Skip to main content

Full text of "A memoir on the Palaeozoic reticulate sponges constituting the family Dictyospongidae"

See other formats 


to/ 


!r 


University  of  the  State  of  New  York 


MEMOIR 


ON    THE 


Palaeozoic  Reticulate  Sponges 

CONSTITUTING   THE    FAMILY 

DICTYOSPONGIDAE 

BY 

JAMES   HALL 

STATE  GEOLOGIST  AND  PALAEONTOLOGIST 

IN   COLLABORATION   WITH 

JOHN   M.  CLARKE 

ASSISTANT  STATE  GEOLOGIST  AND  PALAEONTOLOGIST 


WYNKOOP  HALLENBECK  CRAWFORD  CO. 

STATE    PRINTERS 

NEW  YORK  AND  ALBANY 


PREFACE. 


Tliis  Monograph  of  the  Dictyospongid^  is  the  conclusion  of  an  under- 
taking, the  inception  of  which  is  indicated  by  my  pul>lication  upon  these 
fossils  in  1884.  Up  to  the  year  1863,  only  occasional  expressions  had  been  made 
Avath  reference  to  the  nature  of  such  bodies ;  first,  by  Mr.  T.  A.  Conrad,  after  the 
(!lose  of  his  connection  Avith  the  Geological  Survey  of  the  State  and,  again,  in 
the  same  year  (1842),  by  Mr.  Lardneb  Vanuxem,  in  his  final  report  on  this 
Survey.  The  specimens  which  came  under  the  notice  of  these  observers  were  as 
Avidely  unlike  as  the  limits  of  the  family  permit,  and  while  Conrad  considered 
liis  specimen,  Hydnoceras,  a  cephalopod,  Vanuxem  regarded  the  form 
described  by  him,  Upiiant^nia,  as  a  marine  plant.  In  my  paper  of  1 863,  I  was 
disposed  to  accept  Vanuxem's  views  of  these  bodies  as  nearer  the  truth. 

In  the  years  which  intervened,  from  1863  to  the  date  of  publication  of 
my  second  paper,  in  1884,  specimens  had  begun  to  accumulate  in  the  collec- 
tions of  the  State  Museum,  and  toward  the  latter  part  of  this  period  especial 
(efforts  were  made  to  gather  material  that  Avould  justify  a  more  thorough  study 
of  the  subject,  though  little  anticipating  that  it  would  ever  assume  the 
comprehensive  proportions  with  which  it  appears  to-day.  While  new  material 
Avas  gradually  being  acquired  from  the  Chemung  rocks  of  this  State,  as  one 
result  of  the  activity  of  collectoi's  sent  out  on  behalf  of  the  Palaeontology 
of  New  York,  additional  forms  were  obtained  from  the  WaAcrly  sandstone 
of  Ohio,  during  the  survey  of  that  state  by  the  late  Dr.  J.  S.  Newherry  ; 
and  admirable  specimens  were  being  found  in  the  calcareous  shales  of  the 
Keokuk  group  at  Crawfordsville,  Indiana.  It  was  the  acquisition  and  study 
of  this  last  named  material,  that  led  to  the  determination  by  Mr.  R.  P. 
Whitfield,  in  1881,  of  the  spongous  nature  of  these  fossils. 

In  order  to  show  the  progress  which  had  been  made  in  the  study  of  these 
organisms,  I  issued  in  1882,  four  large  octavo  plates  of  figures  with  explana- 
tions and,  in  1884,  reproduced  these  plates  with  a  brief  descriptive  account 
of  the  fossils  illustrated.  In  1890,  an  addition  to  the  known  species  Avas  made 
by  the  publication  (though  without   illustrations)  of  a  group  of  Chemung 


4  DHrrYc)si>()N(Ui»:. 

81MH.-ie8,  a  paper  prepared  at  ray  request  by  Mr.  John  M.  Clabkk.  AVith  this 
pul.lication,  forty-two  species  of  these  sponges  had  been  described  from  the 
pahieozoic  rocks  of  the  United  States. 

