  very  sparsely  ornamented  with  hairs 
and  refractile  points.  The  urosome  is  longer 
than  the  prosome,  the  ratio  being  1.2  :  1. 

The  egg  sacs  ( Fig.  87 )  are  elongated  and 
slightly  arcuate,  reaching  beyond  the  tips 
of  the  long  caudal  rami.  Each  sac  is  about 
759  X  198  fx,  and  contains  many  eggs  ap- 
proximately 57  iJi  in  diameter. 

The  rostral  area  (Fig.  91)  does  not 
project  fonvard  as  in  the  previous  species. 
Between  the  rostrum  and  the  labnnn  there 
is  a  low  longitudinal  ridge  (between  the 
bases  of  the  second  antennae). 

The  first  antenna  (Fig.  92)  is  7-seg- 
mented,  the  lengths  of  the  segments  (mea- 
sured along  their  posterior  non-setiferous 
margins)  being:  18  (44  /x  along  the  an- 
terior margin),  94,  35,  45,  31,  18,  and  14 
IJi  respectively.  The  formula  for  the  setae 
and  aesthetes  is  the  same  as  in  A',  firmiis. 
All  the  setae  are  naked. 

The  second  antenna   (Fig.  93)  is  4-seg- 


mented  and  fairly  robust.  Each  of  the  first 
2  segments  bears  a  short  distal  inner  seta 
wdth  lamellate  margins.  The  third  segment 
bears  2  such  setae  plus  a  longer  seta.  The 
fourth  segment  carries  7  elements  (includ- 
ing 2  stout  recurved  claws)  much  like 
those  of  N.  firmus. 

The  labrum  (Figs.  91  and  94)  bears 
anterolaterally,  as  in  the  previous  species, 
a  pair  of  ventrally  directed  naked  setae, 
each  55  /x  in  length.  There  is  no  triangular 
sclerotized  area  in  front  of  the  labrum,  such 
as  seen  in  N.  firmus.  Posteriorly,  the  edge 
of  the  labrum  is  deeply  bilobed,  with  each 
lobe  elongated,  rounded,  and  unonia- 
mented,  and  with  a  short  median  process 
between  the  bases  of  the  lobes. 

The  mandible  (Fig.  95),  paragnath,  and 
first  maxilla  (Fig.  96)  resemble  those  of 
N.  firmus:  The  second  maxilla  (Fig.  97) 
is  2-segmented.  The  first  segment  has  a 
broad  sclerotized  bulge  on  its  outer  margin 
and  an  interrupted  crescentic  row  of  spi- 
nules on  its  posterodorsal  surface.  The 
second  segment  is  similar  to  that  in  the 
pre\'ious  species,  but  the  spines  on  the 
distal  prolongation  are  more  numerous  and 
slender.  The  maxilliped  is  very  similar  to 
that  of  N.  firmus,  having  the  same  general 
fonn  and  armature  and  with  the  slender 
clawlike  segment  122  /x  long. 

The  postoral  area  (Fig.  98)  resembles 
generally  that  in  the  previous  species  and 
shows  a  weak  line  between  the  bases  of 
the  maxillipeds.  The  ventral  surface  be- 
tween the  bases  of  the  maxillipeds  and  the 
first  pair  of  legs  is  slightly  protuberant. 

Legs  1-4  (Figs.  99,  100,  101,  and  102) 
resemble  those  of  N.  firmus,  with  the  same 
spine  and  setal  formula  except  for  the 
endopod  of  leg  3  where  the  formula  is  0-1; 
0-2;  1,11,2,  the  outer  marginal  spine  on  the 
last  segment  being  retained.  The  terminal 
spine  on  the  last  segment  of  the  exopods 
of  all  4  legs  is  not  reflexed  at  the  tip.  The 
outer  spine  on  the  last  segment  of  the  endo- 
pod of  leg  1  is  14  ^a  in  length.  The  endopod 
of  leg  4  is  elongated  ( Fig.  102 ) .  The  first 
segment   is   33  X  19   /x   and   bears   a    short 


392         Bulletin  Museum  of  Coinpaiative  Zoology,  Vol.  135,  No.  7 


feathered  inner  distal  seta  22  /x  long.  The 
second  segment  is  74  X  13  /^  and  bears  2 
terminal  slightly  barbed  setae  44  /x  (outer) 
and  77  ^  (inner)  in  length.  The  ornamen- 
tation of  the  endopod  resembles  that  of 
N.  firmus.  The  inner  coxal  seta  of  leg  4  is 
short  (13  fi)  and  naked. 

Leg  5  (Fig.  103)  resembles  that  in  the 
previous  species,  with  the  small  free  seg- 
ment 7  ij.  in  anterior  length,  6  /x  in  posterior 
length,  and  8  /x  in  width  at  the  middle. 

Leg  6  is  probably  represented  by  the  2 
setae  near  the  attachment  of  the  egg  sacs 
(see  Fig.  89). 

The  color  in  life  in  transmitted  light  is 
moderately  translucid,  the  eye  red,  the  egg 
sacs  grav. 

Mole— The  body  (Fig.  104)  is  more 
slender  than  that  of  the  female.  The  length 
(excluding  the  setae  on  the  caudal  rami) 
is  0.94  mm  (0.93-0.96  mm)  and  the  great- 
est width  is  0.19  mm  (0.18-0.19  mm), 
based  on  4  specimens.  The  ratio  of  length 
to  width  of  the  prosome  is  2.1  :  1.  The 
segment  of  leg  1  is  less  distinctly  set  off 
from  the  head  than  in  the  female. 

The  segment  of  leg  5,  40  X  58  /x,  re- 
sembles that  of  the  female.  Ventrally  be- 
tween the  segment  of  leg  5  and  the  genital 
segment  there  is  no  evident  intersegmental 
sclerite.  The  genital  segment  (Fig.  105) 
is  elongated,  151  X  50  /x,  with  its  sides  in 
dorsal  view  slightly  rounded.  The  4  post- 
genital  segments  are  57  X  46,  61  X  40,  47  x 
34,  and  57  X  33  /x  from  anterior  to  posterior. 

The  caudal  ramus  ( see  Fig.  105 )  is  elon- 
gated, 135  X  16  /x,  about  8.4  times  longer 
than  wide.  It  is  a  little  less  tapered  dis- 
tally  than  in  the  female,  but  bears  similar 
armature. 

The  dorsal  surface  of  the  prosome  seems 
to  lack  ornamentation.  The  dorsal  and 
ventral  surfaces  of  the  urosome  are  un- 
ornamented  except  for  a  pair  of  hairs  on 
the  dorsal  surface  of  the  anal  segment. 
As  in  the  female,  the  urosome  is  longer 
than  the  prosome,  the  ratio  being  1.3  :  1. 

The  rostral  area  and  first  antenna  re- 
semble those   of  the   female.    The   second 


antenna  (Fig.  106)  is  more  slender  than 
in  the  female.  The  arrangement  of  the 
spines  and  setae  is  the  same  as  in  the  male 
of  N.  firmus;  with  4  elements  on  the  third 
segment  instead  of  3  as  in  the  female.  The 
2  tenninal  claws  are  distinctly  jointed.  The 
slender  clawlike  seta  on  the  last  segment 
has  a  more  blunt  tip  than  in  the  female. 
The  fine  ornamentation,  not  present  in  the 
female,  consists  of  a  small  patch  of  spinules 
on  the  inner  proximal  surface  of  the  first 
segment  and  a  long  patch  of  small  spinules 
on  the  inner  surface  of  the  second  segment. 

The  labrum,  mandible,  paragnath,  and 
first  maxilla  resemble  those  of  the  female. 
The  second  maxilla  is  also  similar  to  that 
in  the  opposite  sex,  but  the  outer  bulge  on 
the  first  segment  is  much  less  prominent. 
The  maxilliped  (Fig.  107)  resembles  in 
general  form  that  of  N.  firmus:  The  second 
segment  has  on  its  inner  surface  2  small 
naked  setae  and  2  rows  of  spinules.  The 
recurved  claw,  68  /x  in  length  (measured 
along  its  axis),  shows  a  slight  indication  of 
division.  Proximally  the  claw  bears  a  pos- 
teromedial barbed  seta  19  /x  long  and  an 
adjacent  inner  naked  setule  6  /x  long.  The 
concave  surface  of  the  claw  bears  a  row  of 
small  spinules,  instead  of  being  covered 
with  minute  blunt  spinules  as  in  N.  firmus. 

The  postoral  area  is  similar  to  that  in 
the  previous  species. 

Legs  1-4  are  like  those  of  the  female, 
with  the  same  spine  and  setal  formula. 
The  outer  distal  spine  on  the  last  segment 
of  the  endopod  of  leg  1  is  16.5  /x  in  length, 
being  slightly  longer  than  in  the  female. 

Leg  5  resembles  that  of  the  female. 

Leg  6  (Fig.  108)  consists  of  a  postero- 
lateral flap  on  the  ventral  surface  of  the 
genital  segment,  bearing  2  slender  naked 
setae  26  and  31  ^a  in  length. 

The  spermatophore  (Fig.  109),  seen  only 
inside  the  body  of  a  male,  is  elongated, 
113  X  49  ^,  not  including  the  neck. 

The  color  in  life  resembles  that  of  the 
female. 

Comparison  iritJi  other  species. — Naso- 
moliius  Icpfus  may  readily  be  distinguished 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        393 


from  A',  finims  and  N.  cristatus  by  its 
greater  length,  by  its  more  slender  body 
fonn,  and  by  the  much  more  elongated 
caudal  rami.  It  differs  further  from  N. 
firmiis  in  the  nature  of  the  protuberance 
on  the  first  segment  of  the  second  maxilla 
in  the  female  and  in  the  ornamentation 
of  the  claw  on  the  maxilliped  in  the  male. 

Nasomolgus  rudis^  n.  sp. 
Figs.  110-135 

Type  material. — 10  females  and  5  males 
from  one  SabeUastarte  magnifica  (Shaw), 
in  2  m,  at  Ambariotelo,  a  small  island  be- 
tween Nosy  Be  and  Nosy  Komba,  Madagas- 
car. Collected  May  15,  1964.  Holotype  fe- 
male, allotype,  and  10  paratypes  ( 8  females 
and  2  males )  deposited  in  the  United  States 
National  Museum,  Washington;  the  remain- 
ing paratypes  (dissected)  in  the  collection 
of  A.  G.  Humes. 

Other  specimens  (all  from  Sabellastarte 
magnifica,  but  one  host  identification  un- 
certain as  indicated  below). — 10  females 
and  2  males  from  2  hosts,  under  intertidal 
dead  coral,  Antsakoabe,  on  the  northern 
shore  of  Nosy  Be,  November  1,  1963;  11 
females  from  1  host  (only  the  plume  col- 
lected, but  probably  S.  magnifica),  in  14 
m,  Tany  Kely,  a  small  island  south  of  Nosy 
Be,  December  23,  1963;  9  females  and  1 
male  from  1  host,  under  intertidal  rock, 
Antsakoabe,  September  7,  1964  (these  speci- 
mens placed  in  the  Museum  of  Compara- 
tive Zoology);  and  4  females  from  1  host 
in  1  m,  Ambariobe,  near  Ambariotelo,  Oc- 
tober 10,  1960. 

Female. — The  body  (Fig.  110)  is  broad- 
ened in  the  prosomal  region.  The  length 
(excluding  the  setae  on  the  caudal  rami) 
is  0.S7  mm  ( 0.77-0.97  mm )  and  the  greatest 
width  is  0.41  mm  (0.36-0.46  mm),  based 
on  10  specimens.  The  ratio  of  the  length 
to  the  width  of  the  prosome  is  1.45  :  1.  The 
segment   of  leg   1   is   clearly   set   off  from 


^  The  specific  name  nidis  ( from  Latin  —  un- 
affected, simple )  refers  to  the  relatively  uncom- 
plicated  external   anatomy   of   this   species. 


the  head.  The  epimeral  areas  of  the  pedig- 
erous  segments  are  shaped  much  as  in  N. 
Jeptus. 

The  segment  of  leg  5  ( Fig.  Ill )  is  55  X  78 
II,  with  the  fifth  legs  borne  as  in  the  2 
previous  species.  There  is  a  small  inter- 
segmental sclerite  ventrally  between  the 
segment  of  leg  5  and  the  genital  segment. 
The  genital  segment  is  a  little  wider  than 
long  and  in  dorsal  view  is  broadly  expanded 
in  its  anterior  two-thirds  and  constricted  in 
its  posterior  third.  The  length  of  the  seg- 
ment is  99  jx,  the  width  at  the  expansions 
112  IX,  and  the  width  in  the  posterior  part 
of  the  constricted  area  58  ix.  The  areas  of 
attachment  of  the  egg  sacs  are  situated 
dorsolaterally  on  the  posterior  halves  of  the 
expansions.  Each  area  (Fig.  112)  bears  2 
naked  setae  31  and  11  jx  in  length,  a  short 
inner  spinous  process,  and  an  outer  mem- 
branous expansion.  The  3  postgenital  seg- 
ments, \\athout  posterior  rows  of  spinules, 
are  32  x  52,  24  x  45,  and  33  X  43  ^  from 
anterior  to  posterior. 

The  caudal  ramus  (Fig.  113)  is  short,  25 
IX  along  its  inner  margin,  27  jx  along  its  outer 
margin,  and  18  fx  wide  at  the  level  of  the 
outer  lateral  seta.  The  ratio  of  length  to 
width  is  1.44  :  1.  The  outer  lateral  seta, 
located  14  /x  from  the  base  of  the  ramus, 
is  56  IX  long.  The  pedicellate  dorsal  seta 
is  33  IX,  the  outermost  terminal  seta  54  //,, 
the  innermost  terminal  seta  72  jx,  and  the  2 
long  median  terminal  setae  are  265  ix 
(outer)  and  407  ix  (inner)  and  are  basally 
"pegged."  All  the  setae  are  naked.  Two 
minute  hairs  occur  on  the  dorsal  surface 
of  the  ramus. 

The  dorsal  surface  of  the  prosome  is 
almost  devoid  of  ornamentation,  there  being 
only  a  few  hairs  on  the  metasomal  seg- 
ments. The  dorsal  and  ventral  surfaces 
of  the  urosome  have  scattered  hairs  and 
refractile  points.  The  prosome  is  much 
longer  than  the  urosome,  the  ratio  being 
2.53  :  1. 

The  egg  sacs  are  elongated,  extending  to 
the  ends  of  the  ramal  setae.   Each  sac  (Fig. 


394        Bulletin  Mu.scuni  of  Coinpanitwe  Zoology,  Vol.  135,  No.  7 


114)  is  about  462  X  132  ja,  with  luimerous 
eggs  44-47  /x  in  diameter. 

The  rostral  area  (Fig.  115)  resembles 
generally  that  of  N.  Icptus,  and  there  is  a 
low  ridge  between  the  bases  of  the  second 
antennae  as  in  that  species. 

The  first  antenna  (Fig.  116)  is  much 
like  that  in  N.  leptiis,  with  the  same  ar- 
rangement of  setae  and  aesthetes.  The 
lengths  of  the  segments  (measured  along 
their  posterior  non-setiferous  margins)  are: 
13  (38  fji  along  the  anterior  margin),  68, 
26,  34,  27,  17,  and  15  fi  respectively.  All 
the  setae  are  naked. 

The  second  antenna  (Fig.  117)  also 
resembles  that  of  N.  leptus,  but  the  seta  on 
the  first  segment  is  hyaline  and  lacks  the 
lamellate  margins. 

The  labrum  (Figs.  115  and  118)  resembles 
in  general  form  that  of  the  2  previous  spe- 
cies. Each  of  the  2  anterolateral  and  ven- 
trally  directed  setae  is  36  /a  long  and  naked. 
The  2  posterior  lobes  are  unornamented. 
There  is  no  triangular  sclerotized  area  near 
the  front  of  the  labrum,  such  as  seen  in  N. 
firmus. 

The  mandible  (Fig.  119),  paragnath,  and 
first  maxilla  (Fig.  120)  are  similar  to  those 
in  the  2  previous  species.  The  second 
maxilla  (Fig.  121)  resembles  in  general 
fonn  that  of  N.  leptus.  There  is  a  lightly 
sclerotized  bulge  on  the  outer  margin  of 
the  first  segment.  Of  the  several  spines  on 
the  distal  end  of  the  second  segment  the 
first  spine  is  somewhat  larger  than  the 
succeeding  ones.  The  maxilliped  (Fig.  122) 
resembles  that  in  the  2  previous  species, 
the  slender  terminal  segment  being  113  /x 
long. 

The  postoral  area  (Fig.  123)  resembles 
that  of  N.  leptus. 

Legs  1-4  (Figs.  124,  125,  126,  and  127) 
have  the  same  spine  and  setal  formula  as  in 
N.  leptus,  and  closely  resemble  that  species 
in  their  fine  ornamentation.  The  outer  spine 
on  the  last  segment  of  the  endopod  of  leg 
1  is  10  jt.  long.  The  3  spines  on  the  last 
segment  of  th(>  endopod  of  leg  2  are  12, 
9,  and  9  /;,  in  length  from  distal  to  proximal; 


those  on  leg  3  are  16,  10,  and  11  ^u,  respec- 
tively. The  inner  coxal  seta  of  leg  4  is  15  ju, 
long  and  naked.  The  endopod  of  leg  4  is 
less  elongated  (Fig.  127)  than  in  either 
of  the  2  previous  species.  The  first  segment 
is  28  X  14  iJu  and  bears  a  short  feathered 
inner  distal  seta  29  /x  long.  The  second 
segment  is  47  X  12  ^  and  bears  2  terminal 
slightly  barbed  setae  32  /a  ( outer )  and  66  /x 
(inner)  in  length.  The  fine  ornamentation 
of  the  endopod  is  similar  to  that  in  the  2 
previous  species. 

Leg  5  (Fig.  128)  resembles  that  of  N. 
leptus,  with  the  segment  having  similar 
dimensions. 

Leg  6  is  probably  represented  by  the  2 
setae  near  the  attachment  of  the  egg  sacs 
(see  Fig.  112). 

The  color  in  life  in  transmitted  light  re- 
sembles that  in  the  2  previous  species. 

Male. — The  body  (Fig.  129)  is  much 
more  slender  than  in  the  female.  The  length 
(excluding  the  setae  on  the  caudal  rami) 
is  0.52  mm  ( 0.50-0.54  mm )  and  the  greatest 
width  is  0.15  mm  (0.14-0.16  mm),  based 
on  8  specimens  (the  allotype,  4  paratypes, 
2  specimens  collected  at  Antsakoabe  on 
November  1,  1963,  and  one  specimen  col- 
lected at  the  same  locality  on  September 
7,  1964).  The  ratio  of  the  length  to  the 
width  of  the  prosome  is  2.2  :  1.  The  separa- 
tion between  the  segment  of  leg  1  and  the 
head  is  rather  weak. 

The  segment  of  leg  4,  28  X  44  /x,  re- 
sembles that  of  the  female.  The  ventral 
intersegmental  sclerite  between  the  segment 
of  leg  5  and  the  genital  segment  is  absent. 
The  genital  segment  (Fig.  130)  is  elongated, 
90  X  73  /x,  with  its  sides  in  dorsal  view 
nearly  parallel.  The  4  postgenital  segments 
are  22  X  35,  22  X  33,  16  X  31,  and  19  X  30 
II  from   anterior  to  posterior. 

The  caudal  ramus  (see  Fig.  130)  is 
relatively  shorter  than  in  the  female.  The 
length  along  the  outer  edge  is  19  /x,  along 
the  inner  edge  18  /x,  and  its  width  is  14  /x, 
or  about  1.3  times  longer  than  wide. 

The  dorsal  surface  of  the  prosome  is 
unornamented.   The  dorsal  and  ventral  sur- 


New  Copepods  From  Annelids  in  Madagascar  •  Iltuncs  and  Ho        395 


faces  of  the  urosome  bear  a  few  hairs  and  genital  segment,  bearing  2  slender  naked 

refractile    points.     As    in    the    female,    the  setae  both  22  /a  long. 

prosome  is  longer  than   the   urosome,   the  The  spermatophore  (Fig.  135),  attached 

ratio  being  1.54  :  1.  to  the  female,  is  elongated,  74  X  32  [x,  not 

The  rostral  area  resembles   that   of  the  including  the  neck.    Spennatophores   may 

female.    The  ridge  seen  between  the  bases  be  attached  singly  or  in  pairs  with  a  com- 

of  the  second  antennae  in  the  female  ap-  mon  stalk  as  seen  in  N.  firmus. 

pears  to  be  absent.  The  color  in  life  resembles  that  of  the 

The  first  antenna  is  like  that  of  the  fe-  female, 

male.    The  second  antenna  (Fig.  131)  has  Comparison    with   other  species. — Naso- 

the  same  arrangement  of  spines,  setae,  and  molgns  riidis  may  be   separated   from   N. 

claws  as  in  the  preceding  2  species,  with  4  cristatus,   N.  firmus,  and  N.  leptus  by  the 

elements  on  the  third  segment  instead  of  length  of  the  caudal  rami.    In  addition,  it 

3  as  in  the  female.    The  seta  on  the  first  may  be  distinguished  from  N.  jirmus  in  that 

segment    has    lamellate    margins.     The    2  its  body  wall  is  less  strongly  sclerotized,  cer- 

large  claws  are  distinctly  jointed.    The  fine  t^i"    ^etae    on    the    second    antenna    have 

ornamentation  is  like  that  in  N.  leptus.  lamellate  margins,  the  first  segment  of  the 

The  labmm  resembles  that  of  the  female,  f^^"^^  ""f^^^  ^"  *^^^  female  has  a  broad 

1     .    .1       o    1               1.                 -^     ^    1  bulge  rather  than  a  prominent  conical  pro- 

but   the   1   large    setae   are   situated   more  .      .        .i      r          ^    \,.^      ^    ^               \    c 

,     .     ,       r^<^              ^.^  ^                    ,i          ^  icctiou,  the  formula  for  the  last  segment  ot 

posteriorly.    Ihe  mandible,  paragnath,  and  ',          ,         ,.    ,      o  •    t  tt  o     u      T\        tt  o 

i-.    .          .-,,               11      1          .      ?     r        T  the  endopod  in  leg  3  IS  1,11,2  rather  than  11,2, 

hrst  maxilla  resemble  those  in  the  fema  e.  ^j^^  .^^^^^^  ^.^^,^^  ^^^^  ^^  ^^^  ^.^^^  ^^g^^^t 

The  second   maxilla    (Fig.    lo2)    lacks  the  ^^  ^j^^  endopod  in  leg  4  is  much  shorter, 

bulge  on  the  outer  margin  of  the  first  seg-  ^^^^  ^j^^  ornamentation  of  the  claw  of  the 

ment,   but  instead   bears   groups   of  small  maxihiped   of   the   male   is   less   extensive, 

spinules;   the  spines   along  the  distal  end  Yrom  N.  leptus  it  may  also  be  distinguished 

of  the  second  segment  are  more  numerous  by  the  shorter  second  segment  in  the  endo- 

than  in  the  opposite  sex.    The  maxilliped  pod  of  leg  4. 
(Fig.  133)   resembles  generally  that  of  N. 

leptus,   the   recurved   claw   being  50  /x  in  Nasomolgus  parvulus^  n.  sp. 

length    (measured   along  its   axis).  Figs.  136-142 

The  postoral  area  is  like  that  of  the  fe-  Type  materiol— 14  females  from  one 
male.  SabeUostorte  magnifica  (Shaw),  in  2  m, 
Legs  1-4  have  the  same  spine  and  setal  at  Ambariotelo,  a  small  island  betw^een 
formula  as  in  the  female.  The  spines  on  Nosy  Be  and  Nosy  Komba,  Madagascar, 
the  rami  are  somewhat  longer  and  often  Collected  May  15,  1964.  Holotype  and  12 
straighter.  The  outer  spine  on  the  last  paratypic  females  deposited  in  the  United 
segment  of  the  endopod  of  leg  1  is  19  ^  long.  States  National  Museum,  Washington;  the 
The  3  spines  on  the  last  segment  of  the  remaining  paratypes  (dissected)  in  the  col- 
endopod  of  leg  2  are  18,  13,  and  9  /x  long  lection  of  A.  G.  Humes, 
from  distal  to  proximal;  those  on  the  last  Other  speeimens  (all  from  Sahellastaiie 
segment  of  the  endopod  of  leg  3  are  21,  moonifiea).—2  females  from  2  hosts,  under 
11,  and  10  fx  respectively.  The  second  seg-  intertidal  dead  coral,  Antsakoabe,  on  the 
ment  of  the  endopod  of  leg  4  is  relatively  northern  shore  of  Nosy  Be,  November  1, 
a  little  shorter  than  in  the  female,  being  1^63  (these  specimens  placed  in  the  Mu- 
31  X  10  u,  seum  of  Comparative  Zoology);  2  females 
Leg  5  resembles  that  of  the  female.  from  1  host,  under  intertidal  rock,  Ant- 
Leg  6  (Fig.  134)  consists  of  a  postero-  i  The  specific  name  parou/i«  (from  Latin  =  very 
lateral  flap  on  the  ventral  surface  of  the  small)  alludes  to  the  small  size  of  tliis  species. 


396        Bulletin  Museum  of  Coin])aratwe  Zoology,  Vol.  135,  No. 


sakoabe,  September  7.  1964;  and  1  female  erately  elongated,  280  x  128  /x,  and  contain 

from    1    host,    in     1   m,    Ambariobe,    near  fewer  eggs  than  in  N.  Icptus  or  N.  mdis, 

Ambariotelo,  October  10,  1960.  each  egg  being  44-50  fi  in  diameter. 

Female— The    body    (Fig.    136)    has    a  The   rostral   area,   first   antenna,    second 

broad  prosome.    The  length  (not  comiting  antenna,  labrum,  mandible,  paragnath,  and 

the  setae  on  the  caudal  rami)  is  0.57  mm  first   maxilla   resemble   those   of   N.    rudis. 

(0.49-0.54    mm)    and    the   greatest    width  The  second   maxilla    (Fig.    140)    lacks  the 

is  0.23  mm   (0.20-0.26  mm),  based  on   10  bulge  on  the  first  segment  seen  in  N.  rudis, 

specimens.    The  ratio  of  the  length  to  the  and  the  first  spine  in  the  row  on  the  distal 

width  of  the  prosome  is  1.35  :  1.'  The  seg-  part  of  the  second  segment  is  less  enlarged 

ment  of  leg  1  is  separated  from  the  head  than  in  that  species.    The  maxilliped  and 

by  a   dorsal   furrow.     The  epimeral   areas  postoral  area  resemble  those  of  N.  rudis. 

of  the  pedigerous  segments  resemble  those  Legs  1-4  have  the  same  spine  and  setal 

of  N.  leptus  and  N.  rudis.  fonnula  as  in  N.  Icptus  and  N.  rudis.    The 

The  segment  of  leg  5  is  similar  to  that  outer  distal  corner  of  the  first  segment  of 

in  N.  rudis.    Between" this  segment  and  the  the  endopod  of  legs  1  and  2  lacks  a  spinous 

genital   segment  there   is    a   small   ventral  process.    In  leg  3  such  a  process  is  absent 

intersegmental    sclerite.     The    genital    seg-  on  both  first  and  second  segments  of  the 

ment  (Fig.  137)  is  a  little  wider  than  long  endopod.   The  endopod  of  leg  4  (Fig.  141) 

and  in  dorsal  view  is  broadly  expanded  in  resembles  in  general  form  that  of  N.  rudis. 

its    anterior   three-fourths    and    constricted  The  first  segment  is  17  x  9  ^  and  bears  a 

in  its  posterior  fourth.    The  length  of  the  short  feathered  inner  distal  seta  11  ^  long, 

segment  is  72  ^u,  the  width  in  the  expanded  The  second  segment  is  31  X  7  /x  and  carries 

portion   83  ^,   and   the   width   in  the   con-  2  terminal  slightly  barbed  setae  37  ^  (outer) 

stricted  part  52  fi.    The  dorsolateral  areas  and  10  //.  (inner)  in  length.    The  relation- 

of  attachment  of  the  egg  sacs  are  located  ship  between  the  lengths  of  these  2  tenni- 

a   little  more  anteriorly  than  in   N.  rudis.  nal  setae  is  about  4  :  1  rather  than  about 

Each  area  (Fig.  138)  bears  2  naked  setae  2  :  1  as  in  N.  rudis.  The  fine  ornamentation 

29  and  10  ,j.  in  length.    The  3  postgenital  of  the  endopod  is  similar  to  that  in  the  3 

segments,    without    posterior   rows   of   spi-  previous  species, 

nules,  are  25  X  41,  19  X  .33,  and  24  X  31  ^  Leg  5  resembles  that  of  N.  rudis. 

from  anterior  to  posterior.  Leg  6  is  represented  by  the  2  setae  near 

The  caudal  ramus   (Fig.  139)   is  almost  the  attachment  of  the  egg  sacs, 

quadrate,  16  X  14  /x,  the  ratio  of  length  to  The  spermatophore  (Fig.  142),  attached 

width  being  1.14  :  1.   The  outer  lateral  seta  to  the  female  in  pairs,  is  77  X  31  /x,  not  in- 

is  35  IX  long,  the  pedicellate  dorsal  seta  24  eluding  the  neck. 

/x,   the   outermost  terminal   seta  40  /x,   the  The  color  in  life  in  transmitted  light  re- 
innermost   terminal   seta  44   /x,   and  the   2  sembles  that  of  the  3  previous  species, 
long  median  tenninal  setae  are  133  fx  (outer)  Male. — Unknown. 

and  237  /x   (inner)    and  are  basally  "peg-  Comparison    uitli    other  species. — Naso- 

ged."  All  the  setae  are  naked.    A  minute  molp^us    parvidus    may    be    distinguished 

hair  is  bome  on  the  dorsal  surface  of  the  from  all  other  species  in  the  genus  on  the 

ramus.  basis   of   its    small    size,    its   almost   square 

The  dorsal  surface  of  the  prosome  and  caudal   ramus,  and  the  two  very  unequal 

the  dorsal  and  ventral  surfaces  of  the  uro-  terminal   setae   on  the  second  segment   of 

some   bear   a   few   hairs.    The   prosome   is  the  endopod  of  leg  4.   In  addition,  it  differs 

longer  than   the  urosome,  the  ratio  being  from    xV.   firmus   in    having   a   much   more 

2.0  :  1.  weakly  sclerotized  body  wall,  in  the  rela- 

The  egg  sacs    (see  Fig.   136)    are  mod-  tive  lengths  of  the  prosome  and  urosome, 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho 


397 


in  its  stouter  egg  sac,  in  the  absence  of  a 
prominent  conical  projection  on  the  first 
segment  of  the  second  maxilla,  and  in  the 
armature  of  the  last  segment  of  the  endo- 
pod  of  leg  3.  From  N.  Jcptus  it  is  very 
easily  distinguished  by  its  much  less  elon- 
gated body,  in  having  a  shorter  egg  sac 
with  fewer  eggs,  and  in  lacking  the  sclero- 
tized  bulge  on  the  first  segment  of  the  sec- 
ond maxilla.  It  differs  furthennore  from 
N.  rudis  in  the  proportional  lengths  of  the 
prosome  and  urosome,  in  its  shorter  egg 
sac  with  fewer  eggs,  and  in  the  absence  of 
an  outer  bulge  on  the  first  segment  of  the 
second  maxilla. 

REMARKS  ON  THE  GENUS 
NASOMOLGUS 

\\'ith  the  finding  of  these  four  new  spe- 
cies of  Nasomolgus  living  on  a  sabellid 
polychaete,  it  seems  probable  that  mem- 
bers of  this  genus  customarily  live  in  asso- 
ciation with  annelids.  One  may  conjecture 
that  Sewell's  specimen  of  N.  cristatus, 
found  in  dredged  debris,  may  have  been 
dislodged  from  a  pohchaete  host,  but  this 
is  impossible  to  establish. 

More  than  one  species  of  Nasomolgus 
may  live  on  a  single  polychaete.  In  two 
instances  (collections  at  Ambariotelo,  May 
15,  1964,  and  at  Antsakoabe,  September  7, 
1964)  N.  firmiis,  X.  lepfiis,  and  N.  paividus 
occurred  on  a  single  Sabelkistarfc  mag- 
nifica.  In  one  case  (collection  at  Antsa- 
koabe, November  1,  1963)  all  four  species 
of  Nasomolgus  were  recovered  from  t\\'0 
hosts. 

The  exact  region  of  the  ])ody  where  the 
copepods  live  is  not  known.  It  is  possible 
that  each  species  of  copepod  occupies  a 
region  separate  from  the  others,  but  such  a 
supposition  can  only  be  substantiated  by 
careful  observation  of  undisturbed  li\ing 
hosts. 

Since  the  type  species,  N.  cristatus,  is 
now  known  to  possess  in  common  with  the 
four  new  species  from  Madagascar  certain 
fundamental  characters  (i.e.,  the  pair  of 
setae  on  the  anterior  part  of  the  labrum,  the 


3-segmented  maxilliped  with  a  long  slender 
last  segment,  and  the  number  of  spines 
[111,1]  on  the  last  segment  of  the  exopod  of 
leg  1 )  which  were  inadequately  mentioned 
in  Sewell's  original  description,  a  revision 
of  the  diagnosis  of  the  genus  Nasomolgus 
would  be  desirable.  However,  in  view  of 
the  fact  that  we  have  been  unable  to  make 
a  firsthand  study  of  the  single  specimen  of 
N.  crisfatus\  we  are  not  attempting  such  a 
revision  at  present.  There  are  certain  fea- 
tures (especially  the  structure  of  the 
mouthparts)  that  would  be  necessary  to 
clarify  before  undertaking  a  definitive  re- 
vision. For  the  moment,  the  characteristics 
given  by  Sewell,  together  with  the  addi- 
tions and  corrections  just  mentioned,  will 
sen/e  to  characterize  the  genus. 

REFERENCES  CITED 

BocQUET,  C,  AND  J.  H.  Stock.  1957.  Cope- 
podes  parasites  d'invertebres  des  cotes  de 
France.  IVa.  Le  double  parasitisme  de 
Sipunciilus  niidus  L.  par  Mtjzomolgus  stiipen- 
diis  nov.  gen.,  nov.  sp.,  et  Catinia  plana  nov. 
gen.,  nov.  sp.,  copepodes  cyclopoide^s  tres 
remarquables.  Koninkl.  Nederl.  Akad.  We- 
tensch.  Amsterdam,  Proc,  Ser.  C,  60(3): 
410-431. 

Brady,  G.  S.  1880.  A  monograph  of  the  free 
and  semi-parasitic  Copepoda  of  the  British 
Islands.    Ray  Society,  London,  3:   1-83. 

Canu,  E.  1892.  Les  Copepodes  du  Boiilonnais, 
morphologic,  embryologie,  taxonomie.  Trav. 
Lab.  Zool.  Mar.  Wimereux-Ambleteuse  (Pas- 
de-Calais),    6:    1-354. 

Claus,  C.  1889.  Uber  neue  oder  wenig  be- 
kannte  halbparasitische  Copepoden,  insbeson- 
dere  der  Lichomolgiden-  iind  Ascomyzontiden- 
Gruppe.  Arb.  Zool.  Inst.  Univ.  Wien,  8(3): 
327-370. 

Gallien,  L.  1935.  Psciidandiessitts-  nemerto- 
philiis  no\'.  sp.,  copcpode  commensal  de 
Linens  longiss'imus  Sowerby.  Bull.  Soc.  Zool. 
France,   60:    451-459. 

Gooding,  R.  U.  1963  (unpublished).  External 
morphology  and  classification  of  marine 
poecilostome  copepods  belonging  to  the  fami- 
lies Clausidiidae,  Clausiidae,  Nereicolidae, 
Eunicicolidae,  Synaptiphilidae,  Catiniidae, 
Anomopsyllidae,  and  Echiurophilidae.  Ph.D. 
Thesis,  Uni\ersity  of  Washington,  Seattle. 

GuRNEY,  R.  1927.  Zoological  results  of  the 
Cambridge  expedition  to  the  Suez  Canal, 
1924.     XXXIII.      Report    on    the    Crustacea: 


398 


Bulletin  Museum  of  Comparative  Zoology,  Vo].  13S,  No.  7 


Copepoda  (littoral  and  semi-parasitic).  Trans. 
Zool.  Soc.  London,  22(4):  451-477. 

Humes,  A.  G.  1966.  Pscudantliessius  proctirrens 
n.  sp.,  a  cyclopoid  copepod  associated  with  a 
cidarid  echinoid  in  Madagascar.  Breviora, 
Mus.   Comp.   Zool.,   No.   246:    1-14. 

Humes,  A.  G.,  and  R.  F.  Cressey.  1961.  Deux 
nou\ellcs  especcs  de  Pseudanthessiiis  (Cope- 
poda, Cyclopoida)  parasites  des  oursins  a 
Madagascar.  Mem.  Inst.  Sci.  Madagascar, 
Ser.  F,   1959,  3:   67-82. 

Illg,  p.  1950.  A  new  copepod,  Pseiidanthessius 
hitits  (Cyclopoida:  Lichomolgidae )  commen- 
sal w  ith  a  marine  flatworm.  Jour.  Washington 
Acad.  Sci.,  40(4):   129-133. 

LiNDBERG,  K.  1945.  Un  nouveau  copepode 
poecilostome  de  I'lnde  de  la  famille  des 
Lichomolgidae;  PseudontJicssiiis  spinifer,  n. 
sp.    Bull.  Soc.  Zool.  France,  70:  81-84. 

NicHOLLs,  A.  G.  1944.  Littoral  Copepoda  from 
South  Australia  (II).  Calanoida,  Cyclopoida, 
Notodelphyoida,  Monstrilloida  and  Caligoida. 
Rec.  So.  Austral.  Mus.,  8(1):   1-62. 

Sars,  G.  O.  1917.  An  account  of  tlie  Crustacea 
of  Norway  with  short  descriptions  and  figures 
of  all  the  species.  Vol.  6,  Copepoda,  Cyclo- 
poida, pts.  11  and  12,  Clausidiidae,  Lichomol- 
gidae (part),  pp.  141-172.  Bergen  Museum, 
Bergen. 


.      1918.     An  account  of  the  Crustacea  of 

Norway  with  short  descriptions  and  figures 
of  all  the  species.  Vol.  6,  Copepoda,  Cyclo- 
poida, pts.  13  and  14,  Lichomolgidae  (con- 
cluded ) ,  Oncaeidae,  Coiycaeidae,  Ergasilidae, 
Clausiidae.  Eunicicolidae,  Supplement,  pp. 
173-225.    Bergen   Museum,   Bergen. 

Scott,  A.  1909.  The  Copepoda  of  the  Siboga 
Expedition.  Part  I.  Free-swimming,  littoral 
and  semi-parasitic  Copepoda.  Siboga  Exped., 
29a:    1-323. 

Sewell,  R.  B.  S.  1949.  The  littoral  and  semi- 
parasitic  Cyclopoida,  the  MonstriUoida  and 
Notodelphyoida.  John  Murray  Exped.  1933- 
34,   Sci.   Repts.,  9(2):    17-199. 

Stock,  J.  H.,  A.  G.  Humes,  and  R.  U.  Gooding. 
1963.  Copepoda  associated  with  West  In- 
dian invertebrates  IV.  The  genera  Octopicola, 
Pseudanthessius  and  Meomicola  (Cyclopoida, 
Lichomolgidae).  Studies  Fauna  Curagao  and 
other  Caribbean  Is.,   18(77):    1-74. 

Thompson,  I.  C,  and  A.  Scott.  1903.  Report 
on  the  Copepoda  collected  by  Professor 
Herdman,  at  Ceylon,  in  1902.  Rept.  Govt. 
Ceylon  Pearl  Oyster  Fish.  Gulf  of  Manaar,  pt. 
I,  Suppl.  Rept.  No.  7:  227-307. 

Wilson,  M.  S.,  and  P.  L.  Illg.  1955.  The 
family  Clausiidae  ( Copepoda,  Cyclopoida ) . 
Proc.  Biol.  Soc.  Washington,  68:   129-142. 

{Received  December  10,  1965) 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho 


399 


Figures    1-7.      Cotylomolgus    lepidonoti    n.    gen.,    n.    sp.,    female.     1,    dorsal    (A);    2,    lateral    (A);    3,    urosome,    dorsal    (B);    4, 
urosome,    ventral    (B);    5,    area    of   attachment  of   egg    sac,    dorsal    (C);    6,    caudal    ramus,    dorsal    (D);    7,    egg    sac,    dorsal    (A). 


400        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  7 


Figures  8-17.  Cofy/omo/gus  lepidonoti  n.  gen.,  n.  sp.,  female  (continued).  8,  cephalosome,  ventral  (B);  9,  first  an- 
tenna, ventral  (E);  10,  second  antenna,  anterior  (E);  11,  second  antenna,  posterior  |E);  12,  labrum,  anterior  (C);  13, 
mandible,  lateral  (F);  14,  paragnath,  ventral  (F);  15,  first  maxilla,  anterior  (F);  16,  second  maxilla,  posterior  (E);  17, 
second    maxilla,    ventral    (E). 


New  Copepods  From  Annelids  in  Madagascah  •  Humes  and  Ho        401 


Figures  18-23.  Cofylomo/gus  lepidonoti  n.  gen.,  n.  sp.,  female  (continued).  18,  oral  area  posterior  to  labrum,  ventral 
(C);  19,  postoral  protuberance  between  maxillipeds  and  leg  1 ,  ventral  (G);  20,  leg  1  and  intercoxal  plate,  anterior  (E); 
21,    leg    2    and    intercoxal    plate,    anterior    (E);    22,    leg    3,    ventral    (D);    23,    leg    5,    dorsal    (G). 


402        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  7 


25 


28 


29 


Figures   24-29.      Cofy/omo/gus    lepidonofi    n.    gen.,    n.    sp.,    male.      24,    dorsal    (H);    25,     urosome,    dorsal     (B);    26,    urosome, 
ventral    (B);    27,    maxilliped,    posterior    (D);    28,    leg    5,    lateral    (D);    29,    spermafophore,    ventral     (G). 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        403 


Figures  30-38.  Pseudanthessiui  ferox  n.  sp.,  female.  30,  dorsal  (A);  31,  urosome,  dorsal  (B);  32,  urosome,  lateral  (B); 
33,  genital  segment,  ventral  (B);  34,  area  of  attachment  of  egg  sac,  dorsal  (D);  35,  enlargement  of  setae  on  egg  sac 
attachment    area,    lateral    (C);    36,    caudal    ramus,    dorsal    (D);   37,    cepholosome,    ventral    (H);    38,    first  antenna,    ventral    (G). 


404         Bulletin  Museum  of  Coiniuiidlirc  Zoology,  Vol.  135,  No.  7 


Figures  39-48.  Pseudonfhess/us  ferox  n.  sp.,  female  (continued).  39,  second  antenna,  anterior  (G);  40,  labrum,  ventral 
|D);  41,  mandible,  anterior  (D);  42,  first  maxilla,  anterior  (D);  43,  second  maxilla,  posterodorsal  (D);  44,  maxilliped,  an- 
terior (D);  45,  leg  1  and  intercoxal  plate,  anterior  (G);  46,  angular  lamella  on  basis  of  leg  1,  anterior  (C);  47,  leg  2, 
anterior   |G);    48,    leg    3,   anterior   (G). 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        405 


Figures    49-50.     Pseudanthesslus    ferox    n.    sp.,    female    (continued).     49,    leg    4    and    intercoxal    plate,    anterior    (G);    50,    leg 
5,  dorsal  (D). 

Figures    51-59.      Pseudonfhess/us    ferox    n.    sp.,    male.     51,    dorsal  (A);  52,    urosome,    dorsal    (B);    53,    last  segment   of  second 

antenna,     posterior     (C);    54,     second     maxilla,     posterodorsal    (D);  55,  maxilliped,     anterior    or    dorsal     (D);     56,     claw     of 

maxilliped,    anterior    or    dorsal     (D);    57,    leg    1,    anterior    (D);   58,  leg  6,    ventral    (D);    59,    spermatophore   inside    male,    ven- 
tral (E). 


406 


Bulletin  Mti.scuiii  of  Comparative  Zoology,  Vol.  135,  No. 


Figures  60-70.  Nosomo/gus  f'lrmus  n.  sp.,  female.  60,  dorsal  (H);  61,  urosome,  dorsal  (G);  62,  area  of  attachment  of 
egg  sac,  dorsal  (C);  63,  caudal  ramus,  dorsal  (F);  64,  egg  sac  (H);  65,  median  part  of  ceptialosome,  ventral  (E);  66, 
first  antenna,  ventrol  (D);  67,  second  antenna,  anterior  (C);  68,  mandible,  ventral  (C);  69,  paragnath,  ventral  (I);  70, 
first    maxilla,    anteroventral    (F). 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        407 


Figures  ly-ll .  Naiomolgus  firmus  n.  sp.,  female  (continued).  71,  second  maxilla,  posterodorsal  (C);  72,  mcxiiliped, 
anterior  (C);  73,  leg  1  and  intercoxal  plate,  anterior  (D);  74,  leg  2,  anterior  (D);  75,  leg  3,  anterior  (D);  76,  leg  4 
and    intercoxal    plate,    anterior    (D);    77,    leg    5,    dorsal    (F). 


Figures   78-80.      Nasomolgus   firmus    n.    sp.,    male.     78,    dorsal    (B);    79,    urosome,    dorsal    (E);    80,    caudol    ramus,    dorsal 


408        Bulletin  Museum  of  Cotiijxiiative  Zoology,  Vol.  135,  No.  7 


Figures  81-86.  Nasomo/gus  lirmus  n.  sp.,  male  (continued).  81,  second  antenna,  inner  (C);  82,  second  maxilla,  posterior 
(F);  83,  maxilliped,  posteromedial  (C);  84,  endopod  of  leg  1,  anterior  (C);  85,  leg  6,  ventral  (F);  86,  spermatophore, 
attached    to    female,    ventral    (D). 

Figures  87-91.  Nosomolgus  lepfus  n.  sp.,  female.  87,  dorsal  (A);  88,  urosome,  dorsal  (H);  89,  area  of  attachment  of 
egg  sac,  dorsal  (C);  90,  caudal  ramus,  dorsal  (G);  91,  rostral  area  and  labrum,  with  labral  lobes  erected  ventral!/ 
and    thus    foreshortened    in    the    dravving,    ventral    (E). 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho 


409 


Figures  92-101.  Nasomolgus  leptus  n.  sp.,  female  (continued).  92,  first  antenna,  dorsal  (D);  93,  second  antenna,  poste- 
rior (D);  94,  posterior  part  of  iobrum  with  bases  of  two  setae,  ventral  (C);  95,  mandible,  ventral  (C);  96,  first  maxilla, 
inner  (C);  97,  second  maxilla,  posterodorsal  (C);  98,  postoral  area,  ventral  (E);  99,  leg  1  and  intercoxal  plate,  anterior 
(E);    100,    leg    2,    anterior    (E);    101,    leg    3,    anterior    (E). 


410        Bulletin  Mmcuiti  of  Comparative  Zoology,  Vol.  135,  No.  7 


Figures    102-103.      Nosomo/gus    leptus    n.    sp.,    female    (continued).     102,    leg    4    and    intercoxol    plate,    anterior    (E);    103,    leg 
5,    dorsal    (F). 

Figures    104-109.      Nosomo/gus    /ep/us    n.    sp.,    male.     104,    dorsal   (H);  105,  urosome,  dorsal    (B);   106,   second  antenna,    poste- 
rior   (C);    107,    maxilliped,    posteromedial    (C);    108,    leg    6,    ventral    (C);    109,    spermatophore,    inside    male,    dorsal    (E). 

Figures    110-111.      Nosomo/gus    rudis    n.    sp.,    female.      110,    dorsal    (H);    111,    urosome,    dorsal    (G). 


New  Copepods  From  Annelids  in  Madagascar  •  Humes  and  Ho        411 


Figures  112-124.  Nosomo/gus  rudis  n.  sp.,  female  (continued).  112,  area  of  attacfiment  of  egg  sac,  dorsal  (C);  113, 
caudal  ramus,  dorsal  (F);  114,  egg  sac,  dorsal  (H);  115,  rostral  area  and  labrum,  with  labral  lobes  erected  ventrally 
and  thus  foreshortened  in  the  drawing,  ventral  (E);  116,  first  antenna,  dorsal  (D);  117,  second  antenna,  posterior  (D);  118, 
posterior  part  of  labrum,  ventral  (C);  119,  mandible,  dorsal  (C);  120,  first  maxilla,  ventral  (C);  121,  second  maxilla, 
posterior    (C);    122,    maxilliped,    posterior    (C);    123,    postoral    area,   ventral    (E);    124,    leg    1    and    intercoxal    plate,    anterior    (E). 


412        Bulletin  Museum  uf  Comparative  Zoology,  Vol.  135,  No.  7 


Figures    125-128.      Nosomo/gus    rudis    n.    sp.,    female    (continued).     125,    leg    2,    anterior    (E);    126,    leg    3,    anterior    (E);    127, 
leg  4  end  intercoxal  plate,  anterior  (E);   128,  leg  5,  dorsal   (F). 

Figures    129-132.      Nosomo/gus    rudis    n.    sp.,    male.      129,    dorsal    (B);    130,    urosome,  dorsal   (E);    131,  second   antenna,   poste- 
rior (C);   132,   second  maxilla,   anterior  (I). 


New  Copepods  From  Annelids  in  Madagascar  •  Humeri  and  Ho        413 


136 


134 


135 


142 


137 


138 


140 


141 


Figures  133-135.  Nasomolgus  rudis  n.  sp.,  male  (continued).  133,  maxilliped,  posteromedial  (C);  134,  leg  6,  ventral 
(F);    135,    spermatophore,    attached    to    female,    dorsal    (D). 

Figures  136-142.  Nasomolgus  parvulus  n.  sp.,  female.  136,  dorsal  (H);  137,  urosome,  dorsal  (D);  138,  area  of  attach- 
ment of  egg  sac,  dorsal  (C);  139,  caudal  ramus,  dorsal  (I);  140,  second  maxilla,  posterior  (F);  141,  endopod  of  leg  4, 
anterior    (C);    142,    spermatophores,    attached    to    female,    ventral    (D). 


llili^ii;;l^yi::^^'G:';;'-v^^ 


BuLiQtln     OF     THE 

Museum   of 

omparative 
Zoology 


'„-,  vvf'-'';,- 


Proterochampsa    barrionuevoi    and    the    Early 
Evolution    of    the    Crocodilia 


WILLIAM  D.  SILL 


HARVARD   UNIVERSITY  VOLUME    135,   NUMBER   8 

CAMBRIDGE,  MASSACHUSETTS,   U.S.A.  APRIL  24,    1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

Memoris  1864-1938 

JoHNSONLV,  Department  of  Mollusks,  1941- 

OccASioNAL  Papers  on  Mollusks,  1945- 

Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects. $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-list  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(Mollusca:  Bivalvia).    $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.   $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.   ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©  The   President  and    Fellows  of   Harvard    College    1967. 


PROTEROCHAMPSA  BARRIONUEVOI  AND  THE 
EARLY  EVOLUTION   OF  THE  CROCODILIA 


WILLIAM   D.   SILL 


INTRODUCTION 

During  the  months  April  through  June 
of  1958  a  joint  expedition  of  the  Museo 
Argentino  de  Ciencias  Naturales  and  the 
Museum  of  Comparative  Zoology  explored 
continental  deposits  in  the  Province  of  Men- 
doza  and  San  Juan  in  western  Argentina. 
The  last  six  weeks  of  the  field  season  were 
spent  in  Triassic  beds  at  Ischigualasto,  a 
\'alley  in  the  northeastern  part  of  the  prov- 
ince of  San  Juan.  Fossils  at  this  locality 
proved  to  be  so  abundant  and  so  easily 
found  that  Romer  (1962)  has  described  it 
as  a  "paleontologist's  dream."  The  crocodil- 
ians  described  in  this  paper  were  found 
there  by  Professor  Bryan  Patterson  in  the 
upper  third  of  the  Ischigualasto  Formation, 
approximately  160  feet  from  the  base.  The 
formation  consists  of  interbedded  clays, 
shales,  and  some  sandstone,  characterized 
by  the  variegated  green,  white,  brown,  and 
red  colors  t\'pical  of  so  many  fossil  bearing 
continental  deposits.  As  regards  age,  Fren- 
guelli  (1948)  considered  the  formation  to 
be  upper  Keuper;  Romer  (1960,  1962) 
states  that  it  is  certainly  pre-Norian  and 
probably  pre-Carnian.  Gomphodont  cyno- 
donts  and  rhynchosaurs  were  found  in 
abundance  in  the  formation  in  the  same 
general  area  as  the  material  here  discussed. 

The  following  year  the  Instituto  Miguel 
Lillo  of  the  Universidad  de  Tucuman  sent 
two  expeditions  to  the  area.  Under  the 
direction  of  Dr.  Osvaldo  A.  Reig  a  number 
of  specimens  were  found,  some  of  which 


have  already  been  described  (Reig,  1959; 
Casamiquela,  1960;  Bonaparte,  1962,  1963). 
Reig  (1959)  published  a  preliminary  ac- 
count of  the  ancestral  crocodilian  discussed 
here,  giving  it  the  name  Pwterochampsa 
harrionuevoi. 

The  material  discussed  in  this  paper  con- 
sists of  one  complete,  well  preserved  skull 
(but  with  parts  of  the  ventral  area  badly 
fractured),  together  with  13  articulated 
vertebrae  and  ribs,  MCZ  3408,  and  one 
partial  skull,  MACN  18165  (Museo  Argen- 
tino de  Ciencias  Naturales). 

I  am  obliged  to  Arnold  D.  Lewis  for 
preparation  of  the  material,  to  Dr.  Bernhard 
Kummel  for  the  photography,  to  Dr.  Edwin 
Colbert  and  The  American  Museum  of 
Natural  History  for  permission  to  examine 
Protosuchus,  to  Yale  Peabody  Museum  for 
making  available  various  mesosuchians  for 
comparative  study,  and  to  Dr.  K.  A.  Ker- 
mack  of  University  College  London  for  al- 
lowing me  to  examine  the  primitive  croco- 
dilian from  Wales.  The  manuscript  has 
been  read  by  Professors  Bryan  Patterson, 
Ernest  Williams,  and  Alfred  Romer.  To  all 
of  these  people  I  express  my  sincere  thanks. 
The  expedition  was  supported  in  part  by 
the  National  Science  Foundation,  and  in 
part  by  Life  Magazine. 


Bull. 


MORPHOLOGY 

THE  SKULL 

General  remarks:    The  skull  of  Protero- 
champsa  presents   a  remarkable  combina- 

Mus.  Comp.  Zool..  135(8):  415-446,  April,  1967        415 


416        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  8 


tion  ot  priiiiitixe  thecodont,  acKauctd  croc- 
odilian, and  transitional  charactcis.  The 
dorsal  surface  resembles  that  of  a  modern 
crocodile  in  the  highly  sculptured  surface, 
large  dorsally  placed  orbits,  small  supra- 
temporal  fenestrae,  and  external  nares  near 
the  midline  of  the  snout.  The  skull  is  flat, 
relatively  broad,  and  has  a  long  snout.  The 
sculpturing  of  the  cranial  table  is  note- 
\vorth\-  in  that  the  rugose  ridges  usually 
follow  a  pattern,  differing  in  this  respect 
from  the  random  pitted  type  of  sculpturing 
found  in  later  crocodiles.  Most  of  the 
transitional  characters  are  in  the  palatal 
region.  The  presence  of  relatively  large 
antorbital  fenestrae  and  of  long  curved 
teeth  may  be  considered  transitional  or 
primitive.  Sutures  on  the  dorsal  side  of  the 
skull  are  generalh  well  preserved;  on  the 
underside,  however,  it  is  much  more  diffi- 
cult to  determine  the  bone  pattern  due  to 
the  fractured  nature  of  the  region. 

The  occipital  face  of  the  skull  of  Protero- 
cliampsa  resembles  that  of  a  somewhat  flat- 
tened version  of  a  modern  crocodile.  Al- 
though the  skull  is  35  centimeters  from  the 
occipital  condyle  to  the  tip  of  the  snout,  it 
is  only  4  centimeters  high  from  the  base 
of  the  occipital  condyle  to  the  top  of  the 
parietal.  Of  course,  some  allowance  must 
be  made  for  flattening  and  deformation  in 
the  process  of  fossilization,  but  the  general 
aspect  of  length  to  depth  remains  the  same 
(a  modern  crocodile  of  comparable  size 
measured  only  6  centimeters  in  depth  from 
condyle  to  parietal). 

The  ventral  portion  of  the  skull  is  by  no 
means  as  well  preserved  as  the  dorsal; 
fortunate!),  however,  the  position  of  the 
internal  nares  and  the  limits  of  the  second- 
ary palate  are  quite  clear.  The  basisphe- 
noid,  the  pterygoid  flange  and  its  tooth 
row,  and  the  ectoj^terygoid  relationships 
are  clearly  seen.  The  rest  of  the  basicranial 
region  is  crushed  and  distorted.  Th{>  inter- 
pretation presented  here  necessarily  con- 
tains an  element  of  the  speculative. 

Prcmoxilki.  The  prcinaxilla  occupies  the 
anterior  end   ol    the   snout   extending  back 


from  the  "canine  notch"  toward  the  midline, 
forming  an  inverted  V-shaped  suture  with 
the  nasal  and  the  maxilla.  Ventrally,  the 
premaxilla  folds  over  and  joins  at  the  mid- 
line to  form  the  anterior  part  of  the  sec- 
ondar\'  palate.  The  ventral  suture  of  the 
maxilla  and  premaxilla  is  not  visible,  nor 
is  it  possible  to  verify  the  presence  of  an 
incisive  foramen.  The  premaxilla  bears  six 
tc>eth.  The  shape  and  sutural  relations  of 
the  dorsal  side  of  this  bone  are  very  similar 
to  those  of  a  modern  crocodile  (see  Plate 
V). 

Maxilla.  The  maxilla  extends  posteriorly 
and  laterally  from  the  "canine  notch," 
forms  the  anterior  border  of  the  antorbital 
fenestra,  and  joins  the  nasal  medially  and  the 
jugal  posteriorly.  The  maxilla  in  MCZ  3408 
bears  eleven  teeth,  and  that  of  MACN  1S165 
at  least  twelve.  Like  the  premaxilla,  the 
sutural  configuration  of  the  maxilla  is  very 
much  like  that  of  modern  crocodiles  (see 
Plate  V).  \^entrally,  the  maxilla  folds  over 
to  join  \\'ith  the  premaxilla  in  the  forma- 
tion of  the  secondary  palate.  In  this  re- 
gion there  is  no  definite  border  delimiting 
the  maxilla  with  relation  to  the  internal 
nares  and  the  palatine  bones.  There  is, 
however,  a  slight  difference  in  the  color  of 
the  l)one,  \\'hich  is  symmetrical  on  both 
sides  of  the  midline  and  has  been  taken  as 
the  probable  boundary  of  the  maxilla  with 
the  palatine.  The  maxillae  are  compara- 
tively small  ventrally,  extending  clown 
from  the  posterolateral  side  of  the  upper 
jaw,  meeting  at  the  midline  with  the  pre- 
maxillae,  with  a  small  process  entering  into 
the  anterior  border  of  the  internal  nares. 

Nasal.  The  nasal  bone  of  Proterochampsa 
forms  a  horizontal  plate  extending  from  an 
inverted  V-shaped  posterior  border  with 
the  frontal  and  prefrontal  to  an  anterior 
V-shaped  suture  with  the  premaxilla.  A 
prominent,  sculptured  ridge  runs  longitu- 
dinally down  the  medial  side  of  each  nasal. 
Ventrally,  the  relationships  are  obscure,  al- 
though it  appears  that  there  is  a  contact  of 
the  vomer  with  the  nasal  anterior  to  the  in- 
ternal naris.    The  nasal  extends  somewhat 


Proterochampsa   and  crocodile  evolutiox  •  S;7/   417 


Fig.    1.      Dorsal   and  ventral   views  of  skull   of  Proterochampsa  barrionuevoi,   MCZ   3408.     X    '/V    Abbreviations  on   page  436. 


418         BuUctiu  Museum  of  Comparative  Zoology,  Vol  135,  No.  8 


laterally  at  the  antorbital  fenestra,  of  whicli 
it  forms  the  medial  border. 

Prefrontal.  T1k>  suture  of  the  prefrontal 
with  the  nasal  is  not  elearly  visible;  it  is 
inferred  from  the  ehange  in  pattern  of  the 
bone  and  the  orientation  of  the  seulpturing 
in  this  Avca.  The  prefrontal  stands  out  as 
a  highh-  seulptured  triangular  bone  forming 
the  anteromedial  border  of  the  orbit  and 
the  posteromedial  border  of  the  antorl^ital 
fenestra.  At  the  border  with  the  lacrimal, 
the  prefrontal  is  also  marked  by  the  strong 
seulptured  crest  extending  around  all  but 
the  lateral  one-third  of  the  orbit  (see  dis- 
cussion of  orbit  below).  The  prefrontal  is 
bordered  medially  by  the  nasal,  anteriorl\- 
b\-  the  antorbital  fenestra,  laterally  by  the 
lacrimal,  and  posteriorly  In-  the  orbit  and 
a  small  part  of  the  frontal. 

Laerimal.  The  lacrimal  of  Prutcro- 
ehiunpsa  is  smaller  than  in  later  crocodiles. 
It  is  a  triangular,  lightly  sculptured  bone 
bordered  by  the  jugal  laterally,  the  pre- 
frontal medially,  the  antorbital  fenestra  an- 
teriorly, and  the  orbit  posteriorly.  It  is  dis- 
tincti\e  in  not  having  an  orbital  crest  along 
its  border  with  the  orbit.  The  lacrimal  in- 
clines somewhat  laterally;  it  is  primitive  in 
that  it  extends  around  a  large  part  of  the 
lateral  margin  of  the  orbit,  but  advanced  as 
regards  small  size. 

Frontal.  The  frontals  are  fused  at  the 
midline  but  there  is  still  an  indication  of  a 
suture  between  them.  The  sculpture  pat- 
tern of  the  frontals  is  distinctive  in  consist- 
ing of  transverse  ridges  forming  peaks  at 
the  midhne,  slanting  laterally,  then  sweep- 
ing up  at  the  margin  of  the  orbit  to  join  in 
the  orbital  crest.  The  conjoined  frontals 
have  a  triangular  shape  and  look  rather  like 
an  arrowhead  pointing  down  the  snout. 
They  are  bordered  anteriorly  by  the  nasals 
and  the  prefrontals,  posteriorly  by  the  pari- 
etals,  and  laterally  by  the  orbits. 

Parietal.  The  parietals  are  completely 
fu.sed,  ^\  ith  no  trace  at  all  of  a  suture  sep- 
arating them.  They  are  slightly  concave  at 
the  midline  anterior  to  the  supratemporal 
fenestrae,  and  slope  upward  posteriorly  to 


form  part  of  the  occipital  crest.  Small 
ridges,  not  as  prominent  as  those  of  the 
frontals,  radiate  out  from  the  center  of  the 
fused  parietals,  becoming  quite  prominent 
posteriorly  near  the  occipital  crest.  The 
parietals  are  roughly  T-shaped,  with  the 
crossbar  of  the  T  forming  part  of  the  oc- 
cipital crest.  They  are  bordered  postero- 
laterally  by  the  squamosals,  posteriorly  by 
the  supraoccipital  and  the  exoccipitals,  lat- 
erally b\  the  postorbitals  and  anteriorly  by 
the  frontals.  A  very  small  lateral  part  en- 
ters the  orbit,  and  the  entire  posterior 
border  of  the  supratemporal  fenestra  is 
formed  by  the  anterior  edge  of  the  crossbar 
portion  of  the  bone. 

Postorhital.  The  postorbital  is  a  massive, 
heavily  sculptured  bone  forming  the  greater 
part  of  the  posterior  margin  of  the  orbit. 
It  is  bordered  medially  by  the  parietal, 
posteriorly  by  the  squamosal,  and  laterally 
b\'  the  jugal.  The  postorbital  participates 
in  the  formation  of  the  orbit  anteriorly,  the 
supratemporal  fenestra  posteromedially, 
and  forms  the  medial  edge  of  the  infratem- 
poral fenestra  laterally.  Together  with  a 
medial  extension  of  the  jugal  it  forms  the 
postorbital  bar.  Although  the  form  of  the 
bar  is  very  similar  to  that  of  modern  croc- 
odiles, it  remains  entirely  on  the  dorsal  sur- 
face of  the  skull. 

Squamosal.  The  squamosal  forms  the 
principal  part  of  the  occipital  crest.  It  is  a 
strong,  massive  bone  with  highly  sculp- 
tured, very  prominent  ridges  extending  lat- 
erally along  the  occipital  crest.  Other  ridges 
extend  diagonally  from  the  postorbital  in 
front  of  the  supratemporal  fenestra  and 
back  across  the  squamosal  to  the  lateral 
edge  of  the  cranial  table.  Reig  called  spe- 
cial attention  to  this  diagonal  ridge;  on  the 
cast  of  the  type  specimen,  which  he  kindly 
sent  to  the  Museum  of  Comparative  Zo- 
ology, this  crest  is  much  more  prominent 
than  on  the  specimens  here  described.  The 
sc^uamosal  rests  posteriorly  on  top  of  the 
exoccipital;  laterally  it  meets  the  quadrato- 
jugal  nearly  horizontally,  with  strong  sculp- 
turing present  on  both  bones.  A  small  proc- 


Proterochampsa  and  crocodile  evolution  •  Sill        419 


ess  of  the  squamosal  extends  clown  the 
lateral  border  of  the  exoccipital.  Postero- 
lateral!)' the  quadrate  s\\'eeps  up  to  form 
a  steep,  smooth  suture  directly  underneath 
the  squamosal.  The  squamosal  forms  a 
small  part  of  the  lateral  border  of  the  supra- 
temporal  fenestra  and  the  largest  part  of 
the  posterior  border  of  the  infratemporal 
fenestra.  It  is  bordered  by  the  parietal,  ex- 
occipital,  quadrate,  quadratojugal,  and  post- 
orbital. 

Jiigal.  The  position  of  the  jugal  is  quite 
different  from  that  of  the  mesosuchians  and 
eusuchians  and  much  more  like  that  of 
some  of  the  early  thecodonts.  Unlike  croc- 
odiles in  general,  the  jugal  of  Protero- 
champsa fomis  only  a  minor  part  of  the  or- 
bit, being  ventral  to  the  lacrimal  for  most 
of  the  orbital  area.  The  jugal  extends  for- 
ward from  its  border  ^^'ith  the  quadratojugal 
to  form  most  of  the  lateral  border  of  the  in- 
fratemporal fenestra;  it  sends  a  process  half- 
way up  between  the  orbit  and  the  infra- 
temporal fenestra  to  meet  the  postorbital 
and  to  form  the  postorbital  bar,  \\'hich.  as 
just  described,  is  not  displaced  downward 
from  the  cranial  surface  (see  Fig.  2).  The 
jugal  forms  a  small  part  of  the  lateral  bor- 
der of  the  orbit,  meets  the  lacrimal  and 
continues  lateral  to  and  beyond  it  to  form 
the  lateral  border  of  the  antorbital  fenestra. 
The  jugal  is  in  general  more  lightly  sculp- 
tured than  most  of  the  dorsal  skull  bones. 
Together  with  the  lacrimal,  it  foiTns  that 
part  of  the  orbit  which  does  not  ha\'e  a 
prominent  raised  crest.  The  part  that  forms 
the  bar  below  the  infratemporal  fenestra  is, 
however,  quite  massive  and  highly  sculp- 
tured. 

On  the  palate  the  sutures  of  the  jugal 
are  not  fully  discernible.  However,  the 
ectopterygoid  can  be  seen  extending  from 
the  jugal,  and  the  suture  with  the  quad- 
ratojugal can  be  seen.  The  border  with  the 
maxilla  and  the  exact  location  of  the  sutiue 
with  the  ectopteiA'goid  are  not  \isible. 

Quadratoiuiial.  Like  the  jugal,  the  quad- 
ratojugal resembles  that  of  its  thecodont 
ancestors  much  more  closely  than  it  does 


m 


tfi 


I 


)-. 


k  ^ 


^^, 


(< 


t- 


^ 


r^ 


X 


oo 
o 

n 
U 


D 
-Q 


a 

E 

D 


o 
«t 

"o 
'o 

CN 

6> 


420         Bulletin  Museum  of  Coniparativc  Zoology,  Vol.  135,  No.  8 


Fig.  3.      Occipital  view  of  Proterochampsa  barnonuevoi,  MCZ  3408.     X    Vl- 


that  of  later  crocodiles.  It  i.s  miicli  larger 
than  the  modern  crocodilian  ([iiadratojugal 
and  forms  the  posterolateral  corner  of  the 
dorsal  surface  of  the  skull.  Anteriorly,  it 
joins  with  the  jugal,  forms  the  posterolateral 
margin  of  the  infratemporal  fossa,  and 
unites  medially  with  the  squamosal.  Pos- 
teriorly and  medially,  the  quadratojugal  has 
a  smooth  face  which  joins  with  the  c^uad- 
rate. 

On  the  palatal  surface,  the  suture  of  the 
quadratojugal  and  the  jugal  runs  diagonally 
forward.  Its  other  ventral  contact  is  with 
the  quadrate,  on  which  it  sits  like  a  cap 
with  the  lateral  edge  folded  under.  The 
dorsal  and  lateral  parts  of  the  bone  are 
sculptured  to  about  the  same  extent  as  the 
jugal,  being  more  massive  along  the  infra- 
temporal bar.  The  posterior  face  of  the 
bone  is  smooth  and  quite  steeply  inclined 
down  to  its  jimction  with  the  quadrate. 

Quadrate.  In  its  posterior  aspect  the 
quadrate  of  Frotcrochampsa  resembles  that 
of  advanced  crocodilians  in  extending  up- 
ward and  medially  with  a  smooth  sloping 
face  to  join  the  exoccipital  and  the  squamo- 
sal; it  is  not,  howeven-,  overlain  by  the  lat- 
ter as  in  eusuchians.  Further,  it  is  quite 
unlike  Recent  crocodiles  in  having  the  ex- 
ternal auditor)'  meatus  in  the  form  of  a 
groove  going  into  the  inner  ear  along  the 
posterodorsal  part  of  the  quadrate,  anterior 
to  the  exoccipital  and  underneath,  but  not 
enclosed  by,  the  scjuamosal  (see  discussion 
of  ear  region  below).  The  configuration 
of  the  articulating  condyles  of  the  quadrate 
is  similar  to  that  of  some  of  the  mesosuchian 


crocodiles  [Teleosauridae,  Libycosuchidae, 
and  Metriorhynchidae  (Kiilin,  1955)]  in 
the  presence  of  two  condyles,  the  medial 
larger  than  the  lateral.  This  is  in  contrast, 
on  the  one  hand,  to  the  usual  thecodont 
condition  of  only  one  condyle  and,  on  the 
other,  to  the  Recent  crocodilians,  which 
have  two  condyles  but  the  lateral  larger 
than  the  medial. 

Ventrally,  the  posterior  edge  of  the  quad- 
rate foiTns  a  vertical  ridge  running  diago- 
nally from  the  articulating  surface  medially 
to  join  the  posterior  border  of  the  ptery- 
goid. The  lateral  border,  covered  dorsally 
by  the  quadratojugal,  remains  horizontal 
and  gives  rise  to  the  vertical  ridge  previ- 
oush'  mentioned.  In  size,  the  quadrate  is 
more  like  that  of  the  thecodonts  than  of 
the  crocodilians,  but  in  shape  it  is  transi- 
tional between  the  two. 

Stipraoccipital.  The  supraoccipital  in 
Frotcrochampsa  is  a  subtriangular,  smooth 
bone  lying  vertically  on  the  posterior  face 
of  the  skull,  just  under  the  fused  parietals, 
much  as  in  the  modern  crocodiles.  It  is 
bordered  dorsally  by  the  parietals,  laterally 
by  the  exoccipitals,  and  ventrally,  just 
above  the  foramen  magnum,  by  a  thin  ex- 
pression of  the  exoccipital. 

Exoccipital.  The  exoccipital  forms  all  but 
the  most  ventral  portion  of  the  border  of 
the  foramen  magnum;  it  extends  outward 
to  underlie  the  lateral  part  of  the  parietal 
and  the  posterior  part  of  the  squamosal. 
The  small  occipital  process  of  the  squamosal 
forms  the  lateral  liorder  of  the  exoccipital. 
Laterally     and    ventrally    the     exoccipital 


Proterochampsa  and  crocodile  evolution  •  Sill        421 


B 


Fig.    4.      External    ear  structure   in  A,   Chosmofosourus    (proterosuchian   thecodont) 
suchian  crocodile);    and   D,   Crocody/us. 


B,  Proterochampsa;  C,   Pe/ogosourus   (meso- 


joins  with  the  quadrate.  Ventrally,  on  the 
posterior  face  of  the  skull,  it  joins  with  the 
basioccipital. 

Basioccij)ital.  The  basioccipital  forms  the 
entire  occipital  condyle  and  the  most  ven- 
tral border  of  the  foramen  magnum.  Ven- 
trally, it  is  fused  completely  with  the  basi- 
sphenoid,  has  a  subrounded  shape,  resem- 
bling the  thecodont  rather  than  the  eusu- 
chian  condition  and  is  \\'ithout  the  basioc- 
cipital processes  common  in  mesosuchians. 

Ear  structure.  Frotcrochampsa  shows  the 
beginning  of  the  acquisition  of  the  otic 
notch  in  crocodiles,  and,  with  the  meso- 
suchians, provides  a  phylogenetic  sequence 
for  the  e\'olution  of  the  unusual  crocodilian 
ear  structure  (Fig.  4).  In  Frotcrochampsa 
there  is  a  meatal  groove  running  from  the 


border  of  the  quadrate  and  squamosal, 
passing  anterior  to  the  exoccipital  and  into 
the  inner  ear.  Haughton  ( 1924 )  mentions 
a  similar  groove  in  Notochampsa.  In  Froto- 
suchus  there  is  no  evidence  of  either  a 
groove  or  an  otic  notch. 

In  the  earliest  thecodonts,  there  is  no 
indication  of  an  otic  notch  ( later  thecodonts 
do  possess  one,  although  it  is  usually  formed 
completely  within  the  squamosal),  and  the 
quadrate  is,  as  a  rule,  completely  vertical. 

In  mesosuchians  the  otic  notch  is  usuallv 
quite  prominent,  fomied  by  the  squamosal 
projecting  out  over  the  quadrate  (see  Fig. 
4C).  In  modern  crocodiles  the  squamosal 
has  extended  backward  and  downward 
onto  the  quadrate  to  close  the  otic  notch 
and  gain  a  broad  posterior  contact  \\\\h  the 


422         Bulletin  Museum  of  Coinparativc  Zoology,  Vol.  135,  No.  8 


Fig.    5.      Comparison    of    the    external    auditory    meatus    in    Crocody/us     (/eft)     (X-ray,     after     Edinger,     1938),     and     Protero- 
zhampsa    [right]. 


quadrate,  forming,  M'ith  the  exoccipital, 
the  deep  pocket  of  the  external  and  middle 
ear  structures.  This  closing  of  the  otic 
notch  has  been  accompanied  by  a  displace- 
ment of  the  tympanic  cavity  laterally  from 
the  braincase.  This  is  already  shown  by 
some  late  Cretaceous  mesosuchians  of  the 
family  Notosuchidae  (Kalin,  1955),  in 
which  the  quadrate  has  become  more  pos- 
teriorly inclined  (and  the  skull  much  more 
flattened).  In  this  respect  the  quadrate  of 
Proteroc]uunj)sa  resembles  that  of  meso- 
suchians in  both  inclination  and  height. 

Although  the  acciuisition  of  the  otic 
notch  appears  to  iollow  the  developmental 
sequence  outHned  above,  it  should  be  em- 
phasized that  this  sequence  is  based  more 
on  external  form  than  on  a  detailed  com- 
parative^ study  of  the  osteology  of  the  ear 
region  of  fossil  crocodilians. 

Dorsal  opemn^s  of  the  skull.  The  external 
nares  lie  on  either  side  of  the  midline,  sep- 
arated at  least  in  part  (and  probably  en- 
tirely) by  the  slender  tip  of  the  paired 
nasals.  They  are  appioximately  five  cen- 
timeters long,  two  and  a  hall  centimeters 
across,  and  oval  shaped.  They  are  com- 
pletely enclosed  by  the  premaxillae,  except 
for  the  nasal  process  which  separates  them. 


The  antorbital  fenestrae  are  oval-.shaped 
openings,  somewhat  wider  posteriorly  than 
anteriorly.  Like  the  external  nares,  they  are 
completely  dorsal.  They  are  bordered  by 
the  maxilla,  nasal,  prefrontal,  lacrimal,  and 
jugal  (as  previously  described). 

The  orbits  of  Froicrochampsa  are  quite 
like  those  of  modern  crocodiles  in  being 
almost  completely  within  the  horizontal 
plane  of  the  skull  (there  is  a  small  lateral 
angle  downward  as  in  modern  crocodiles). 
However,  as  in  thecodonts  and  early  croc- 
odilians, the  postorbital  bar  is  still  at  the 
dorsal  surface  of  the  skull  and  the  lacrimal 
bone  plays  an  important  part  in  the  border 
of  the  orbit.  The  orbit  is  bordered  by  a 
very  strong  crest  that  extends  around  all 
but  the  anterolateral  one-third  of  the  cir- 
cmnference.  Anteriorly,  the  crest  stops  at 
the  lateral  edge  of  the  prefrontal  part  of 
the  border  to  run  down  the  prefrontal  to 
the  antorbital  fenestra.  Posteriorly,  the  en- 
circling crest  stops  at  the  edge  of  the  post- 
orbital,  halfway  down  the  postorbital  bar, 
and  another  crest  goes  off  diagonally  at 
the  beginning  of  the  postorbital  bar  across 
the  postorbital  and  squamosal  bones. 

The  supratemporal  openings  are  consid- 
erably smaller  than  the  infratemporal  open- 


Proterochampsa  and  crocodile  evolution  •  Sill        423 


ings,  as  is  the  case  in  Recent  Crocodilia. 
This  condition  is  probably  primitive  and, 
as  indicated  by  Colbert  and  Mook  (1951), 
was  lost  in  the  mesosnchians  and  reacquired 
in  the  eusuchians.  In  Proterochampsa  these 
openings  are  nearly  horizontal,  slanting 
slightly  downward  towards  the  orbits  from 
the  prominent  crests  of  the  parietal. 

The  infratemporal  openings  of  Protero- 
champsa are  much  larger  than  those  of 
Recent  crocodiles,  being  about  as  large 
relative  to  the  orbits  as  in  thecodonts.  In 
the  latter,  however,  they  are  vertical, 
whereas  in  Proterochampsa  they  lie  at  about 
a  45°  angle. 

Secondary  palate.  The  secondary  palate 
of  Proterochampsa  is  of  great  interest.  As 
mentioned  above,  it  consists  of  the  pre- 
maxillae,  the  maxillae,  and  possibly  a  small 
part  of  the  palatines.  The  premaxillae  and 
the  maxillae  simply  extend  over  to  the  mid- 
line, and  the  internal  nares  open  at  the 
posterior  border  of  the  maxillae.  They  are 
bordered  laterally  by  the  palatines  and  sep- 
arated by  the  vomers.  This  arrangement 
provides  an  almost  perfect  transition  be- 
tween the  thecodont  type  of  palate  and  the 
mesosuchian,  in  which  the  internal  nares 
have  moved  back  to  the  posterior  border 
of  the  palatines. 

Palatine.  The  palatine  bones  cannot  be 
clearly  distinguished  in  the  two  skulls  avail- 
able for  study.  However,  on  the  basis  of 
a  slight  color  and  textural  difference  of  the 
bone  which  follows  the  general  osteological 
pattern  of  crocodiles  and  is  symmetrical  on 
both  sides  of  the  midline,  the  medial  bound- 
aries are  tentatively  placed  at  the  lateral 
borders  of  the  internal  nares.  If  this  is 
correct,  the  palatines,  in  addition  to  border- 
ing the  internal  nares,  form  part  of  the 
anterolateral  margins  of  the  pterygoid 
fenestrae. 

Vomer.  The  vomers  diverge  from  the 
midline  between  the  internal  nares,  emerge 
on  the  dorsal  side  of  the  secondary  palate 
and  extend  posteriorly  and  laterally  to  re- 
veal a  V-shaped  exposure  of  the  pterygoids 
above  and  between  them  (Plate  VII).   The 


vomers  form  the  medial  border  of  the  in- 
ternal nares. 

Pterygoid.  The  pterygoids  are  the  larg- 
est of  the  basicranial  elements,  consisting 
of  an  anterior  plate,  a  flange,  and  a  poste- 
rior process  from  the  base  of  the  flange. 
In  the  available  specimens  the  anterior  part 
of  the  pterygoid  is  badly  fractured,  but  a 
row  of  eight  to  twelve  very  small  teeth  is 
nevertheless  visible  extending  from  just 
behind  the  internal  nares  to  the  base  of 
the  pterygoid  flange.  At  the  midline  the 
pterygoids  are  separated  by  an  interptery- 
goid  fenestra  situated  between  the  internal 
nares  and  the  basisphenoid.  Here  the  ptery- 
goid border  is  formed  by  a  prominent  ridge 
extending  from  the  vomer  to  the  pterygoid 
flange.  Laterally,  the  ectopteiygoid  joins 
the  pterygoid  at  the  anterior  margin  of  the 
pterygoid  flange.  This  flange  in  Protero- 
champsa is  similar  to  that  of  modern  Croc- 
odilia in  its  flat,  blade-like  shape.  It  dif- 
fers from  that  of  Recent  crocodiles  in  not 
being  appreciably  extended  downward  be- 
yond the  level  of  the  jugal.  The  posterior 
edge  of  the  flange  curves  medially  and 
dorsally  to  meet  the  medial  process  of  the 
quadrate. 

Ecto))terij^oid.  The  ectopterygoid  forms 
a  bar  between  the  pterygoid  fenestra  and 
the  open  area  around  the  pterygoid  flange 
(it  is  well  preserved  and  is  a  simple  straight- 
forward bone).  It  is  relatively  short  and  is 
bordered  medially  by  the  pterygoid  and 
laterally  by  the  jugal. 

Basisphenoid.  The  basisphenoid  bears 
no  resemblance  at  all  to  that  of  modern 
crocodiles.  It  is  small,  sub-rounded  and 
completely  fused  with  the  basioccipital. 
On  its  anterior  face  are  two  flangelike  proc- 
esses with  a  median  cleft  between  them 
(which  may  be  the  eustachian  tube).  Just 
behind  these  processes  are  two  prominent 
openings  assumed  to  be  the  carotid  foram- 
ina. The  area  immediately  around  the  basi- 
sphenoid is  badly  fractured,  making  its  re- 
lationship with  the  pterygoid  somewhat 
obscure.   It  appears,  however,  to  be  in  con- 


424         Bulletin  Museum  of  Coiujxiidtivc  Zoo/og;/,  Vol.  135,  No.  8 


tact  w  itli  the  postciior  l)oi"(.kr  ol  thr  ptc'r\- 
goid,  sliglitK  \(-iitral  to  it. 

Ventral  opciiiiiL^s  of  the  skull.  It  is  not 
pos.sil)K'  to  deterniine  the  si/e  ol  the  inci- 
sive loiunien  cine  to  th(>  nianchble  ])eing 
pnshed  np  into  the  sknll  and  obscnring  the 
anterior  portion  of  the  palate. 

The  internal  nares  are  approximately  five 
centimeters  long  by  one  and  a  lialt  centi- 
meters wide  and  extend  slightly  lateral 
from  the  midhne,  being  bordered  as  pre- 
\  iously  described. 

The  interpterygoid  fenestra  is  a  triangn- 
lar-shaped  opening  separating  the  ptery- 
goids at  the  midline.  It  is  seven  centimeters 
long  b\-  three  centimeters  wide  at  its  base. 

The  pterygoid  fenestrate  are  o\'al-shaped 
openhigs.  The)  luc  formed  on  the  lateral 
side  of  the  pterygoid  and  are  approximately 
five  and  a  haU  centimeters  long  by  two 
centimeters  wide.  They  are  inclined  slightly 
toward  the  midline  and  are  narrower  ante- 
riorK'  than  posteriorK'. 

MANDIBLE 

The  mandible  is  present  in  MCZ  3408, 
bnt  is  crnshed  np  into  the  sknll,  lea\ing 
only  the  ventral  surface  visible.  Although 
it  is  not  possible  to  give  a  complete  descrip- 
tion of  the  jaw,  there  are  some  interesting 
features  to  be  noted.  The  anterior  part  of 
the  mandil)le  is  quite  like  that  of  Crocodij- 
lus\  with  a  relatively  narrow,  oval-shaped 
ramus,  the  symphysis  extending  to  about 
the  fourth  tooth,  and  not  including  the 
splenial.  Posterior  to  the  dentary,  how- 
ever, the  mandible  flares  out  to  a  broad 
articular  surface  at  only  a  slight  angle,  with 
\\  hat  is  jiresumed  to  be  the  prearticular  be- 
coming (juite  thin  laterally.  The  articular 
is,  so  far  as  can  be  determined,  much  larger 
than  in  other  crocodiles  or  in  thecodonts, 
forming  the  entire  ventral  surface  of  the 
articular  n  gion.  A  distinctive  feature  is 
the  complete  lack  of  a  retroarticular  proc- 
ess. The  angular  is  relatively  small,  lying 
ventral  to  the  articular  and  not  extending 
beyond  the  maximum  curvature  of  the 
angle.    The  adductor  fossa  is  (juite  shallow 


and  elongate,  the  floor  apparently  formed 
largely  by  the  articular.  A  small  elongate 
external  fenestra  is  present;  the  articular 
l)artieipates  in  its  posterior  border. 