So  rapid  lias  been  the  growth  of  these  collections  and  so  generous  the 
disiMisition  of  all  collectors  and  students  toward  the  perfecting  of  this 
monograph,  that  to-day  we  present  descriptions  of  about  thrice  the 
numl)er  of  specific  fonns  recognized  as  such,  seven  years  ago.  This 
wonderful  increase  is  not  especially  a  matter  for  congratulation,  save  that  it 
serves  to  show  the  amazing  diversity  of  these  siliceous  sponges  in  late 
Devonian  and  early  Carboniferous  faunas.  The  species  and  genera  are 
established  in  accordance  with  the  best  standards  attainable  for  these 
variable  organisms.  Future  collections  and  observations  may  show  that  these 
divisions  are  not  in  all  cases  sharply  defined  ;  they  will  undoubtedly  show 
that  the  present  account  of  this  organic  element  in  these  ancient  rocks  is 
far  from  being  complete. 

In  1892,  Mr.  Clarke  undertook  a  revision  of  all  previous  work  upon 
these  fossils  anil  this,  with  the  elaboration  of  new  material,  he  has  carried 
forward  to  its  present  state. 

During  the  preparation  of  the  work,  many  favors  have  been  shown  by 
geologists  and  others  having  specimens  of  Dictyospongidje  in  their  collections. 
Acknowledgments  are  due  to  them  all,  but  our  obligation  to  some  recpiires 
particular  mention. 

Tliis  monograph  could  never  have  been  made  what  it  now  is,  were  it  not 
for  the  enthusiastic  cooperation  of  Mr.  Edwin  B.  Hall,  of  Wellsville,  N.  Y. 
Mr.  Hall  has  demonstrated  a  wonderful  abundance  of  the  Dictto 
SP0NC4ID^E  in  the  upper  Chemung  rocks  of  New  York.  His  collections,  which 
are  constantly  growing  under  his  zealous  explorations,  are  remarkably  rich  in 
this  material  and  have  been  at  all  times  accessible  for  use  in  this  work. 
It  is  to  be  remarked,  with  some  regret  in  vie^v  of  the  issuance  of  this 
book,  that  Mr.  Hall's  more  recent  collections  afford  evidence  that  as  yet 
we  are  far  fi-om  having  a  full  representation  of  the  hexactinellid  sponge  fauna  of 
this  geologic  period,  not\vithstanding  that  more  than  sixty  species  of 
this  ancient  date  are  here  illustrated.  With  our  present  knowledge, 
the  middle  and  upper  beds  of  the  Chemung  group  in  Allegany  and  Catta- 
raugus counties  appear  to  have  been  the  seat  of  extensive  sponge 
plantations,  more  prolific  in  diverse  forms  and  individuals  than  the 
region  immediately  to  the  east  and  Avest.  This  impression  may,  however, 
be  a   mistaken   one  and  due  to  the   fact   that   Chautauipia,  Chemung  and 


Pkkfack.  5 

Broome  counties  have  yet  to  be  as  carefully  explored  for  these  fossiln  an 
have  been  Allegany,  Cattaraugus  and  Steuben  counties. 

To  thelatelleverend  Jonathan  Allkn,  former  president  of  Alfred  Univer- 
sity at  Alfred,  Allegany  county,  an  early  and  interested  collector  of  these 
sponges,  I  owe  many  favors.  From  him,  by  loan  and  purchase,  I  obtained  some 
interesting  specimens,  and  at  one  time  employed  his  son  to  make  collections 
from  the  localities  in  and  about  Alfred. 

Mr.  T).  Dana  Litthku,  both  before  and  during  his  engagement  tipon  the 
geologic  work  of  this  depai'tment,  has  supplied  a  number  of  intei-esting 
forms  from  the  lower  beds  of  the  Chemung  group  in  Ontario,  Yates  and  Steuben 
counties,  and  has  I'ecently  discovered  DicTYospoNoiiKii;  in  some  variety  in  the 
shales  and  sandst(mes  of  the  Portage  group,  in  Wyoming  and  Ontario  counties; 
a  formation  from    which    they   had   not  been    previously   obtained. 