This  would  appear  to  be  a  rather  weak 
jaw  for  an  aggressive  carnivore,  and,  taking 
into  account  the  relatively  small  number  of 
teeth  ( 17  total,  with  a  third  of  these  prob- 
ablv  imdergoing  replacement  at  any  one 
tiiue)'  the  mandible  may  be  considered 
[irimitive  or  may  indicate  a  specialized  diet, 
perhaps  lish  or  carrion. 

DENTITION 

There  is  a  slight  amount  of  dental  differ- 
entiation in  Protcrochampsa  as  shown  by 
the  smaller  alveoli  at  the  anterior  end  of 
the  snout.  A  slight  constriction  is  present 
across  the  snout  in  the  region  of  the  pre- 
maxilla,  which  may  have  served  to  accom- 
modate larger  teeth  in  the  lower  jaw,  but 
this  is  not  a  canine  notch  in  the  usual  sense 
of  the  word.  The  maxillary  tooth  row  is 
distinctive  in  extending  back  only  as  far 
as  the  anterior  border  of  the  orbit  and  con- 
taining only  II  teeth.  About  one-third  of 
these  appear  to  have  been  undergoing  re- 
placement, giving  the  animal  only  8  or  9 
operating  maxillary  teeth.  The  teeth  are 
relatively  long,  slightly  curved  posteriorly, 
and  slightly  ovoid,  the  largest  ones  lying  in 
the  center  of  the  maxilla,  resembling  those 
of  the  early  mesosuehians.  The  mandibular 
teeth  were  not  \'isible  in  either  of  the  speci- 
mens examined. 

The  palatal  teeth  of  Proterochampsa  are 
extremely  small,  the  largest  being  about 
two  millimeters  in  diameter.  They  form  a 
row  along  the  length  of  the  pterygoid,  each 
row  possessing  from  eight  to  twelve  teeth. 
It  is  difficult  to  imagine  these  teeth  func- 
tioning as  either  grasping  or  chewing  mech- 
anisms; they  were  probably  vestigial.  The 
palatal  teeth  do,  however,  provide  a  possi- 
ble link  to  the  primitive  proterosuchid 
thecodonts. 

VERTEBRAE  AND  RIBS 

The   postcranial   material   consists   of    13 


Proterochampsa  and  crocodile  evolution  •  S/7/        425 


articulated  vertebrae,  and  most  of  the  as- 
sociated ril3s  ot  MCZ  3408,  which  were 
found  in  series  and  in  articulation  with  the 
skull.   All  except  the  first  bear  ribs. 

Cervical  vertebrae.  There  are  seven  or 
eight  cervical  \ertebrae  preserved  in  Pro- 
terochampsa. Except  for  the  loss  of  the 
upper  part  of  some  of  the  neural  spines,  all 
are  well  preserved. 

The  atlas-axis  complex  is  similar  to  that 
of  crocodiles  in  general;  the  proatlas,  if 
present,  was  not  presei-ved.  The  atlas  con- 
sists of  the  two  sides  of  the  neural  arch 
surrounding  what  appears  to  be  the  odon- 
toid process.  The  axis  resembles  the  other 
cer\ical  \'ertebrae,  differing  only  in  having 
a  wider  ventral  keel,  small,  diagonally 
placed  parapophyses  and  the  characteris- 
tically larger  neural  spine.  The  rib  of  the 
axis  is  a  normal  rib,  differing  from  the 
other  cervical  ribs  only  in  articulating  more 
anteriorly  on  the  centrum.  This  is  in  con- 
trast to  the  highly  modified  splint-like  rib 
of  the  axis  of  modern  crocodiles. 

The  remaining  cervical  vertebrae  are 
rather  lightly  constructed  with  quite  thin 
and  comparati\ely  long  neural  spines.  The 
centra  are  strongly  amphicoelous,  slightK 
longer  than  high,  oval  in  cross  section  but 
with  prominent  ventral  keels.  The  neural 
arch  lies  relatively  low  on  the  centrum, 
with  the  diapophyses  extending  straight 
down  the  sides  to  just  below  the  neurocen- 
tral  suture.  The  diapophysis  appears  to 
angle  slightly  posteriorly  in  the  more  pos- 
terior cervicals.  The  base  of  the  diapophy- 
sis is  very  strong,  extending  like  an  inverted 
triangle  the  entire  length  of  each  of  the 
neural  arches.  The  parapophysis  is  a  small, 
flattened,  oval-shaped  process  projecting 
outward  from  the  base  of  the  centrum  just 
below,  and  slightly  anterior  to  the  diapoph- 
ysis; that  of  the  last  two  cervicals  projects 
somewhat  posteriorly.  Unlike  modern  croc- 
odiles, the  capitular  facets  face  straight  out- 
ward instead  of  downward. 

The  anterior  zygapophysis  projects  for- 
ward from  the  sub-triangular  body  of  the 
neural   arch  as   a  blade-like  process,  with 


the  articular  surface  facing  dorsally  and 
medially.  In  all  of  the  cervicals  it  over- 
hangs considerably  the  anterior  edge  of 
the  centrum. 

The  posterior  zygapophysis  extends  to 
the  midline  above  the  centrum  (see  Plate 
IX).  It  does  not  overhang  the  posterior 
face  of  the  centrum.  The  articular  facet 
lies  just  under  the  neural  spine,  facing 
downward  and  slightly  outward.  The  ar- 
ticular surface  of  both  zygapophyses  is  con- 
siderably more  near  the  horizontal  than 
that  of  modern  crocodiles. 

The  neural  spine  is  a  narrow  plate  about 
twice  the  height  of  the  centrum,  arising 
from  the  posterior  part  of  the  neural  arch 
just  above  the  posterior  zygapophysis  and 
curving  slightly  backward.  The  neural 
spine  becomes  somewhat  broader  in  the 
posterior  cervical  vertebrae. 

Anterior  dorsal  vertebrae.  Like  the  cerv- 
icals the  dorsal  vertebrae  are  strongly  am- 
phicoelous, with  the  centrum  slightly  longer 
than  high.  The  transverse  process  projects 
out  and  down  from  the  neural  arch  for  a 
distance  equaling  the  length  of  the  cen- 
trum, reaching  to  about  the  level  of  the 
neuro-central  suture.  It  is  a  relatively  wide, 
blade-like  process  similar  to  that  of  modern 
crocodilians.  The  parapophysis,  situated 
anteriorly  and  ventrally  on  the  side  of  the 
centrum  is  very  short  and  faces  straight 
out,  thus  differing  considerably  from  the 
more  advanced  crocodilian  condition  in 
which  both  heads  of  the  rib  articulate  on 
the  transverse  process.  The  dorsal  centra 
resemble  those  of  the  cervicals  in  being 
strongly  keeled,  but  are  more  heavily  con- 
structed. The  anterior  zygapophysis  is  sim- 
ilar in  general  to  that  of  the  cer\  ical  verte- 
brae, although  somewhat  shorter  and  stur- 
dier. The  posterior  zygapophysis  is  also 
very  similar  to  that  of  the  cervical,  but 
somewhat  stronger  and  has  a  more  promi- 
nent median  cleft.  The  neural  spines  of  the 
dorsal  vertebrae  are  wider  and  ht^avier  than 
in  the  cervicals,  and  arise  from  the  posterior 
edge  of  the  neural  arch. 

Ribs.   All  the  ribs  preserved  are  bicipital 


426        BiiUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  8 


and  articulati'  doisalK  and  \(.'iilrall\-  (not 
on  the  same  level,  as  do  those  of  modern 
crocodiles).  The  cer\ical  ribs  are  well 
developed,  differing  onl\  slighth  from 
those  of  ensuchians.  The  first  cer\  ieal  rib 
is  a  slender,  t\\o-headed  element  attached 
to  the  normal  diapophysis  and  to  a  very 
small  parapophysis,  which  is  nothing  more 
than  an  articular  facet  facing  laterally  and 
anteriorl)  on  the  bottom  of  the  centrum 
near  the  anterior  edge.  The  remaining 
cervical  ribs  are  relati\t>ly  slender  and  pro- 
ject posteriori).  They  become  longer  and 
sturdier  posteriorly  and  thus  grade  into  the 
size  of  the  dorsal  ribs.  The  articulation  of 
the  tuberculum  with  the  diapophxsis  on 
the  transverse  process  is  considerably  larger 
than  in  later  crocodiles,  occupsing  the  en- 
tire face  of  the  process.  The  capitulum  ar- 
ticulates ventrally  and  slightly  anteriorly  to 
the  tuberculum.  These  ribs  possess  what 
appear  to  be  the  beginnings  of  uncinate 
"bulges,"  located  about  one-third  of  the 
way  down  the  rib  body  and  bearing  prom- 
inent ridges  on  the  anterior  edges  of  their 
dorsal  surfaces. 

DISCUSSION 
HISTORICAL  REVIEW 

The  affinities  of  primitive  crocodilians 
are  uncertain  due  to  the  small  number  of 
early  forms  so  far  known.  In  addition  to 
the  Middle  Triassic  Proterocliamjisa,  these 
are:  Pwtosuchus  of  North  America,  and 
Notochampsa  and  Enjthrochamp.sa  of  South 
Africa.  (A  questionable  crocodilian  two 
centimeters  long,  without  skull  or  limbs, 
was  described  by  Young  hi  1951  from  the 
Upper  Triassic  of  China;  also  an  unde- 
scribed  crocodilian  from  the  Upper  Triassic 
of  Wales  has  been  reported  by  Kennack 
11956].)  Of  these  three,  Pwto.siichiis  is 
known  from  a  skeleton,  the  other  two  from 
much  less  complete  material.  All  three  are 
late  Triassic  or  earliest  Jurassic  in  age. 

Notocliamp.sa  istedana  was  described  by 
Broom  (1904)  from  the  impressions  of  the 
undersides  of  most  of  the  roofing  bones  of 


a  skull,  most  of  the  dorsal  armor,  a  scapula, 
a  coracoid,  parts  of  a  humerus,  radius  and 
ulna,  part  of  a  femur,  and  parts  of  a  tibia 
and  fibula.  A  second  specimen  from  the 
same  general  area  (Rarkly  East,  Cape 
Province,  South  Africa )  was  named  by  him 
Notochampsa  longipes.  This  specimen  con- 
sisted of  a  well  preserved  pelvis,  a  femur, 
tibia  and  fibula,  some  foot  bones,  and  part 
of  the  dorsal  armor.  Roth  specimens  were 
found  in  the  upper  part  of  the  Stormberg 
series.  Rroom  placed  the  two  species  in  the 
same  genus  on  the  basis  of  the  similarity  of 
the  dorsal  armor. 

Haughton  ( 1924 )  excluded  Notochampsa 
istedana  from  the  Crocodilia  and  grouped 
it  with  Pedcticosaurus  in  a  family  Noto- 
champsidae,  which  he  referred  to  the  the- 
codont suborder  Pseudosuchia.  He  con- 
sidered the  family  to  be  intermediate  be- 
tween the  aetosaurian  thecodonts  and  the 
crocodiles.  The  other  species,  N.  loiifi^ipes, 
he  separated  from  Notochampsa,  placing  it 
within  the  Crocodilia  but  not  in  any  family, 
giving  the  following  reason  (1924:369): 

"If  R room's  N.  lon^ipes  is  to  be  kept  in 
the  genus  Notochampsa  then  the  genus 
must  be  considered  to  be  characterised  by 
the  possession  of  a  skull  differing  from  that 
of  a  true  Crocodile  and  of  a  typically  croco- 
dilian pelvis.  This  is  not  impossible;  but 
until  more  is  known  of  these  forms  it  would 
seem  best  to  separate  the  two  forms  from 
one  another,  classing  istedana  as  one  of 
the  higher  Pseudosuchians  and  erecting,  as 
above,  a  new  genus  Erythrochampsa  for  the 
more  truly  Crocodilian  Erythrochampsa 
lonii.ipcs."  This  explanation  reflects  the 
then  current  phil()soph\'  of  graded  rather 
than  mosaic  evolution  and  may  not  be  justi- 
fiable. 

The  following  year  von  Huene  ( 1925 ) 
studied  the  material  and  came  to  the  con- 
clusion that  both  genera  were  pseudosuch- 
ian  thecodonts,  forming  the  end  members 
of  an  evolutionary  seciuence  leading  towards 
the  crocodiles  from  the  pseudosuchians. 
The  sequence  he  proposed  consisted  of  the 
following  forms:   Erpetosuchus,  Aetosaurus, 


Proterochampsa  and  crocodile  evolution  •  Sill       427 


Stegomosuchus,  Sphenosuchtis,  Pcdetico- 
saurus,  ISJotochampsa  and,  finally,  Enjthro- 
champsa. 

Broom  returned  to  the  subject  of  Noto- 
champsa  and  Enjthrochampsa  in  1927.  He 
reviewed  the  work  of  Haughton  and  of 
von  Huene  and  agreed  to  the  generic  sep- 
aration of  Notocho77ipsa  and  Enjthrochamp- 
sa,  but  remained  firm  in  his  conviction  that 
they  are  closely  related  and  are  true  croco- 
diles. He  modified  the  classifications  of 
Haughton  and  von  Huene  by  placing  both 
genera  in  the  Notochampsidae  and  refer- 
ring the  family  to  the  order  Crocodilia. 

No  further  information  regarding  the  an- 
cestry of  crocodiles  appeared  until  1933, 
when  Brown  reported  the  well  preserved 
crocodilian  from  the  Upper  Triassic  or 
Lower  Jurassic  rocks  of  Arizona,  to  which 
he  ga\e  the  name  Piotosiicluis  richarclsoni, 
erecting  for  it  the  family  Protosuchidae. 
Unfortunately,  as  in  the  other  early  croco- 
diles, nothing  is  preserved  of  the  palatal 
region  of  Protosiichiis. 

The  following  year  ( 1934 )  Mook  pre- 
sented a  classification  of  the  Crocodilia  in 
which  the  Protosuchidae  was  placed  in  a 
new  suborder,  the  Protosuchia,  and  no 
mention  at  all  was  made  of  either  Noto- 
champsa  or  Enjthrochampsa  as  members  of 
the  Crocodilia.  Romer  (1945)  combined 
Protosuchus,  ISJotochampsa,  Enjthrochampsa 
and,  questionably,  Pcdcticosaurus  in  the 
family  Notochampsidae.  In  1951  Colbert 
and  Mook  published  a  thorough  descrip- 
tion of  Protosuchus,  and  placed  Noto- 
champsa  and  Enjthrochampsa  in  the  sub- 
order Protosuchia  as  members  of  the  family 
Protosuchidae. 

Kiilin  ( 1955 )  modified  Colbert  and 
Mook's  classification  b\-  removing  Noto- 
champsa  and  Enjthrochampsa  from  the 
Protosuchidae  and  reuniting  them  in  the 
Notochampsidae.  He  recognized  the  sub- 
order Protosuchia  and  referred  the  Noto- 
champsidae to  it.  The  current  classification 
of  ancestral  crocodilians  is  as  follows: 

Protosuchia 
Protosuchidae 


Fig.  6.  Comparison  of  dorsal  view  of  skull  in  A,  Protosuchus; 
B,  Pro/eroc/iampso;  C,  Notochampio;  not  drawn  to  scale. 
(A,    after    Colbert    and    Mook;    C,    after    Broonn.) 

Protosuchus 
Notochampsidae 
Notochampsa 
Enjtlirochampsa 

The    discovery    of    Proterochampsa    re- 
quires a  reappraisal. 

THE  AFFINITIES  OF 
PROTEROCHAMPSA 

In  comparing  Proterochampsa  with  these 
early  crocodilians  and  with  members  of  the 


428        BuUefin  Museum  of  Comparative  Zoolof^y,  Vol.  135,  No.  8 


Fig.    7.      Profile    comparison    of    skulls    of    A,    Profosuchus,    B,    Proferochampso;    C,    Nofochompso;    not    drawn    to    scale.      (A, 
after    Colbert    and    Mock;    C,    after    von    Huene.) 


Thecodontia,  it  becomes  evident  that  F/o- 
fcroehampsa  represents  an  excellent  ex- 
ample of  mosaic  evolution  and  provides 
man\-  of  the  characters  expected  in  a  transi- 
tional form.  The  snout  and  the  dorsum  of 
the  skull  generally  are  indistinguishable 
from  those  of  a  nKxleni  crocodile  (except 
for  the  primitive  antorbilal  tenestrae).  On 
the  other  hand,  the  posterior  part  is  re- 
markably like  that  of  a  primitive  thecodont, 
except  for  its  flatness  and  reduced  supra- 
temporal  fenestrae.  The  palatal  area  is 
again  quite  like  that  of  the  primitive  theco- 


donts except  for  the  transitional  features  of 
the  secondary  palate  and  the  development 
of  the  pterygoid  flanges. 

In  trying  to  place  Proterochampsa  within 
the  classification  of  the  early  Crocodilia 
we  are  faced  as  usual  with  the  problems 
presented  by  inadequate  material.  Of 
Proterochampso  we  ha\e  only  the  skull  and 
a  few  vertebrae  and  ribs.  ProtosucJuis  is 
known  from  a  skeleton,  but  the  skull  is 
imperfect,  especially  as  regards  the  ven- 
tral surface;  Not(H-]taiyi))sa  is  represented  by 
a  very  poor  skull  and  some  postcranial  ma- 


Proterochampsa  and  crocodile  evolution  •  Sill        429 


terial;  and  Erythrochatnpsa  is  known  only 
from  postcranial  elements.  The  main  basis 
for  relating  Profostichus,  Notocliamj)sa,  and 
Enjthrochampsa  to  each  other  has  been 
the  character  of  the  postcraninm:  prin- 
cipally the  similarities  of  the  pectoral  girdles 
of  Notochampsa  and  Pwtosuchus  and  the 
pelvic  girdles  of  ErytJirocliam))s(i  and  Pro- 
tosucJiiis. 

When  the  sknll  of  P rotcrochampsa  is  com- 
pared with  that  of  ProtosiicJius  it  is  imme- 
diately apparent  that  they  do  not  resemble 
one  another  sufficiently  to  be  considered  as 
members  of  the  same  phylogenetic  line. 
The  skull  of  Protosuchus  is  short  relative 
to  width  and  considerably  deeper  than 
either  Proterochampsa  or  'N otochampsa  (see 
Figure  10).  The  orbits  of  Protosuchus  are 
nearly  on  a  vertical  plane,  facing  outward 
and  forward,  while  those  of  Proterochampsa 
are  on  a  horizontal  plane,  facing  upward. 
The  snout  of  Protosuchus  is  short  (less  than 
half  the  total  length  of  the  skull )  and  lacks 
antorbital  fenestrae,  while  that  of  Protero- 
champsa is  long  (over  half  the  total  skull 
length),  and  has  prominent  antorbital  fe- 
nestrae. Protosuchus  has  a  very  lightly 
sculptured  skull  compared  to  the  heavih' 
sculptured  skull  of  Proterochampsa.  In  Pro- 
tosuchus the  external  nares  are  small, 
clearly  separated,  and  at  the  very  tip  of  the 
snout.  Those  of  Proterochampsa  are  quite 
the  opposite,  being  elongate,  separated  only 
by  a  thin  nasal  process  and  situated  con- 
siderably back  from  the  tip  of  the  snout. 
In  Protosuchus  the  squamosal  is  large  and 
overlies  the  quadrate  and  quadratojugal 
completely;  in  Proterochampsa  the  squamo- 
sal is  relatively  small  and  overlies  none  of 
the  (quadratojugal  and  only  a  part  of  the 
quadrate. 

The  resemblances  between  the  tvvo  forms 
are  not  impressive.  Both  have  sculptured 
skulls  (although  different  sculpturing), 
small  supratemporal  fenestrae,  amphicoe- 
lous  vertebrae,  and  a  relatively  small  atlas. 
Both  Protosuchus  and  Proterochampsa  are 
crocodilians,  but  on  the  basis  of  skull  mor- 
phology   it    would    appear    that    Protero- 


champsa is  closer  to  the  main  line  of  croc- 
odilian evolution  than  is  Protosuchus. 

A  much  closer  resemblance  exists  be- 
tween Proterochampsa  and  Notochampsa 
as  regards  the  skull.  In  Notochampsa  this 
is  relatively  long  compared  to  width  (the 
snout  occupies  more  than  half  the  length  of 
the  skull),  is  relatively  flat,  and  the  orbits 
are  in  the  horizontal  plane,  all  of  these 
features  being  in  common  with  Protero- 
champsa. Both  Proterochampsa  and  Noto- 
champsa possess  an  auditoiy  canal  on  the 
posterior  face  of  the  skull.  Regarding  the 
presence  of  antorbital  fenestrae  in  Noto- 
champsa, Broom  says  there  are  none,  von 
Huene  says  there  are,  and  Haughton  was 
unable  to  decide;  the  specimen  is  too  im- 
perfect for  a  definite  decision. 

The  skulls  of  Notochampsa  and  Protero- 
champsa quite  evidently  resemble  each 
other  more  than  either  of  them  resembles 
Protosuchus.  This  presents  the  problem  of 
the  taxonomic  position  of  Proterochampsa 
and  indeed  requires  a  re-evaluation  of  Pro- 
tosuchus as  an  ancestral  crocodile  and  of 
the  role  of  the  pseudosuchian  thecodonts  as 
possible  ancestors. 

Proterochampsa  does  not  belong  in  the 
suborder  Protosuchia  on  the  basis  of  most 
taxonomic  characters  now  used  to  define 
that  group.  The  possibilities  of  phyloge- 
netic placement  then  are  the  following:  the 
suborder  Protosuchia  may  be  redefined  as  a 
group  including  all  pre- Jurassic  crocodilians 
without  special  regard  to  moi-phological 
similarities,  and  Proterochampsa  placed  in 
it;  or  Proterochampsa,  Notochampsa,  and 
Erythrochampsa  may  be  grouped  into  a 
separate  suborder  leading  to  the  Meso- 
suchia,  and  the  Protosuchia  retained  as  an 
aberrant  lineage  arising  from  the  early  croc- 
odilian stock. 

On  the  basis  of  skull  comparison,  it  ap- 
pears likely  that  Protosuchus  is  an  aberrant 
offshoot  from  the  line  which  gave  rise  to  the 
Crocodilia,  possessing  some  characters  of 
both  thecodonts  and  crocodiles.  It  may  be 
argued  that  Proterochampsa  cannot  be  le- 
gitimately compared  to  Protosuchus  until 


430         Bulletin  Mtisctim  of  Conqnirative  Zoology,  Vol.  135,  No.  8 


Fig.   8.      Profile   comparison  of  skulls  of  A,   Proferochompso;   B,   Crocodylus;   C,   Profosuchus;    not  drawn   to  scale. 


something  i.s  known  about  the  pectoral  and 
pelvic  girdles  of  the  fonner.  However,  in 
this  respect  there  are  only  three  possibilities 
for  the  girdles  of  Proterochampsa:  they  may 
be  the  same  as  Pwto.suchus,  more  crocodil- 
ian, or  less  crocodilian.  None  of  these  possi- 
bilities alters  the  fact  that  Protostichus-  has 
strayed  considerably  from  the  earlier  line  of 
more  typical  crocodilians  represented  by 
Proterochampsa. 

If  Protosuchus  is  regarded  as  representing 
an  aberrant  group,  one  not  on  the  direct 
ancestral  line  leading  to  the  mesosuchians. 


the  possibility  that  Proterochampsa  and 
Notocliampsa  actually  belong  within  the 
suborder  Mesosuchia  must  be  examined. 

The  suborder  Mesosuchia  is  defined  as 
having  a  secondary  palate  formed  by  the 
premaxillae,  the  maxillae,  and  the  palatines, 
the  pubis  excluded  from  the  acetabulum, 
the  postorbital  bar  at  the  dorsal  surface  of 
the  skull,  and  the  vertebrae  amphicoelous 
or  platycoelous.  To  redefine  the  suborder  to 
include  Proterocham))sa  and  the  Noto- 
champsidae  it  would  be  necessary,  so  far, 
only  to  modify  the  definition  with  regard  to 


Proterochampsa  and  crocodile  evolution  •  Sill        431 


the  secondary  palate.  However,  to  redefine 
the  Mesosuchia  in  this  way  would  be  to  dis- 
rupt the  classification  of  what  appears  to  be 
a  natural  group,  or  at  least  a  fairly  uniform 
evolutionary  grade.  The  acquisition  of  the 
secondary  palate  and  internal  nares  of  the 
Mesosuchia  marks  a  significant  phyloge- 
netic  stage,  and  is  remarkably  consistent 
throughout  the  nine  families  and  thirty- 
seven  genera  of  the  suborder.  In  addition 
to  the  more  primitive  condition  of  the  palate 
( at  least  in  Proterochampsa ) ,  Nofochampsa 
and  Proterochampsa  also  possess  a  more 
primitive  condition  of  the  external  auditory 
meatus,  while  the  mesosuchians  are  consis- 
tent in  the  transitional  nature  of  this  char- 
acter (see  discussion  of  the  ear).  In  gen- 
eral, the  Mesosuchia  presents  the  appear- 
ance of  a  well  established  group,  greatly 
dixersified,  into  which  Proterochampsa  and 
the  Notochampsidae  would  fit  less  consis- 
tently than  do  any  of  the  other  families 
within  this  suborder.  The  remaining  alter- 
native is  to  place  Proterochampsa  and  the 
Notochampsidae  in  a  separate  suborder, 
recognizing  that  the  Notochampsidae  are 
\  ery  poorly  known  and  may  later  prove  not 
to  be  that  closely  related  to  Proterochampsa. 
However,  the  elongate  skull  and  dorsal 
orbits,  particularly  the  latter,  indicate  that 
Notochampsa  had  acquired  the  aquatic 
habitus  of  the  Crocodilia. 

EARLY  HISTORY  OF  THE 
CROCODILIA 

The  evolutionary  trends  within  the  Croc- 
odilia only  become  relatively  well  docu- 
mented after  the  late  Jurassic,  although 
specialized  marine  mesosuchians  are  known 
from  the  early  and  middle  parts  of  this 
period.  The  early  and  middle  Jurassic  were 
probably  times  of  great  divergence  within 
the  order,  but  non-marine  representatives 
are  practically  unknown,  due  to  the  lack  of 
continental  sediments  of  these  ages.  The 
primitive  Triassic  members  of  the  order,  as 
previously  mentioned,  are  few  in  number 
and  most  of  them  are  poorly  preserved. 
Thus,  there  is  a  gap  in  knowledge  from  the 


late  Triassic  to  the  late  Jurassic,  coupled 
with  a  dearth  of  material  from  the  earlier 
Triassic. 

Protosuchus  was  the  first  reasonably  well 
preserved  early  crocodilian  found.  Although 
not  closely  resembling  later  crocodiles,  it 
possessed  a  number  of  crocodilian  charac- 
ters, especially  in  the  postcranial  skeleton. 
The  strongly  crocodilian  coracoid  and  pubis, 
together  with  the  more  thecodont-like  skull, 
suggested  a  pattern  of  gradual  acquisition 
of  crocodilian  characters  from  a  pseudo- 
suchian  ancestiy.  Notochatnpsa  and  Erijth- 
rocliampsa,  then  as  now,  were  too  poorly 
known  to  contribute  evidence  of  any  great 
value.  All  of  these  forais  were  of  very  latest 
Triassic  age,  with  Protosuchus  considered 
as  more  or  less  the  prototype  of  later  croco- 
diles. Yet  by  the  earliest  Jurassic  there 
existed  good  mesosuchian  representatives; 
the  order  was  by  then  well  differentiated 
and  was  undergoing  rapid  radiation.  This 
would  leave  very  little  time  between  the 
rather  thecodont-like  Protosuchus  and  the 
earliest  mesosuchians.  Although  such  rapid 
evolution  and  radiation  as  this  view  would 
imply  might  not  be  impossible  it  is  rather 
unlikely. 

The  discovery  of  Proterochampsa  changes 
all  this,  however,  by  demonstrating  that  a 
number  of  "modern"  cranial  characters  of 
the  Crocodilia  were  already  in  existence  by 
the  late  Middle  Triassic.  The  conclusion 
that  the  Crocodilia  became  differentiated 
relatively  early  in  the  history  of  the  Theco- 
dontia  rather  than  being  an  "end  product" 
of  that  group  seems  inescapable.  Protero- 
champsa is  certainly  a  crocodile,  and  at 
present  is  the  best  known  representative 
of  the  primitive  members  of  that  group.  It 
is  of  course  possible  that  it  is  not  the  ances- 
tor of  the  later  crocodiles,  but  it  appears  to 
be  closer  to  such  an  ancestor,  at  least  mor- 
phologically, than  any  of  the  other  presently 
known  early  forms. 

This  being  so,  the  non-crocodilian  fea- 
tures of  Protosuchus  might  be  explained  in 
one  of  several  ways:  1 )  the  Crocodilia  arose 
from  a  non-pseudosuchian  group  of  aquatic 


432        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  8 


Fig.  9.      Comparison  of  ventral  view  of  skull  in  A,  Chasmatosaurus;  and  B,  Proferochompso.     X    /3- 


thecodonts,  in  which  case  the  Protosuchia  Crocodiha  arose  from  a  primitive  group  of 
could  be  regarded  as  forms  that  secondarily  terrestrial  thecodonts,  possibly  early  pseu- 
became  adapted  for  terrestrial  life;  2)  the      dosuchians,  in  which  case  the  Protosuchia 


PROTEROCHAMPSA    AND   CROCODILE   EVOLUTION    •    5/7/  433 


could  be  regarded  as  having  retained  the 
primiti\e  terrestrial  features  of  the  transi- 
tional group;  3)  the  Protosuchia  were  not 
true  crocodilians  and  independently  evolved 
crocodilian  characters.  Of  these  possibilities 
the  last  seems  to  be  the  least  likely  on  pres- 
ent evidence,  although  the  other  two  are 
almost  equally  uncertain.  Any  one  of  these 
possibilities,  however,  could  explain  the 
existence  of  groups  that  possessed  a  few 
good  crocodilian  characters  but  were  more 
thecodont  in  habitus. 

The  primitive  crocodilian  recently  dis- 
covered in  the  Triassic  of  Wales  has  been 
characterized  by  Dr.  K.  A.  Kennack  (pers. 
comm. )  as  a  "crocodile  trying  to  be  a  dino- 
saur." This  description  might  be  applied  to 
some  of  the  other  archosaurs  that  appear  to 
be  in  the  "fringe  area"  of  the  Crocodilia, 
such  as  Pedeticosaurus,  Sphenosiichiis,  Platy- 
ofi,nathus,  and  perhaps  Hcsperosuchus.  In 
this  context  Pwfosiichus  might  be  con- 
sidered as  less  successful  in  "becoming  a 
dinosaur"  than  the  sphenosuchians,  and 
therefore  as  looking  more  like  a  t>pical 
crocodile. 

The  solution  to  the  question  of  crocodilian 
origins  naturally  lies  within  the  Thecodon- 
tia,  but  unfortunately  this  group  is  not  well 
understood  at  present.  Among  early  theco- 
donts, Chasmatosaurm  somewhat  resembles 
Protcrochampsa  in  the  palatal  area.  Al- 
though Chasmafosaunis  is  extremely  prim- 
itive, it  may  nevertheless  represent  the 
group  of  thecodonts  from  which  the  Croco- 
dilia arose. 

CLASSIFICATION  OF  THE 
EARLIEST  CROCODILIA 

On  the  basis  of  this  study  it  is  proposed 
that  the  primitive,  Triassic  crocodilians  be 
di\  ided  into  two  groups,  the  suborder  Pro- 
tosuchia, characterized  by  the  Protosuchi- 
dae,  and  including,  questionably,  the  sphe- 
nosuchid  thecodonts,  and  a  new  suborder, 
the  Archaeosuchia,  for  the  Proterochamp- 
sidae  and,  provisionally,  the  Notochamp- 
sidae. 


ARCHAEOSUCHIA  new  suborder 

The  Archaeosuchia  may  be  defined  as 
follows:  Crocodilia  with  orbits  in  dorsal 
plane  of  skull,  cranial  table  sculptured, 
snout  long  relative  to  width,  palatines  not 
participating  in  secondary  palate,  postorbi- 
tal  bar  at  surface  of  skull,  auditory  canal  on 
posterior  face  of  skull,  vertebrae  amphicoe- 
lous;  pubis  elongate,  nearly  or  completely 
excluded  from  the  acetabulum. 

NOTOCHAMPSIDAE  Haughton  1924 

The  family  Notochampsidae,  although 
erected  in  1924,  has  never  been  defined. 
The  following  definition  is  proposed  for  it: 
premaxillae  small,  external  nares  divided, 
squamosals  large,  forming  most  of  lateral 
border  of  cranial  table,  frontals  not  fused, 
participating  in  border  of  supratemporal 
fenestrae;  coracoid  enlarged,  similar  in 
shape  to  scapula.  The  type  genus  of  the 
famih'  is  Nofochampsa. 

PROTEROCHAMPSIDAE  new  family 

Protcrochampsa  differs  from  the  noto- 
champsids  to  a  degree  sufficient  to  \\'arrant 
the  erection  of  a  family  for  its  reception. 
This  may  be  defined  as  follows:  Archaeo- 
suchia with  external  nares  united  at  midline, 
premaxillae  large,  frontals  small,  fused,  not 
participating  in  border  of  supratemporal 
fenestrae,  squamosals  small,  limited  to  pos- 
terior border  of  skull.  Protcrochampsa  is 
designated  as  the  type  genus  of  the  family. 
This  would  result  in  the  follo\\'ing  classi- 
fication: 

PROTOSUCHIA  Mook  1934 
Protosuchidae  Brown  1933 
Protosuch  us  Brown  1933 
PROTOSUCHIA  incertae  sedis 
Sphenosuchidae  Haughton  1924 
Sphenosuchus  Haughton  1915 
Sphenosuchidae  incoiac  scdis 
Pedeticosaurus  Van  Hoepen  1915 
Platyognafhus  Young  1944 
ARCHAEOSUCHIA  new  suborder 
Proterochampsidae   new  family 
Protcrochampsa  Reig  1959 


434        Bulletin  Museum  of  Coinparativc  Zoology,  Vol.  135,  No.  8 


Fig.    10.      Diagrammatic    comparison    of    skull    relationships    of    A,    Pro/osuchus;    B,    Proferochampso;    C,    Crocodylus;    D,    Nofo- 
champso.     Lengffi  of  tlie  skull   is  reduced  to  unity. 


Notochampsidae  Haughton  1924 
Notochampsa  Broom  1904 
Ertjthrochampsa  Haughton  1924 

SUMMARY 

Protcrochampsa  harrionucvoi  represents 
a  late  Middle  Triassie  line  of  croeodilians 
showing  many  of  the  "progressive"  features 
characteristic  of  later  members  of  the  order. 
The  dorsum  of  the  skull  is  almost  identical 
with  that  of  the  modern  Cwcodyhis  except 
for  the  presence  of  an  antorbital  fenestra 
and  the  lateral  position  of  the  quadratojugal. 
Ventrally,  a  rudimentary  secondary  palate 
has  evolved,  consisting  only  of  the  premax- 
illa  and  the  maxilla,  while  the  very  small 
pterygoid  teeth  and  an  interpteiygoidal 
vacuity  are  retained.  Tlie  posterior  surface 
of  the  skull  shows  a  meatal  groove  begin- 
ning at  the  ventral  tip  of  the  squamosal 
and  passing  anterior  to  the  exoccipital.  This 
is  possibly  the  beginning  of  the  reacquisition 
of  an  otic  notch  in  the  Crocodilia.  The  man- 
dible is  distinctive  in  the  absence  of  a  retro- 
articular  process,  the  large  size  of  the  articu- 
lar bone,  and  the  slight  angle.  In  general 
the  jaw  appears  to  have  been  a  relatively 
weak  structure.  The  marginal  teeth  are 
slightly  ovoid,  rather  slender,  and  slightly 
curved  posteriorly,  fitting  the  typical  the- 


codont pattern.  The  small  number  of  these 
teeth  ( 17 )  may  be  a  primitive  character  or 
may  indicate  a  specialized  diet,  perhaps  fish 
or  carrion.  Tlie  pterygoid  teeth  are  so  small 
that  it  is  difficult  to  believe  they  were  of 
any  great  use.  The  skull  of  Proterochampsa 
shows  an  interesting  combination  of  primi- 
tive, transitional,  and  advanced  characters; 
it  provides  an  excellent  example  of  mosaic 
evolution. 

The  postcranium  is  represented  only  by 
the  anterior  vertebrae  and  ribs.  The  verte- 
brae are  strongly  amphicoeleous  and  have 
prominent  keels.  The  ribs  are  all  bicipital, 
with  small  uncinate  processes  present  on 
those  of  the  thoracic  region. 

The  Crocodilia  have  long  been  considered 
an  "end  product"  that  arose  from  late  Trias- 
sie thecodonts  by  the  gradual  acquisition  of 
distinctive  characters  and  an  aquatic  habi- 
tus. P voterochampsa  provides  evidence  that 
the  major  features  of  crocodilian  skull  struc- 
ture were  in  existence  by  the  latter  part  of 
the  Middle  Triassie.  A  re-evaluation  of  the 
known  primitive  croeodilians  suggests  that 
there  were  apparently  two  lines  of  evolu- 
tion during  the  Triassie.  On  one  of  these 
lines,  crocodilian  characters,  most  of  which 
are  shown  in  the  skull  of  Proterochampsa, 
were  evolved,  while  in  the  other,  character- 
ized bv  Protosuchns,   the  trend   led  away 


Proterochampsa  and  crocodile  evolution  •  Sill        435 


from  the  crocodilian  way  of  life  toward  a 
more  terrestrial  habitat  and  acquisition  of 
the  necessary  morphologic  features  for  suc- 
cessful competition  with  its  thecodont 
relatives. 

On  the  basis  of  this  study  the  family 
Proterochampsidae  is  proposed,  and  the 
primitive  crocodilians  are  divided  into  two 
suborders,  the  Protosuchia,  consisting  of 
Protosuchidae  and,  questionably,  the  sphe- 
nosuchoidean  thecodonts,  and  a  new  sub- 
order, the  Archaeosuchia,  for  the  Protero- 
champsidae and,  provisionally,  the  Noto- 
champsidae. 

RESUMEN 

Proterochampsa  barrionuevoi  representa 
una  linea  de  cocodrilos  del  Triasico  medio 
que  muestra  muchos  de  los  rasgos  "progres- 
ivos"  que  caracterizan  a  los  miembros  mas 
avanzados  del  orden.  La  superficie  dorsal 
del  craneo  es  casi  igual  a  la  de  Crocodyhis 
de  la  actualidad,  pero  retiene  los  caracteres 
primitivos  de  las  fosas  anteorbitarias  y  de 
la  posicion  lateral  del  cuadrado-yugal.  Por 
el  lado  ventral,  muestra  un  paladar  secun- 
dario  rudimentario  fonnado  por  los  premax- 
ilares  y  los  maxilares,  y  a  la  vez  retiene  los 
pcquenos  dientes  pterigoideos  y  la  fosa  in- 
terpterigoidea  que  son  mas  bien  caracteris- 
ticas  de  los  tecodontes  primitivos.  La  super- 
ficie posterior  del  craneo  posee  un  surco 
meatal  que  comienza  en  el  punto  ventral  del 
escamoso  y  pasa  por  delante  del  exoccipital. 
Esto  puede  indicar  el  comienzo  de  la  adqui- 
sicion  de  la  muesca  otica  en  el  orden  Croco- 
dilia.  La  mandibula  se  destaca  por  la  falta 
del  proceso  retroarticular,  por  el  tomailo 
grande  del  articular,  y  por  la  pequenez  del 
angulo.  En  general  la  mandibula  parece 
haber  sido  una  estructura  relativamente 
debil.  Los  dientes  marginales  son  ligera- 
mente  ovoides,  delgados,  y  algo  recurvados 
hacia  atras,  siendo  su  aspecto  similar  al  de  los 
tecodontes.  El  pequeiio  numero  de  estos 
dientes  (17)  puede  ser  un  caracter  primitive  o 
quizas  un  indicio  de  una  dieta  especializada, 
que  podn'a  haber  consistido  en  paces  o  car- 


rona.  Los  dientes  pterigoideos  son  tan  redu- 
cidos  que  resulta  dificil  creer  que  fuesen  de 
utilidad  alguna.  El  craneo  de  Protero- 
champsa muestra  una  combinacion  suma- 
mente  interesante  de  caracteres  primitivos, 
transicionales,  y  avanzados,  por  lo  que  pro- 
porciona  un  excelente  ejemplo  de  evolucion 
mosaico. 

I^a  region  post-craneana  de  Protero- 
champsa esta  unicamente  representada  por 
las  vertebras  y  las  costillas  anteriores.  Las 
vertebras  son  biconcavas  y  tienen  quillas 
prominentes.  Las  costillas  son  todas  bicipi- 
tales  y  las  de  la  region  toracica  poseen  pe- 
queiios  procesos  uncinados. 

Tradicionalmente  se  ha  considerado  el 
orden  Crocodilia  como  el  "producto  final" 
de  una  cepa  de  tecodontes  del  Triasico 
superior,  diferenciandose  por  la  adquisicion 
gradual  de  caracteres  tipicos  y  un  habito 
acuatico.  Proterochampsa  demuestra  que 
los  principales  rasgos  diagnosticos  del  orden 
Crocodilia  ya  existian  en  el  Triasico  medio. 
Una  reevaluacion  de  los  cocodrilos  primi- 
tivos conocidos  hasta  ahora  sugiere  la  posi- 
bilidad  de  que  durante  el  Triasico  habia  dos 
lineas  de  evolucion.  En  una  de  estas  lineas 
se  desarrollaron  los  rasgos  tipicos  del  orden 
Crocodilia,  la  mayor  parte  de  los  cuales  se 
ven  en  el  craneo  de  Proterochampsa.  La 
otra  linea,  caracterizado  por  Protosuchus,  se 
aparto  del  ambiente  tipico  de  los  cocodrilos 
hacia  una  vida  mas  terrestre,  con  la  subsig- 
uiente  adquisicion  de  los  rasgos  necesarios 
para  competir  con  tecodontes  del  mismo 
habito. 

Como  resultado  de  este  estudio  se  pro- 
pone la  creacion  de  la  familia  Protero- 
champsidae y  la  division  de  los  cocodrilos 
primitivos  en  dos  subordenes:  Protosuchia, 
constituida  por  Protosuchus  y,  presunta- 
mente,  los  tecodontes  esfenosucoideos,  y  un 
nuevo  suborden,  Archaeosuchia,  para  la 
Proterochampsidae  y,  provisionalmente,  la 
Notochampsidae. 