As  early  as  1883,  Professor  Ciiakles  Barkois,  of  the  University  of  Lille, 
demonstrated  the  existence  of  Dictyospongid^  in  the  Psammites  du  Condroz  of 
Brittany.  The  forms  which  were  described  by  hina  are  noteworthy  in  their  simi- 
larity to  those  occurring  in  our  equivalent  geologic  horizon,  the  Chemung  group. 
Professor  Barrois,  Avith  most  courteous  liberality,  has  placed  these  specimens, 
together  with  some  later  collections  from  the  same  horizon,  in  my  hands  for 
comparative  study,  with  full  permission  to  illustrate  and  describe  them.  As 
no  other  part  of  the  world  has  yet  furnished  species  so  closely  allied  to  the 
American  Dictyosponges,  we  have  gladly  availed  ourselves  of  this  privilege  and 
these  French  species  will  be  foxmd  fully  described  and  illustrated  in  this 
volume. 

Further, — to  the  following  gentlemen  and  institutions,  acknowledgment 
is  here  made  for  their  courtesy  in  the  loan  of  specimens. 

Sir  AViLLiAM  Dawson,  of  Montreal,  whose  important  descriptions  of 
various  beautiful  forms  of  Protospongid^  from  the  slates  at  Little  Metis  are 
well  known,  and  who  has  kindly  loaned  a  nearly  complete  series  of  his  species 
for  comparative  study. 

Mr.  A.  S.  Tiffany,  of  Davenport,  Iowa,  for  the  loan  of  specimens  from 
Crawfordsville,  Indiana,  representing  the  new  genus  Lkbedictya,  and  specimens 
of  Clatiirospongia  both  from  the  locality  mentioned  and  from  the  Waverly 
group  of  Ohio.  ,, 

Mr.  J.  F.  Carll,  of  Pleasantville,  Pennsylvania,  who  has  supplied  speci- 
mens from  the  lower  Carboniferous  rocks  of  Pennsylvania,  upon  which  we 
have  established  the  species  Cidathospongia  Carlli. 


6  DlClYOBl'ONdlDyJi. 

Through  the  kindness  of  Prof.  I.  C.  Russell,  of  the  University  of  Michigan, 
was  ol)taine(l  the  use  of  the  original  specimen  of  Lyrodictya  Romingeri  and  a 
few  othei's  belonging  to  the  museum  of  that  institution. 

From  the  American  Museum  of  Natural  History,  through  Prof.  11.  P. 
Wun'FiELD,  was  received  on  loan  all  the  material  representing  the  Dictyo- 
spoNGiD^  in  the  Hall  collection  together  with  that  utilized  bj'  this  author 
as  the  basis  of  his  published  determinations. 

Through  Prof.  J.  F.  Kemp,  of  Columbia  University,  the  specimens  from 
the  loAver  Carboniferous  rocks  t)f  Ohio,  collected  by  the  late  Dr.  J.  S. 
Newberry  and  belonging  to  the  museum  of  that  institution,  were  kindly 
loaned. 

Interesting  specimens  from  the  Chemung  and  Waverly  groups,  which 
have  been  utilized  in  our  illustrations,  -were  obtained  from  the  Peabody 
Museum  of  Yale  University  through  the  favor  of  Prof.  C.  E.  Beecher. 

Some  important  specimens  from  the  Utiea  slate,  fully  discussed  herein, 
have  been  received  from  the  National  Museum  at  Washington,  by  the  courtesy 
of  Mr.  Charles  Schuciiebt  and  the  late  Dr.  G.  Brown^  Goode. 