LITERATURE  CITED 

Bonaparte,  J.  F.      1962.     Descripcion  del  craneo 
y  mandibula  de  Exaeretodon  frengiiellii,  Ca- 


436         Bulletin  Museum  uf  Comparative  Zuulo^y,  Vol.  135,  No.  8 


brera.     Pub.    Nhis.    .\hinic.    Cienc.    Xat.   Trad. 

Mar   del   Plata,    1:    135-202. 
.      1963.     Promu.stodon.sciitni.s    hclhnani.    ii. 

g.  et  n.  sp.,  capitosaurido  tlt-l  Triasict)  Medio 

de  Argentina.    Amegbiniana,  3:  67-77. 
Broom,   R.      1904.     On   a   new   croeodilian   genus 

( Notocliatnpsa )    from    tbe    upper    Stormberg 

beds   of   Soutb    Africa.      CJeol.    Mag.,    Dec.    5, 

1:  582-584. 
.      1927.      On  Splicmi.nicliu.s  and  tlie  origin 

of   the   crocodiles.      Proc.   Zool.    Soc.    London, 

1927    (7):    359-370. 
BiiowN,  B.      1933.     An  ancestral  crocodile.    Anier. 

Mus.  Novit,  No.  638:   1-4. 
CAS.\.\nyuELA,    R.    M.      1960.      Notieia    preliminar 

sobre  dos  nuevos  estagonolepoideos  argentinos. 

Amegbiniana,  2:   3-10. 
COLBEIRT,    E.    H.    AND   C.    C.    MooK.      1951.      Tbe 

ancestral  crocodilian  Protosuchus.    Bull.  Amer. 

Mus.  Nat.  Hist.,  97:   149-180. 
Edinceh,    T.      1938.     Uber    Steinkerne    von    Hirn 

und  Obr-Holilen  der  Mesosucbier  GoniophoU.s 

und     Pliolidvsauni.s     aus     deni     Buckeburger 

Wealden.     Acta   Zoologica,    19:    467-504. 
Frenguelli,    J.      1948.      Estratigrafia    y    edad   del 

llaniado    Retico    en    la    Argentina.     Gaea,    8: 

159-309. 
H.^UGHTON,  S.  H.      1915.     A  new  tbeeodont  from 

tbe     Stormberg     beds.      Ann.     Soutb     African 

Mus.,    12    (3):    65-106. 
.      1924.     The    fauna    and    stratigraphy    of 

tile    Stormberg    series.     Ann.    South    African 

Mus.,  12   (8):    323-497. 
HuENE,  F.  VON.      192.5.      Die  Bedeutung  der  Splic- 

WMUc/iu.s-Gruppe  fiir  den  Ursprung  der  Krok- 

odile.     Zeitsebr.     Indukt.      Abstamminigs-und 

\'ererbunglebre,   38:    307-320. 
Kalix,  J.      1955.     Crocodilia.    /n:     Traite   de   Pa- 

leontologie,    ed.    J.    Piveteau,    vol.    5.     Paris, 

Masson  et  Cie,  pp.  695-787. 
Kermack,    K.    a.      1956.     An    ancestral    crocodile 


from  South  Wales.      Proc.  Linn.  Soc.  London, 
Session  166:   1-2. 

MooK,  C.  C.  1934.  Tbe  evolution  and  classifica- 
tion of  the  Crocodilia.  Jour.  Cleol.,  42:  295- 
304. 

Reig,  O.  a.  1959.  Nuevos  reptiles  arcosanrios 
del  Triasico  de  Ischigualasto.  Rev.  Asoc.  Geol. 
Argentina,  13:    257-270. 

RoMER,  A.  S.  1945.  Vertebrate  paleontology. 
Second  edition.  Chicago:  University  of  Chi- 
cago Press,  pp.  1-677. 

.        1960.        Vertebrate-bearing   continental 

Trias.sic  strata  in  Mendoza  region,  Argentina. 
Bull.  Geol.  Soc.  America,  71:   1279-1294. 

1962.     Tbe  fossiliferous  Triassic  deposits 


of    Ischigualasto,    Argentina.      Breviora,    Mus. 

Comp.  Zool.,  No.  156:    1-7. 
\'an   Hoepen,   E.   C.   N.      1915.      Contributions  to 

the  knowledge  of  tbe   reptiles  of  tbe   Karroo 

Formation.     Ann.  Transvaal  Mus.,   5:   83—87. 
Young,  C.   C.      1944.     On  a  supposed  new  pseu- 

dosuchian  from  the  Upper  Triassic  saurischian- 

bearing  red  beds  of  Lufeng,  Yunnan,  China. 

Amer.  Mus.  Novit.,  No.  1264:   1-4. 
.      1951.     Tbe  Lufeng  sauriscbian  fauna  in 

China.     Paleont.    Sinica,    134:    19-95. 
(  Received  4  ]amianj,  1966. ) 


ABBREVIATIONS 

am   auditory  meatus 
l)o   basioccipital 
bs   basisphenoid 
ec  ectopterygoid 
ex   exoccipital 
f  frontal 
i   jugal 

I  lacrimal 
m   maxilla 

II  nasal 

op  opisthotic 
p   parietal 


pa  paraoccipital 
pi   palatine 
pm   prcmaxilla 
po  postorbital 
prf  prefrontal 
pt  pter\goid 
q  quadrate 
qj   quadratojugal 
so   supraoccipital 
sq  squamosal 
\-  \omer 


Proterochampsa  and  crocodile  evolution-  •  Sill        437 


Table  I 
Table    of    Measurements    in    Millimeters 


MCZ      MACN 

3408       18165 


Total  length  of  the  skull,  from  the 
posterior  border  of  the  quadrate 
to  the  end  of  the  snout  at  the 
midline  395         440 

Length  of  the  skull  from  the  occip- 
ital crest  to  the  end  of  the 
snout  at  the  midline  325         375 

Length  of  the  skull  from  the  occip- 
ital condyle  to  the  end  of  the 
snout     350  - — 

Width   of  skull   between   external 

borders  of  (luadratojugals  270         — 

Width   of   skull  between   external 

borders  of  the  orbit 130         — 

Maximum  diameter  of  the  orbit  -      45  45 

Width  between  the  lateral  borders 

of  the  supratemporal  fenestrae  _      84         — 

Maximum  diameter  of  the  supra- 
temporal   fenestra   28  42 

Maximum   diameter  of  the  antor- 

bital    fenestra    38  44 

Maximmn  diameter  of  the  infra- 
temporal fenestra 71  89 

Width  between  the  medial  borders 

of    the    infratemporal    fenestrae   124         — 


MCZ    MACN 

3408      18165 


the    supratem- 


108 


oo 


Distance    between 

poral  fenestrae 30 

Distance  between  the  orbits  41 

Width  of  the  skull  between  the 
antorbital  fenestrae  58 

Width  of  the  snout  at  the  anterior 
border  of  the  antorbital  fenestrae 

Width  of  the  snout  at  the  "canine 
notch"    

Length  of  the  snout  from  the  an- 
terior border  of  the  orbit  to  the 
tip  of  the  premaxilla  238 

Length  of  the  snout  from  the  an- 
terior border  of  the  antorbital 
fenestra  to  the  tip  of  the  snout  ..   183 

Distance  from  the  anterior  border 
of  the  internal  nares  to  the  tip  of 
the  premaxillae  — 

Maximum  diameter  of  each  inter- 
nal naris  — 

Distance  from  the  posterior  border 
of  the  internal  nares  to  the  oc- 
cipital   condyle    144 

Length  of  maxillarv  tooth  row  .___    155 


109 
59 

280 

208 

184 
13 

207 


P — primitive 


Character 


Table  II 
Comparison  of  Characters  in  Crocodiles 


T — transitional 


A — advanced 


X — absent 


Development  of 

secondary  palate 
Position  of 

postorbital  bar 
Orientation  of  the 

orbits 
Orientation  of  the 

external  nares 
Dental  differentiation 
Relative  length  of  the 

snout 
Presence  of  palatal  teeth 
Sculptured  cranial  table 
Skull  height 
Antorbital  fenestra 
Auditory  canal 
\'ertcbral  structure 
Rib  articulations 


A 

P 

A 
P 
A 
A 
T 
P 
P 
P 


P 
P 

P 

? 

P 
P 

A 
X 
P 
T 


A 
T 

A 
A 
A 

A 
T 
T 
T 
A 


Archaeosuchia   Protosuchia       Mesosuchia         Eusuchia 


A 

A 

A 
A 

A 
A 
A 
A 
A 
A 
A 
A 


438        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  8 


Plate    I.      Dorsal   view  of   skull    of  Prolerochompsa  barrionuevol,  MCZ    3408.      X     '/2    approx. 


Proterochampsa  and  crocodile  evolution  •  Sill       439 


Plate    II.      Ventral    view   of  skull   of  Proterochampsa  barrionuevoi,    MCZ    3408,    showing    mandible    crushed    into    skull.    X     Vl 
approx. 


440        Bulleiln  Museum  of  Coui])aiatwe  Zoology,  Vol.  135,  No.  8 


Plate   III.      Inset  from  Plate   II;  arrows  show  pterygoid  teeth,  MCZ    3408.      X  2. 


Proterochampsa  and  crocodile  evolution  •  Sill        441 


#       ^'*%-«fc      -.ISA        £ 


"S. 


^ 


Plate   IV.     Ventral   aspect  of  skull   and   mandible   in  stereoscopic  view,  MCZ  3408.     X    '/s 


442        BiiUetin  Miiscimi  of  Comparative  ZooJogij,  Vol.  135,  No.  8 


t0 


• 


Plate  V.      Comparison   of  Proterochampso  barrionuevoi,   MCZ  3408    (top),    and   Crocod//us   n/loficus    (bottom),    in    dorsal    view 
X  % 


Proterochampsa   and  crocodile  evolution  •  S(7/   443 


Plate   VI.      Dorsal    view    of    Proterochampsa    barrionuevol ,    MACN    18165,    porfialiy    restored.      X     '/2    approx. 


444        Bulletin  Mu-scuDi  of  Comparative  Zoology,  Vol.  135,  No.  8 


Plate  VII.     Ventral   view  of  Proterochampsa  barrionuevoi,   MACN    18165,   showing   secondary  palate   and   internal    nares. 


Proterochampsa  and  crocodile  evolution  •  Sill        445 


^ 


K. 


%. 


'«&'' 


B 


Plate  VIM.      A,    B,    D,    Fourth  cervical   vertebra  in  dorsal,  ventral,  and  anterior  views,   respectively.     C,   Longitudinal   section  of 
fifth    cervical    vertebra,    MCZ    3408.      X  2. 


446        Bulletin  Miificiiiii  of  Comparative  Zoology,  Vol.  135,  No.  8 


■k^m 


f 


♦JSi 


B 


Plate   IX.     A,  Atlos,   axis,   and  third  cervicol   in  ventral   view,   anterior  end  to  the  left.     B,  Side  view  of  atlas,   axis,   and  third 
cervical,   anterior  end  to  the  right.     C,   and   D,   Fifth  cervical   in   side  view  and   longitudinal    section,    MCZ  3408.     X  2. 


Bulletin  OF  THE 

Museum   of 

Comparative 

Zoology 


Silicified    Silurian    Trilobites    from    Maine 


H.  B.  WHITTINGTON  AND  K.  S.  W.  CAMPBELL 


HARVARD   UNIVERSITY  VOLUME    135,   NUMBER  9 

CAMBRIDGE,  MASSACHUSETTS,   U.S.A.  JUNE  28,   1967 


PUBLICATIONS  ISSUED 

OR  DISTRIBUTED  BY  THE 

MUSEUM  OF  COMPARATIVE  ZOOLOGY 

HARVARD  UNIVERSITY 

Bulletin  1863- 

Breviora  1952- 

Memoirs  1864-1938 

JoHNSONiA,  Department  of  Mollusks,  1941- 

OccASioNAL  Papers  on  Mollusks,  1945- 

Other  Publications. 

Bigelow,  H.  B.  and  W.  C.  Schroeder,  1953.    Fishes  of  the  Gulf  of  Maine. 
Reprint,  $6.50  cloth. 

Brues,  C.  T.,  A.  L.  Melander,  and  F.  M.  Carpenter,  1954.  Classification  of  In- 
sects.  $9.00  cloth. 

Creighton,  W.  S.,  1950.  The  Ants  of  North  America.  Reprint,  $10.00  cloth. 

Lyman,  C.  P.  and  A.  R.  Dawe  (eds.),  1960.    Symposium  on  Natural  Mam- 
malian Hibernation.   $3.00  paper,  $4.50  cloth. 

Peters'  Check-hst  of  Birds  of  the  World,  vols.  2-7,  9,  10,  15.    (Price  list  on 
request. ) 

Turner,  R.  D.,  1966.  A  Survey  and  Illustrated  Catalogue  of  the  Teredinidae 
(MoUusca:  Bivalvia).    $8.00  cloth. 

Whittington,  H.  B.  and  W.  D.  I.  Rolfe  (eds.),  1963.  Phylogeny  and  Evolution 
of  Crustacea.   $6.75  cloth. 

Proceedings  of  the  New  England  Zoological  Club  1899-1948.    ( Complete  sets 
only. ) 

Publications  of  the  Boston  Society  of  Natural  History. 

Publications  Office 

Museum  of  Comparative  Zoology 

Harvard  University 

Cambridge,  Massachusetts  02138,  U.  S.  A. 


©   The   President   and    Fellows   of   Harvard    College    1967. 


SILICIFIED  SILURIAN  TRILOBITES  FROM  MAINE 


H.   B.   WHITTINGTON'  AND  K.   S.   W.   CAMPBELL- 


CONTENTS 

Introduction  and  acknowledgments   447 

Locality,   correlation   and   age  of   fauna   448 

Summary    of    morphological    and    taxonomic 

findings    ..- 449 

Systematic     paleontology     450 

Superfamily    Proetacea    450 

Family   Proetidae    451 

Proetus   pluteus   n.    sp.    451 

Family  Otarionidae 458 

Rhinotarion    n.    gen.    458 

R.  sentosum  n.  sp.  458 

Otarion    460 

O.   megalops    (M'Coy)    461 

O.   instita  n.   sp.   461 

O.   plautum   n.   sp.   46.3 

Otarion  sp.   ind.   463 

Otarionid   hypostome   464 

Otarionid  thoracic  segments  and  pygidium, 
type  A  464 


Otarionid  thoracic  segments  and  pygidium, 

type   B  465 

Otarionid  pygidium,  type  C  465 

Otarionid  pygidium,  type  D  465 

Otarionid    transitory    pygidia    465 

Comparisons  with  other  Silurian  species  ..  466 

Family  Scutelluidae  466 

Scutelluid    gen.    ind.    466 

Family   Encrinuridae   467 

Fragiscutum    n.    gen.    467 

F.    rhytium    n.    sp.    468 

Family  Dalmanitidae   473 

Dalmanites    puticulifrons   n.    sp.    473 

Family    Odontopleuridae    477 

Leonaspis    cf.    ivillUwm    477 

Xanionurus   n.    gen.   __  478 

X.   boucoti  n.  sp.   478 

Transitory    pygidium,    undetennined    481 

References   ..__  481 

Explanation  of  the   Plates  483 


INTRODUCTION  AND 
ACKNOWLEDGMENTS 

The  trilobite  exoskeletons  described  in 
this  account  are  silicified,  and  so  can  be 
freed  from  the  enclosing  rock  with  acid. 
They  are  not  only  the  best-preserved  speci- 
mens known  from  Silurian  rocks,  but  also 
the  only  ones  which  include  a  variety  of 
growth  stages.  They  provide  a  wealth  of 
new  information,  and  have  enabled  us  to 
refine  generic  and  familial  definitions,  as 
well  as  to  make  new  suggestions  regarding 
relationships  to  trilobites  of  older  and 
younger  systems.    It  is  a  fortunate  if  un- 

^  Sedgwick  Museum,  Cambridge,  England. 
-  Australian      National      University,      Canberra, 
A.C.T.,  Australia. 


expected  circumstance  that  such  a  fauna 
should  be  discovered  in  the  strongly  folded 
and  slightly  metamoq^hosed  rocks  of  north- 
western Maine.  The  very  excellence  of  the 
preservation,  however,  militates  against 
comparisons  with  less  well-preserved  and  in- 
complete specimens.  The  study  of  Silurian 
trilobites,  in  North  America  as  in  other 
continents,  has  been  neglected.  There  are 
no  modem,  critical  accounts  of  the  mor- 
phology and  stratigraphical  occurrence  of 
trilobites  from  the  type  Silurian  Series  in 
Britain,  nor  from  the  classical  regions  of 
North  America  such  as  northern  New  York 
State  and  the  central  states.  The  fauna 
from  Maine  is  peculiar  in  composition,  lack- 
ing the  illaenids,  calymenids,  cheirurids  and 
lichids  that  typify  earlier-described  Silurian 

Bull.  Mus.   Comp.  Zool.,   135(9):    447-483,  June,   1967       447 


448        BtiUetin  Miiscinn  of  Coinparafwe  Zoology.  Vol.  135,  No.  9 


faunas,  and  rich  in  proetids  and  otarionids 
that  are  the  least  well-known  elements  in 
such  North  American  faunas.  All  these  fac- 
tors combine  to  make  it  difficult  to  assess 
the  age  more  precisely  than  Middle  to 
Upper  Silurian,  and  rule  out  meaningful 
consideration  of  faunal  affinities. 

Terms  used  in  the  systematic  section  are 
those  defined  by  Harrington,  Moore  and 
Stubblefield  {in  Moore,  1959),  glabella 
being  used  to  include  the  occipital  ring. 
In  describing  Proetus,  the  anterior  subdivi- 
sion of  the  axial  ring  is  called  the  pre- 
annulus  (Richter  and  Richter,  1956;  cf. 
Moore,  1959,  fig.  292 ) .  Other  special  tenns 
and  notations  are  explained  in  Figures 
1,  7-10. 

Blocks  from  Baker  Pond  were  first  col- 
lected by  Professor  Arthur  J.  Boucot  in 
1952,  and  subsequently  prepared  by  Dr.  A. 
R.  Palmer,  U.S.  Geological  Survey.  Whit- 
tington  is  grateful  to  Dr.  Palmer  for  invit- 
ing him  to  study  this  material,  and  to  Profes- 
sor Boucot  for  guiding  him  to  the  locality  and 
helping  to  make  a  large  additional  collec- 
tion in  1959.  This  material  was  prepared  at 
the  Museum  of  Comparative  Zoology,  and 
Campbell's  visit  in  1965  gave  us  the  op- 
portunity to  study  it.  We  express  our  thanks 
to  the  National  Science  Foundation,  Grant 
GB-3577,  for  haxing  made  Campbell's  visit 
possible,  and  for  paying  the  costs  of  print- 
ing the  plates;  Grant  GB-1807  provided 
technical  assistance  to  Whittington.  Text- 
figures  have  been  drawn  by  Mr.  Arnold 
Clapman.  Mrs.  Marjorie  Korringa  has  pre- 
pared all  the  photographic  prints,  lettered 
the  figures,  and  helped  in  many  other  ways. 
The  major  part  of  the  collection,  including 
all  type  and  figured  specimens,  is  deposited 
in  the  U.S.  National  Museum  (abbreviated 
below  as  USNM). 

Dr.  David  L.  Bruton  kindly  gave  us  un- 
published information  on  certain  odonto- 
pleurid  species,  Dr.  J.  S.  Jackson  lent  type 
specimens  from  the  National  Museum  of 
Ireland,  and  Dr.  A.  Martinsson  lent  the 
type  specimen  of  Pwetiis  concinnus  from 
the     Palaeontological     Institute,     Uppsala, 


Sweden.  We  are  also  indebted  to  Mr.  R. 
P.  Tripp  for  discussing  with  us  the  new 
encrinurid  genus  and  notation  of  glabellar 
tubercles,  and  to  Dr.  Robert  R.  Hessler 
for  comments  on  certain  morphological  as- 
pects of  these  trilobites,  but  we  assume 
responsibility  for  interpretations  expressed 
herein. 

LOCALITY,  CORRELATION,  AND  AGE 
OF  THE  FAUNA 

The  blocks  of  grey,  silty  limestone  con- 
taining the  trilobite  fauna  were  selected 
from  those  lying  on  the  central  part  of  the 
east  shore  of  Baker  Pond,  centre  of  Spencer 
Quadrangle,  Somerset  County,  Maine. 
These  limestones  are  different  in  appearance 
and  fossil  content  from  others  that  are 
present  to  the  northeast  and  southwest,  all 
included  within  the  outcrop  of  the  Hard- 
wood Mountain  Formation  of  Boucot  (1961, 
pi.  34).  Consideration  of  the  regional  ge- 
ology (James  B.  Thompson,  Jr.,  personal 
communication)  suggests  that  the  blocks 
cannot  have  been  derived  from  the  north- 
west, but  that  they  came  from  a  part  of  the 
formation  now  concealed  by  glacial  de- 
posits. 

When  dissolved  in  dilute  hydrochloric 
acid  the  blocks  yielded  a  residue  of  silici- 
fied  trilobite  exoskeletons,  ostracode  valves, 
rare  bryozoans  and  brachiopods.  The  re- 
placement of  the  trilobite  exoskeletons 
preserves  in  remarkable  detail  the  original 
surfaces.  In  large  specimens  the  replace- 
ment is  in  the  form  of  two  layers,  one  at 
the  original  outer  surface,  the  other  at  the 
original  inner  surface  (e.g.  Pi.  1,  figs.  1-3; 
PI.  10,  fig.  11;  PI.  11,  fig.  20;  Pi.  13,  fig.  17). 
These  two  layers,  separated  by  a  narrow 
space,  are  readily  apparent  in  damaged 
specimens.  In  the  smaller  specimens  a  single 
layer  replaces  the  entire  thickness  of  the  exo- 
skeleton.  Silt  grains  adhering  to  the  sur- 
face are  difficult  or  impossible  to  remove 
without  damaging  the  specimen,  and  may 
be  seen  in  many  photographs  (e.g.  Pi.  1, 
figs.  5,  23,  24,  30,  31).  Table  1  lists  the 
trilobite  faima  and  gives  a  measure  of  its 


Silurian  Trilobites  •  Whittinfiton  and  Campbell        449 


relative  abundance.  Dr.  Jean  M.  Berdan 
{in  Boucot,  1961,  p.  181)  named  four 
ostracode  genera  from  the  formation.  How- 
ever, Dr.  Berdan  informs  us  ( personal  com- 
munication) that  ostracodes  from  the  silici- 
fied  blocks  are  different  from  those  obtained 
from  other  outcrops  of  the  formation,  and 
show  some  resemblance  to  ostracodes  from 
the  Henryhouse  Fonnation,  Oklahoma.  Pro- 
fessor Boucot  (personal  communication) 
notes  that  the  brachiopod  fauna  from  the 
Baker  Pond  locality  is  rather  different  from 
that  at  other  localities  assigned  to  the  Hard- 
wood Mountain  Formation.  He  concludes 
that  the  age  is  within  the  span  of  Wenlock 
to  early  Ludlow,  and  in  tenns  of  the  North 
American  sequence  could  be  as  old  as  the 
Waldron  Shale.  The  trilobite  fauna  is  dom- 
inated by  proetids  and  otarionids,  Ameri- 
can Silurian  species  of  which  are  poorly 
and  incompletely  known,  and  less  common 
is  the  new  encrinurid  genus  Fmgiscntujii. 
Encrinurids  are  not  known  to  be  present  in 
rocks  of  Devonian  age,  so  that  the  Silurian 
age  of  the  fauna  seems  beyond  question. 
North  American  mid-continental  Silurian 
trilobite  faunas  include  illaenids,  cheirurids, 
calymenids  and  lichids  (cf.  Weller,  1907; 
Raymond,  1916;  Walter,  1927),  but  these 
families  are  not  represented  in  the  Baker 
Pond  collection.  Difficulties  in  correlation 
stem  from  this  lack,  and  from  the  lack  of 
recent  work  on  Silurian  trilobites.  Campbell 
(in  press)  has  studied  trilobites  of  the 
Henryhouse  Formation,  Oklahoma,  and 
while  no  one  species  is  in  common  with  the 
Baker  Pond  fauna,  those  of  Proetus, 
Fragiscutum  n.  gen.,  and  Dalmanites  are 
alike.  The  Henryhouse  Formation  has  not 
yielded  otarionids,  and  only  a  poorly  pre- 
served free  cheek  of  Leonaspis.  The  Baker 
Pond  species  of  Leonaspis  is  compared  to 
younger.  Lower  De\'onian  species,  simply 
because  these  are  the  only  well-known 
American  species  to  which  comparisons  can 
be  made.  Species  of  Leonaspis  are  known 
to  be  long-ranging,  and  this  single  com- 
parison cannot  be  taken  to  imply  that  the 
Baker  Pond  fauna  is  early  Dexonian  in  age. 


The  balance  of  the  evidence  suggests  that 
it  is  of  about  the  same  age  as  the  Henry- 
house— that  is,  near  the  Wenlock-Ludlow 
boundary.  This  age  is  consistent  with  com- 
parisons between  Baker  Pond  and  Bohe- 
mian, British,  and  Swedish  species  made  in 
the  systematic  section,  and  with  age  deter- 
minations based  on  brachiopods  and  ostra- 
codes. Trilobites  of  Lower  Silurian  (  Lland- 
overy Series)  age  are  not  well  known,  but 
such  comparisons  as  can  be  made  do  not 
suggest  that  the  Baker  Pond  fauna  is  early 
Silurian. 

SUMMARY  OF  MORPHOLOGICAL  AND 
TAXONOMIC  FINDINGS 

The  new  morphological  information, 
combined  with  data  revealed  by  the  growth 
stages,  gives  clues  to  relationships  with 
older  and  younger  groups.  Some  of  the 
main  points,  elaborated  in  the  systematic 
section,  are: 

1.  Ordovician  to  Devonian  proetids, 
typified  here  by  Proetus,  appear  to  be  an- 
cestral to  Carboniferous  genera  like  Paladin. 
A  triangular  rostral  plate  and  the  apparent 
absence  of  a  sutural  junction  between  the 
hypostome  and  the  remainder  of  the 
cephalon  characterise  Proetus.  In  Paladin, 
however,  the  rostral  plate  is  subquadrangu- 
lar  in  outline  and  the  hypostome  is  joined 
to  the  remainder  of  the  cephalon  by  a 
suture. 

2.  Otarion  (Ordovician  to  Devonian) 
also  has  a  triangular  rostral  plate.  Growth 
stages  are  remarkably  like  those  of  the 
Ordovician  Dimeropygc,  and  suggest  der- 
ivation from  this  older  group.  The  sup- 
posed brachymetopid  genus  Cordania  ap- 
pears likely  to  be  derived  from  otarionids. 

3.  Within  the  Proetacea,  as  presently 
conceived,  there  are  thus  two  main  phyletic 
lines,  one  leading  through  proetids  to  cer- 
tain Carboniferous  genera,  the  other  be- 
ing the  dimeropygid-otarionid-brachymeto- 
pid  line. 

4.  Three  species  of  Otarion  and  one 
made  the  type  of  a  new  otarionid  genus 
are  described,  based  on  cephala.    Isolated 


450        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Table  1.  Trilobite  fauna  of  the  Hardwood 
Mountain  Formation  at  Baker  Pond,  Somer- 
set County,  Maine.  Indication  of  relative 
abundance  crv'EN    BY    total    numbers    of   cra- 

NIDIA  of  all  sizes  (  INCXUDING  FRAGMENTARY 
specimens)     in    the    SAMPLE    DISSOLVED. 

Pivctus  pJutcus  n.  sp.  150 

Rhinotarion  sento.stiin   n.   gen.,  n.   sp.  8 

Otarioti  in.stita  n.  sp.  148 

Otarioii   ))l(mtuni   u.   sp.  44 

Otarion   sp.   iiid.  38 

Scutclluid   gen.   iiul.  2 

Fni^i.scutuni   rhytiiiin   n.   gen.,  n.   sp.  18 

Dalmanitcs  puticulifrons  n.  sp.  9 

Xanionurus  houcoti  n.   gen.,  n.  sp.  8 

Leonaspis  cf.  tvilliamsi  Whittington,  1956  10 


segnient.s  and  pygidia  can  only  in  a  few 
cases  be  assigned  to  these  species.  Whether 
these  are  four  distinct  species,  or  whether 
one  fonn  may  be  a  sexual  dimorph  of  an- 
other, is  an  open  question. 

5.  The  type  material  of  M'Coy's  Harpi- 
(lella  me^alops  is  redescribed,  and  it  is 
concluded  that  the  generic  name  is  a  sub- 
jective synonym  of  Otarion. 

6.  A  new  encrinurid  genus  is  based  on 
superbly-preserved  material,  including  the 
first  described  developmental  stages.  These 
show  that  each  ring  of  the  pygidial  axis 
represents  a  segment.  Anterior  bands  of 
the  thoracic  pleurae  are  reduced  to  articu- 
latory  flanges,  concealed  in  dorsal  aspect. 
The  pleural  ribs  of  both  thorax  and 
p>gidium  are  posterior  bands.  Canals, 
diminishing  in  diameter  dorsally,  traverse 
these  bands  and  the  axial  rings,  but  appear 
not  to  open  on  either  surface. 

7.  Incomplete  developmental  stages  of 
Dalmanitcs,  beginning  with  the  late  pro- 
taspis,  reveal  the  remarkable  similarity  to 
similar  stages  of  the  ancestral  Ordovician 
Dahnanitina. 

S.  l)cvelopm(>ntal  stages  of  two  odonto- 
pleurid  species  show  the  same  major  spine 
pattern  as  species  of  various  Ordovician 
genera.  One  species,  type  of  a  new  genus, 
appears  to  lie  on  a  phyletic  line  between 
the  Ordovician  Diacanthaspis  and  the  De- 
vonian   Radia.spis. 


SYSTEMATIC  PALEONTOLOGY 
Superfamily   PROETACEA   Salter,    1864 

Discussion.   The  discovery  by  Dr.  W.  T. 
Dean  (personal  communication)  of  a  spe- 
cies   of   Otarion    and    a    proetid   from   the 
Arenig  of  southern  France  shows  the  early 
separation   between   these   two   groups.    A 
third    group,    the    dimeropygids,    is    also 
present  in  the  early  Ordovician  (Whitting- 
ton,   1963:     45-50).     Phaseolops   from   the 
Llanvirn    of   Newfoundland    (Whittington, 
1963:  36-40)  is  considered  to  be  a  proetid, 
and  among  other  characters   it  exhibits  a 
triangular   rostral   plate,    though   the   axial 
rings  of  the  thorax  do  not  have  the  pre- 
annulus.    In  the  later  Ordovician  and  the 
Silurian,  proetids  and  otarionids  are  more 
abundant    and    widespread.     The    present 
material  of  Otarion  shows  the  characteristic 
triangular  rostral  plate,  the  narrow  pander- 
ian   notch   situated   close   to   the   posterior 
margin  of  the  segmental  doublure,  the  ab- 
sence of  the  pre-annulus,  the  characteristic 
long  median  spine  of  one  thoracic  segment, 
and  a  development  of  both  cephalon  and 
pygidium  which  recalls  that  of  Dimeropijge 
(Whittington   and  Evitt,   1954).    The  spe- 
cies of  Proetus  described  here  reveal  the 
triangular    rostral    plate,    the    typical    pre- 
annulus  of  the  segments,  and  the  V-shaped 
panderian  notch  which  is  medially  situated 
in    the    doublure    of    the    segments.     The 
hypostome   is    characteristic   in    shape  but 
appears  not  to  be  attached  at  a  hypostomal 
suture  to  the  rostral  plate  and  doublure  of 
the  free  cheeks.    The  early  developmental 
stages    of    Proetus    differ    from    those    of 
otarionids,  notably  in  the  absence  of  paired 
spines.    A  relationship  between   otarionids 
and  dimeropygids  is  suggested  not  only  by 
the  development  but  also  by  many  aspects 
of    holaspid     morphology,     including     the 
single  median  thoracic  spine  characteristic 
of  many  species. 

Relationships  between  Silurian  otarionids 
and  proetids  and  Devonian  and  younger 
proetaceans  present  many  problems.  For 
example,  the  Lower  Devonian  genus  Cor- 


Silurian  Trilobites  •  Whittitifyfon  and  Campbell        451 


dania  ( Whittington,  1960)  has  an  otarionid-  and  possibly  of  FhiUipsia  (1963,  pi.  61,  figs, 
like  ceplialon,  a  wide  (tr. )  sub-triangular  3,  4)  a  pre-annulus  is  present  on  the  first 
rostral  plate,  and  the  thoracic  segments  axial  ring  of  the  pygidium,  while  in  other 
lacking  the  pre-annulus,  but  the  pygidium  species  he  describes  this  ridge  is  not  de- 
is  unlike  the  otarionid  pattern  in  that  it  is  veloped.  Use  of  this  thoracic  and  pygidial 
large  and  has  some  ten  axial  rings.  In  re-  character,  as  well  as  ventral  cephalic 
cent  publications  Coidania  has  been  placed  characters,  may  help  to  disentangle  the 
in  the  Brachymetopidae  (Whittington,  I960;  relationships  of  Carboniferous  and  younger 
Amos,  Campbell,  and  Goldring,  1960;  trilobites  to  each  other  and  to  different 
Hessler,  1962a;  Hahn,   1964),  a  Devonian-  Devonian  groups. 

Carboniferous  group  embracing  fomis  with  This  discussion  suggests  that  from  early 

the  above  characters,  except  that  two  of  its  Ordovician  onward  there  may  have  been 

Carboniferous     members,     Brachymctopus  two  main  phyletic  lines  within  Proetacea — 

and    Anstralosutum    have    a    rostral    plate  a  proetid  line  leading  to  various  Carbonif- 

that  is  not  sub-triangular  but  extends  close  erous  groups,  and  a  dimeropygid-otarionid 

to  the  genal  angle.    We  suggest  that  Cor-  line  from  which  at  least  some  brachymeto- 

dania  was  derived  from  an  otarionid  rather  pids  may  be  derived, 
than    a   proetid,    but   more    information   is 

needed  before  lines  of  descent  of  the  Car-  Family   PROETIDAE   Salter,    1864 

boniferous    Brachymetopidae    can    be    dis-  Subfamily  PROETINAE  Salter,  1864 

cemed.  Genus  PROETUS  Steininger,  1831 

Silicified  material  of  the   Carboniferous  Proefus  pluteus  n.  sp, 

genus  Paladin   (Whittington,  1954)   shows  Plates  1,  2;  Plate  3,  figs.  6-8,  10,  13-16; 

that  it  is  proetid-like  in  the  shape  and  ar-  Figs.    1-5,  6C. 

rangement  of  the  glabellar  furrows,   pres-  Holofypc.   USNM  154457,  cranidium  and 

ence  of  the  shallow  panderian  notch  in  the  fj.^^  cheek. 

cephalic  doublure,  the  form  of  the  seg-  Description.  Glabella  widening  from  the 
ments,  nature  of  articulating  arrangements,  posterior  edge  to  a  maximum  width  across 
outline  of  the  panderian  notch  on  the  midpoint  of  lateral  occipital  lobes,  narrow- 
thoracic  segments,  and  the  presence  of  the  j^g  abruptlv  fonvard  to  the  anterior  margin 
pre-annulus.  On  the  other  hand,  however,  ^f  ^his  lobe,  less  abruptlv  inside  the  pal- 
the  hypostome  of  Pahdin  has  a  wing  p^bj-^i  lo^^s,  to  the  well  rounded  frontal 
process  on  the  large  anterior  wing  which  \q\^q  Axial  furrow  lightly  impressed,  except 
appears  to  rest  in  a  depression  on  the  posteriorly;  preglabellar  and  anterior  bor- 
surface  of  the  anterior  boss,  and  there  is  a  jer  furrows  confluent  medially,  deeper, 
hvpostomal  suture  linking  it  to  the  rostral  Occipital  furrow  with  an  almost  vertical 
plate  and  the  doublure  of  the  free  cheeks,  anterior  slope  and  more  gentle  posterior 
The  rostral  plate  is  transverse,  subtrape-  slope,  the  course  gently  convex  forward 
zoidal  in  outline.  The  shape  of  the  plate  medially,  swinging  fonvard  and  outward 
and  mode  of  attachment  of  the  hypostome  and  deepening  in  front  of  the  lateral  lobe; 
distinguish  Paladin  from  Proetus  but  the  latter  moderately  convex,  extending  back 
features  in  common  indicate  that  Paladin  almost  to  posterior  margin  of  ring,  sepa- 
and  its  allies  may  have  been  derived  from  rated  from  ring  by  a  furrow  which  becomes 
the  proetids.  faint   distally.     Glabellar   furrows   may   be 

Hessler    (1962b;    1963;    1965)    has    de-  faintly  impressed  on  external  surface  and 

scribed  and  discussed  Lower  Carboniferous  are  indicated  by  smooth  areas;  shape  and 

trilobites  which  he  refers  to  the  Proetidae.  arrangement  shown  in  Figure  1 A  ( compare 

His    figures    suggest    that    in    species    of  Pi.    1,   fig.   8).    On   the   inner  surface   the 

Griffithidcs   (1962b,  pi.   176,   figs.   12,   14)  muscle  areas  may  also  appear  faintly  im- 


452        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Figure  1.  Proetus  pluteus  n.  sp.  A)  Muscle  areas  of  glabella  and  notation  of  points  on  facial  suture  (after  Richter  and 
Richter,  1940)  based  on  original  of  Plate  1,  figure  8.  B)  Restoration  of  rostral  plate,  bosed  on  originals  of  Plate  2, 
figure   2.     Abbreviations:     cs,    connective    suture;    r    pi,    rostral   plate;    rs,    rostral    suture. 


pressed,  or  the  replacing  silicification  may 
be  broken  over  these  areas  (Pi.  2,  fig.  2). 
Eye  lobe  elongate,  gently  convex  eye 
snrface  steeply  sloping,  apparently  smooth 
externally  and  internally — the  preservation 
does  not  reveal  any  facets.  Palpebral  lobe 
flattened.  Outside  eye  lobe,  cheek  curves 
down  to  borders;  posterior  border  widen- 
ing outward  and  curving  posterolaterally 
into  base  of  short,  pointed  fixigenal  spine. 
Posterior  border  furrow  moderately  deep, 
curving  out  on  to  base  of  fixigenal  spine; 
lateral  and  anterior  borders  broad  and 
gently  convex,  separated  from  the  cheek 
by  a  broad  shallow  border  furrow.  Dou- 
blure (PI.  2,  figs.  2,  3)  of  approximately 
same  width  as  lateral  and  anterior  borders, 
gently  convex  ventrally,  the  inner  edge 
curved  up  beneath  the  groove  formed  by 
the  border  furrows.  Beneath  posterior 
border,  doublure  extends  in  to  the  fulcrum 
(PI.  1,  fig.  4);  inside  here,  edge  of  exoskele- 
ton  has  well-developed  recess  for  articulat- 
ing flange  of  first  thoracic  segment;  oc- 
cipital doublure  very  short  ( exs. )  behind 
the  lateral  occipital  lobes,  but  lengthening 
to  about  three-quarters  length  of  occipital 
ring  in  midline.  Shallow  panderian  notch 
in  margin  of  lateral  border  in  front  of 
genal  angle  ( PI.  2,  fig.  2 ) .  Course  of  sutures 
shown     in     Figure     1      (compare     Pi.     2, 


fig.  2).  Anterior  branch  runs  inward  and 
forward  from  /?  to  a  over  edge  of  an- 
terior border  and  continues  inward  across 
doublure  as  the  connective  suture,  the 
two  sutures  meeting  in  the  midline  at 
the  inner  margin  of  the  doublure.  The 
rostral  suture  runs  along  the  doublure 
a  short  distance  in  from  the  anterior  mar- 
gin, and  thus  isolates  a  triangular  rostral 
plate.  Isolated  examples  of  this  plate  have 
not  been  found,  but  the  outline  of  the 
doublure  of  the  free  cheek  leaves  no  doubt 
as  to  its  shape. 