Prof.  Albert  A.  Wright,  of  Oberlin  College,  generously  loaned  speci- 
mens from  the  collections  of  that  institution,  among  them  a  very  large  and 
interesting  example  of  the  I'are  genus  Hypiiant^nia. 

Through  the  favor  of  Prof.  E.  W.  Claypole,  of  Akron,  Ohio,  ^vas  obtained 
the  original  specimen  of  Oleodictija  Olaypolei. 

The  Rev.  J.  H.  McKee,  of  Olean,  has  permitted  the  examination  and  use  of 
Chemung  sponges  in  his  own  collection  and  that  of  the  Olean  Society  of 
Natural  Science. 

The  delineation  of  the  accompanying  figures  has  been  largely  done  by 
Mr.  George  B.  Simpson,  who  has  produced  them  in  a  highly  skillful  and  artistic 
manner.  Some  of  the  earlier  figures  were  made  by  Mr.  Ebenezer  Emmons. 
The  illustrations  of  spicules  are  from  camera  drawings  by  Mr.  Clarke. 

The  lithographic  work  has  been  admirably  executed  by  Mr.  Philip  Ast. 

In  the  correction  of  the  proof  sheets  much  assistance  has  been  rendered 
by  Mr.  Jacob  Van  Deloo. 

To  the  persons  mentioned  and  to  all  others  who  have  been  of  assistance 
to  the  work,  I  offer  the  assurance  of  my  obligation  and  consideration. 

JAMES  HALL. 


Professor  JAMES  HALL  died  on  August  7th,  1898,  at  the  great  age  of 

nearly   eighty-seven  years.     At   the    time  of  this   lamentable  event  only  a 

few  pages   of  this   memoir  had  been  printed,  but  his  lively   interest  in  the 

work  which  owes  to  him  its  inception,  and  his  pride  in  its  accomplishment, 

will  always  associate  his  name  with  present  and  future  knowledge  of  these 

remarkable  organisms,  the  Diotyospongid^. 

John  M.  Clarke. 


CONTENTS. 


rAos 

Title 1 

Preface a 

Contents 9 

The  Palaeozoic  Reticulate  Sponges  constituting  the  Family  Dictyospongidse 11 

Introductory 13 

Preliminary  remarks 13 

General  Observations  on  the  Sponges 14 

Anatomy 14 

Form 16 

Individuality 16 

Mode  of  Attachment 16 

The  Siliceous  sponges.     Skeletal  structure 17 

Taxonomy 20 

Symbiosis 21 

Affinities  of  the  Dictyospongidse 22 

Family  Dictyospongidse 29 

Structure  of  the  Skeleton 29 

Variations  and  Genetic  Relations  of  form 33 

Modes  of  Preservation 36 

Habitat  and  Mode  of  Occurrence 37 

Historical ;  Bibliography 41 

Classification 52 

Synoptic  List  of  Genera  and  Species 54 

Tables  showing  the  Geologic  Distribution 00-62 

Descriptions  of  Genera  and  Species 63 

Species  of  the  Silurian , 63 

Species  of  the  Devonian 64 

Species  of  the  Lower  Helderberg  Group 64 

Species  of  the  Hamilton  Group 65 

Species  of  the  Portage  Group ,.    68 

Species  of  the  Chemung  Group ....  72 

Species  of  the  Psammites  du  Condroz   147 

Species  of  the  Carboniferous 1 51 

Species  of  the  Waverly  Group 151 

Species  of  the  Burlington  limestone 1 65 

Species  of  the  Keokuk  Group 166 


THE  PALAEOZOIC  RETICULATE  SPONGES 


CONSTITUTING  THE  FAMILY 


DICTYOSPONGID^. 


INTRODUCTORY. 