Hypostome  with  strongly  convex  central 
body  partly  subdivided  by  short,  deep, 
backwardly-directed  middle  furrows  into  a 
large  anterior  and  a  small,  crescentic  poste- 
rior lobe;  in  midline  at  anterior  margin  of 
anterior  lobe  is  a  triangular,  flattened  or 
gently  concave  area  (Pi.  1,  fig.  13).  Border 
furrows  well  defined  except  beside  anterior 
wing;  anterior  border  narrow,  convex,  the 
edge  ( which  presumably  faces  forward  and 
downward)  having  a  deep  slot  along  the 
median  portion  (Pi.  1,  figs  13,  18).  Cres- 
centic posterior  body  gently  inflated,  es- 
pecially at  the  tip  where  the  oval  macula  is 
faintly  defined  by  its  convexity  (Pi.  2,  figs. 
1,  4).  Lateral  border  widening  posteriorly, 
posterior  border  flattened  and  bearing  a 
short,    blunt    spine    at    the    posterolateral 


Silurian  Trilobites  •  Whitimp,ton  and  Campbell        453 


angle.  Anterior  wing  consists  of  an  upward  the  anterior  margin;  in  centre  of  this  socket 

and   outwardly   directed   extension    of   the  is  the  small  axial  process    (Pi.  2,  fig.  23) 

edge    of    the    doublure,    subtrapezoidal    in  which  fits  into  a  corresponding  axial  socket, 

outline,  without  a  wing  process;  posterior  The  inner  anterior  edge   of  the   pleura  is 

wing  small,  subtriangular,  directed  upward  rounded,    and    fits    into    a    groove    on    the 

and    slightly    inward;    doublure    between  posterior   edge    (Pi.    2,    figs.    24,    25,    31). 

wings  narrow,  widest  posterolaterally.  Doublure  of  ring  similar  in  fomi  to  that 

External  surface  of  cephalon  ( PI.  1,  figs,  beneath  the  occipital  ring,  medially  extend- 

8,  11,  13,  14,  17,  18;  PL  2,  figs.  1,  4)  inside  ing  forward  as  far  as  articulating  furrow; 

borders   bearing   fine,    evenly   and    closely  anterior  edge  of  articulating  halfring  with 

spaced  granules;  these  granules  absent  in  the  a  deep  slot  along  its  whole  width   ( Pi.  2, 

furrows  and  on  the  inner  part  of  the  pal-  fig.  23). 

pebral  lobe;  posterior  border  apparently  Pygidium  with  axis  ill-defined  at  tip; 
smooth  along  the  crest;  outer  part  of  an-  first  ring  stands  markedly  higher  than  suc- 
terior  and  lateral  borders  and  doublure  ceeding  seven  or  eight  rings.  Pleural  re- 
bearing  slightly  irregular  terrace  lines  gions  with  inner  part  adjacent  to  axis 
running  subparallel  to  the  margin,  these  horizontal,  outer  part  sloping  steeply  down 
terrace  lines  continuing  on  to  the  genal  to  margin,  no  border  furrow.  Three  or  four 
spine.  On  hvpostome,  terrace  lines  on  an-  pleural  and  inteipleural  furrows  visible  on 
terior  lobe  of  middle  body  run  subparallel  pleural  regions,  dying  out  distally  so  that 
to  margins,  fanning  out  anteriorly  beside  border  is  smooth,  especially  posteriorly 
the  flattened  triangular  area.  Posterior  (PL  2,  figs.  26,  27).  Broad,  gently  convex 
lobe  including  macula  apparently  smooth;  doublure  extends  inward  for  about  one- 
terrace  lines  along  borders,  subparallel  to  third  the  width  of  the  pleural  regions, 
margins,  becoming  more  widely  spaced  and  External  surface  of  rings  and  inner  parts 
curving  on  the  outer  surface  of  the  wings,  of  pleurae  of  thorax  and  pygidium  with 
Number  of  thoracic  segments  unknown,  fine  granulation;  terrace  lines  on  pygidial 
Ring  subdivided  by  intra-annular  furrow  doublure  and  around  dorsal  margin; 
which  curves  for\vard  distally  and  joins  the  branches  from  these  lines  curve  fonvard 
more  deeply-incised  articulating  furrow  and  inward  across  the  pleural  region  sub- 
some  distance  in  from  the  axial  furrow;  parallel  to  the  pleural  and  interpleural 
inner  part  of  pleura  flat,  outer  part  steeply  furrows  ( PL  2,  fig.  26 ) . 
bent  down;  inner  part  relatively  narrow  Variation  and  development  of  cranidium. 
'  (tr. )  on  anterior  segments  (PL  2,  figs.  7,  Measurements  of  a  size  series  of  cranidia 
8)  and  the  outer  part  deflected  backward.  (Figs.  2,  3)  show  some  aspects  of  the 
Pleural  furrow  runs  diagonally  out  beyond  variability,  and  that  this  variability  is  con- 
fulcrum  and  along  edge  of  broad  facet,  tinuous.  A  size  series  of  cranidia  (PL  1, 
Doublure  extends  along  posterior  margin  of  figs.  5-7,  9,  10,  15,  16,  19-21,  23-31;  Fig. 
outer  part  of  pleura  and  beneath  tip,  with  6c),  ranging  from  a  length  (sag.)  of  1.3 
a  broad,  deep  panderian  notch  (PL  2,  fig.  mm  upward,  shows  the  relatively  minor 
25).  The  anterior  edge  of  this  notch  is  changes  that  take  place.  Most  noticeable 
raised  to  fomi  a  stop  during  enrollment;  in  is  the  change  in  longitudinal  convexity — 
anterior  segments  this  anterior  part  of  the  from  a  steep  slope  in  front  of  the  eye  lobe 
doublure  is  narrow  and  not  raised  ( PL  2,  to  a  lesser  slope,  combined  with  an  increase 
fig  24).  Articulation  between  anterior  seg-  in  the  inflation  of  the  anterior  border 
ments  of  the  thorax,  the  cephalon,  and  the  which  gives  a  deeper  border  furrow, 
pygidium  is  facilitated  by  the  ring  process  Variation  and  development  of  pygidium. 
situated  at  the  distal  posterior  margin  of  A  series  of  specimens  from  transitory 
the  axial  ring,  which  fits  into  a  socket  on  pygidia  (PL  3,  figs.  10,  13-16)  to  the  small- 


454        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


10-1 


9- 


8- 


7- 


6- 


5- 


4- 


3- 


2- 


CRANIDIA 


E 
E 

CO 

I 

CO 


•  •  •• 


Length    (mm) 


T 


T 


-I 

10 


T 


T 


T 


Figure  2.  Proefus  plufeus  n.  sp.  Dimensions  of  50  cronidio;  length  is  sagittal;  5-5  is  width  across  palpebral  lobes  at 
widest  point.  Measured  cranidia  include  USNM  154458-60,  154464-6,  154475,  154478,  figured  specimens,  remainder  in- 
cluded under  154489. 


est  true  pygidia  (Pi.  3,  figs.  6-8)  to  larger 
ones  (PI.  2,  figs.  17,  20,  26-30)  shows  that 
there  is  little  change  in  the  series,  apart 
from  a  tendency  to  become  slightly  broader. 
At  a  small  size,  the  first  ring  stands  higher 
than  the  others.  The  specimens  are  vari- 
able through  a  considerable  size  range,  as 
shown  in  Figures  4,  5.  There  is  no  discon- 
tinuity within   this   variation   and  the  ma- 


terial all  appears  to  belong  to  one  species. 

Discu.ssion.  The  present  material,  which 
is  of  dissociated  exoskeletal  parts  except 
for  one  cephalon,  is  regarded  as  a  single 
species  because  of  the  continuous  range  of 
variation.  This  range  is  wide,  but  the 
material  is  from   a  single   locality. 

Deep,  smooth-edged  slots  run  along  the 
edge  of  the  rostral  suture  of  the  cranidium 


Silurian  Trilobites  •  \Vhiffinp,tou  and  Campbell        455 


1.5  -I 

CRANIDIA 


1.0- 


0.5- 


£ 
E 


O 
J2 


03 

c 
o 


CD 

c 
<1> 


Length      (mm) 


1 — 1 1 1 \ 1 1 1 1 1 

1  23456789  10 

Figure    3.      Proefus    pluteus    n.    sp.      Dimensions    of    49    cranidia,    both    lengths    sagittal.     Measured    specimens    the    some    as 
Figure    2. 


(PI.  1,  fig.  4;  PL  2,  fig.  2),  around  the  edge  not  clear  what  the  function  of  this  uncalci- 
of  the  articulating  halfring  of  thoracic  seg-  fied  integument  may  have  been.  Between 
ments  (PI.  2,  fig.  23)  and  the  pygidium,  and  the  cephalic  doublure  and  the  hypostome 
along  the  sutural  margin  of  the  hypostome  its  function  may  have  been  articulatory, 
(PI.  1,  fig.  13;  PL  2,  fig.  1).  The  smooth  for  not  only  is  the  sutural  margin  of  the 
edges  of  the  slots,  and  their  consistent  form  hypostome  rounded  and  slotted  ( PL  2,  fig. 
in  many  specimens,  argue  against  these  1),  but  also  the  margin  of  the  doublure  of 
features  being  the  result  of  silicification,  for  the  cheek  adjacent  to  the  rostral  plate  ( PL 
example  as  being  a  gap  between  a  silicified  2,  fig.  3 ) ;  these  edges  did  not  adjoin  along 
layer  on  the  external  and  internal  surfaces  flat  surfaces.  It  is  notable  also  that  the 
of  the  exoskeleton.  Such  double  layering  is  margin  of  the  anterior  wing  of  the  hy- 
seen  (PL  1,  figs.  1-3,  10),  but  where  it  is  postome  is  thin  and  rounded,  that  the  wing 
broken  through  at  the  edge  of  a  specimen  bears  no  wing  process,  and  thus  that  there 
the  margins  of  the  two  layers  are  not  is  no  evidence  of  a  close  link  between  this 
straight  but  ragged  and  irregular.  It  ap-  wing  and  an  anterior  pit;  indeed,  there  ap- 
pears to  us  that  the  slots  represent  parts  pears  to  be  no  anterior  pit  in  this  species, 
of  the  integument  that  were  uncalcified  in  Thus  the  hypostome  of  Procfus  was  not 
life.  In  the  thorax,  where  such  integument  rigidly  fitted  to  the  remainder  of  the  cepha- 
extended  forward  from  the  margin  of  the  Ion,  as  it  appears  to  be  in,  for  example, 
articulating  halfring  to  the  doublure  of  the  Fragiscutum  n.  gen.  (see  below),  and  may 
ring  in  front,  it  must  have  been  articulatory  have  been  movable, 
in  function.    Along  the  rostral  suture  it  is  So  far  as  we  are  aware,  the  hypostome  of  a 


456        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


12-1 


11- 


10- 


9- 


8- 


7- 


6- 


5- 


4- 


3- 


2- 


1- 


PYGIDIA 


E 
E 


•  < 


Length     (mm) 


1 


Figure  4,  Proetus  p/ufeus  n.  sp.  Dimensions  of  40  pygidia; 
length  is  sagittal  (excluding  articulating  halfring);  width 
is  maximum.  Measured  pygidia  include  USNM  154482-5, 
figured    specimens,    remainder    included    under    154490. 


proetid  has  not  been  found  in  place  in  any 
specimen.  This  evidence  supports  the  view 
that  there  was  no  sutural  Hnk  between  the 
hypostome  and  the  rest  of  the  cephalon, 
but  rather  that  the  hypostome  was  kept  in 
position  liy  uncalcified  integument  and 
muscles,  w  hich  decayed  after  death  so  that 
the  hypostome  was  not  retained  in  its 
original  position.  The  attachment  of  the 
hypostome  to  the  remainder  of  the  cephalon 
has  been  discussed  in  Dimeropyge  (Whit- 


tington  and  Evitt,  1954:  38-41,  text-fig.  8), 
and  in  that  genus  there  is  no  sutural  junction 
between  the  inner  edge  of  the  doublure 
and  the  anterior  margin  of  the  hypostome. 
Other  points  of  comparison  between  Dim- 
cropijiie  and  the  present  species  are  the 
doublure  of  the  outer  part  of  the  pleurae, 
the  panderian  notch,  and  the  form  of  the 
stop  to  enrollment,  all  of  which  appear  to 
be  similar  ( compare  Plate  2,  figure  25,  \\\ih 
Whittington  and  Evitt,  1954,  text-fig.  10). 
New  illustrations  are  given  here  of  the 
type  species  of  Vroetus,  P.  concinmis  (PI. 
3,  figs.  4,  5,  9,  11,  12),  from  the  Wenlock 
Series,  Gotland,  Sweden.  The  broken  an- 
terior border  shows  on  the  right  side  of  the 
specimen  the  mould  of  the  doublure,  and 
on  the  left  side  the  course  of  the  anterior 
branch  of  the  suture,  which  is  continued  by 
the  connective  suture  to  the  midline  at  the 
inner  edge  of  the  doublure.  The  rostral 
plate  was  thus  triangular  in  shape  (cf. 
Loven,  1845:  49,  pi.  1,  fig.  2b).  The  type 
species,  P.  pluteus,  and  a  species  from  the 
Henryhouse  Formation  of  Oklahoma  being 
described  by  Campbell  (in  press),  are  ex- 
ceedingly similar.  These  similarities  include 
the  detailed  fonn  of  the  cephalon  ( shape  of 
the  glabella,  fomi  of  the  muscle  areas,  posi- 
tion of  the  eye  lobe,  shape  of  the  borders, 
and  presence  of  a  triangular  rostral  plate), 
and  the  form  of  the  thoracic  segments  and 
pygidium.  The  latter  lacks  a  clearly  defined 
border  and  has  the  pleural  furrows  of  the 
first  one  or  two  segments  extending  close 
to  the  margin.  The  fine  granulation  of  the 
external  surface  of  the  glabella,  the  weak 
pitting  on  the  free  cheek  inside  the  border, 
and  the  presence  of  terrace  lines  on  the 
border  (which  on  the  pleural  regions  of 
the  pygidium  curve  forward  and  inward) 
are  characters  common  to  all  these  species. 
These  close  similarities  might  well  be  used 
to  limit  the  subgenus  Proctiis  (ProeUis)  to 
a  group  of  Middle  and  Upper  Silurian 
species.  The  latter  would  probably  include 
P.  fletcheri,  described  by  Reed  (1901:11- 
14,  pi.  1,  figs.  5,  6)  from  the  Wenlock 
Limestone    of    Britain,    and    P.    mofinensis 


Silurian  Tbilobites  •  Whiftington  and  Campbell        457 


5—, 


4  — 


3- 


2  — 


PYGIDIA 


E 
E 


0) 


•  •        •         • 


•  •        • 


•    •  • 


Length  of  axis     (mm) 


T 

3 


1 
5 


Figure   5.      Proetus  pluteus   n.    sp.     Dimensions  of   40   pygidia,   length    of    axis     is    sagittal,     Ineigfit     is     maximum.      Measured 
specimens    the    same    as    Figure    4. 


Pfibyl  ( 1960:  204-206,  pi.  2,  figs.  1-5)  from 
the  Kopanina  Shale,  Lower  Ludlow,  of 
Bohemia.  However,  examination  of  speci- 
mens of  Proetus  ctivieri,  from  the  Middle 
Devonian  of  the  Eifel  district  of  West 
Gennany  (Pi.  3,  figs.  1-3;  Richter  and 
Richter,  1956),  shows  that  this  species  is 
like  the  type  P.  concinnus.  The  rostral  plate 
is  not  the  relatively  wide   ( tr. )   and  short 


(sag.  and  exs.)  plate  figured  by  Richter 
and  Richter  (1956,  pi.  5,  figs.  33b,  c,  d), 
but  the  doublure  of  the  cephalon  (the  ex- 
ternal mould  is  preserved  in  the  original 
of  Plate  3,  figure  2)  curls  upward  and  in- 
ward so  that  the  inner  edge  lies  beneath 
the  border  furrow.  The  same  specimen  has 
the  edge  of  the  left  connective  suture  pre- 
served, and  it  runs  inward  and  backward 


458        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


to  meet  the  right  suture  at  the  inner  margin 
of  this  doubkire,  isolating  a  rostral  plate 
that  is  triangular  in  outline  and  sharply 
flexed  in  the  longitudinal  direction.  The 
differences  between  holaspid  cuvicri  and 
concinntts  are  apparently  in  minor  features 
—proportions  of  glabella,  length  of  genal 
spine,  presence  of  strong  tubercles  on  the 
external  surface  of  some  specimens  of 
cuvieri,  and  lack  of  incurved  terrace  lines 
on  the  pygidium  of  cuvieri.  These  do  not 
seem  to  justify  a  subgeneric  distinction  be- 
tween the  Middle  and  Upper  Silurian 
species  and  this  Middle  Devonian  species. 
From  P.  concinnus  the  new  species  may 
be  distinguished  by  its  less  globose  glabella, 
relatively  longer  (sag.  and  exs.)  anterior 
l)()rder,  shallower  axial  furrow  between  the 
palpebral  lobe  and  the  glabella,  coarser 
granulation  on  the  glabella,  and  shape  and 
number  of  spines  on  the  posterior  border 
of  the  hvpostome  (Lindstrom,  1901,  pi.  6, 
fig.  21).' 

Family  OTARIONIDAE  Richter  and  Richter, 
1926 
Discussion.  The  present  material  of 
Otarion  and  Rhinotarion  n.  gen.  suggests 
that  to  the  characters  of  this  family  given 
by  Richter,  Richter,  and  Schmidt  (in  Moore, 
1959:  O  40.3-404)  may  be  added  the 
convergence  backward  of  the  connective 
sutures  to  meet  at  the  inner  edge  of  the 
doublure,  the  absence  of  a  panderian  notch 
in  the  doublure  of  the  cheek,  the  absence 
of  the  pre-annulus  on  the  axis  of  the  thorax, 
the  extension  of  the  median  pleural  furrow 
out  close  to  the  tip  of  the  segment,  and 
the  narrow  panderian  notch  situated  close 
to  the  posterior  edge  of  the  segment.  The 
hvpostome  (Richter,  1914;  cf.  Prantl  and 
Piibyl,  1951:  443)  is  unlike  that  of  proetids 
ill  Iiaving  the  middle  furrow  complete  and 
a  more  prominent,  crescentic,  po.sterior  lobe 
of  the  middle  body  on  which  the  macula 
is  not  distinguishable.  In  addition,  in  the 
present  examples,  there  is  a  tiny  wing  pro- 
cess and  the  anterior  wing  itself  is  smaller 
and  differently  .shaped. 


RHINOTARION  n.  gen. 

Type  species.— R/i/nofar/on  sentosum  n.  sp. 

Diagnosis.  Differs  from  Otarion  in  that 
median  part  of  anterior  border  is  drawn 
forward  into  a  projection;  rostral  suture 
traverses  ventral  side  of  projection  close 
to  outer  margin,  hence  rostral  plate  is 
inverted  "T"  shape  rather  than  triangular 
in  outline. 

Distribution.  No  other  species  that  we 
would  assign  to  this  genus  have  been  re- 
corded. However,  specimen  17729  in  the 
collection  of  the  Australian  National  Uni- 
versity, from  Unit  4  of  the  Dargile  Beds, 
Locality  47,  Parish  Heathcote,  Victoria,  is 
almost  certainly  a  member  of  the  genus. 
The  only  reason  for  doubt  is  the  poor  pres- 
ervation of  the  basal  glabellar  lobes. 

Rhinotarion  sentosum  n.  sp. 

Plate  4;  Plate  5,  figures  1-6,  9-11,  13- 
16,  18. 

Holotype.  USNM  154211,  cephalon  and 
two  segments. 

Description.  Glabella  moderately  convex, 
expanding  forward  to  maximum  width 
across  basal  glabellar  lobes,  slight  but 
abrupt  contraction  at  lateral  furrows  Ip, 
frontal  portion  rounded;  maximum  width 
slightly  less  than,  or  equal  to,  length  (sag.). 
Occipital  ring  widest  in  midline,  steep  pos- 
terior slope  and  more  gentle  anterior  slope 
to  the  deep  occipital  furrow;  prominent, 
slightly  backwardly-curved,  blunt,  median 
spine.  Lateral  furrow  Ip  situated  slightly 
in  front  of  mid-length,  narrow  and  deep, 
directed  inward  and  backward  and  reach- 
ing the  occipital  furrow,  deepest  close  to 
axial  furrow;  basal  lobe  lachrymate  in  out- 
line, moderately  convex.  Lateral  furrow 
2p  situated  opposite  anterior  margin  of  eye 
lobe,  a  smooth  oval  area  situated  a  short 
distance  in  from  the  axial  furrow.  Convex 
cheek  slopes  steeply  outward,  cheeks  joined 
by  preglabellar  field  that  is  only  slightly 
wider  (sag.  exs.)  than  the  anterolateral 
border;  borders  defined  by  deep,  well 
marked  border  furrows.  Posterior  border 
narrow    (exs.)    adjacent   to   occipital   ring, 


Silurian  Trilobites  •  Whittiiifiton  and  Campbell        459 


widening  rapidly  outside  the  fulcrum  and  vincular    notch    or    panderian    notch,    and 

merging  with  the  base  of  the  genal  spine,  narrows  rapidly  inward  to  disappear  at  the 

Lateral  and  anterolateral  border  with  flat-  fulcrum;  along  edge  of  exoskeleton  inside 

tened  upper  surface  which  is  outward  slop-  fulcrum  (PI.  5,  fig.  4)  is  a  recess  to  accept 

ing,    maximum    width    of    border    antero-  the  articulating  flange  of  the  first  thoracic 

laterally;  medially,  borders  drawn  forward  segment.    External  surface  covered,  except 

into  a  blunt,  rounded  projection,  which  is  in  furrows,  by  large  rounded  tubercles  and 

of  width  (tr. )  similar  to  the  width  of  the  smaller  thomlike   spines;   these  irregularly 

anterolateral     border.      Preglabellar     field  scattered   but  sparse   on   the   upper,   inner 

gently  inflated.    Genal  spine  long,  curved  surface    of    the    cephalic    borders,    absent 

and    tapering.     Eye    lobe    relatively    large  distally  on  the  genal  spines,  and  few  on  the 

( exsagittal  length  greater  than   Mi  sagittal  posterior  border.   On  ventral  facing  surface 

length  of  glabella)  and  situated  at  highest  of  border  and  doublure  are  prominent  ter- 

point  of  inflated  inner  area  of  cheek;  cheek  race  lines,  these  not  extending  on   to  the 

slopes  steeply  in  all  directions  downward  smooth  underside  of  the  genal  spine.    Gla- 

from  margin  of  eye  lobe.    Palpebral  lobe  bellar   furrows    not   extended   ventrally   as 

curves  inward  and  downward,  without  rim;  apodemes.  Hypostome  unknown, 

eye   surface   convex  externally,   apparently  Number  of  thoracic  segments  unknown, 

smooth.    Broad,  low,  eye  ridge  runs  inward  Certain  isolated  segments  in  the  collection, 

and  forward  to  axial  furrow  opposite  frontal  including   one   with   a   long   median   spine 

glabellar  lobe   (PI.  5,  figs.   1,  4).    Anterior  (Pi.  5,  figs.  9,  10,  13-16,  18),  have  a  median 

branch  of  suture  runs  forward  and  slightly  tubercle  and  three  additional  pairs  on  the 

outward  to  border  furrow,  curves  over  bor-  axial  ring,  and  one  to  two  tubercles  scat- 

der  and  runs   inward  along  outer  vertical  tered  along  the  posterior  pleural  band,  and 

slope  to  meet  rostral  suture  at  a  very  oblique  a  group  at  the  posterolateral  tip.    In  these 

angle;  rostral  suture  runs  along  forward  and  respects   they   resemble  the  two   segments 

downward  facing  surface  of  border  close  attached  to  the  holotype  cephalon,  and  dif- 

to  the  margin  of  the  projection.    Posterior  fer  from  isolated  segments  of  type  A   (Pi. 

branch  of  suture  runs  directly  outward  and  8,  figs.   13-15,  21 )   or  type  B   ( PI.  9,  figs, 

backward   across   the   cheek   and   posterior  1-3,   8),    and   hence   are    assigned   to   this 

border,    down   the    posterior   slope   of   the  species. 

latter  and  curves  across  the  doublure  Convex  axial  ring  with  distal  part  curved 
beneath  the  base  of  the  genal  spine.  At  slightly  forward;  anterior  edge  slopes  gently 
margins,  surface  of  the  border  curves  evenly  to  articulating  furrow  which  rises  sharply 
around  to  inner  ventral  side,  where  exo-  to  the  articulating  halfring.  Long  median 
skeleton  is  flexed  up  almost  vertically  and  axial  spine  on  one  segment.  Pleurae  flexed 
extended  as  the  doublure  close  up  to  the  down  at  fulcrum,  subdivided  by  U-shaped 
impression  of  the  border  furrows.  A  shallow  pleural  furrow  which  runs  straight  outward 
impression  in  the  doublure  immediately  be-  and  slightly  backward  and  extends  close  to 
hind  the  anterior  projection;  latter  is  oval  the  tip;  convex  posterior  band  the  widest 
in  transverse  section.  Connective  sutures  close  to  the  axial  furrow,  anterior  band 
run  inward  and  backward  to  inner  margin  widest  distally.  Anterior  segment  has  outer 
of  the  doublure  approximately  in  line  ( exs. )  part  of  pleurae  bent  slightly  back,  and 
with  the  lateral  margin  of  the  anterior  pro-  facetted  so  that  anterior  band  is  cut  off; 
jection.  Rostral  plate  (PI.  4,  figs.  10-12)  subsequent  segments  have  the  pleurae  trans- 
is  thus  an  inverted  "T"  shape,  with  the  verse  and  parallel-sided,  the  tip  rounded, 
posterior  edge  indented  and  bent  to  slope  Articulating  ring  process  on  posterior  mar- 
shaqoly  upward  and  inward.  Doublure  con-  gin,  axial  socket  immediately  above  it,  on 
tinues  beneath  genal  angle  without  either  anterior  margin  corresponding  ring  socket 


460        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


and  axial  process;  on  inner  part  of  pleura  types  of  cephala.  The  procedure  adopted  is 

anterior  articulating  flange  which  fits  into  to  describe  the  two  most  abundant  types  of 

recess  on  posterior  margin   of  segment  in  cephala   as   two   distinct   species,    and   the 

front;   marked   fulcral   articulating   process  third  as  a  possible  but  unnamed  additional 

and  socket.    Doubkue  extends  along  pos-  species.    Then  follow  descriptions  of  a  hy- 

terior  margin  of  outer  part  of  pleurae  and  postome  that  probably  belongs  within  this 

beneath   tip,   narrow,   V-shaped   panderian  genus,    segments,    pygidia,    and    transitory 

notch  (PI.  5,  figs.  13,  18)  close  to  posterior  pygidia.     Whether    or    not    two    or    three 

margin,  anterior  edge  raised  to  form  a  stop  species    are    represented    is    problematical, 

during  enrollment.    Doublure  not  extended  and  there  is  also  the  possibility  that  two 

along  anterior  margin  of  outer  part.    Distal  of  them  are  sexual  dimorphs  of  one  species, 

part  of  articulating  furrow  broadened  and  or  even  that  one  of  them  is  a  sexual  dimorph 

deepened,   but   not   extended   ventrally    as  of  Rhinotarion  scntosum  n.  gen.,  n.  sp. 

an   apodeme.    External    surface   bearing   a  Developmental   series   of   the   cranidium 

fine  granulation  in  addition  to  the  elongate  and  pygidium  are  described  below.    There 

tubercles  described  above.  is    a    remarkable    similarity    between    the 

Pygidium  not  known  with  certainty.  small  cranidia  and  pygidia  of  Otarion  and 

The    original    of   Plate    5,    figures    5,    6,  those    of    the    Ordovician    Dimeropijiie,    a 

appears  to  be  the  cranidium  of  a  malformed  greater  similarity  than  that  between  small 

indi\  idual  of  this  species.    The  part  of  the  stages    of   Otarion    and    Proctus   described 

glabella    in    front    of   the    lateral    lobes    is  here.  The  paired  spines,  present  in  Otarion 

relatively  longer  than  in  other  specimens,  but  not  in  Proetiis  (Fig.  6),  are  the  major 

the   preglabellar  field   is   absent  medially,  feature  in  which   Otarion   resembles  Dim- 

and  the  projection  on  the  anterior  border  eropijge.     This    resemblance    in    develop- 

is  smaller  and  upwardly  directed.    The  tip  mental  stages  may  be  taken  to  suggest  a 

of  the  projection  is  either  broken  ( if  so  the  closer  relationship  between  otarionids  and 

break  is   remarkably  even)    or  tenninated  dimeropygids  than  between  otarionids  and 

by   the  rostral  suture  which   runs   up   the  proetids. 

sides  and  over  the  top  of  the  tip.  Recent  descriptions  and  diagnoses  (Prantl 

and    Pfibyl,    1951;    Richter,    Richter,    and 
Genus  OTARION  Zenker,    1833  Schmidt  in  Moore,   1959)   do  not  give  in- 
Type   species.— Ofar/on   diffractum   Zenker,  formation  on  the  rostral  plate  in  this  genus. 
1833.  A  topotype   specimen   of  the  type  species 
Siinoui/tn    (subjective).     Harpidella   M'Coy,    1849.  is  here  illustrated    (PI.    10,  figs.   12,   14-16) 

Discussion.     The    tvpe    species    of    Har-  ^^^  comparison   with   the   Maine   material, 

imh'Ua,    H.    mcoalops    (M'Coy,    1846),    is  ^"'^    '^   ^^""^   ^'^^"    prepared   to   reveal   the 

redescribed  below   and  reasons  are   given  ^^^^^^^  P^'^*^  (P^-  1^'  ^^8-  ^^^-    ^he  rostral 

for  regarding  Harpidella  as  a  synonym  of  su*"^^   runs   parallel   to   the   outer  margin 

Otarion.  oi^^   ^^^c   doublure,    and   is   situated   a   short 

Specimens  from  Maine  referred  to  Of«no/j  distance  in   from   it.    The   anterior  branch 

are  distinguished  from  those  of  Rhinotarion  ^i  the  suture  (Pi.  10,  fig.  16)  runs  inward 

in    that   the   projection    from   the    anterior  ^"f^  forward  over  the  edge  of  the  border, 

border  is  absent.    Cephala   ( excluding  the  '^"c^  is  continued  inward  across  the  doublure 

h\ postome)   are  readily  divisible  into  two  by  the  connective  suture.    The  inner  edge 

kinds,  and  a  possible  third.    Only  one  ex-  of  the  doublure  is  curled  upward  beneath 

ample  of  the  latter  has  a  thoracic  segment  the  border  furrow,  but  clearly  the  connec- 

articulated  with  it,  so  that  the   segments,  tive  sutures  are  close  together,  or  meet  at 

pygidia    and    their    developmental    stages,  the  inner  margin.    The  rostral  plate  is  thus 

cannot  readily  be  matched  with  these  three  triangular    in    outline,    and    flexed    in    the 


Silurian  Trilobites  •  Whitting,ton  and  Campbell        461 


longitudinal  direction.  The  rostral  plate 
in  the  xMaine  species  (PI.  5,  fig.  26;  PI.  6, 
figs.  2,  9;  PI.  7,  figs.  6,  7)  is  similar  in  shape. 

Ofarion   megalops    (M'Coy,    1846) 
Plate   19,  figures   1-14,   16. 

Holotijpe.  National  Museum  of  Ireland, 
internal  mould  of  incomplete  cranidium 
(PI.  19,  figs.  1-4),  from  Boocaun,  near 
Cong,  County  Galvvay,  Eire.  The  strata  at 
this  locality  are  of  Upper  Llandovery  age 
(Whittard,  1938:  101-102;  Haq^er,  1949: 
54). 

Other  material.  Two  topotype  cranidia, 
and  a  cranidium  from  strata  of  the  same  age 
at  Tonlegee,  near  Cong. 

Description.  The  glabella  is  widest  across 
the     elongate-oval,     inflated     basal     lobes, 
rounded  anteriorly;  lateral  furrow  Ip  curv- 
ing inward  and  backward,  shallowing  pos- 
teriorly  before   it   merges   with   the   broad 
occipital    furrow;    furrow    2p    is    a    short, 
shallow  depression  running  directly  inward 
from  the  axial  furrow.  The  cheek  is  highest 
posteriorly,  the  large  palpebral  lobe  curv- 
ing up  from  the  summit   (PI.   19,  figs.   12, 
13);    the    line    S-8    runs    across    the    basal 
glabellar  lobe  at  about  one-third  the  length. 
Inside  the  palpebral  lobe  the  cheek  is  con- 
vex and  descends  steeply  to  the  axial  fur- 
row.   In   front   of  the   palpebral   lobe   the 
cheek  descends  at  first  steeply,  then  there 
is    a    break    to    a    gentler   slope   before    it 
curves    steeply    down    to    become    vertical 
adjacent  to  the  border  furrow.    The  break 
in  slope  mentioned  runs  from  immediatelv 
in  front  of  the  palpebral  lobe,  forward  and 
inward  to  meet  the  axial  furrow  opposite 
the  most  anterior  part  of  the  glabella;  this 
break    in    slope    has    been    interpreted    by 
M'Coy  and  later  authors  as  an  eye  ridge. 
The    preservation    as    internal    moulds    in 
medium-grained    sandstones   makes   it   un- 
certain that  this  change  in  slope  is  truly 
an  eye  ridge — no  distinct  ridge  can  be  seen. 
Preglabellar    field    is    convex,    descending 
vertically,  anterior  border  jutting  forward. 
Anterior  branches   of  sutures   are   straight 
and  diverge  forward,  on   the  crest  of  the 


anterior  border  curving  to  run  inward.  The 
posterior  branch  runs  backward  and  shghtly 
outward,  so  that  the  posterior  part  of  the 
fixed  cheek  (PI.  19,  figs.  1,  13,  16)  is  short 
(exs.)  and  narrow  (tr.).  Median  occipital 
tubercle  on  posterior  margin  of  occipital 
ring. 

Di.scusmon.     The   cranidium    of   MCoy's 
species  differs  from  that  of  the  type  species 
of  Otarion  (PI.  10,  figs.  12,  14,  16)  in  the 
presence   of  a   distinct   lateral   furrow   2p, 
in  the  larger  palpebral  lobe  which  is  situated 
farther    back,    in    the    relative    narrowness 
( tr. )  of  the  posterior  part  of  the  fixed  cheek, 
and  the  relatively  shorter   (sag.  and  exs.) 
preglabellar    field.     These    differences    do 
not   seem   to   be   worthy   of   generic   rank. 
The  two  new  species  and  one  indeterminate 
species    of   Otarion,    described    below,    are 
like    me^alop.s  in    the   presence    of   lateral 
glabellar  furrow  2p,  the  size  and  position  of 
the  eye  lobe,  and  the  narrow  ( tr. )  posterior 
part  of  the  fixed  cheek.  A  Middle  Devonian 
species  having  a  large  eye  lobe  and  prob- 
ably of  this  type  is  O.   ung,uloides  ranun- 
culum  Erben,  1953.   Pfibyl  ( 1960:  218-220, 
pi.  3,  figs.  5,  6)   has  used  HarpideUa  for 
Barrande's  species  novella  and  certain  North 
American     species.      A    cranidium    of    O. 
novella    (MCZ  8552)    from   the   Kopanina 
beds.  Lower  Ludlow,  at  Kopanina,  shows 
that  while   lateral   glabellar   furrow   2p   is 
present,  and  there  is  a  distinct  eye  ridge, 
the  palpebral   lobe  is   small   and   similarly 
situated  to  that  of  O.  diffractum,  and  the 
posterior  part  of  the  cheek  is  as  wide  as  in 
the  latter  species.    O.  novella  is  thus  inter- 
mediate between  diffractum  and  megalop.s\ 
and    we    consider    HarpideUa    should    not 
be  used  as  Pribyl  suggests.  Re-investigation 
of  many  species  is  necessary  as  well  as  more 
complete  material,  before  any  subdivision 
of  Otarion  can  be  placed  on  a  sound  foot- 
ing. 

Otarion  instifa   n.  sp. 

Plate  5,  figure  24;  Plate  6;  Figure  6a. 

Holotijpe.  USNM  154220,  cranidium  with 
left  free  cheek. 


462        Bullet  in  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Description.  Glabella  is  subparallel-sided, 
bluntly  rounded  anteriorly,  and  moderately 
convex;   occipital   furrow   deep   and   trans- 
verse  behind   median    lobe,    curving   back 
behind    basal    lobe.      Latter    isolated    by 
straight  furrow  Ip  which  runs  from  opposite 
midpoint  of  palpebral  lobe  diagonally  in- 
ward  and   backward   to   occipital   furrow- 
basal  lobe  convex,  length  ( exs. )  about  one- 
quarter  of  sagittal  length  of  glabella.   Con- 
spicuous  smooth  muscle   area    (Pi.   6,   fig. 
4)  runs  in  from  axial  furrow  along  anterior 
side  of  furrow  Ip  for  about  half  its  length; 
similar   but   shorter    (tr. )    muscle   area  2p 
runs  directly  inward  from  axial  furrow  in 
line  with  anterior  end  of  eye  lobe.    Cheeks 
inside  border  furrows  united  by  broad  (  sag. 
and   exs.)    preglabellar  field   which   slopes 
gently    forward    and    at   its    margin    drops 
abruptly    vertically   to   the    inner   edge    of 
the  border.   Border  broadest  anteriorly  and 
anterolaterally,   flattened  upper  surface  is 
horizontal   anteriorly  but  outward   sloping 
laterally.    Posterior   border   narrow    (exs.) 
between  axial  furrow  and  fulcrum,  beyond 
here   widening   rapidly   and  merging   with 
broad  lateral  border  at  genal  angle;  genal 
spine    long,    gently    tapering    and    curved. 
Large  eye  lobe  of  length    (exs.)    approxi- 
mately one-third  sagittal  length  of  cephalon; 
convex  eye  surface  apparently  smooth  ex- 
ternally, l)ut  internal  surface  showing  many 
minute  facets  (PI.  5,  fig.  24).    Median  pit 
in    palpebral    lobe.     Anterior   branches    of 
suture  moderately  divergent,  crossing  bor- 
der in  line  (exs.)  with  midpoint  of  palpebral 
lobe,     posterior    branch     running     straight 
outward    and    backward    across    border    a 
short  distance  inside  base  of  genal  spine. 
Doublure   flattened  on   under  surface,  ex- 
tending inward  to  lateral  and  anterior  bor- 
der furrows,  curved  up  at  the  inner  edge 
so   that   this   edge   lies   close   beneath   the 
furrow.    Doublure  extends  inward  beneath 
posterior  border  as  far  as  fulcrum.    Rostral 
sutiue  runs  along  vertical  face  of  anterior 
])order;  connective  sutures  converge  back- 
ward  to  isolate  a  triangular  rostral   plate 
(PI.  6,  figs.  2,  9),  the  innermost  portion  of 


which  is  flexed  upward  and  indented.  Ex- 
ternal surface  bearing  closely-spaced  tuber- 
cles on  glabella  except  in  furrows;  large 
median  occipital  tubercle;  similar  tubercles 
along  anterior  portion,  and  steeply  sloping 
edge  of  preglabellar  field  and  anterior  part 
of  cheek;  elsewhere,  external  surface  ap- 
parently smooth,  except  for  terrace  lines 
on  edge  of  border,  genal  spine  and 
doublure. 

Development.  A  size  series  of  cranidia 
(PI.  6,  figs.  8,  9,  13-16,  18-24;  Fig.  6a)  has 
been  picked  out,  the  smallest  example  ap- 
proximately 1  mm  in  length  ( sag. ) .  In  this 
specimen  the  glabella  as  well  as  the  cra- 
nidium  as  a  whole  is  more  convex  than  in 
larger  examples;  the  anterior  border  is 
relatively  narrower  (sag.  and  exs.)  and 
less  flattened  on  the  upper  surface;  the 
basal  glabellar  lobe  is  present  and  of  ap- 
proximately the  same  relative  size  as  in 
larger  examples.  Most  striking  are  the 
spines,  a  median  occipital,  three  pairs  on 
the  glabella  (of  which  the  posterior  is  the 
longest  and  thickest),  a  median  pair  on 
the  preglabellar  field  and  the  anterior  bor- 
der, and  accessory  pairs  on  the  fixed  cheeks, 
palpebral  lobe  and  borders.  With  increase 
in  size  there  is  a  rapid  reduction  of  these 
spines,  and  a  loss  of  the  symmetrical  ar- 
rangement, followed  by  a  gradual  assump- 
tion of  the  pattern  of  the  large  examples. 
A  size  series  of  the  free  cheek  is  difficult 
to  pick  out,  but  small  examples  which  prob- 
ably belong  (PI.  6,  fig.  17)  bear  many 
short  spines  which  are  rapidly  reduced 
and  disappear  (PI.  6,  figs.  10-12). 

The  form  of  the  small  cranidia,  and 
particularly  the  arrangement  of  paired 
spines,  resembles  that  of  small  cranidia 
of  the  Ordovician  Dime  ropy  ge  (Whitting- 
ton  and  Evitt,  1954:  44-46;  and  compare 
PI.  6,  figs.  18,  19,  24  with  Whittington  and 
Evitt,  1954,  pi.  3,  figs.  16,  17,  21-26;  pi.  22, 
figs.  1-10).  In  the  Ordovician  genus  there 
is  an  increase  in  convexity  of  the  cranidium 
during  development,  there  is  no  basal 
glabellar  lobe,  the  palpebral  lobe  is  smaller, 


Silurian  Trilobites  •  Whittington  and  Campbell        463 


but  the  spines  are  reduced  in  size  and  lose 
the  symmetrical  arrangement,  as  in  Otarion. 

Otarion   plautum   n.   sp. 

Plate  7,  figures  1-9,  11-15,  17-19,  23- 
25:    Figure  6b. 

Holotijpe.  USNM  154231,  cephalon  lack- 
ing hypostome. 

Description.  Much  less  abundant  than 
O.  insfita  are  cephala  and  cranidia  of  this 
type,  distinguished  by  the  glabella  having 
a  width  across  the  base  approximately 
equal  to  the  length  (sag.),  the  flatter 
transverse  profile  of  both  glabella  and 
cephalon,  the  more  divergent  anterior 
branches  of  the  suture,  the  narrower  border 
anteriorly  and  anterolaterally,  the  stronger 
backward  flexure  of  the  posterior  border 
outside  the  fulcrum,  and  the  shorter,  more 
rapidly-tapering  genal  spine.  A  faint  eye 
ridge  (Pi.  7,  figs.  6,  8)  runs  inward  and 
forward  to  the  axial  furrow.  The  doublure 
is  narrow,  fomiing  with  the  border  a  tube- 
like structure,  the  rostral  plate  triangular 
(PI.  7,  figs.  6,  7)  and  relatively  short  (sag.). 
The  external  surface  is  tuberculate,  and 
there  is  a  slightly  larger  median  occipital 
tubercle.  Tubercles  are  irregularly  but 
closely  spaced  on  the  glabella,  fixed  cheek 
inside  the  eye  lobe,  and  preglabellar  field. 

A  developmental  series  of  cranidia  has 
been  picked  out  (PI.  7,  figs.  8,  11-14,  17- 
19,  23-25;  Fig.  6b),  which  is  distinguished 
from  that  of  O.  instita  (Pi.  6,  figs.  8,  9,  13- 
24 )  by  the  consistently  steeper  slope  of  the 
preglabellar  area,  as  well  as  the  more 
divergent  anterior  branches  of  the  suture 
and  the  stronger  backward  flexure  of  the 
posterior  border.  The  smallest  example 
(PI.  7,  figs.  19,  24,  25)  is  extremely  like 
that  of  O.  instita  (PI.  6,  figs.  18,  19,  24), 
and  bearing  spines  of  similar  relative  size 
and  paired  arrangement.  The  glabella  is 
subparallel-sided  and  bluntly  rounded  an- 
teriorly with  a  small,  convex  basal  lobe. 
With  increasing  size  there  is  a  general  re- 
duction in  convexity,  and  the  glabella 
gradually  assumes  the  broader,  relatively 
shorter  appearance.    Spines  are  rapidly  re- 


duced and  the  irregular  arrangement  of 
tubercles  assumed.  As  in  the  case  of  O. 
instita,  attention  is  drawn  to  the  similarity 
between  this  developmental  series  and  that 
of  Dimeropyge  (Whittington  and  Evitt, 
1954). 

Otarion  sp.  ind. 

Plate  5,  figures  7,  8,  12,  17,  19-23,  25, 
26. 

Description.  This  type  of  cephalon  is 
almost  as  abundant  as  that  of  O.  plautum. 
It  is  intermediate  between  O.  instita  and 
O.  plautum  in  the  convexity  of  the  cephalon, 
the  outline  of  the  glabella,  and  the  width 
of  the  anterior  border.  The  outline  of  the 
anterior  margin  is  bluntly  and  obliquely 
angulate,  as  is  the  course  of  the  anterior 
border  furrow;  these  outlines  are  more 
angulate  than  those  of  O.  instita,  while 
those  of  O.  plautum  are  rounded.  On  the 
other  hand,  the  cephalon  resembles  that 
of  O.  plautum  in  having  the  short  genal 
spines  and  stronger  backward  flexure  of 
the  outer  part  of  the  posterior  border. 
Doublure  is  similar  to  that  of  the  other 
species,  and  there  is  a  triangular  rostral 
plate  (PI.  5,  fig.  26).  No  very  complete 
developmental  series  of  cranidia  has  been 
recognized,  because  of  the  obvious  dif- 
ficulty of  picking  out  this  intermediate  type. 
The  smallest  cranidium  that  appears  to 
belong  to  it  (PL  5,  figs.  22,  23)  is  of  length 
(sag.)  1.5  mm.  The  external  surface  bears 
elongate,  irregularly  scattered  tubercles 
and  paired  spines  on  the  glabella,  the  pos- 
terior pair  in  line  with  the  posterior  part 
of  the  palpebral  lobe  being  notably  longer 
and  thicker. 