Preliminary  Memarhx.  Tlie  purpose  of  this  Avork  is  to  present  an 
account  of  some  palaeozoic  forms  of  an  extinct  group  of  organisms,  repre- 
senting the  ''glass-sponges"  of  existing  seas.  Certain  simple  structures 
known  as  Protospongia  which  occur  in  the  Cambrian  and  early  Silurian 
faunas,  have  for  many  years  been  looked  upon  as  examples  of  such  siliceous 
sponges,  notwithstanding  the  fact  that  their  skeletons  have  been  wholly 
replaced  by  other  substances,  but  until  a  comparatively  recent  period  the 
interesting  division  of  these  fossils  here  designated  as  a  distinct  family,  was 
unknoAvn  or  unrecognized  as  sponges  in  palaeozoic  faunas.  The  earliest  form 
recorded  was  looked  ujjon  by  its  discoverer,  Mr.  T.  A.  Conrad,  as  a  cephalopod 
shell,  and  ^vas  so  described  in  1842,  under  the  name  Hydnoceras  tuherosuni. 
This  sj)ecimen  was  from  the  Chemung  rocks  of  Steuben  county,  N.  Y.  At 
the  same  date,  a  wholly  distinct  form  from  the  same  geologic  fomiation  was 
described  by  Mr.  Laudneu  Vanuxeh  as  Upliantceaia  Chemungensis  and  was 
regarded  by  him  as  a  marine  plant.  This  interpretation  of  the  nature  of  such 
bodies  became  current  for  the  time,  and  was  expressed  in  the  term  Diciyo 
pJiyton,  which  was  subsequently  applied  to  several  species  from  the  upper 
Devonian  and  Waverly  horizons.  Although  the  species  which  had  been 
described  up  to  the  year  1880,  manifested  considerable  diversity  of  form,  their 
condition  of  preservation  did  not  readily  suggest  their  real  nature.  In  all 
instances  the  delicate  skeleton  of  the  sponge  had  been  dissolved  and  had 
disappeared,  leaving  only  internal  and  external  casts  of  the  bodies 
with  more  or  less  distinct  impressions  of  the  spicular  network.  Such 
markings  were  believed  to  be  comparable  to  those  occurring  in  some  of  the 
living  marine  algae  and  this  similarity,  in  the  absence  of  any  recognized  agree- 
ment with  other  known  organisms,  was  the  basis  of  the  interpretation  of  their 
nature  as  then  made.  The  discovery  of  bodies  of  a  similar  character  in  the 
calcareous  shales  of  lower  Carboniferous  age  at  Crawfordsville,  Indiana, 
revealed  a  spicular  skeleton,  all  the  parts  of  which  had  been  changed  to  iron 
pyrites,  but  -were  not  otherwise  materially  modilied.  The  study  of  this 
material  elicited  the  first  definite  evidence  of   relationship  of  these  bodies  to 


,  •      .  DlCTY08P0N<lIDiK. 

14 

the  reticulate  siliceous  sponges,  an  opinion  which  all  later  investigations  have 

tended  to  confinn* 

All  that  is  now  known  with  reference  to  this  family  of  the  silice- 
ous sponges,  the  Dictyospongid^.,  has  been  brought  together  in  this 
volume  The  last  ten  years  have  vastly  increased  the  number  of  known  forms 
and  demonstrated  that  in  certain  late  Devonian  and  early  Carboniferous 
faunas  they  were  abundant  in  individuals,  species  and  genera.  The  accounts 
of  the  species  previously  described  have  been  for  the  most  part  brief  and 
either  insufficiently  illustrated  or  entirely  without  illustration.  In  addi- 
tion to  the  representation  here  given  of  the  general  aspect  of  these  forms, 
there  is  added,  where  the  material  permitted,  a  microscopic  analysis  of 
the  spicular  skeleton. 

General  Observations  on  the  Sponges. 
The  Sponges,  wth  the  exception  of  a  few  forms  inhabiting  fresh  waters, 
are  marine  animals,  and  among  the  lowest  and  simplest  in  the  scale  of 
life.  Less  simple  than  the  Protozoa,  they  are  also  of  less  complicated 
anatomy  than  typical  Coelenterata  (anemones  and  corals),  and  there  is  an 
inclination  among  some  later  investigators  to  recognize  them  as  a  distinct 
sub-kingdom  of  the  animal  world. 