One  specimen  (PI.  5,  figs.  19,  25,  26) 
has  the  anterior  thoracic  segment  linked 
to  the  cephalon.  The  axial  ring  bears  a 
number  of  tubercles;  the  outer  part  of  the 
pleural  region  is  facetted  so  that  only  the 
posterior  band  extends  to  the  tip,  and  bears 
on  the  external  surface  tubercles,  including 
a  small  group  at  the  tip. 


464        Bulletin  Museum  of  Comparative  Zoology,  Vol  135,  No.  9 


E 
O 


E 
E 

O 


Figure  6.  Smallest  cranidia  in  size  series  showing  some  paired  spines.  A)  Ofor/on  inslita  n.  sp.  (original  of  PI.  6, 
figs.  18,  19,  24).  B)  Olarion  plautum  n.  sp.  (original  of  PI.  7,  figs.  19,  24,  25).  C)  Proefus  plufeus  n.  sp.,  USNM  154491, 
sligfitly  smaller   tfian  original   of  Plate    1,    figs.    23,    24,   30,    31. 


Otarionid    Hypostome 

Plate  7,  figures  10,  16,  20-22,  26,  27. 

Description.  The  small  number  of  speci- 
mens shows  a  variety  of  fomi — some  rela- 
tively long  (PI.  7,  figs.  10,  16),  others  re- 
latively broader  (PL  7,  figs.  21,  22).  In 
both  types  the  anterior  lobe  of  the  middle 
body  is  moderately  convex,  separated  by  a 
complete  middle  furrow  from  the  inflated, 
crescentic  posterior  lobe  of  the  middle  body. 
Macula  not  discernible.  Shallow  lateral 
and  posterior  border  furrows,  borders  nar- 
row, distinct  shoulder  at  a  point  beyond 
the  mid-length,  short  spine  at  posterolateral 
angle.  Anterior  wing  triangular,  directed 
upward  and  outward,  tip  rounded,  on  an- 
terioi  margin  near  tip  a  small  process 
directed  forward.  Doublure  commences 
behind  anterior  wing  and  posterolaterally 
is  of  similar  width  to  the  border;  posterior 
wing  a  pointed  process  directed  upward 
and  inward  from  the  margin  of  the  doublure 
beneatli   the   shoulder. 

Smaller  examples  are  similar,  the  smallest 
(PI.  7,  fig.  20)  distinguished  by  the  rela- 
tively larger  posterior  lobe  of  the  middle 
body,  the  relatively  wider  posterolateral 
border  bearing  a  number  of  short  spines, 
and  the  downwardly  flexed  anterior  edge. 

Otarionid  hypostomes  have  rarely  been 
found;  the  present  examples  are  similar  to 
that  of  the  Middle  Devonian  Otarion 
ceratophtlmhnus  jiortrayed  by  Richter  (1914, 
text-fig.  1). 


Otarionid  Thoracic  Segments  and  Pygidium, 
Type  A 

Plate  8,  figures  1-6,  9,  10,  13-15,  19,  21- 
26. 

Description.  These  are  the  most  abun- 
dant types  as  isolated  specimens,  including 
segments  with  a  median  spine,  and  in  three 
examples  with  several  posterior  segments 
articulated  with  a  pygidium.  The  segments 
are  typically  otarionid,  the  narrow,  shallow, 
pleural  furrow  running  out  almost  to  the 
tip  before  dying  out.  Inner  part  of  pleura 
relatively  narrow  (tr. ),  with  anterior  flange 
and  posterior  recess;  doublure  commences 
outside  this  recess  and  runs  along  the  pos- 
terior edge  and  beneath  the  tip  of  the  outer 
part.  Narrow  panderian  notch  (Pi.  8,  fig. 
15)  situated  beneath  posterior  margin  of 
pleural  furrow,  anterior  edge  raised  to  form 
a  stop  during  enrollment.  Pygidium  trans- 
verse, broad  axis  extends  back  to  inner 
margin  of  border  and  is  bluntly  rounded. 
Only  first  ring  indicated  by  faint  ring 
furrow.  Pleiual  regions  subdivided  by  first 
interpleural  furrow,  and  one  to  three  pleural 
furrows,  the  second  and  third  extremely 
faint.  Narrow  border  developed  as  a  faint 
convexity  without  border  furrow.  Doublure 
is  widest  laterally,  behind  tip  of  axis  be- 
coming narrow  and  strongly  convex  (Pi. 
8,  fig.  10).  External  surface  of  both  seg- 
ments and  pygidium  finely  granulate,  the 
axial  ring  in  larger  examples   (PI.  8,  figs. 


Silurian  Trilobites  •  Whittington  and  Campbell        465 


13,    14,   23,   25)    bearing   numerous  tuber-  the  presence  of  the  tubercles  or  short  spines 

cles.  on  axial  rings  and  pleural  regions.    Three 

The   tubercles   on   the   axis   and   lack   of  axial    rings    and    a    faint    fourth    ring    are 

tuberculation   on   the   pleural   regions   sug-  marked  out  by  the  first  two  ring  furrows 

gest  that  these  segments  and  pygidia  may  and    bands    of    tubercles;    on    the    pleural 

belong    to    either    Otarion    instita    or    O.  regions   the    first   interpleural   furrow   and 

phiittiim,   their   abundance  indicating  that  three  pleural  furrows  may  be  distinguished, 

they  may  belong  to  the  fonner.    The  more  posteriorly  as  smooth  bands  between  rows 

posterior     segments,     particularly     in     the  of  tubercles.    The   doublure   is   similar  to 

smallest  example  (Pi.  8,  figs.  3,  6,  9)  have  that  of  type  A,  narrow  and  convex  behind 

the  posterior  tip  bluntly  pointed   and   di-  the  axis, 
rected  backward.    This  may  be  a  feature 

associated  with  small  segments,  for  in  the  Otarionid  Pygidium,  Type  C 
largest  example  the  tips  (Pi.  8,  figs.  24,  25)  Plate  8,  figures  7,  8,  11,   12. 

are  rounded  on  all  segments;  on  the  other  Description.  One  example  of  this  distinc- 

hand,  this  may  be  a  specific  difference.  ^ive   pygidium    is    known,    the    outer   part 

of  the  pleural  regions  steeply  sloping  and 

Otarionid  Thoracic  Segments  and  Pygidium,  ^yith   a  marked   angle   between   outer  and 

Type  B  inner  parts  along  the  anterior  margin.   The 

PI        Q    f.  1_14  doublure  is  of  approximately  constant  width, 

and  there  is  no  border.   The  first  axial  ring 

Description.     These    segments    are    dif-  j^  distinct,   as  is  the  first  pleura,   with   a 

ferentiated  chiefly  by  the  external  suriace  ^i^.^i^^^^  pleural  furrow.    The  general  form 

which  is  tuberculate  on  the  axial  ring  and  ^^^  ^j-,^  external   granulation  suggest  that 

the  posterior  bands,  with  a  group  of  small,  ^j^j^  pygidium  may  belong  with  Rhinotorion 

shori:  spines  projecting  backward  and  out-  sentosiim,  particularly  if  the  angle  between 

ward    from    the    posterolateral    margin    of  ^j^^  ^^^^^j.  ^j^j   q^^^^j.  ^.^^^^   ^^  ^j^^   pleural 

the  tip.    In  front  of  this   group   of  spines  regions  is  compared  with  that  of  the  most 

the  edge  of  the  tip  is  slightly  excavated,  posterior  segment  referred  to  this   species 

giving    a    characteristic    scalloped    outline  ( compare  PI.  8,  figs.  7,  12,  with  PI.  5.  figs. 

(PI.  9,  figs.  1,  8).    The  occipital  ring  and  9  14). 
posterior  border  of  the  cephalon  of  Otarion 

sp.  ind.   (PI.  5,  figs.  19,  25)  bear  scattered  Otarionid  Pygidium,  Type  D 
tubercles  and  short  spines,  as  does  the  axial  Plate  8,  figures  16,  17,  18,  20. 

ring  and  posterior  pleural  band  of  the  at-  Description.    This  pygidium  is  quite  like 

tached    segment.     The    latter    is    strongly  ^^p^   ^    ^^^    -^   distinguished   by    a   more 

facetted,  but  the  posterolateral  tip  appears  ^.^^-^^^  tapering  axis,  the  less  distinct  ring 

to  bear  a  few  outwardly  directed  tiny  spines.  ^^^  pj^^j..^!  f^j-rows,  and  the  lack  of  dis- 

It  appears  that  these  segments  may  belong  jj^^t  narrowing  of  the  doublure  posteriorly, 
with  this  cephalon,  and  the  axis  is  notably 

more  convex  than  in  type  A.  Otarionid   Transitory   Pygidia 

No  pygidium  is  known  articulated  with  Plate  9,  figures   15-27. 

these  segments;   less  abundant  than  those  Description.     Size    series    are    known    of 

of  type   A   are   those  placed  here    (Pi.   9,  ^,0,-,-,^    ^f   the    tvpes    of   pygidia    described 

figs.  4-7,  9-14).    Distinctive  of  the  largest  above,  and  in  addition  there  are  transitory 

specimen    (compare  Pi.  9,   figs.   4-7,  with  pygidia  such   as   those   shown   in   Plate  9, 

PL  8,  figs.   1,  4,   10)   is  the   slightly  more  figures    1.5-26,    which    appear    to    fonn    a 

prominent   axis,   the   more   triangular   out-  series.     The    relatively    narrow    axis    bears 

line,   the  lack  of  any  distinct  border   and  median  spines  on  the  rings,  one  of  which 


466        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


is  much  stouter  and  longer  than  the  others. 
This  stout  spine  mav  be  on  the  third  axial 
ring  (PI.  9,  figs.  15,'  16,  19,  20,  23-25),  or 
the  first  (PI.  9,  figs.  17,  21,  26),  or  be  ab- 
.sent  (PI.  9,  figs.  18,  22).  It  thus  appears  to 
progress  forward,  and  its  absence  suggests 
that  the  segment  bearing  it  has  been  re- 
leased into  the  thorax.  The  pleural  regions 
slope  gently  outward  from  the  axis,  more 
steeply  distally.  In  larger  specimens  pleural 
and  inteipleural  furrows  are  well-marked 
(PI.  9,  figs.  17,  21).  At  the  distal  change  in 
slope  there  is  a  spine  on  the  posterior  band 
of  each  segment,  and  a  second  spine  fur- 
ther inward  (Pi.  9,  fig.  17).  These  same 
rows  of  spines  are  present  on  the  example 
lacking  median  axial  spines  ( Pi.  9,  figs.  18, 
22).  The  external  surface  is  granulate  be- 
tween the  spines.  The  doublure  is  flat  and 
of  even  width. 

A  different  type  of  transitory  pygidium, 
without   prominent    axial   spines    and   with 
less    conspicuous    spines    on    the    posterior 
bands  of  the  pleurae,  having  well  marked 
pleural  and  interpleural  furrows,  and  with 
the   ends   of   the   segments    extended    into 
short  backward   pointing  spines,  is  shown 
in  Plate  9,  figure  27.   Possibly  it  is  an  earlier 
stage  of  type  A,  in  which  the  posterolateral 
tip  of  the  segment  is  also  bluntly  pointed 
in  small  examples   (Pi.  8,  figs.  3,  5,  6,  9). 
Discussion.    These  transitory  pygidia  are 
considered    to   belong   to   Otarion    because 
]Daired  spines  are  also  conspicuous  in   the 
early    development    of    the    cranidium    of 
Otarion,  and  because  of  the  presence  of  the 
median  axial  spine.    It  has  not  been  pos- 
sible to  make  any  sjx'cific  separations  be- 
tween tin  in.    The  outline,  the  median  axial 
spine    and    its    progress    forward,    and    the 
spines  on  the  posterior  pleural  bands  (most 
conspicuous  distally)    are  points  of  strong 
resemblance  between   these   transitory  py- 
gidia and  those  of  the  Ordovician  Dimero- 
py^c   ( \\  hittington  and  Evitt,  19.54,  pi.  3, 
figs.  13-15,  18-20;  pi.  23,  figs.  18-25). 

Comparisons  with  other  Silurian  species 
Few  species  have  been   described  from 


Silurian  rocks,  and  either  the  descriptions 
are  old  and  inadequate  or  the  more  re- 
cently described  specimens  are  fragmen- 
tary. Material  in  the  Museum  of  Compara- 
tive Zoology,  limited  in  quantity  and 
vaguely  localized,  has  enabled  us  to  make 
the  following  comments: 

Otarion  sp.,  presumably  O.  eJegantula 
Loven,  1845,  MCZ  8595,  enrolled  exoskele- 
tons  from  Gotland,  Sweden.  Preglabellar 
field  is  short  (sag.  and  exs.),  steep,  and 
the  anterior  border  flat,  but  not  as  broad 
(sag.  and  exs.)  as  in  O.  instita.  Cephalon 
appears  to  range  in  outline  between  that  of 
O.  pJautum  and  O.  sp.  ind.,  but  the  flat 
border  is  distinctive.  Thorax  of  12  seg- 
ments, and  there  is  no  median  axial 
thoracic  spine. 

Otarion  sp.,  from  Wenlock  Limestone, 
Dudlev,  England  (Salter,  1853;  Whittard, 
1938:  '102-103),  MCZ  8597.  The  narrow 
preglabellar  field  and  border  are  of  length 
( sag. )  about  one-third  that  of  the  glabella, 
and  the  eye  lobe  is  relatively  small  and 
high.  The  outline  of  the  glabella  resembles 
that  of  O.  sp.  ind.,  but  the  form  of  the 
cephalon  is  not  the  same.  There  are  11 
thoracic  segments,  and  a  prominent  median 
axial  spine  on  the  6th  segment. 

Otarion  christyi  (Hall,  1879),  MCZ  8596, 
cephala  and  complete  exoskeletons  from 
the  Waldron  Shale,  Waldron,  Indiana. 
Cephala  range  in  outline  and  convexity  be- 
tween that  of  O.  phiiitum  and  O.  sp.  ind., 
and  like  these  forms,  the  eye  lobe  is  rela- 
tively large.  There  are  12  thoracic  seg- 
ments, and  no  median  axial  spine.  Of  the 
three  species  discussed  here,  O.  christyi 
is  most  like  O.  plautum  and  O.  sp.  ind.  from 
Maine.  Pending  the  redescription  of  ade- 
quate Waldron  material,  the  Maine  species 
are  regarded  as  distinct. 

Family  SCUTELLUIDAE  Richter  and  Richter, 

1955 
Scutelluid  gen.  ind. 

Plate  10,  figures  4,  6-8,  13. 

Description.  Two  incomplete  fragmen- 
tarv   cranidia   onlv  have  been   found,  and 


Silurian  Trilobites  •  Whittiuf^foii  and  C(im})]jcJJ        467 


Table  2.     Arrangement  of  glabellar  tubercles  in  11  specimens  of  Fragiscutuin  rhytiuin,  listed 
IN  order  of  decreasing  size,  the  arrangement  expressed  in  the  formula  of  Tripp  (1957,  1962). 


USNM   154272   (Pi.  11,  fig.  2): 

USNM   154273   (PI.  12,  fig.  8): 

USNM  154275   (Pi.  12,  fig.   1:  Fig.  7h) 

USNM   154290; 

USNM   1.54291: 

USNM   1.54277   (PI.  12,  fig.  4): 

USNM   154278   (PI.  12,  fig.  6): 

USNM  154279   (PI.  12,  fig.  9): 
USNM  154292: 

USNM   L54280   (PI.  12,  fig.   15;  Fig.  7a; 

USNM   154281    (Pi.  12,  fig.   18): 


ii-1;  II-l,  2;  III-l,  3;  iv-1;  IV-0*,  1,  2,  3;  v-O,  1,  2;  V-0*, 
1,  2,  3;   \T-(). 

Il-O*,  1,  2;  Ill-O*,  1,  2,  3;  iv-0,  1;  IV-OI,  1,  2,  3;  V-OI, 

1,  2,  3;  VI-1,  2. 

ii-1;    Il-l,   2;    III-l,   3;    iv-1,   2;    IV-O,    1,    2,   3;    v-O,    1,   2; 
V-O,   1,  2,  3;   VI-1,  2. 

ii-0;  II-l,  2;  III-l,  3;   iv-1;  IV-O,   1,  2,  3;  v-O,   1;   V-O,   1, 
2    3-   VI-1    2 

ii-0;  II-l,  2;  III-l,  3;  iv-Oj:,   i,  2;  IV-O,  1,  2,  3;  V-O,  1, 

2,  3;   VI-1. 

ii-0*;  II-l,  2;  iii-0*;  III-l,  3;  IV-O,  1,  2,  3;  v-1,  2;  V-O,  1, 
2,  3;  VI-1,  2. 

II-l,  2;   III-l,  3;   iv-1;   I\-0|,   1,  2,  3;  v-1,  2;   V-1,   2,  3. 

ii-1;  II-l,  2;  III-l,  3;  iv-Os,  1;  IV-1,  2,  3;  V-1,  2,  3. 

ii-0;  II-l,  2;  III-l,  3;  iv-0;  IV-1,  2;  v-O;  V-O,  1,  2,  3;  VI-0. 

II-l,  2;   III-l,  3;   iv-1;   IV-O*,   1,   2,  3;  v-1,  2;   V-1,  2,  3; 
VI-1. 

II-l,  2;  III-l,  3;  iv-0;  IV-1,  2,  3;  V-OI,  1,  2,  3. 


both  have  a  large,  curved,  backw  ardly- 
directed  median  occipital  spine,  and  a 
small  spine  projecting  back  from  the  pos- 
terior margin  of  the  palpebral  lobe.  The 
glabella  expands  forsvard  and  is  moderately 
convex,  reaching  to  the  anterior  margin 
where  it  merges  with  the  anterior  border. 
The  palpebral  lobe  is  placed  far  back 
and  in  line  (exs.)  with  the  outermost  part 
of  the  glabella,  the  anterior  branches  of 
the  suture  divergent.  Glabellar  furrows 
cannot  be  distinguished.  On  the  external 
surface  there  are  terrace  lines  running  con- 
centrically on  the  anterior  slope  of  the 
glabella. 

Discussion.  Species  of  several  genera 
described  by  Snajdr  ( 1960 )  exhibit  both 
the  occipital  and  palpebral  spines,  and 
without  information  on  the  glabellar  fur- 
rows it  is  not  possible  to  be  sure  to  which 
genus  these  specimens  may  belong.  Spe- 
cies of  Kosovopeltis,  DecowscutcIIum,  and 
SpinisciitcUum    are    present    in    the    upper 


Silurian  of  Bohemia.  Species  of  Kosovopel- 
tis exhibit  occipital  and  palpebral  spines  in 
the  smaller  specimens,  but  they  are  absent 
in  the  larger  ones;  species  of  Decorosciitel- 
lum  commonlv'  have  two  spines  on  the 
palpebral  lobe,  and  species  of  Spiniscutcl- 
lum  have  a  wider  anterior  border  than  in 
the  present   specimens. 

Family  ENCRINURIDAE  Angelin,  1854 
FRAGISCUTUM  n.  gen. 

Type  species.    Fragiscufum  rhytium  n.  sp. 

Diafinosis.  Basic  tubercle  pattern  (  Table 
2)  includes:  II-(l).  2;  III-l,  3;  iv-1;  IV- 
(0),  1,  (2),  3;  V-(0),  1,  2,  3;  VI-1;  lateral 
glabellar  lobe  Ip  reduced  to  a  small  lateral 
remnant,  so  that  apodeme  Ip  lies  almost 
in  the  same  transverse  line  as  the  median 
part  of  the  occipital  furrow.  True  and 
"false"  preglabellar  furrows  weak,  "false" 
anterior  border  short  (sag.  and  exs.)  and 
with  weak  tuberculation.    Extremely  small 


468        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


fixigenal  spine.  Rostral  plate  trapezoidal, 
widest  at  hypostomal  margin;  hypostome 
with  median  lobe  not  extending  beyond 
anterior  margin,  macula  inconspicuous. 
\'incular  furrow  extends  along  anterolateral 
part  of  ventral  surface  of  border.  Granula- 
tion on  borders,  crests  of  large  tubercles, 
and  between  the  bases  of  these  tubercles. 
Ten  thoracic  segments,  none  with  a  median 
axial  spine.  Axis  of  pygidium  with  less 
than  20  axial  rings,  weak  median  smooth 
band  commencing  behind  second  ring. 
Pleural  region  rounded  posteriorly,  where 
pleural  bands  curve  inward. 

Discussion.     Various    authors    have    dis- 
tinguished   groups    of    species    within    the 
genus  Encrinurus  (Reed,  1928;  Rosenstein, 
1941;  Tripp,  1957,   1962),  but  while  there 
is   agreement  between   these   authors   that 
the  groups  center  around  certain   species, 
the   composition   of  particular   groups   has 
been  disputed.    Tripp  used  a  combination 
of  cephalic  and  pygidial,  but  not  thoracic, 
characters    in    defining    particular    groups. 
We    have     followed     his     approach,     and 
elected  to  give  generic  status  to  the  new 
group.     The   "species-group"   most    closely 
allied  to  Fragiscutum  is  that  of  Encrinurus 
variolaris    (Reed,     1928;    Temple,     1956). 
Temple    referred    to    this    group    as    "ad- 
vanced" and  noted  the  reduction  of  the  pre- 
glabellar  furrow.    Other  characters  which 
this  group  has  in  common  with  Fragiscutum 
are  the  absence  of  a  median  axial  spine  on 
the  thorax,  the  rounded  pygidial  tennina- 
tion  and  the  relative  fewness  of  the  axial 
rings  of  the  pygidium.    The  E.   variolaris 
group   is    distinguished   from    Fragiscutum 
by  the  pattern  of  tubercles  on  the  glabella 
(Tripp,  1962,  pi.  65,  figs.  17-20),  particu- 
larly in  the  presence  of  a  distinct  row   I, 
the  11  thoracic  segments  and  the  absence 
of  a  median  band  on  the  pygidium. 

Fragiscutum  rhytium  n.  sp. 

Plate  10,  figure  11;  Plates  11-13;  Figures 
7,   8. 

IIoloii/iH'.     USNM     154272,    incomplete 
exoskelcton. 


Description.   Occipital  ring  widest  (sag.) 
medially.    Lateral   glabellar  lobe   Ip   short 
( exs. )    and   limited   to  the  lateral  part   of 
the    glabella   immediately   in   front   of   the 
outermost  part  of  the  occipital  ring,  so  that 
furrow  Ip  and  occipital  furrow  are  united 
medially.   Furrow  2p  commences  as  a  deep 
pit  in  the  axial  furrow  and  extends  inward 
as  a  shallow  depression,  the  furrows  unit- 
ing medially  to  isolate  a  2p  glabellar  ring. 
Furrow  3p,  situated  opposite  the  anterior 
margin    of   the    eye   lobe,    is    shallow    and 
short.     Curved    apodemes,    expanded    dis- 
tally,    present    at   outer   ends    of    occipital, 
Ip  and  2p  furrows   (PI.   11,  fig.  3;  Pi.   12, 
figs.    2,    3).     Anterior   margin    of    glabella 
(Pi.  12,  fig.  5)  faintly  defined  laterally  by 
shallow  preglabellar  furrow  which  dies  out 
medially.     Axial    furrow    deep,    wide    and 
steep-sided,  with  a  U-shaped  cross-section 
on  cranidium,  continued  on  free  cheek  (Pi. 
11,  figs.  16,  18)   as  a  much  shallower  de- 
pression   which   dies    out   toward   the    an- 
terior margin.   Anterior  pit  (Pi.  11,  fig.  13) 
a  deep  depression  immediately  inside  the 
margin  of  the  cranidium.    Highest  point  of 
cheek  bearing  the  large  eye  lobe,  the  mid- 
point   of   which    is    situated    in    line    with 
lateral  glabellar  lobe  3p.    Posterior  border 
furrow  proximally  as  deep  as  axial  furrow, 
shallowing  distally  where  it  curves  fonxard 
inside  the  genal  angle  and  continues  as  the 
shallow  lateral  border  furrow  (Pi.  11,  fig. 
1 ) .    Posterior  border  narrow  and  strongly 
convex  in  the  inner  part,  outer  part  broader 
and    less    convex;    tiny    fixigenal    spine    in 
largest  specimens   (Pi.  10,  fig.   11).    Ante- 
rior border  furrow  shallow  and  ill-defined 
(PI.   11,   figs.   16,   18),  running  parallel  to 
anterior  margin  and  closer  to  this  margin 
than    to    suture    line.     Anterior   branch    of 
suture    runs    fon\'ard    and    inward    across 
cheek  and  axial  furrow,  curves  around  sub- 
parallel  to  preglabellar  furrow,  and  makes 
an  oblique  angle  with  the  short  (tr.)  rostral 
suture   (PI.   11,  fig.   13).    Posterior  branch 
of    suture    runs    outward    and    backward, 
curves  over  the  border  at  the  genal  angle 
to  reach  the  posterior  margin  immediately 


Silurian  Trilobites  •  Whittiufiton   and  Campbell        469 


outside  the  tiny  fixigenal  spine,  and  crosses  forward  and  produced  into  a  slender 
the  doubkn-e  (Pi.  12,  figs.  2,  3).  Connec-  process  with  a  concave  inner  side  which 
tive  suture  runs  downward  and  slightly  probably  lies  against  the  anterior  pit  of  the 
outward  (Pi.  12,  fig.  5),  so  that  rostral  cephalon;  wing  process  does  not  extend  as 
plate  was  evidently  trapezoidal  in  outline  high  as  above  mentioned  process  and  lies 
and  widest  at  the  anterior  margin.  Don-  outside  it.  Transverse  slit-like  pit  in  outer 
blure  of  cephalon  extends  from  immedi-  face  of  wing  corresponds  with  this  process; 
ately  inside  genal  angle  forward  beneath  elongate  boss  on  outer  face  of  wing  dorsad 
the  cheek  border,  and  is  widest  postero-  of  pit.  Posterior  wing  an  inwardly  and 
laterally  \\'here  it  extends  inward  as  far  as  upwardly  directed  subrectangular  projec- 
the  border  furrow.  Anterolaterally  the  tion,  situated  about  halfway  between  the 
rolled  margin  is  indented  by  a  shallow  anterior  wing  and  the  rear  of  the  central 
vincular  furrow  (Pi.  12,  fig.  3),  which  dies  body;  doublure  wide  between  wings,  be- 
out  before  reaching  a  point  in  line  with  hind  posterior  wing  narrow,  but  widening 
the  axial  furrow.  Anteriorly,  doublure  is  beneath  posterior  border  though  it  does  not 
narrower  where  it  is  bounded  by  the  hypo-  extend  inward  as  far  as  the  lateral  or  pos- 
stomal  suture.  On  the  free  cheek  (Pi.  11,  terior  border,  except  medially  where  there 
fig.  19)  this  suture  may  be  seen  forming  is  a  small  cusp  which  extends  forward  al- 
the  inner  edge  of  the  doublure,  mnning  most  to  beneath  the  border  furrow, 
from  the  antero\entral  margin  inward  to  External  surface,  except  in  furrows,  bear- 
meet  the  inner  margin  of  the  doublure  im-  ing  large  scattered  tubercles.  The  arrange- 
mediately  beneath  the  anterior  pit.  From  ment  of  these  on  the  glabella  is  shown  in 
this  posterior  end  of  the  hypostomal  suture  Figure  7,  and  in  Table  2.  Arrangement  of 
a  flexure  (PI.  12,  fig.  3)  runs  diagonally  tubercles  on  cheek  and  border  is  shown 
across  the  doublure  of  the  free  cheek.  in  the  photographs;  notable  is  the  single 
Hypostome  of  width  approximately  equal  row  on  the  cranidium  between  the  pre- 
to  length  (sag.),  elongate-oval  to  diamond  glabellar  furrow  and  the  sutural  margin, 
shaped  in  outline.  Central  body  strongly  the  median  and  faint  additional  tubercles 
inflated  and  with  a  pronounced,  narrow  present  on  the  occipital  ring,  and  low 
anterior  median  lobe  which  projects  for-  tubercles  on  the  posterior  border.  A  fine 
ward  below  the  anterior  l^order  furrow,  granulation  is  present  on  and  between  the 
Latter  shallow,  separated  from  the  sutural  tubercles  (PI.  U,  fig.  20),  on  the  ventrally 
margin  by  a  narrow  band.  Shallow  furrow  facing  part  of  the  border,  and  on  the  hypo- 
at  side  of  steep  slope  of  median  lobe  iTins  stome.  This  granulation  is  not  present  in 
backward  and  slightly  outward  to  die  out  the  deeper  parts  of  the  furrows, 
level  with  the  anterior  wing.  Lateral  bor-  Thorax  of  10  segments;  axis  about  one- 
der  narrow  and  convex,  border  furrow  third  total  width  (tr. )  at  the  fifth  segment; 
deep  and  inflated  middle  body  overhangs  axial  furrows  sHghtly  impressed;  axial  rings 
this  border;  posterior  border  widest  medi-  with  a  faint  elongate  swelling  distally.  In- 
ally,  flattened,  separated  from  the  middle  ner  part  of  pleura  horizontal,  outer  part 
body  by  a  shallow  border  furrow,  and  flexed  steeply  down;  broad,  convex  pleural 
forming  a  flat  posteriorly-directed  projec-  band  is  four  times  the  width  ( exs. )  of  the 
tion.  Macula  (Pi.  11,  fig.  21)  a  faint  low  narrow,  flattened  anterior  flange,  the  two 
swelling  at  posterolateral  margin  of  the  separated  by  a  sharp  change  in  slope  rather 
middle  body.  Anterior  wing  of  hypostome  than  a  pleural  furrow.  Lateral  to  the  ful- 
(Pl.  11,  figs.  6,  7,  12,  13)  greater  in  height  crum,  anterior  flange  expands  to  form  a 
than  the  central  body;  outer  tip  of  wing  broad  flattened  facet;  the  pleural  band 
twisted  and  deflected  outward  and  back-  tapers  slightly,  and  is  curved  in  an  ante- 
ward;  inner,  dorsal  edge  of  wing  deflected  riorly  concave  arch,  the  tip  extended  as  a 


470        BiiUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


.  0  111  1)1 


VI 

-  1,2 

V 

-  0,1-3 

V 

-  0,1,2 

IV 

-  0,1-3 

Iv 

-  1,2 

III 

-  1,3 

11 

-  1,2 

Figure  7.  Fragiscutum  rhytium  n.  gen.,  n.  sp.  Two 
cranidia  showing  notation  (after  Tripp,  1957,  1962)  of 
glabellar  tubercles.  A)  Original  of  Plate  12,  figure  15. 
B)  Original  of  Plate  12,  figure  1.  Abbreviatons:  Oa,  la, 
2a,  positions  of  occipital,  first  and  second  apodemes;  ap, 
anterior    pit;    pgf,    preglabellar    furrow. 

short,  blunt  spine.  Apodeme  (Pi.  11,  fig. 
3;  PI.  12,  figs.  13,  17)  is  curved  inward  and 
downward,  and  situated  a  short  distance 
in  trom  the  axial  furrow.  Ventral  surface 
of  pleura  does  not  reflect  dorsal  surface 
(Fig.  8b);  anterior  half  is  convex  down- 
ward, and  there  is  a  deep  groove  which 
cur\(s  outward  just  inside  the  narrow 
doublure;  latter  widens  and  extends  across 
the  pleura  at  the  tip.  The  various  struc- 
tures that  facilitate  articulation  between 
the  segments  and  limit  enrollment  are 
shown  in  Figure  8a  (compare  PI.  11,  figs. 
1-3;  PI.  12,  figs.  13,  14,  17,  20).  The  down- 
ward-projecting ring  process  fits  into  a  ring 
socket  that  surrounds  the  axial  process. 
The  anterior  flange  on  the  inner  part  of  the 
pleura    fits    beneath    the   narrow   posterior 


Figure  8.  Fragiscutum  rhytium  n.  gen.,  n.  sp.  A)  Ventral 
view  of  part  of  a  segment.  Compare  Plate  12,  figures  13, 
14,  17,  20.  B)  Section  tfirougfi  inner  part  of  pleura,  in 
exsagittal  plane.  Solid  black  is  replaced  outer  portion  of 
exoskeleton.  Abbreviations:  afl,  anterior  flange;  ap,  apo- 
deme; axp,  axial  articulating  process;  axs,  axial  articulat- 
ing socket;  d,  doublure  of  axial  ring;  fp,  projection  at 
anterior  margin  of  facet;  pb,  pleural  band;  pis,  pleural 
spine;  pr,  posterior  recess;  rp,  ring  articulating  process; 
rs,  ring  articulating  socket;  s,  silicified  lining  of  canal, 
appearing  as  a  hollow  cone  (compare  PI.  13,  fig.  17); 
vf,   vincular    furrow. 


recess  so  that  in  dorsal  aspect  only  the 
pleural  bands  are  visible.  Along  the  pos- 
terior edge  of  the  outer  part  of  the  dou- 
blure is  the  vincular  furrow,  which  receives 
the  anterior  edge  of  the  pleural  band  of 
the  succeeding  segment.  A  limit  to  enroll- 
ment occurred  when  the  outer  parts  of  the 
pleural  bands  moved  against  each  other 
and  the  thickened  projections  of  the  facets 
approached  each  other.  Anterior  two  tho- 
racic segments  have  outer  parts  of  pleurae 
slightly  shorter  ( tr. )  than  succeeding  seg- 
ments; posterior  border  of  cephalon  with 
structures  for  articulation  corresponding  to 
those  along  posterior  margin  of  segment. 
Cross  section  of  inner  part  of  pleura  of 
segment  ( Fig.  8b )  shows  exceptional  thick- 
ness of  exoskeleton  in  this  region,  and  the 
exoskeleton  of  both  the  axial  ring  and  outer 
part  of  the  pleura  must  have  been  similarly 
thickened.    As  certain  specimens  show  (PI. 


Silurian  Trilobites  •  Whittington  and  Campbell        All 


10,  fig.  11;  PI.  13,  fig.  17),  along  the  trans-  behind  here  it  is  flattened  on  the  ventral 
verse  midline  of  the  axial  ring  and  pleura,  side;  the  projecting  anterior  part  is  shaped 
the  exoskeleton  was  partially  traversed  by  to  fit  inside  the  doublure  of  the  cephalon 
canals,  preserved  in  the  silicified  material  during  enrollment.  There  is  a  deep  notch 
as  hollow  cones  extending  from  the  inner  in  the  doublure  beneath  the  tip  of  the  axis; 
toward  the  outer  surface  of  the  exoskeleton.  this  notch  received  the  projecting  median 
These  canals,  however,  did  not  open  on  lobe  of  the  hypostome  during  enrollment, 
either  the  inner  or  the  outer  surface  of  the  External  surface  of  thorax  and  pygidium 
exoskeleton,  so  far  as  the  silicified  speci-  granulated,  except  in  the  furrows.  The 
mens  show  (PI.  12,  fig.  20;  Pi.  13,  figs.  11,  granulation  extends  over  the  ventral-facing 
14-17).  Similar  structures  have  been  seen  part  of  the  pygidial  border,  but  not  on  to 
in  specimens  in  which  the  exoskeleton  is  the  inward  facing  doublure.  The  median 
preserved  as  calcium  carbonate  (Campbell,  axial  tubercles  of  the  pygidium  are  present 
in  press),  and  when  weathered  the  canals  in  both  small  and  large  specimens,  but 
may  be  seen  as  perforations,  but  again  they  some  specimens  show  no  other  tubercles  on 
are  apparently  not  visible  on  unweathered  either  axial  rings  or  pleural  bands.  In  some 
surfaces  and  it  is  uncertain  whether  or  not  specimens  four  or  five  extremely  faint 
they  traversed  the  outemiost  layers  of  the  tubercles  may  be  recognized  amid  the 
exoskeleton.  granulation  on  the  pleural  bands;  these  are 
Pygidium  triangular  in  outline,  width/  analogous  to  the  very  faint  tubercles  which 
length/height  ratio  approximately  5/4/2.5.  may  sometimes  be  recognized  on  the  pos- 
Axis  with  flattened  profile  in  cross  section  terior  border  of  the  cephalon. 
(similar  to  that  of  axial  rings  of  thorax).  Development.  Characteristic  of  the  small 
larger  specimens  (PI.  13,  fig.  11)  with  18  exoskeleton  is  the  spininess — the  larger 
axial  rings,  ring  furrows  deeply  incised  tubercles  are  elongated  as  blunt  spines, 
laterally;  behind  second  ring  these  furrows  and  smaller  tubercles  as  thorn-like  spines, 
weaken  so  that  a  smooth  median  track  runs  There  is  a  curving  fixigenal  spine  and  the 
posteriorly  along  the  axis.  Four  or  five  posterior  bands  of  both  thorax  and  pygid- 
large  median  axial  tubercles,  the  first  on  ium  are  extended  as  spines.  On  the  small- 
ring  3  or  4,  the  second  on  ring  6  or  7,  the  est  cranidium  (Pi.  12,  figs.  12,  18;  length 
third  on  ring  10  or  11,  the  fourth  and  fifth  1.3  mm)  the  arrangement  of  tubercles  on 
present  close  to  the  tip  of  the  axis.  Pleural  the  glabella  ( Table  2 )  is  like  that  in  larger 
regions  curve  steeply  down,  subdivided  by  specimens,  showing  that  the  main  outlines 
deep  furrows  into  eight  pleural  bands,  the  of  this  pattern  are  established  at  an  early 
posterior  pair  curving  inward  distally  and  stage.  Tubercles  additional  to  the  median 
merging  behind  the  tip  of  the  axis;  inside  are  present  on  the  occipital  ring,  and  also 
this  pair  there  is  a  ninth  pair  and  a  faint  on  the  posterior  border  and  base  of  the 
median  strip  visible  on  larger  specimens,  fixigenal  spine.  The  external  surface  of  the 
The  first  five  furrows  extend  to  the  margin,  free  cheek  (PL  12,  fig.  11)  also  is  spinose 
and  the  tips  of  the  pleural  bands  project  rather  than  tuberculate.  The  small  hypo- 
and  are  slightly  expanded  (PI.  13,  figs.  14,  stome  (Pi.  11,  figs.  11,  14,  15,  17)  has  the 
15);  behind  here  the  pleural  furrows  die  forward  projection  of  the  middle  body  less 
out  before  reaching  the  margin.  Anterior  prominently  developed,  and  the  posterior 
margin  of  pygidium  like  that  of  anterior  border  relatively  narrower.  This  border 
margin  of  segment,  with  anterior  flange  bears  a  median  posterior  and  three  pairs  of 
and  facet,  the  latter  crossed  by  a  shallow  tiny  spines  on  the  margin.  With  increase 
(pleural?)  furrow  (Pi.  12,  fig.  23;  Pi.  13,  in  size  the  main  changes  in  the  cephalon 
fig.  15).  Below  tips  of  first  five  pleurae  are  reduction  of  the  spines  to  tubercles 
border    of   pygidium    projects    downward;  and  a  rapid  relative  reduction  of  the  fixi- 


472        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


genal  spine  beyond  a  length   (sag.)  of  1.5  pygidium  is  also  a  reduced  remnant  of  an 

mm  (PI.  12,  figs.  1,  4,  6,  9,  15,  18).  axial  spine.    The  median  tubercles  in  the 

The  smallest  transitory  pygidium  ( PI.  13,  holaspid  pygidium  are  not  situated  oppo- 

figs.  4,  9,  13)  includes  at  least  six  segments,  site  every  third  ring  as  they  appear  to  be 

the  axis  tapering  rapidly  back\\ard  to  the  in    the   transitory    pygidia.     This    arrange- 

rounded  tip,  the  third  ring  bearing  a  long  ment  may  result  from  the  crowding  of  the 

upwardly  and  backwardly  directed  median  rings    in    the    axis,   which    evidently   takes 

spine  and  a  shorter,  more  backwardly  di-  place  at  a  developmental  stage  subsequent 

rected  spine  behind  this,   possibly  on  the  to  that  of  these  transitory  pygidia. 
fifth   segment.    The  pleural   regions   curve  Discussion.   A  second  species  of  the  new 

down  steeply  and  extend  in  a  narrow  band  genus,  from  the  Henryhouse  Formation  of 

behind  the  tip  of  the  axis;  pleural  bands  Oklahoma,  is  being  described  by  Campbell 

are  extended  as  spines,  the  longest  on  the  (in  press).    Other  American  Silurian  spe- 

first   band    directed   backward,    successive  cies  (e.g.,  Raymond,  1916;  Best,  1961)  are 

spines     directed     slightly     inward.      Each  either  unlike  the  present  one  or  known  only 

pleural   band   bears   a   prominent  tubercle  from  such  fragmentary  material,  including 

at    about    the    mid-length,    these    tubercles  internal  moulds,  that  comparisons  are  not 

fonning  a  line  that  curves  back  subparallel  possible. 

to    the    axial    furrow.     Doublure    narrow,         Several    authors    (Rosenstein,    1941:    57, 

curled    under,    narrowest    medially    where  pi.  2,  fig.  2;  Temple,  1954;  Tripp,  1962,  pi. 

the  margin  is  arched  upward  in  posterior  67,  figs.  2,  9b,  10;  \\'hittard,  1938:   120,  pi. 

view,  but  lacking  the  median  notch  of  large  4,  fig.  7 )  have  described  the  hypostome  of 

specimens  (Pi.  13,  fig.  14).    The  lack  of  a  Silurian  species  of  Encrimirus  and  the  way 

notch  corresponds  with  the  lower  convexity  in  which  it  was  attached  to  the  remainder 

of  the  median  hypostomal  lobe  at  this  stage  of  the  cephalon.    The  preservation  of  the 

(PI.  11,  figs.  14,  17).  present  material  shows  clearly  the  forni  of 

Larger  transitory  pygidia  (Pi.  13,  figs,  the  anterior  wing  of  the  hypostome  (Pi. 
1-3,  5-8,  12)  contain  more  segments,  but  11,  fig.  7)  and  how  it  was  related  to  the 
are  generally  similar  in  form.  Up  to  three  cranidium  (Pi.  11,  fig.  13).  The  anterior 
median  axial  spines  may  be  present,  these  and  posterior  wings  are  joined  in  a  single 
spines  apparently  situated  on  every  third  structure  by  a  broad  portion  of  the  dou- 
ring.  The  border  spines  of  the  posterior  blure;  the  anterior  wing  is  the  larger  and 
pleural  bands  are  relatively  shorter  but  is  twisted  so  that  its  distal  cross  section  is 
similarly  directed.  There  may  be  two  rows  U-shaped,  the  open  end  of  the  'U'  facing 
of  tubercles  on  the  pleural  bands,  and  anteriorly.  From  the  inner  surface  projects 
there  is  also  a  fine  granulation  on  the  rings  the  wing  process,  and  there  is  a  pit  cor- 
and  bands.  As  a  small  segment  shows  (Pi.  responding  to  this  process  on  the  outer, 
12,  figs.  21,  22),  these  segments  with  a  posterior  surface.  On  the  outer  surface  of 
median  axial  spine  and  spines  on  the  the  wing,  inside  this  pit,  there  is  an  elon- 
pleural  band  are  released  into  the  thorax  gate  projection  from  the  surface  of  the 
during  development.  Apparently  the  me-  wing — this  apparently  corresponds  to  the 
dian  spines,  as  well  as  the  additional  knob  described  by  Temple.  Inside  this 
tubercles  on  axial  ring  and  pleural  bands,  knob,  the  inner  extremity  of  the  wing  is 
are  reduced  and  disappear  in  the  larger  extended  fonvard  as  a  long  flange,  project- 
stages;  the  same  is  true  in  the  pygidium,  ing  directly  anteriorly  (Pi.  11,  fig.  12).  On 
though  in  rare  specimens  tubercles  on  the  the  inner  surface  of  the  cranidium  the 
pleural  bands  may  be  distinguished.  It  axial  furrow  fomis  a  broad  prominent 
seems  probable  that  each  of  the  five  tuber-  ridge,  and  just  inside  the  sutural  margin  is 
cles   along  the   median   axial  band   of  the  the  boss  reflecting  the  anterior  pit  in  the 


Silurian  Trilobites  •  Whittington  and  Campbell        473 


external  surface.  When  the  hypostome  was 
in  position,  presumably  the  tip  of  the  wing 
process  was  close  to  the  backward-facing 
slope  of  the  anterior  boss.  There  does  not 
appear  to  be  a  distinct  pit  in  this  backward- 
facing  slope  that  received  the  tip  of  the 
wing  process,  similar  to  that  seen  for  ex- 
ample in  CerauiincUa  (Whittington  and 
Evitt,  1954,  pi.  12,  fig.  30).  It  is  clear  that 
the  curved  extremity  of  the  wing  is  so 
shaped  as  to  fit  around  close  to  the  slopes 
of  the  axial  furrow,  the  flange  on  the  inner 
tip  of  the  wing  extending  beside  the  steep 
inner  slope  of  this  furrow.  When  the  hypo- 
stome was  in  place,  with  the  relatively 
broad,  flat  surfaces  along  the  hypostomal 
suture  in  contact  with  the  rostral  plate  and 
free  cheeks,  the  anterior  wing  wrapping 
around  the  anterior  boss,  it  is  difficult  to 
imagine  that  any  relative  movement  was 
possible  between  the  hypostome  and  the 
cephalon. 

In  Whittington's  (1965:  420-421)  recent 
diagnosis  of  Encrinuridae,  reference  is 
made  to  anterior  and  posterior  pleural 
bands  of  the  thorax.  In  the  present  species, 
no  pleural  furrows  and  anterior  bands  are 
visible  in  the  articulated  thorax  (Pi.  11, 
fig.  2;  PI.  13,  fig.  17 ) .  What  appears  to  be 
an  anterior  band  in  an  isolated  segment 
(Pi.  12,  fig.  14)  is  a  narrow  strip,  here 
termed  the  articulating  flange,  which  fits 
beneath  the  posterior  recess  (Fig.  8)  of 
the  segment  in  front.  The  only  pleural 
furrow  visible  is  that  on  the  facet  of  the 
pygidium  (PI.  12,  fig.  23;  PI.  13,  fig.  15). 
We  have  thus  referred  to  pleural  bands  of 
thorax  and  pygidium,  but  the  pleural  fur- 
row on  the  first  segment  of  the  pygidium 
makes  clear  that  these  bands  are  posterior 
bands  and  that  the  anterior  band  has  been 
reduced.  The  position  of  the  pleural  fur- 
row is  at  the  foot  of  the  slope  of  pleural 
band  down  to  articulating  flange.  Tripp's 
(1962:  466)  description  of  the  thorax  of 
Encrinunis  punctatus  suggests  that  the 
condition  may  be  the  same  in  that  species. 
Other  encrinurid  species  (Tripp,  1962,  pi. 
67,  fig.  3;  1957,  pi.  11,  fig.  17;  pi.  12,  figs. 


11,  16,  17)  show  clearly  anterior  bands  and 
pleural  furrows  on  both  thorax  and  pygid- 
ium; i.e.,  in  some  species  the  anterior  band 
is  not  so  reduced  as  in  F.  rJnjtium.  The 
transitory  pygidia  show  an  extremely  nar- 
row anterior  band  (or  articulating  flange) 
on  only  the  first  segment,  not  succeeding 
segments. 

Family   DALMANITIDAE   Vogdes,    1890 
Genus  DALMANITES  Barrande,  1852 
Dalmanites  puticulifrons   n.   sp. 

Plates  14,  15;  Plate  19,  figures  15,  17. 

Holotype.  USNM  1.54302,  cranidium,  an- 
terior border,  and  part  of  left  free  cheek. 

Description.  Cephalon  of  width  about 
twice  the  length  (sag.);  longitudinal  pro- 
file of  glabella  low,  highest  point  at  occipi- 
tal ring,  the  profile  descending  gradually 
forward  to  the  back  of  the  anterior  glabella 
lobe  which  is  gently  inflated;  occipital  ring 
more  markedly  convex,  particularly  trans- 
versely. Occipital  furrow  broad  and  well 
rounded  medially,  but  narrowing  rapidly 
into  the  deep,  slot-like  apodeme.  Glabellar 
furrow  Ip  transverse,  shallow  adjacent  to 
the  axial  furrow,  deepening  inward;  fur- 
row 2p  directed  in\\'ard  and  slightly  for- 
ward, extremely  shallow  adjacent  to  the 
axial  furrow  but  deepening  at  the  apo- 
deme; furro\\'  3p  directed  inward  and 
backward,  widest  at  the  axial  furrow,  nar- 
rowing and  deepening  inward  but  not  ex- 
tended from  the  ventral  surface  as  an 
apodeme.  Occipital  and  Ip  apodemes  (Pi. 
19,  fig.  15)  triangular  in  cross  section 
proximally,  distally  becoming  blade-like, 
and  twisted  so  that  the  flat,  blade-like  por- 
tion is  directed  diagonally;  apodeme  2p 
slimmer,  blade-like,  and  not  twisted.  Me- 
dian glabellar  lobe  narrower  (tr. )  than 
lateral  lobes,  anterior  lobe  diamond-shaped 
with  a  broad,  shallow  median  pit  in  the 
posterior  portion.  Axial  furrow  shallow, 
curving  inward  at  the  anterior  edge  of  the 
occipital  ring,  then  running  forward  and 
slightly  outward,  rising  and  gradually  shal- 
lowing to  the  mid-length  of  lateral  lobe  .3p, 
in    front   of   here   dropping   into   a   broad. 


474        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


well-rounded  furrow  which  cur\es  around 
the  extremity  of  the  anterior  lobe;  pre- 
glabellar  furrow  shallow.  Eye  lobe  situated 
in  the  inner  corner  of  the  cheek,  the  ante- 
rior margin  abutting  against  the  axial  fur- 
row, the  posterior  margin  a  short  distance 
in  front  of  the  posterior  border  furrow. 
Palpebral  lobe  rises  moderately  steeply 
from  the  axial  furrow,  palpebral  furrow 
much  deeper  in  its  anterior  than  its  poste- 
rior half,  the  palpebral  rim  standing  high 
abo\e  the  crest  of  the  lobe,  flattened  on 
the  crest,  with  a  narrow  marginal  band; 
the  rim  asymmetrical,  the  anterior  part 
being  the  larger.  Eye  surface  steeply  slop- 
ing, facets  large  and  arranged  in  diagonal 
lines  (PI.  19,  fig.  15).  Distinct  furrow 
around  the  anterolateral  margin  of  the  eye 
lobe,  on  the  outside  of  which  is  a  low 
ridge,  most  prominent  anteriorly;  outside 
this  ridge  the  cheek  slopes  gently  outward 
and  downward  to  the  broad  lateral  border 
furrow.  Lateral  border  with  a  flattened, 
outward-sloping  upper  surface,  which  is 
continuous  with  that  of  the  anterior  border; 
latter  widest  medially,  forming  a  blunt 
projection  (Pi.  14,  fig.  1).  Posterior  bor- 
der furrow  sabre-like  in  outline  with  its 
anterior  slope  steeper  than  the  posterior; 
posterior  and  lateral  furrows  do  not  merge 
at  the  genal  angle,  but  the  posterior  border 
terminates  slightly  above  and  inside  the 
shallow  lateral  furrow.  Posterior  border 
widening  (exs.)  rapidly  outside  the  ful- 
crum, and  running  out  into  the  base  of  the 
genal  spine.  Posterior  branch  of  suture 
curves  around  the  posterolateral  margin  of 
the  eye  and  then  runs  straight  out  and 
slightly  backward  across  the  cheek  to  the 
margin;  the  points  e-t  and  oj-oj  are  thus  in 
the  same  transverse  line.  Anterior  branch 
of  suture  runs  forward  in  a  sweeping  curve 
a  short  distance  outside  the  axial  furrow 
and  over  the  upper  surface  of  the  anterior 
border  close  to  the  inner  edge,  the  two 
branches  meeting  in  a  smooth  curve. 
Points  fi-/i  in  line  with  the  maximum  width 
of  the  glabella  across  the  anterior  lobe. 
Doublure    of    cephalon    beneath    lateral 


border  of  same  width  as  border,  wider  an- 
teriorly (  PI.  14,  fig.  4 ) .  Beneath  posterior 
border  doublure  extends  from  base  of 
genal  spine  to  fulcrum  but  not  beneath 
inner  part  of  border.  At  the  inner  margin 
the  lateral  doublure  is  bent  shaiply  dorsad; 
this  flexure  dies  out  anteriorly  and  is  ab- 
sent along  the  hypostomal  suture.  Hypo- 
stome  unknown. 

External  surface  finely  granulated  (Pi. 
19,  fig.  17)  except  in  axial,  glabellar,  and 
palpebral  furrows  and  at  base  of  median 
glabellar  pit. 

Three  thoracic  segments  known,  axial 
rings  evenly  arched  transversely,  longitudi- 
nally flattened;  articulating  furrow  with 
slopes  equally  steep;  axial  furrow  im- 
pressed only  at  the  rear  of  each  segment. 
Pleura  having  anterior  band  much  higher 
than  posterior  band,  with  a  vertical  or 
slightly  undercut  posterior  slope  down  to 
the  pleural  furrow;  anterior  slope  of  an- 
terior band  gentle,  passing  out  beyond  the 
fulcrum  into  the  broad  facet;  faint  articu- 
lating flange  along  anterior  edge  of  pleura, 
dying  away  distally.  Pleural  furrow  running 
in  a  slightly  sigmoid  course  from  inner  an- 
terior corner  to  the  posterior  border  of  the 
tip,  dying  out  at  a  point  directly  above  the 
inner  edge  of  the  doublure;  furrow  shallow 
and  rounded  near  the  axis,  broader  and 
with  a  flat  base  near  the  fulcrum,  narrow- 
ing distally;  posterior  band  of  pleura  with 
a  gentle  anterior  slope  throughout.  Dou- 
blure extends  along  the  posterior  margin  of 
segment  halfway  in  to  the  fulcrum,  but  is 
not  so  extended  along  the  anterior  side. 

Axis  of  pygidium  with  12  complete  rings 
and  terminal  portion,  behind  eighth  ring 
axis  tapers  more  gently  and  the  ring  fur- 
rows become  shallower;  in  this  differenti- 
ated posterior  part  of  the  axis  the  tip  is 
rounded  and  prominent,  the  ring  furrows  are 
present  only  on  the  outer  part,  shallow 
and  outwardly  and  slightly  backwardly 
directed.  Articulating  halfring  three-fifths 
the  length  of  the  remainder  of  the  first 
segment;  excavation  in  posterior  margin  of 
first    axial    ring    of    similar    length    (sag.). 


Silurian  Trilobites  •  Whittington  and  Campbell        475 


much  smaller  excavations  in  the  posterior  pleural  band.    Doublure  granulated  along 

margin    of   the   succeeding   two   rings.     In  the  outer  portion,  the  granulation  diminish- 

longitudinal    profile    all    except    most   pos-  ing  and  disappearing  inward, 

terior  rings  with  a  much  steeper  posterior  Development.     One    protaspis    (Pi.    15, 

than  anterior  slope.    The  outer  quarter  of  figs.  5,  6,  10,  14,  17),  lacking  the  free  cheeks 

the  ring  furrow  is  deep,  and  on  the  inner  and   hypostome,   has   the   cephalic  portion 

surface  there  is  an  apodeme  on  the  inner  of   length    ( sag. )    0.66   mm.     The   form   is 

part  of  this  deepened  region,  the  apodemes  extremely  like  that   of  the  protaspis   from 

present  on  the  articulating  furrow  and  the  the    Devonian    described    by    Whittington 

next  eight  ring  furrows    (PL   14,  fig.    14).  (1956a:    105-106,  pi.  24,  figs.  1-5;  text-fig. 

Proximally    the    apodemes    are    triangular  1 ) .    Glabella  shows  a  similar  division  into 

in  section,  the  acute  edge  on  the  outer  side,  an    extremely    short    ( sag. )    occipital    ring 

tapering  to  a  blade-like  tip  directed  down-  with   a   prominent  median  tubercle,   three 

ward  and  slightly  forward.    Pleural  regions  glabellar     rings     of     approximately     equal 

curving  downward  from  the   axial  furrow  length,  and  a  wider  anterior  lobe  which  has 

to  the  narrow,  flattened,  and  steeply  slop-  the  lateral  portions  faintly  set  off  from  the 

ing  border  on  which  pleurae  are  not  marked,  median  region.    On   the  preglabellar  area 

The  first  seven  interf)leural  furrows  shaq^ly  there  are  two  pairs  of  spines;  the  palpebral 

incised,    the    eighth    indicated    only    by    a  lobe  is  situated  on  the  anterolateral  margin 

smooth    band    in    the    granulation;    pleural  of  the  shield,  its  position  and  the  course  of 

furrows  narrow  ( exs. )  adjacent  to  the  axial  the   sutures   as   in   the   Devonian   example, 

furrow    then    widening    rapidly    and    with  The    narrow    posterior   border    is    directed 

a  broad,  flat  floor  extending  out  to  the  in-  straight  outward  and  turns  vertically  down- 

ner  edge  of  the  border;  on  all  segments  an-  ward  on  the  flank,  where  the  specimen  is 

terior    slope    steep,    posterior    slope    much  broken.   The  protopygidium  is  also  broken, 

more  gentle.    Anterior  band  well  rounded  but  shows  the  convex  axis,  and  apparently 

on  all  segments,  each  standing  higher  than  two  segments,  having  backwardly  directed 

the  posterior  band  immediately  in  front  of  spines   on   the   outer   part   of  the   pleurae, 

it,  these  posterior  bands  narrow  and  flat-  The  external  surface  of  the  fixed  cheek  is 

tened.    At   posterior  tip,   border   extended  pitted   (as  in  the  Ordovician   Dolmanitina 

into  a  median  spine  which  is  flattened  on  protaspis  of  Temple,   1952,  pi.  10,  fig.  6), 

the    under   surface    (Pi.    14,   figs.    15,    16).  and  there  are  paired  granules  on  the  cheek. 

Doublure  of  about  the  same  width  as  the  including    one   halfway    across    the    cheek, 

border,  the  inner  edge  bent  up  vertically  situated  immediately  in  front  of  the  pos- 

(Pl.  14,  fig.  14).  terior     border     furrow.      Six     conspicuous 

External  surface  of  thorax  and  pygidium  granules  on  the  median  occipital  tubercle. 

(PI.    14,  fig.   12)    bearing  granules,   which  Cranidia  of  length  1.0  mm  to  1.3  mm  are 

are  present  on  the  axial  rings,  in  the  articu-  meraspides    (Pi.   15,   figs.  7,   11,   15).    The 

lating    furrow    and    adjacent    edge    of    the  frontal  lobe  of  the  glabella  is  expanded  and 

articulating  halfring  of  the  thorax,  and  are  bent   down   more   steeply   with   increasing 

coarsest  along  the  posterior  margin  of  the  size — the    width    becoming    three-quarters 

ring;  the  rings  of  the  pygidium  are  similar,  the  length   ( sag. )   of  the  cranidium  rather 

but  the  ring  furrow  is  smooth.   The  coarsest  than  less  than  half  as  in  the  protaspis.   The 

granulation   on   the   pleural   regions   is    on  palpebral  lobe  moves  relatively  backward 

the    crests    of   the    anterior    and    posterior  as  well  as  inward  as  shown  by  the  position 

bands,    with    a    finer    granulation    in    the  of  the  line  8-8.    In  the  protaspis  this  line 

pleural  furrows,  on  the  facets,  pygidial  bor-  runs   just    behind   the    anterior   margin    of 

der  and  spine.  An  extremely  narrow  smooth  the  third  glabellar  ring,  one-third  the  length 

strip  on  the  anterior  margin  of  the  anterior  ( sag. )    of   the   glabella    from    the   anterior 


476        Bulletin  Miisemu  of  Comparative  Zoolugy,  Vol.  135,  No.  9 


marpn.  In  progres.si\  el\'  larger  specimens 
the  line  8-8  moves  to  about  the  mid-length 
(PI.  15,  fig.  7)  of  the  glabella,  thence  to  a  posi- 
tion about  two-thirds  the  length  (Pi.  15,  fig. 
1),  where  it  crosses  the  second  gla])ellar 
ring.  Not  only  the  frontal  glabellar  lobe 
i)ut  lobes  2p  and  3p  become  relatively  wider 
so  that  the  axial  furrows  are  strongly  diver- 
gent forward  (Pi.  15,  figs.  1,  4),  and  glabel- 
lar furrow  3p  ceases  to  be  transverse,  and 
becomes  inclined  to  the  transverse  line. 
On  the  inner  surface  of  the  small  cranidia 
there  are  rounded  projections  adjacent  to 
the  outer  ends  of  the  occipital  furrow  and 
glabellar  furrows  l-3p,  but  strongly  pro- 
jecting apodemes  are  developed  only  in 
large  cranidia. 

In  the  original  of  Plate  15,  figure  1  (5.2 
mm  in  length  [sag.]),  granulation  is  pres- 
ent, the  median  occipital  tubercle  is  al- 
most completely  reduced,  and  the  pit  is 
present  in  the  frontal  glabellar  lobe. 

The  original  of  Plate  15,  figures  9,  12, 
13,  16,  a  transitory  pygidium,  shows  the 
completely  formed  articulating  halfring  of 
the  second  segment  beneath  the  first  ring. 
The  spines  on  the  ends  of  the  pleurae  ap- 
l^ear  to  be  the  combined  tips  of  the  anterior 
and  posterior  bands,  which  bands  are  more 
equal  in  width  (exs.)  than  in  larger  stages. 
Scattered  short  spines  are  present  on  the 
axial  rings  and  pleural  bands,  as  well  as 
granulation;  short  apodemes  are  present. 
A  specimen  of  the  same  size  ( Pi.  15,  fig.  8 ) 
does  not  have  the  articulating  halfring  of 
the  second  segment  developed,  and  may  be 
a  small  holaspid,  but  in  dorsal  aspect  is 
very  similar  to  the  other  example. 

Discussion.  Numerous  American  Silurian 
species  of  Dalmanites  have  been  described 
(Delo,  1940:  .37-52)  but  only  ten  of  them 
are  known  from  both  cephalon  and  py- 
gidium. The  form  of  cephalon  and  pygid- 
ium, the  depressed  profile,  the  shape  of 
the  genal  spines,  shape  of  the  posterior 
border  and  its  relation  to  the  lateral  bor- 
der furrow,  and  the  relatively  uniform  na- 
ture of  the  granulation,  ally  this  species 
with  D.  limuJunts  (Green,  1832)  from  the 


Rochester  Shale.  However,  D.  puticiiUfrons 
has  a  relatively  larger  eye  lobe,  no  nodes 
at  the  extremities  of  the  thoracic  rings, 
deeper  pleural  furrows  terminating  at  the 
rear  edge  of  each  thoracic  segment,  more 
numerous  segments  in  the  pygidium,  a 
shorter  terminal  spine,  and  coarser  granula- 
tion on  the  external  surface. 

A  new  species  of  Dalmanites  being  de- 
scribed by  Campbell  (in  press),  from  the 
Ilenryhouse  Fonnation  of  Oklahoma,  has 
many  features  in  common  with  D.  puticuJi- 
frons  and  D.  Umtihirus,  the  cephalon  being 
very  like  that  of  D.  putictiUfwns.  It  may 
be  distinguished  by  having  a  pygidium  of 
different  shape,  with  fewer  rings  and 
pleurae,  and  the  granulation  which  varies 
more  markedly  in  density  and  size  on  dif- 
ferent parts  of  the  cephalon.  The  British 
Silurian  species  D.  mijops  ( see  Dean,  1960), 
from  the  Wenlock  Series,  belongs  to  the 
same  species  group  and  examination  of 
specimens  in  the  Museum  of  Comparative 
Zoology  suggests  that  it  is  distinct  from  D. 
piificuUfrons  in  having,  among  other  char- 
acters, a  shorter  palpebral  lobe,  the  facial 
suture  lying  in  the  preglabellar  furrow, 
larger  tubercles  on  the  surface  of  the 
glabella,  and  differently  shaped  pleural 
furrows  on  both  the  thorax  and  pygidium. 

According  to  Richter,  Richter  and  Struve 
{in  Moore,  1959:  O  471-2),  Odontochile 
differs  from  Dalmanites  in  that  the  anterior 
branches  of  the  suture  line  are  farther  from 
the  frontal  glabellar  lobe,  in  the  number  of 
spines  on  the  margin  of  the  hypostome  and 
in  the  greater  number  of  segments  in  the 
pygidium.  The  course  of  the  anterior 
branches  of  the  suture  in  D.  putieulifrons 
is  like  that  in  Odontochile,  but  the  hypo- 
stome is  unknown.  Delo  (1940:  55)  used 
nimiber  of  segments  in  the  pygidium  as  the 
main  criterion  in  placing  various  American 
Devonian  species  in  Odontochile.  The 
value  of  this  single  character  is  open  to 
question.  Much  more  needs  to  be  known 
of  Silurian  and  Devonian  dalmanitids  be- 
fore generic  criteria  can  be  clarified,  and 


Silurian  Trilobites  •  Whittington  and  Campbell        477 


particularly    the    use    of    Dalmanites    and 
Odontochile. 

Family     ODONTOPLEURIDAE     Burmeister, 
1843 

Genus  LEONASPIS  R.  and  E.  Richter,  1917 

Leonaspis  cf.  williamsi  Whittington,  1956 
Plate    16,   figures    1-14,    16-18,    20-22; 
Plate  17,  figures  1-12,  16;  Figure  9. 

Description.  The  material  is  excellently 
preserved,  and  while  it  shows  that  there  is 
individual  variation  (compare  Pi.  16,  figs. 
1,  2,  with  PI.  16,  figs.  5,  6),  it  is  not  adequate 
to  show  the  range  of  this  variation.  The 
exoskeletal  parts  are  like,  but  apparently 
not  identical  with,  those  of  the  Lower 
Devonian  species  Leonaspis  tuherculatus 
from  New  York  and  those  of  the  species 
L.  uiUiamsi  from  Oklahoma  (Whittington, 
1956b:  507-510,  pi.  57,  pi.  58,  figs.  1-4,  6,  7). 
The  uncertainty  regarding  identity  is  partly 
because  the  New  York  and  Oklahoma  speci- 
mens are  less  perfectly  preserved,  and  partly 
because  the  material  from  any  one  of  the 
localities  is  limited.  In  these  circumstances 
we  have  chosen  to  compare  the  Maine  form 
to  williamsi,  as  the  better-known  of  pre- 
viously described  species.  Some  of  the  dif- 
ferences between  the  Maine  specimens  and 
those  from  the  other  localities  are: 

1 )  median  occipital  spine  is  longer  than 
that  of  williamsi,  not  as  long  as  that  of 
tuherculatus; 

2)  eye  lobe  higher  than  that  of  williamsi; 
eye  lobe  of  available  specimens  of  tuher- 
culatus is  broken; 

3)  number  of  thoracic  segments  is  un- 
known; tuherculatus  has  9,  williamsi  has  8; 

4)  in  such  details  as  the  number  of 
tubercles  on  the  anterior  border  of  the 
cranidium,  number  of  spines  on  the  outer 
edge  of  the  border  of  the  free  cheek  and 
spacing  of  these  spines,  curvature  and 
length  of  the  posterior  pleural  spines  of  the 
thorax,  and  border  spines  of  the  pygidium,  it 
is  exactly  like  williamsi.  The  Maine  speci- 
mens are  unlike  tuherculatus  in  the  shorter 
genal  spines  and  length  of  pleural  thoracic 


spines.  In  williamsi  (Whittington,  1956b, 
pi.  58,  figs.  1,  2)  there  are  five  tubercles 
along  the  posterior  border  of  the  cephalon, 
several  tubercles  at  the  base  of  the  poste- 
rior pleural  spine,  as  well  as  two  prominent 
tubercles  on  the  pleural  ridge  of  the  pygid- 
ium. In  the  Maine  specimens  there  are  only 
two  tubercles  on  the  posterior  border  of  the 
cranidium,  large  tubercles  are  lacking  at 
the  base  of  the  posterior  pleural  spines,  and 
only  one  tubercle  is  present  on  the  pleural 
ridge  of  the  pygidium. 

These  differences  are  in  details,  and  it 
is  difficult  to  assess  their  taxonomic  signifi- 
cance. 

The  specimens  from  New  York,  Okla- 
homa and  Maine  (Pi.  17,  figs.  5,  8-12) 
exhibit  the  anterior  pleural  spines,  which 
are  fused  at  the  base  and  branch  distally. 
The  present  material  shows  (PI.  17,  figs. 
1,  2,  4)  that  on  the  first,  and  probably 
second,  thoracic  segments  the  anterolateral 
portion  is  beveled  and  facetted,  so  that  on 
these  segments  there  is  only  a  short,  rapidly 
tapering  posterior  pleural  spine.  The  an- 
terior pleural  spine  is  not  developed.  More 
posterior  segments  of  the  thorax  show  the 
narrow  doublure  which  lies  immediately 
inside  the  base  of  the  anterior  pleural  spine 
(PI.  17,  fig.  5).  The  anterior  comer  of  this 
doublure  projects  forward  as  an  articulat- 
ing process;  the  posterior  end  shows  a 
notch  to  receive  the  articulating  process  of 
the  following  segment,  this  notch  lying  be- 
neath the  base  of  the  posterior  pleural 
spine. 

The  two  small  cranidia  (PL  16,  figs.  7, 
9,  10,  12-14)  are  the  only  known  develop- 
mental stages  of  a  species  of  this  genus. 
In  the  smallest  the  palpebral  lobe  is  far 
back,  in  line  with  the  anterior  edge  of  the 
occipital  furrow,  and  the  glabella  has  the 
specific  outline.  It  also  shows  typical 
features  of  a  small  odontopleurid  in  that 
the  rounded  tubercles  of  the  larger  stages 
are  represented  by  thorn-like  spines,  and 
the  median  occipital  spine  is  much  longer 
than  in  larger  stages.  The  major  spines 
(Fig.  9)    are  typical   of  the  cranidium   of 


478        BiiUeiin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


1.0  mm 


W\-/}r/^A:yv-/2_ 


Figure  9.  Leonaspis  cf.  williamsi  Whittington,  1956.  Major 
paired  spines  of  a  small  cronidium,  original  of  Plate  16, 
figures    12-14,    lettered    following    Whittington,    1956c,    text- 

f.g.  1. 


an  early  odontopleurid  developmental  .stage. 
These  include  pains  2  to  5  on  the  glabella, 
Ai^a  on  the  fixed  cheek,  and  the  spine  on 
the  eye  ridge  (compare  Whittington,  1956c, 
text-figs.  1,  6,  9,  22,  for  corresponding 
spines  in  develo]:)mental  stages  of  other 
genera ) . 

XANIONURUS  n.  gen. 

Type  species.    Xanionurus  boucoti  n.  sp. 

Diagnosis:  Differs  from  Radia.s))is  in 
that: 

1 )  lateral  glal^ellar  lobes  are  not  fused, 
but  separated  by  deep  Ip  furrow; 

2)  there  is  a  distinct,  shallow  occipital 
furrow  and  the  occipital  ring  is  relatively 
much  shorter   (sag.); 

3)  second  axial  ring  of  pygidium  low, 
faintly  divided  by  median  longitudinal 
furrow  and  bearing  pair  of  spines; 

4)  border  of  pygidium  with  14  (not  16) 
spines,  the  posterior  band  of  the  anterior 
segment  running  into  the  base  of  the  5th 
(not  the  6th)  border  spine; 

5 )  besides  the  main  paired  spines  on  the 
glabella  there  are  many  additional  spines 
and  granules;  similarly,  on  the  cheek  there 
are  spines  on  the  upper  surface  of  the  bor- 
der, and  many  scattered  spines  and  gran- 
ules as  on  the  pygidium.  In  Radiaspis  the 
exoskeleton  is  smooth  between  the  main 
spines. 

Geological  range.  Upper  Wenlock  to 
Ludlow. 


Discussion.  Bruton  (personal  communi- 
cation )  points  out  that  Radiaspis  is  like 
Diacanthaspis  in  lacking  major  border 
spines  on  the  pygidium  and  in  the  ra- 
diating arrangement  of  the  border  spines. 
We  agree  with  this,  and  add  that  the  early 
developmental  stages  of  the  cranidium  of 
the  two  genera  appear  to  be  similar,  and 
the  thoracic  segments  are  alike.  As  in 
Acidaspis  and  Dudleyaspis  there  is  a  stout 
posterior  sutural  ridge  in  the  new  genus 
and  a  steeply  inclined  row  of  spines  on  the 
lower  edge  of  the  cheek  border.  However, 
Xanionurus  lacks  the  characteristic  infla- 
tion of  the  posterior  band  of  the  thoracic 
segments  at  the  fulcrum,  seen  in  these  two 
genera,  and  the  pygidia  are  unlike.  It  is 
possible  that  there  is  a  line  of  descent  from 
Diacanthaspis  to  Xanionurus  to  Radiaspis, 
independent  of  the  possible  line  leading 
from  Primaspis  to  Acidaspis  and  Dudleyas- 
pis and  of  the  line  from  Primaspis  to 
Leonaspis  and  Odontopleura.  If  these 
views  are  correct  they  would  suggest  some 
modification  of  the  phylogeny  suggested 
by  Whittington   (19.56c,  te.xt-fig.  3). 

Xanionurus  boucof/  n.  sp. 

Plate  16,  figures  15,  19;  Plate  17,  figures 
13-15,  17-26;  Plate  18,  figures  1-9, 
11-15;  Figure  10. 

Holotype.  USNM  154449,  incomplete 
cephalon. 

Description.  Glabella  moderately  convex 
transversely  and  longitudinally,  widest  at 
occipital  ring,  in  front  of  lateral  lobes  Ip 
tapering  forward  to  rounded  anterior  mar- 
gin. Occipital  ring  about  three  times  as 
wide  as  long  (sag.),  bearing  a  pair  of  curved 
posterodorsally  directed  spines  which  arise 
from  the  rear  edge;  much  shorter  median 
tubercle  directed  almost  vertically,  the  tip 
exhibiting  four  small  pits  arranged  at  the 
corners  of  a  square  ( Pi.  18,  fig.  6 ) .  Occipi- 
tal lobe  poorly  defined  on  inner  surface, 
occipital  furrow  shallow,  well  rounded 
medially,  distally  passing  into  deep  pit-like 
apodeme.  Median  glabellar  lobe  standing 
higher    than    lateral    lobes,    parallel-sided; 


Silurian  Trilobites  •  Whittington  and  Campbell        479 


frontal  lobe  of  similar  width.  Lateral  lobes  cephalon  bearing  spines  of  various  sizes, 
Ip  and  2p  oval  in  outline,  independently  between  which  is  a  fine  granulation.  Three 
convex,  Ip  one  and  one-half  times  longer  main  pairs  of  spines  (2-4;  see  Fig.  10)  are 
than  2p,  lateral  furrow  Ip  diagonalK'  di-  visible  on  the  frontomedian  glabellar  lobe, 
rected  and  deepening  adaxially  into  an  with  pair  2a  situated  close  together  im- 
apodemal  pit;  lateral  furrow  2p  similar  in  mediately  in  front  of  the  occipital  furrow, 
fonn.  Third  lateral  lobes  not  developed.  Spines  A,  .■>  visible  on  inner  part  of  cheek; 
Axial  furrow  moderately  deep.  Cheek  quar-  prominent  spine  B  on  posterior  border 
ter-circle  in  outline,  outward-sloping,  eye  distally;  row  of  6  spines  on  crest  of  antero- 
lobe  situated  at  highest  point  and  in  line  lateral  border,  the  posterior  situated  above 
with  anterior  part  of  lateral  glabellar  lobe  the  base  of  the  third-from-the-last  down- 
Ip;  prominent  eye  ridge  curves  inward  and  wardly  directed  border  spine,  and  the  an- 
forward  to  shallow  axial  furrow  opposite  terior  above  the  outer  part  of  the  rostral 
extremity  of  frontal  glabellar  lobe.  Cheek  plate.  Granulation  is  absent  only  from  the 
inside  eye  inflated,  shallow  furrow  runs  deepest  parts  of  furrows, 
along  inner  edge  of  eye  ridge,  posterior  Number  of  thoracic  segments  unknown, 
sutural  ridge  runs  outward  from  eye  lobe  Convex  axial  ring  with  pair  of  prominent 
to  cur\e  back  and  merge  with  swollen  base  sharp  spines,  articulating  halfring  long 
of  librigenal  spine.  Anterior  branch  of  (sag.),  equal  in  length  to  the  ring;  articu- 
suture  runs  along  upper  surface  of  eye  lating  furro\\'  with  anterior  slope  under- 
ridge  and  over  narrow  anterior  border  in  cutting  articulating  halfring;  apodemes 
line  with  axial  furrow;  posterior  branch  of  blunt  and  short.  Inner  part  of  pleura  ex- 
suture  runs  along  sutural  ridge  to  cross  tends  out  horizontally,  divided  by  shallow 
posterior  border  between  spine  B  and  base  pleural  furrow  into  anterior  band  which  is 
of  librigenal  spine.  Posterior  border  widens  one-third  the  width  (exs.)  of  the  posterior 
rapidly  distally,  curving  for\\  ard  slighth'  to  band  and  lies  below  the  convex  posterior 
base  of  genal  spine.  Gently  convex  antero-  band.  Two  prominent  spines  on  the  poste- 
lateral  border  wider  than  anterior  border,  rior  band  (Pi.  17,  figs.  15,  22,  25),  the 
edge  curled  under  and  directed  upward  inner  situated  at  about  half  the  width  ( tr. ) , 
and  inward  as  the  narrow  doublure.  Outer  the  outer  situated  at  the  fulcrum  and  di- 
margin  of  border  bears  10  closely-spaced,  rected  upward  and  outward.  Anterior 
down\\'ardly  and  outwardly  directed  spines  pleural  spine  generally  stubby,  but  becom- 
which  diminish  progressively  in  length  an-  ing  progressively  longer  (tr. )  toward  the 
teriorly;  marked  gap  between  posterior  rear,  and  directed  out^vard  and  downward 
spine  and  base  of  curved  genal  spine;  throughout;  posterior  pleural  spine  much 
shallow  antennal  notch  immediately  inside  more  prominent  and  longer,  directed  down- 
anterior  spine.  Rostral  plate  not  biown  but  ward  and  back\\'ard — the  medial  segments 
rostral  suture  runs  transversely  along  outer  having  this  spine  directed  more  steeply 
edge  of  anterior  border;  connective  suture  do\^'nward  and  more  directly  outward  than 
runs  inward  and  upward  across  doublure  in  the  more  posterior  segments,  in  which 
(PI.  18,  fig.  2),  hence  rostral  plate  is  short  the  posterior  pleural  spine  is  quite  strongly 
(sag.  and  exs.)  and  wide,  trapezoidal  in  curved  (compare  Pi.  17,  figs.  15,  17,  18,  22, 
outline.  Hypostome  not  known.  Inner  with  PI.  17,  figs.  23-25).  The  first  one  or 
surface  of  cephalon  shows  short,  blunt  two  segments  (Pi.  17,  figs.  13,  14)  differ 
apodemes  formed  by  inner  end  of  deepest  in  that  the  pleura  curves  outward  to  end  in 
part  of  occipital  furrow  and  lateral  glabel-  a  short,  blunt,  posterior  pleural  spine,  the 
lar  furrows  Ip  and  2p.  Anterior  pit  not  anterior  pleural  spine  being  merely  a  small 
developed  on  outer  surface,  nor  as  apo-  projection;  the  segment  is  beveled  to  fit 
deme  on  inner  surface.    External  surface  of  under  the  cephalon  and  as  a  consequence 


480        Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Er- 


PI 


Figure  10.  Xonionurus  boucofi  n.  gen.,  n.  sp.  Major 
paired  spines  of  small  cranidium,  original  of  Plate  18, 
figures    8,    9,    11,    12.     Spines    lettered    as    in    Figure    9. 


the  outer  spine  on  the  posterior  pleural 
band  is  not  developed.  External  surface 
bearing  granules  of  various  sizes  on  axial 
rings,  single  row  of  granules  along  rear 
edge  of  articulating  halfring;  flattened  strip 
behind  posterior  pleural  band  smooth; 
smooth  areas  on  crest  of  posterior  band 
between  the  spines  and  between  the  inner 
spine  and  the  axial  furrow;  upper  surfaces 
of  both  pleural  spines  smooth,  apart  from 
scattered  granules  on  distal  half  of  poste- 
rior spine;  a  band  of  elongate,  sharp  gran- 
ules runs  along  the  leading  and  trailing 
edges  of  both  pleural  spines,  those  on  the 
posterior  spine  being  larger;  tubercles  pres- 
ent along  anterior  band.  The  doublure  is 
present  only  at  the  extremity  of  the  seg- 
ment, and  there  is  a  deep  embayment  for 
articulation  between  the  doublure  and 
upper  surface  on  both  anterior  and  poste- 
rior edges. 

Pygidium  (PI.  17,  figs.  19-21, 26),  exclusive 
of  spines,  of  \\'idth  3^->  to  4  times  length 
(sag.).  Axis  consisting  of  two  segments; 
axial  furrow  shallow  beside  first  ring  but 
very  deep  beside  second  ring  except  at  ex- 
treme tip,  where  it  shallows.  Articulating 
halfring  like  those  of  the  segments;  first 
axial  ring  prominent  with  one  pair  of 
spines.  Second  ring  depressed  below  level 
of  anterior  ring  and  having  a  faint  median 
longitudinal  depression,  the  lateral  parts 
bearing  one  pair  of  small  thorn-like  spines. 
Pleural  region  subdivided  by  triangular 
depression  of  pleural  furrow  into  a  trans- 
verse anterior  band  and  a  diagonally  di- 
rected pleural  ridge;  border  convex  and 
bearing  seven  pairs  of  border  spines,  none 


of  which  is  more  prominent  than  any  other, 
directed  more  steeply  downward  anteriorly 
than  posteriorly,  the  arrangement  radial. 
Fifth  border  spine  from  anterior  is  opposite 
pleural  ridge.  Border  curls  under  on  ven- 
tral side  to  form  a  narro\\'  doublure  in 
which  there  is  a  distinct  median  embay- 
ment. Upper  surface  of  border  bearing 
spines  as  prominent  as  those  on  axial  rings, 
situated  at  base  of  fourth  to  seventh  border 
spines,  that  at  the  base  of  the  fifth  larger 
than  the  others.  Elsewhere  external  surface 
granulate,  except  in  deepest  depressions, 
one  or  two  larger  spines  on  anterior  band 
of  first  segment. 

Development.  Two  small  cranidia  (Pi. 
18,  figs.  8,  9,  11-15;  Fig.  10)  show  the  rela- 
tively large  paired  spines  typical  of  odonto- 
pleurid  developmental  stages.  In  outline, 
glabellar  lobation,  position  of  the  palpebral 
lobe,  as  well  as  spine  pattern,  these 
cranidia  are  like  those  of  the  Ordovician 
Diocanthaspi.s  ( Whittington,  1956c,  pi.  4, 
figs.  12-14).  The  long,  curved  occipital 
spines,  general  shape  and  spine  patteni  are 
also  like  those  of  the  Ordovician  Apianuius 
(Whittington,  1956c,  pi.  19,  figs.  14-16; 
text-fig.  22f ),  but  the  small  spines  are  thick 
and  straight  in  Xanionum.s  (not  slim  and 
curved  as  in  Apianurus)  and  the  posterior 
part  of  the  fixed  cheek  is  wider  ( tr. ) . 

Diseussion.  Prantl  and  Vanek  (in  Homy, 
Prantl,  and  Vanek,  1958:  265-266,  pi.  3, 
fig.  5)  described  an  incomplete  pygidium 
from  the  Upper  Wenlock  of  Czechoslovakia 
which  is  like  that  of  the  present  species  in 
the  number  of  i^airs  of  border  spines,  rela- 
tion of  the  fifth  spine  to  the  pleural  ridge, 
and  appearance  of  the  second  axial  ring. 
They  refer  this  species,  formosa,  with  ques- 
tion to  Radia.spis.  Bruton  (personal  com- 
munication) has  identified  an  entire  speci- 
men of  this  species  (Pi.  18,  fig.  10), 
flattened  in  shale.  It  is  quite  like  the 
present  one,  differing  in  that  the  genal 
spine  is  relatively  much  longer  (reaching 
back  to  the  pygidium),  the  posterior  pleural 
thoracic  spines  are  longer,  particularly  on 
the  anterior  segments  where  they  are  di- 


Silurian  Trilobites  •  Whittiugton  and  Campbell        481 


rected  slightly  forward,  the  pygidial  border 
spines  are  slimmer,  the  external  surface 
appears  to  bear  fewer  and  more  scattered 
spines  and  to  be  granulated.  We  regard 
formosa  as  congeneric  with  our  species  and 
differing  from  the  type  species  of  Radiaspis 
in  the  characters  given  above  in  the  generic 
diagnosis. 

Transitory   pygidium,    undetermined 
Plate  10,  figures  1-3,  5,  9,  10. 

Discussion.  The  two  examples  placed 
here  may  represent  the  same  species,  and 
are  characterized  by  the  axis  tapering 
genth'  backward,  the  ring  furrows  deep 
distall)-,  the  pleural  regions  horizontal  and 
subdivided  by  inteipleural  furrows,  the 
pleural  bands  bearing  prominent  border 
spines  which  are  outwardly  and  back- 
\\'ardly  directed,  more  strongly  backwardly 
directed  posteriorly.  The  tip  is  bent  down 
(PL  10,  fig.  3),  and  the  pleural  border 
spines  alternate  in  size,  this  latter  feature 
particularly  shown  by  the  larger  example 
(PL  10,  figs.  1,  5).  External  surface  of 
rings  and  pleural  bands  finely  granulate,  no 
larger  tubercles  or  spines. 

The  shape  of  this  pygidium,  including 
the  flat  pleural  regions  and  bent-down  tip, 
is  reminiscent  of  that  of  the  cheirurid 
CemurincUa  (Whittiugton  and  Evitt,  1954, 
pi.  12,  figs.  1-3,  9-11;  pi.  28,  figs.  1-14). 
The  alternation  in  size  of  the  pleural  bor- 
der spines  distinguishes  the  present  ex- 
amples, as  does  the  lack  of  a  pleural  furrow. 
No  other  cheirurid  material  has  been  iden- 
tified among  the  present  collection;  the 
lack  of  paired  axial  or  pleural  spines  ap- 
pears to  exclude  this  pygidium  from  odonto- 
pleurids,  and  the  lack  of  a  pleural  furrow 
from  the  other  groups. 

REFERENCES 

Amos,  A.  J-,  K.  S.  W.  Campbell,  and  R.  Gold- 
ring.  1960.  Australosufura  gen.  nov.  (Tri- 
lobita )  from  the  Carboniferous  of  Australia 
and  Argentina.  Palaeontology,  3:  227-236, 
pis.   39-40. 

Best,  R.  V.  1961.  Intraspecific  variation  in 
Encrinurus  orrmtus.  Jour.  Paleont.,  35:  1029- 
1040,  pi.   124. 


BoucoT,  A.  J.      1961.     Stratigraphy  of  the  Moose 

River  synclinorium,   Maine.     Bull.   U.S.   Geol. 

Surv.,  No.  1111-E:   153-188,  pi.  34. 
Campbell,  K.  S.  W.      (In  press).       Trilobites  of 

the  Henryhouse  Fomiation,  Oklahoma.    Bull. 

Okla.  Geol.  Surv. 
Dean,  W.  T.     1960.     The  Silurian  trilobite  Dal- 

manites    mijops    (Konig).     Palaeontology,    2: 

280. 
Delo,  D.  M.      1940.     Phacopid  trilobites  of  North 

America.     Spec.   Pap.   Geol.   Soc.   Amer.,   No. 

29:   1-135,  pis.   1-13. 
Erben,    H.    K.      195.3.     tJber    einige    Otarionidae 

(Tril.)     aus     dem     Mittcl-Devon     der     Eifel. 

Senckenbergiana,    34:    73-80. 
Green,  J-      1832.     A  monograph  of  the  trilobites 

of  North  America.    Philadelphia.    94  p. 
H.\HN,  G.      1964.     Trilobiten  der  unteren  Pcricij- 

clu.s — Stufe    (  Unterkarbon )    aus   dem   Kohlcii- 

kalk  Belgiens.    Part  2.    Senckenbergiana,  45: 

347-379. 
Hall,     J.      1879.     The     fauna     of     the     Niagara 

Group,   in   central   Indiana.     New   York   State 

Mus.,  Ann.  Rept.  28:   98-203,  pis.  3-34   [an 

earlier,  incomplete  edition  in  1875]. 
Harper,  J.  C.      1949.     The  Ordovician  and  Silu- 
rian rocks   of  Ireland.    Proc.   Liverpool   Geol. 

Soc,   20:   48-67. 
Hesslek,  R.  R.      1962a.     The  Lower  Mississippian 

genus   Proetides    (Tril.).    Jour.   Paleont.,   36: 

811-816,  pi.  119. 
.      1962b.      Secondary   segmentation   in   the 

thorax  of  trilobites.    Jour.  Paleont.,  36:   130.5- 

1312,   pi.    176. 
.      1963.     Lower  Mississippian  trilobites  of 

the    familv    Proetidae    in    the    United    States, 

Part    I.     jour.    Paleont.,    37:     543-563,    pis. 

59-62. 

1965.     Lower  Mississippian  trilobites  of 


the  familv  Proetidae  in  the  United  States, 
Part  II.  lour.  Paleont.,  39:  248-264,  pis. 
37-40. 

Horny,  R.,  F.  Prantl,  and  J.  Vanek.  1958. 
On  the  limit  between  the  Wenlock  and  the 
Ludlow  in  the  Barrandian.  Sbornik,  Ust. 
Ust.  Geol.,  24:  217-278,  pis.  1-11  (Czech 
and  English  text,  Russian  summary). 

LiNDSTROM,  G.  1901.  Researches  on  the  visual 
organs  of  the  trilobites.  Kongl.  S\enska 
Vetenskaps-Akad.  Handl.,  34(8):  1-86,  pis. 
1-6. 

Loven,  S.  L.  1845.  Svenska  Trilobiter.  Ofv. 
Kongl.  Vetenskaps-Akad.  Forhandl.,  Stock- 
holm. 2(3):  46-56,  pi.  1. 

M'CoY,  F.  1846.  A  synopsis  of  the  Silurian  fos- 
sils of  Ireland.    Dublin.    72  p.,  5  pis. 

.      1849.     On    the    classification    of    some 

British  fossil  Crustacea,  with  notices  of  new 
fonns  in  the  University  collection  at  Cam- 
bridge.   Ann.    Mag.   Natur.   Hist.,    Ser.   2,   4: 


482 


Bulletin  Museum  of  Coiuparative  Zoology,  Vol.  135,  No.  9 


161-179,    330-335,    392-414,    15    figs,    (not 

numbered). 
MooRK,   R.   C.    (Editor).      1959.     Treatise  on   in- 

\ertebratc  paleoiitolog)-.    Part   O,   Arthropoda 

1.     Geol.    Soc.    America    and    Univ.    Kansas. 

xix  +  560  pp.,  415  figs. 
Prantl,  F.,  and  a.  Pribyl.      1951.     A  revision  of 

die   Bohemian    representatives    of   the   family 

Otarionidac   R.    and   E.   Richtcr    ( Trilobitae ) . 

Stat.     Geol.     Cst.     Ceskoslov.      Repub.,    17: 

353-512.   pis.   1-5    (Czech  and   English   text, 

Russian  summary). 
Pribyl,  A.      1960.     Neuc  Erkenntnisse  iiber  Mit- 

telbohniische      Trilobiten      der      superfamilie 

Proetacea.      Sbornik,     tJst.     Ust.     Geo!.,    25: 

177-221,  pis.  1-4  (Czech  and  Gennan  text). 
Raymond,    P.    E.      1916.     New   and    old    Silurian 

trilobites    from    southeastern    Wisconsin,    with 

notes  on   the  genera  of  the   Illaenidae.     Bull. 

Mus.  Comp.  Zool.,  60:    1-41,  pis.  1^. 
Reed,    F.    R.    C.      1901.     Woodwardian    Museum 

notes:    Salter's  undescribed  species.    2.    Geol. 

Mag.,  N.  S.,  Dec.  4,  8:  5-14,  pi.  1. 
.     1928.     Notes   on   the   family   Encrinuri- 

dae.    Geol.  Mag.,  65:  51-77. 
RiCHTER,     R.      1914.      iJber    das    Hypostom     und 

einige   Arten    der   Gattung    Ctjpha.spis.     Cen- 

tralbl.    Nrineral.,    Geol.,    Palaont.,    Jahr.    1914: 

306-317. 
RiCHTER,       R.       AND       E.        RiCHTER.       1940.      Die 

Sdtikiartda-Siuie       von       Andalusicn.        Abh. 

Scnckcnb.    Naturf.    Gesell.,    450:     1-88,    pis. 

1-5. 
.      1956.     Annular-teilung     bei     Trilobiten 

am     Beispiel    besonders    von    Proetus     ( Pr. ) 

ciwieri  und  jnox.    Senck.  Leth.,  37:  343-381, 

pis.    1-6. 
RosENsTtrN,  E.      1941.     Die  £/irn»K/n/.v-Arten  des 

estlandischen  Silurs.   Geol.    Inst.    Univ.   Tartu, 

No.  62:  49-80,  pis.   1-4. 
Sai.tkh,   J.   W.     1853.     Figures    and    descriptions 

illustrative  of  British  organic  remains.    Mem. 

Geol.  Surv.  U.  K.,  Dec.  7,  pis.  1-10. 
Snajdr,    M.      1960.     Studie   o   celedi   Scutelluidac 

(Trilobitae).    Rozpravy,  Ust.  tJst.  Geol.,  26: 

1-264,  37  pis.,  61  figs.   (Czech  text,  English 

summary ) . 
Temple,  J.  T.      1952.     The  ontogeny  of  the   tri- 

lobite    Dahuanitina    olini.     Geol.    Mag.,    89: 

251-262,  pis.  9,  10. 


.      1954.     The    hypostome     of    Encrimirus 

variolaris    and    its    relation    to    the    cephalon. 

Geol.  Mag.,  91:  315-318. 
.      1956.     Notes    on    the    Cheiruracea    and 

Phacopacea.    Geol.  Mag.,  93:  418-430. 
Tripp,     R.     P.     1957.     The    trilobite    Encriminis 

Diultisegnientatus       (Portlock)       and       allied 

Middle      and      Upper      Ordovician      species. 

Palaeontology,  1:  60-72,  pis.  11,  12. 
.      1962.     The  Silurian  trilobite  Encriminis 

punctatus    (Wahlenberg)    and   allied   species. 

Palaeontology,  5:   460-477,  pis.  65-68. 
Walter,    O.    T.      1927.     Trilobites    of    Iowa    and 

some    related    Paleozoic    forms.     Iowa    Geol. 

Surv.,  31:   17.3-400,  pis.  10-27. 
Weller,     S.      1907.     The     paleontology     of     the 

Niagaran    Limestone    in    the    Chicago    area. 

The  Trilobita.    Chicago  Acad.  Sci.,  Bull.   No. 

4,   (II):    163-281,  pis.  16-25. 
Whittard,   W.   F.      1938.     The   Upper  Valentian 

trilobite    fauna    of    Shropshire.      Ann.     Mag. 

Natur.  Hist.,  Ser.  11,  1:  85-140,  pis.  2-5. 
Whittington,   H.  B.      1954.     Two  silicified  Car- 
boniferous trilobites  from  West  Texas.    Smith- 
son.   Misc.  Coll.,   122:    1-16,  pis.   1-3. 
.     1956a.     Beecher's      supposed      odonto- 

pleurid  protaspis  is  a  phacopid.    Jour.  Paleont., 

.30:   104-109,  pi.  24. 
.      1956b.     Tyi5e     and     other     species     of 

Odontopleuridae    (Trilobita).     four.    Paleont., 

30:   504-520,  pis.   57-60. 
.      1956c.     Silicified      Middle      Ordovician 

trilobites:     the    Odontopleuridae.     Bull.    Mus. 

Comp.   Zool.,   114:    155-288,   pis.    1-24,  figs. 

1-25. 
.      1960.     Coidaiiia  and  other  trilobites  from 

the     Lower     and     Middle     Devonian.      Jour. 

Paleont.,  34:    405-420,  pis.  51-54. 
.      1963.     Middle  Ordovician  trilobites  from 

Lower    Head,    western    Newfoundland.     Bvdl. 

Mus.    Comp.    Zool.,    129:     1-118,   pis.    1-36, 

figs.   1-6. 
.      1965.     Trilobites      of      the      Ordovician 


Table  Head  Fomiation,  western  Newfound- 
land. Bull.  Mus.  Comp.  Zool,  132:  27.5-442, 
pis.  1-68. 
Whittington,  H.  B.,  and  W.  R.  Evitt.  1954. 
Silicified  Middle  Ordovician  trilobites.  Mem. 
Geol.  Soc.  Amer.,  No.  59:  1-137,  pis.  1-33, 
27  figs. 

(Received  6  Ajuil  nJ66) 


Silurian  Trilobites  •  Whittington  and  Campbell 


Explanation  of  Plates  1-19. 
To  make  the  photographs,  the  silicified  specimens  were  mounted  on  insect  pins  with  gum  orabic;  these  pins  have  been 
blacked  out,  otherwise  the  photographs  have  not  been  retouched.  Specimens  preserved  in  limestone  were  coated  with 
a  dilute  "opaque  '  to  give  a  dull  surface.  All  specimens  have  been  given  a  light  coating  of  ammonium  chloride  be- 
fore being  photographed.  It  is  arbitrarily  decided  that  the  plane  running  through  the  posterior  margin  of  the  occipital  or 
axial  ring  shall  run  in  the  dorsoventral  direction,  and  views  are  described  accordingly.  Exterior  views  have  been  taken 
in  the  direction  lying  in  the  sagittal  plane  to  give  the  fullest  possible  view  of  the  exoskeletal  surface.  Oblique  views  have 
been  taken  in  directions  at  an  angle  to  the  sagittal  plane.  The  initials  USNM  refer  to  the  U.S.  Notional  Museum,  Wash- 
ington, D.C.,  and  the  initials  MCZ  to  the  Museum  of  Comparative  Zoology.  These  initials  are  followed  by  the  catalogue 
numbers. 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Figure 
1-3 


12 

13, 

14, 

17,   11 

15, 

16 

19, 

20 

21, 

28 

22 

23, 

24, 

30,  3 

25, 

26 

27, 

29 

PLATE  1 

Proefus  pluteus  n.   sp. 

Hardwood  Mounfain  Formation,  Silurian,  Maine 


Holotype,    cranidium    and    incomplete    left   free  cheek,     USNM    154457,    dorsal,     left    lateral,     anterior    views, 

X  3. 
4^     5  Cranidium,    USNM    154458,    sfiowing    complete  doublure    of    posterior    border    on    right    side,    ventral,    dorsal 

views,    X  4.5. 
6,     7  Cranidium,    USNM   154459,   dorsal,   left   lateral  views,    X  3. 

8  Part   of   cranidium,    USNM    154475,    oblique   view,    to    show    occipital    lobe    and    smooth    muscle    areas    (com- 

pare   Fig.    lA),    X  15. 
9,    10  Cranidium,    USNM    154460,    dorsal,    left  lateral   views,    X    3. 

1  Free  cheek,    USNM    154461,    exterior   view,    X  4,5. 

Free  cheek,    USNM    154463,    exterior   view,    X  9. 

Hypostome,    USNM    154470,    anterior,    posterior,    exterior,    left    lateral    views,    X  4.5. 

Cranidium,    USNM    154464,    dorsal,     left    lateral   views,    X  4.5. 

Cranidium,    USNM    154465,    dorsal,     left    lateral    views,    X  6. 

Cranidium,     USNM    154468,    dorsal,     right    lateral    views,    X  9. 

Free   cheek,    USNM    154462,    exterior   view,    X  6. 

Cranidium,    USNM    154469,    anterior,    ventral,   right    lateral,    dorsal    views,    X  15. 

Cranidium,    USNM     154466,    dorsal,    right    lateral    views,    X  6. 

Cranidium,    USNM    154467,     dorsal,     right    lateral    views,    X  6. 


Silurian  Trilobites  •  Whittingion  and  Campbell 


PLATE 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  2 

Proetus  pluteus  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 
1,     4-6,    16       Hypostome,    USNM    154471,    two    oblique    exterior  views,    X  9;   oblique    interior,   right   lateral,    exterior  views, 

X  4.5. 
2  Cranidium    (USNM    154478)    and    left    free    cheek  (USNM  154476)  placed  close  together,  ventral  view,  to  show 

doublure  and  course  of  connective  suture  (compare  Fig.  IB),  X  4.5.  Portions  of  the  muscle  areas  appear 
dork    where    the    innermost    silicified    layer    is  broken. 

Inner  edge  of  doublure  of  free  cheek,  USNM  154476,  showing  that  this  edge  is  rounded  and  not  a  flattened 
hypostomal     sutural     margin.      Oblique    interior  view,    X    9. 

Anterior  segment,  USNM  154479,  dorsal,  right  lateral  views,  X  5;  oblique  interior  view  showing  narrow 
doublure    at    tip,     X  9. 

Small    pygidium,    USNM    154484,    left    lateral,    dorsal,  ventral   views,    X  6. 

Segment,  USNM  154480:  figs.  10,  14,  31,  32,  left  lateral,  dorsal,  posterior,  anterior  views,  X  5;  figs.  23, 
25,  oblique  interior  views,  showing  slot  along  anterior  edge  of  articulating  halfring  and  panderian  opening, 
X  9. 

Hypostome,    USNM    154473,    exterior,    interior,  right    lateral    views,    X  6. 
Hypostome,    USNM    154472,    exterior,    right    lateral    views,    X  6. 
Posterior   segment,    USNM    154481,   dorsal   view,  X  5. 

Smallest    hypostome,    USNM    154474,    exterior,    left   lateral    views,    X  15. 
Pygidium,    USNM    154483:     oblique   exterior  view,    X  9;  ventral  view,    X  4.5. 
Pygidium,   USNM   154482,   dorsal,    posterior,    right   lateral    views,    X  4.5. 


3 

7, 

8, 

24 

9, 

17, 

20, 

10, 

14, 

23 

25, 

31, 

32 

11, 

18, 

19 

12, 

13 

15 

21, 

22 

26, 

30 

27- 

29 

Silurian  Trilobites  •  Whittington  and  Campbell 


PLATE  2 


Bulletin  Mmeum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  3 

Proefus  cuvieri  Steininger,   1831 

Eifel  District,  Middle  Devonian,  Western  Germany 

Figure 

1,     3  Incomplete    cepholon    with    thoracic    segments    articulated,   anteroventral,  dorsal   views,    X  4.5.     Schultze  collec- 

tion, MCZ  5924/2. 

2  Incomplete   cephalon,   anterior   view,    X  6.     Anterior   border   is  broken  off  to   reveal   on    left  side  of  photograph 

mould  of  external  surface  of  doublure,  and  in  center  the  exoskeleton  adjacent  to  the  connective  sutures. 
Schultze  collection,  MCZ  5924/1. 

Proetus  conc/'nnus  (Dolman,  1827) 

4,     5,     9,      Holotype,    incomplete    partially    enrolled    individual,    Mulde    Beds,    Wenlock    Series,     Djupvik,     parish    of    Eksta, 
11,    12  Gotland,    Sweden.     Pal.    Instit.    Uppsala    G    733.     Fig.   4,    oblique   view   of   cephalon    photographed    under   alco- 

hol, to  show  muscle  areas  as  dark  patches  on  glabella,  X  6.  Figs.  5,  11,  12,  dorsal  view  of  cephalon, 
dorsal  view  of  pygidium,  left  lateral  view,  X  4.5.  Fig.  9,  anterior  view  of  cephalon  showing  mould  of 
doublure    on    right    side    and    position    of    connective   suture    on    left   side,    X  6. 

Proefus  plufeus  n.  sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

6,  10,   13      Transitory    pygidium,    USNM    154485,    left    lateral,  ventral,    dorsal    views,    X  15. 

7,  8  Transitory    pygidium,    USNM    154486,    dorsal,    right   lateral   views,    X  15. 
14,   15  Transitory    pygidium,    USNM    154487,    dorsal,    ventral    views,    X  15. 

16  Transitory   pygidium,    USNM   154488,   dorsal    view,  X  30. 


SiLUBiAN  Trilobites  •  Whittingtoti  and  Campbell 


PLATE  3 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  4 

Rhinotarion  sentosum  n.  gen.,  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 

1-7         Holotype,    USNM   154211:     figs.    1 ,    3,   4,   5,    dorsal,   anterior,    posterior,    left    lateral    views,    X  6;    figs.    2,    6,    7,    two 
oblique    exterior   and    a    ventral    view,    X  9. 

8,     9       Free    cfieek,    USNM    154213,    exterior,    interior   views,     X  6. 
10-12       Isolated    rostral     plate,    USNM    154212,    interior,    exterior,    oblique    interior   views,    X  15. 


Silurian  Trilobites  •  Whittington  and  Campbell 


PLATE  4 


Bulletin  Mitficiim  of  Comparative  Zoologij,  Vol.  135,  No.  9 


PLATE  5 

Rhinotarion  sentosum  n.  gen.,  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 

1,  2,     4  Cranidium,    USNM   154214,  dorsal,    oblique   exterior,    ventral    views,    X  9. 

3  Free  cheek,    USNM    154215,  exterior  view,    X  6. 

5,  6  Malformed    cranidium,    USNM    154217,    dorsal,  right    lateral    views,    X  6. 

9,  10,   13,    14     Two  segments  articulated  together,  USNM  154218,  dorsal,   left  lateral,  oblique  ventral,   anterior  views,    X  9. 

11  Small    free   cheek,   USNM    154216,    exterior  view,    X   15. 

15,  16,    18  Segment  with   median   axial   spine,    USNM   154219,    dorsal,    right   lateral,    oblique    ventral    views,    X  9. 

Otarion  sp.  ind. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

7,     8,    12,    17     Cranidium  with  right  free  cheek  attached,  USNM   154240,   right  lateral,   anterior,  dorsal,   oblique  views,    X  6. 

19,  25,  26  Cranidium    with    left    free    cheek    and    first    thoracic   segment  attached,    USNM   154241;     fig.    19,   oblique   view 

showing    outer   port   of    segment,    X  15;    figs.    25,    26,    dorsal,    ventral   views,    X  9. 

20,  21  Cranidium,   USNM    154242,   dorsal,    left   lateral    views,    X  6. 

22,  23  Small    cranidium,    USNM    154243,    dorsal,    right   lateral    views,    X  15. 

Otarion  imtila  n.  sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

24  Interior   view    of   eye    surface    (original    of    PI.    6,    figs.    7,    10)    showing    minute    facets,    X  30. 


Silurian  Trilobites  •  Whittington  and  Campbell 


PLATE  5 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Fig 

ure 

1- 

•5 

6 

7, 

10 

8 

9, 

15, 

11, 

12 

13, 

14 

17 

18, 

19, 

20, 

21 

22, 

23 

PLATE  6 

Otarion  instiia  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Holotype,    cranidium    with    left    free    cheek    attached,   USNM   154220:     figs.   1,   2,   3,  5,   anterior,   oblique  ventral, 

dorsal,   left   lateral    views,    X  6;    fig.   4,    oblique   exterior  view,    X  8. 

Cranidium,  USNM  154221,  dorsal  view,   X  6. 

Free  cheek,   USNM    154228,   interior,    exterior  views,    X  ^-5. 

Cranidium,    USNM    154222,    dorsal    view,    X  6. 
5,    16     Cranidium  with   rostral   plate  attached,   USNM   154224:     fig.    9,    oblique    ventral    view,    X9;    figs.    15,    16,    dorsal, 

right   lateral    views,    X  6. 

Free    cheek,    USNM    154229,    interior,    exterior   views,   X  15. 

Cranidium    with    left   free    cheek    attached,    USNM    154223,   dorsal,    left   lateral   views,    X  6. 

Small    free   cheek,    USNM    154230,    exterior   view,    X  30. 
9,   24     Smallest  cranidium,    USNM    154227,    right    lateral,  dorsal,    anterior   views,    X  30. 

Cranidium,    USNM    154225,    dorsal,    right    lateral  views,    X  9. 

Cranidium,    USNM    154226,    left   lateral,   dorsal   views,   X     15. 


Silurian  Trilobites  •   Whittingtun  and  Campbell 


PLATE  6 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  7 

Otarion  plautum  n.  sp. 

Hardwood  Mountain   Formation,  Silurian,  Maine 

Figure 

1,     2,     4,     6     Holotype    cephalon    lacking    hypostome,     USNM   154231:     figs.    1,    2,    4,    dorsal,    left    lateral,    anterior   views, 
X  9;    fig.    6,   oblique   ventral    view,    X  10. 

3,     5  Cranidium,     USNM    154232,     left     lateral,    dorsal  views,    X  9. 

7  Cranidium    with    incomplete    right    free    cheek    and   rostral    plate,    USNM    154233,    ventral    view,    X  9. 

8,   11,    12  Cranidium,    USNM    154234:      fig.    8,    oblique  exterior   view    showing    smooth    muscle    areas,    X  15;    figs.    11, 

12,    dorsal,    left    lateral    views,    X  9. 

9  Free   cheek,   USNM   154238,    interior  view,    X  9. 

13,   14  Cranidium,    USNM    154235,   dorsal,    left   lateral  views,    X  15. 

15  Small    free  cheek,    USNM    154239,    exterior   view,    X  15. 

17,    18,   23  Cranidium,    USNM    154236,    dorsal,    left    lateral,    ventral    views,    X  15. 

19,   24,   25  Cranidium,     USNM     154237,     left    lateral,    dorsal,    anterior    views,    X  30. 

Otarionid  hypostome 
Hardwood  Mountain  Formation,  Silurian,  Maine 

10,  16  Left    lateral,    exterior    views,     X  15.      USNM  154244. 

20  Smallest    example,    exterior    view,     X  30.      USNM    154247. 

21,  22  Left    lateral,    exterior    views,    X  15.      USNM  154245. 

26,  27  Left    lateral,    exterior    views,     X  20.      USNM  154246. 


Silurian  Trilobites  •  Whitthigton  and  Campbell 


PLATE  7 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  8 

Otarionid  Thoracic  Segments  and  Pygidium,  Type  A 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 

1,  4,  10             Large    pygidium,    USNM    154248,    dorsal,    postsrior,    ventral    views,    X  15. 

2  Small    pygidium,    USNM    154249,    dorsal    view,    X  15. 

3^  5,     6,     9     Three  articulated  segments  and  pygidium,  USNM  154250,    posterior,  ventral,   dorsal,   left   lateral   views,    X  15. 

13,  15,  21              Segment    with    median    axial    spine,    USNM    154253,    dorsal,    ventral,    left   lateral    views,    X  9. 
14                             Two    segments,    USNM    154254,    dorsal    view,     X  4.5. 

19,  22,  23,   26     Four   segments    and    pygidium,    USNM    154251,  right    lateral,    posterior,    two    dorsal    views,    X  15. 

24,  25                    Five  segments   and   pygidium,   USNM   154252,    two   dorsal   views,    X  9. 

Otarionid  Pygidium  Type  C 
Hardwood  Mountain  Formation,  Silurian,  Maine 

7,     8,   11,   12     Dorsal,     right     lateral,    ventral,     posterior    views,    X  15.     USNM   154255. 

Otarionid  Pygidium  Type  D 
Hardwood  Mountain  Formation,  Silurian,  Maine 

16,    17,    18,   20     Posterior,    right    lateral,    dorsal,    ventral    views,  X  15.     USNM    154256. 


Silurian  Trilobites  •  Whittington  and  Campbell 


PLATE  8 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  9 

Otarionid  Thoracic  Segments  and  Pygidium  Type  B 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 

1,     2  Six    thoracic   segments,    USNM    154257,    left    lateral,   dorsal   views,    X  15. 

3,     8  Six   thoracic    segments,    USNM    154258,   dorsal,  left   lateral   views,    X  15. 

4-7  Pygidium,    USNM    154259,    dorsal,    right    lateral,    posterior  views,    X  30;    ventral    view,    X  15. 

9,    10,  12,    13     Pygidium,    USNM    154260,    dorsal,    ventral,    posterior,    right    lateral    views,    X  15. 
11,    14  Pygidium,    USNM    154261,    dorsal,    posterior   views,    X  30. 

Otarionid  Transitory   Pygidium 
Hardwood  Mountain  Formation,  Silurian,  Maine 

15,  16  USNM  154263,   right  lateral,   dorsal   views,    X  30. 

17,  21,  26  Specimen    with    ma|or   spine    on    first    axial    ring,    USNM    154265,    dorsal,    ventral,    posterior   views,    X  30. 

18,  22  Specimen   without   major   spine   on   axial    rings,  USNM    154266,   dorsal,    posterior  views,    X  30. 

19,  20  Example   with    major    spine   on    third    axial    ring,   USNM    154262,   dorsal,    right  lateral    views,    X  30. 

23-25  Specimen    with    major    spine    on    first    and    third    axial    rings,    USNM    154264,    dorsal,     right    lateral,    ventral 

views,     X  30. 

Otarionid  Transitory  Pygidium 
Hardwood  Mountain  Formation,  Silurian,  Maine 

27  USNM  154267,   dorsal   view,    X  30. 


Silurian  Trilobites  •   Whittinfiton  and  Campbell 


PLATE  9 


Bulletin  Museiinj  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  10 

Unidentified  Transitory  Pygidium 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 

1-3,     5  USNM    154268,    dorsal,    ventral,    left   lateral,    posterior  views,    X  30. 

9,        10  USNM    154269,    dorsal,    posterior   views,    X  30. 

Scutelluid   gen.    ind. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

4,     6-8  Incomplete    cranidium,    USNM    154270,    right    lateral,    ventral,    dorsal,    anterior    views,    X  6. 

13  Incomplete    cranidium,    USNM    154271,    dorsal    view,    X  15. 

Frogiscufum   rhytium   n.  gen.,   n.   sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

n  Holotype,   USNM   154272,   dorsal  view  of  posterior  margin  of  cephalon  and  tfie  first  five  tfioracic  segments;  on 

inner   part  of   pleurae    upper   silicified    layer    is    broken    away,    showing    cones    projecting    up    from    midline    of 
inner  surface    (cf.    Fig.    8b),    X  9. 

O/or/on  diffractum  Zenker,   1833 
Kopanina  Beds,  Ludlow,  Upper  Silurian,  St.   Ivan, 
Czechoslovakia.     Krantz  Collection,  MCZ  8542a. 

12,        14,    15,      Enrolled    incomplete    exoskeleton,    figs.    12,    14,    16,     dorsal,     right     lateral,     anterior    views,     X  4.5;     fig.     15, 
16  ventral    view    showing    connective    sutures    and    rostral    suture,    X  6. 


Silurian  Trilobites  •   \Vhittmfi,fou  and  Campbell 


PLATE  10 


Bulletin  Museum  of  Comparative  Zoo/o^'iy,  Vol.  135,  No.  9 


PLATE  11 
Fragiscutum  rhytium  n.  gen.,  n.  sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 
Figure 
1-3  Holotype,    USNM    154272,    oblique    exterior,    dorsal,    ventral    views,     X  3. 

4,     5  Hypostome,   USNM   154282,    posterior,   exterior  views,    X  4.5. 

6,     7,    12     Hypostome,    USNM    154276;    figs.    6,    12,    anterior,    interior  views,    X  4.5;    fig.    7,    oblique    exterior  view   of    right 

wings,    X  15. 
8-10  Hypostome,    USNM    154285,    exterior,    posterior,    righl    lateral    views,    X  6. 

11,    15  Hypostome,    USNM    154283,    exterior,     right    lateral   views,    X  15. 

13  Cranidium    (USNM    154273)    and    hypostome    (USNM   154276)    in    juxtaposition,   oblique  view,    X  6. 

14,    17  Smallest    hypostome,    USNM    154284,    exterior,    left  lateral    views,    X  15. 

16  Free    cheek,    USNM    154288,    exterior    view,    X  6. 

18,    19  Free    cheek,    USNM    154287,    exterior,    interior    views,    X  4.5. 

20  Part  of  cephalon  of  holotype  (USNM  154272),  dorsal  view  to   show   granulation   on   and   between   tubercles,    X  9. 

21  Hypostome,    USNM    154286,   oblique  view  to   show  granulation   and    faintly-defined   macula,    X  9. 


SiLUHiAx  Trilobites  •   \V7u7///i^'fo;i  and  Campbell 


PLATE  11 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  12 

Fragiscutum  rhytium  n.  gen.,  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Cranidium,    USNM    154275,    dorsal     (cf.     Fig.  7b),   ventral    views,    X  4.5. 

Cranidium     (USNM    154273)    and    free    cheek  (USNM     154274)     in     juxtaposition,    oblique    interior,     anterior, 

dorsal    views,     X  4.5. 

Cranidium,    USNM    154277,    dorsal    view,     X  4.5. 

Cranidium,   USNM    154278,    dorsal   view,    X  6. 

Cranidium,    USNM    154279,    right    lateral,    dorsal,   ventral    views,    X  15. 

Small    free    cheek,   USNM    154289,   exterior  view,    X  15. 

Cranidium,    USNM     154281,    anterior,    dorsal  views,    X  15. 
13,    14,    17,   20     Thoracic    segment,    USNM    154294,    anterior,    cJorsal,    posterior,    ventral    views,    X  4.5. 
15,    16,    19  Cranidium,    USNM    154280,    dorsal     (cf.    Fig.  7a],    left   lateral,    anterior   views,    X  15. 

21,   22  Small    segment,    USNM    154295,    showing    median    axial    spine    and    tubercles    on     pleural     bands,    posterior, 

dorsal    views,    X  15. 
23  Pygidium,    USNM    154296,    right    lateral    view,    X  4.5. 


Figi 

ure 

1, 

2 

3, 

5, 

4 

6 

7, 

9, 

11 

12, 

18 

Silurian  Trilobites  •   Whittington  and  Campbell 


PLATE  12 


BuUetin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  13 

Fragiscutum  rhytium  n.  gen.,  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 

1-3  Transitory    pygidium,    USNM    154298,    dorsal,    ventral,    right    lateral    views,    X  15. 

4,     9,  13       Transitory    pygidium,    USNM    154301,    dorsal,    posterior,    ventral    views,     X  30. 

5-7  Transitory    pygidium,    USNM    154299,    dorsal,    posterior,    right    lateral    views,     X  15. 

8,   12  Transitory    pygidium,    USNM    154300,    dorsal,    oblique   views,    X  30. 

10,    11,  14,      Pygidium,    USNM    154296,    posterior,    dorsal,    ventral    views,     X  4.5;    oblique    view    showing     pleural     ribs    and 

15  outer    port    of   first    pleural    furrow    on    facet,    X   15. 

16  Pygidium,    USNM    154297,   dorsal    view,    X  9. 

17  Three    thoracic    segments,    USNM    154293,    dorsal   view   showing   through    broken  outer  silicified   layer  cones    pro- 
jecting   upward    from    the    lower    layer,    X  15. 


Silurian  Trilobites  •  Whittington  and  Campbell 


PLATE  13 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


Fig 

ure 

1, 

4 

2, 

3 

5 

6, 

7 

8, 

14 

9- 

•11, 

12 

15, 

16 

PLATE  14 

Dalmanites  puticulifronz  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Holotype    USNM    154302,    incomplete    cronidium    with    part    of    left    free    cheek    and    anterior    border    attached, 
dorsal,    ventral    views,    X  2. 

Incomplete    cronidium,    USNM    154304,    dorsal,    right    lateral   views,    X   1- 
Latex  cast  of  external   mould  of  cronidium,   USNM  154303,    dorsal    view,     X  1- 
Cronidium,     USNM     154305,     dorsal,     anterior     views,    X  1- 

Incomplete    pygidium,    USNM    154432,    showing    apodemes,   ventral    view,    X  1;    oblique  view,    X  3. 
13     Segment,    USNM   154430,    ventral,    anterior,   dorsal,   right    lateral    views,    X   1-7. 
Pygidium,    USNM    154431,    dorsal     view    showing  granulation,    X  3. 
Incomplete    pygidium,    USNM    154433,    showing    terminal    spine,    dorsal,    ventral    views,    X  4.5. 


Silurian  Trilobites  •   Wliittington  and  Campbell 


PLATE  14 


Bulletin  Museum  of  Comparatwc  Zoology,  Vol.  135,  No.  9 


PLATE  15 

Dalmanifes  puf/cu//7rons  n.  sp. 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 
1-3  Incomplete    cranidium,    USNM    154306,    dorsal,    anterior,    left    lateral    views,    X  6. 

4  Incomplete    cranidium,    USNM    154307,    dorsal   view,    X  9. 

5,     6,  10,    14,    17      Incomplete  protaspis,  USNM  154310,  right  lateral,  oblique  exterior,  anterior,  dorsal,  posterior  views,  X  50. 

7  Incomplete    cranidium,    USNM    154308,    dorsal   view,    X  30. 

8  Transitory    pygidium,    USNM    154435,    dorsal    view,    X  15. 

9,    12,    13,    16  Transitory    pygidium,    USNM    154434,    oblique    exterior,    left    lateral,    ventral,    dorsal    views,    X  15. 

11,   15  Incomplete   cranidium,    USNM    154309,    dorsal,   right  lateral   views,    X  25. 


Silurian  Trilobites  •  Whittington  and  Campbell 


PLATE  15 


Bulletin  Museum  of  Comparative  Zoolofnj,  Vol.  135,  No.  9 


PLATE  16 

Leonasph  cf.  williamsi  Whittington,  1956 

Hardwood  Mountain  Formation,  Silurian,  Maine 

Figure 
1-4,     8,   20     Cranidium,    USNM    154437,    dorsal,    anterior,    right    lateral,    posterior,    oblique    exterior    views,     X  4.5;    ante- 
rior view   of    part   of   cranidium    to    sfiow    external    surface,    X15. 
Cranidium,   USNM   154436,   dorsal,   anterior   views,    X  4.5. 
Small    cranidium,    USNM    154438,    left    lateral,   dorsal,    anterior   views,    X  15. 
Free    cfieek,    USNM    154440,    dorsal,    exterior,  interior    views,    X  4.5. 
Small    cranidium,    USNM    154439,    dorsal,    anterior,   left  lateral  views,    X  15. 

fHypostome,    USNM    154441,    silicified    layer    replacing    external    surface    broken    away    except    anteriorly,    ante- 
rior,   exterior,    interior    views,    X  6. 

Xanionurus  faoucof/'  n.  gen.,  n.  sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

15,       19  Cepfialon    lacking    fiypostome,    USNM    154450,  dorsal   view,    X  7.5;   oblique   exterior  view,    X  10. 


5, 

6 

7, 

9, 

10 

11, 

21, 

22 

12- 

-14 

16- 

-18 

Silurian  Tkilobites  •   W}iittmp,ton  and  Campbell 


PLATE  16 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  17 

Leonaspis  cf.  williamii  Whiftington,  1956 

Hardwood  Mountain  Fornnation,  Silurian,  Maine 

Anterior   thoracic   segment,    USNM    154442,    anterior,    dorsal    views,    X  10- 

Anterior    thoracic    segment,    USNM    154443,    ventral    view,    X  4.5. 

Pygidium,     USNM    154447,     right    lateral,    ventral   views,    X  6;    dorsal   view,    X  15. 

Segment    from    posterior    part    of    thorax,    USNM    154445,    oblique    Interior    view,     X  9;    left    lateral,    dorsal 

views,     X  4.5. 

Incomplete    pygidium,    USNM    154448,    dorsal  view,   X  6. 

Medial    thoracic    segment,    USNM    154444,    dorsal,    anterior    views,    X  10. 

Posterior    thoracic    segment,     USNM     154446,   dorsal   view,    X   10. 

Xanionurus  boucoti  n.  gen.,  n.  sp. 
Hardwood  Mountain   Formation,  Silurian,  Maine 

13,   14  Anterior  thoracic  segment,    USNM    154453,    right    lateral,    dorsal    views,    X   15. 

15,   17,    18,  22     Medial    thoracic   segment,    USNM    154454,    posterior,    ventral,    right    lateral,    anterior   views,    X  6. 
19-21,     26  Pygidium,    USNM    154456,    right    lateral,    ventral,    posterior    views,     X  4.5;    dorsal    view,     X  9. 

23-25  Posterior    thoracic    segment,     USNM     154455,  dorsal,    left   lateral,    anterior  views,    X  9. 


Figure 

1,     4 

2 

3,     7, 

16 

5,     9, 

11 

6 

8,   10 

12 

Silurian  Trilobites  •  Whittington  and  Campbell 


21 


fm^^^^ 


,I„._.ljg^ 

l^^f 

m 

(II 

PLATE  17 


Bulletin  Museum  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  18 
Xanionurus  boucoti  n.  gen.,  n,  sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 
Figure 

1-6  Holotype   cephalon    lacking   rostral    plate   and  hypostotne,    USNM   154449,  dorsal,    posterior,  ventral,   anterior, 

left    lateral    views,     X  10;    occipital    ring,    paired  spines  and  median  spine,  latter  with   pits  at  the  tip,    X  30. 
7  Cephalon  lacking  rostral  plate  and  hypostome  (see  PI.   16,  figs.  15,  19),  USNM  154450,    anterior  view,    X  7.5. 

8,     9,    11,    12     Small   cranidium,    USNM    154451,    dorsal,    left    loferal,   anterior,   oblique   exterior  views,    X  15. 
13-15  Smallest    cranidium,     USNM     154452,    dorsal,   right    lateral,    anterior  views,    X  15. 

Xonionurus  formosus  (PrantI  and  Vanek,  1958) 

10  Incomplete    extended    exoskeleton,    MCZ   4166,  dorsal    view,     X  9.      Lodenice,    Liten    Beds    (Upper    Wenlock), 

Czechoslovakia. 


Silurian  Trilobites  •  Whiftinfi,ton  and  CcwrphcU 


PLATE  18 


Bulletin  Miiscutn  of  Comparative  Zoology,  Vol.  135,  No.  9 


PLATE  19 
Otarion  megalops   (M'Coy,   1846) 

Beds   of    presumed   Upper   Llandovery   age,    in    the    neighbourhood   of  Cong,   County  Galway,    Eire.     Griffith   Collection,    Na- 
tional   Museum    of    Ireland. 
Figure 
1-4  Holotype,    internal    mould    of   incomplete   cranidium,    original    of   M'Coy,    1846,    pi.    4,    fig.    5,    from    Boocoun, 

near   Cong.     Dorsal    view,    X   10;    left    lateral  view,    X  6;  oblique  exterior  view,    X  9;   anterior  view,X  6. 
5_     8,     9  Topotype,    internal    mould    of    incomplete    cranidium,    dorsal,    anterior,    left    lateral    views,    X  6. 

6_     7_   10,  11      Topotype,  internal   mould  of  incomplete  cranidium,  left  lateral,  oblique  exterior,  dorsal,  anterior  views,    X  6. 
12-14,     16  Internal  mould  of  cranidium,  anterior,  dorsal,    left  lateral,  oblique  exterior  views,  X  6.    Tonlegee,  near  Cong. 

Dalmanites  puticulifrons  n.  sp. 
Hardwood  Mountain  Formation,  Silurian,  Maine 

15  Interior    of    counterpart    silicified    exoskeleton  of  original    of    Plate    14,    figure   5,    interior   view   showing    apo- 

demes  of  occipital   ring,   Ip  and  2p,  facets  of  eye  surface,    X  3. 
17  Exterior  view  of   part   of  original    of   Plate   14,  figure  6,    showing    granulation    on    external    surface,    X  4.5. 


Silurian  Trilobites  •  Wliitfington  and  Campbell 


PLATE  19 


i 


Date  Due 


Harvard  MCZ  Librar 


3   2044  066  303  967