A7U,tomy.  A  sponge  is  an  aggregation  or  colony  of  simple  cells;  a 
mass  of  connective  tissue  lined  without  and  within  by  obscurely  differentiated 
layers.  Together,  these  usually  form  a  sac-like  body,  attached  at  the  lower 
or  closed  end,  and  to  its  interior  cavity  (paraffaster)  the  water  has  free  access 
by  means  of  canals,  extending  from  the  exterior  surface  and  opening  upon 
the  walls  of  the  paragaster.  The  course  of  the  water  currents  is,  thus, 
centripetal  over  the  general  surface  of  the  sponge,  and  centrifugal  from  the 
paragaster  through  the  osculum  or  mouth  of  the  sac. 

Among  the  cells  of  the  outer  epithelial  layer  which  lines  all  the  canals 
leading  into  the  internal  cavity,  are  scattered  numerous  flagellate  cells,  by  the 
action^  of  whose  long  "  whip-lashes "  the  flow  of  nutrient  water  currents  is 
produced   and   determined.     Such    flagellate    cells   are    aggregated   in   little 

•  Previous  to  1881  there  ha.l  boon  only  vague  suggestions  of  the  sponge-nature  of  these  fossils.  The 
.low  to  their  real  character  was  first  furnished  by  the  introduction  into  the  museum  '"•»«';t'°"«  "^  I'"" 
country  of  the  exquisite  glass-sponge  kno^vn  as  the  "Venus  flower-basket "  and  described  by  the  late 
Richard  Owen  as  EnpUctella  aspergiUum.  Although  this  description  dates  as  far  back  as  1841,  he 
species  was  a  very  great  rar.ly  in  collections  here  until  fifteen  or  twenty  yearsago.  About  that  t.me  the 
sponge  fishers  of  the  Philippine  Islands,  the  seas  of  which  are  the  favorite  habitat  of  the  spec.es,  «  nttcd 
the  market,  so  that  in  England,  according  to  Sir  Wyvillb  Thomson,  the  price  of  good  specimens  tell  at 
once  from  five  pounds  to  ten  BbiUings  or  less  ( rft«  Atlantic,  p.  136,  1878). 


Introductory. 


15 


chambers  within  the  fleshy  part  of  the  sponge  wall ;  to  them  run  the  inhalent 
canals  (epirrhysa)  from  the  external  surface  of  the  sponge,  and  from  them 
depart  the  exhalent  canals  (aporrhysd)  emptying  into  the  paragastral  cavity, 
through  apertures  termed  ontia.  All  skeletal  structures  belong  to  the 
connective  tissue,  that  is,  the  median  layer. 

Such,  in  brief,  is  the  essen- 
tial structure  qi  the  typical 
sponge,  and  from  this  type  there 
are  no  departures  so  extreme 
as  to  obscure  comparison  with 
this  standard.  In  the  simplest 
sponges,  as  the  Ascones,  the  wall 
is  thin  and  there  are  no  special 
chambers  where  the  ciliary  cells 
are  aggregated.  The  entire  sur- 
face of  the  paragaster,  however, 
is  covered  with  these  flagella. 
It  seems  highly  probable  that 
a  similar  condition  prevailed 
among  the  ancient  thin-walled 
hexactinellids,  which  are  here 
under  consideration.  Sciiulze 
states  that  ^vhile  he  could  not  de- 
termine the  existence  of  flagella 
in  the  hexactinellids  on  account 
of  the  smallness  of  the  cells,  he 
believes  that  they  exist.  (Chal- 
lenger Expedition,  loo.  cit.  ]).  2'?).) 

Fixation  is  an  not  invariable  character.  There  are,  for  example,  certain 
ball-shaped  lithistid  sponges,  like  the  palaeozoic  Astylospongia  and  Hindia, 
which  appea?'  to  have  been  imattached  and  the  sport  of  the  water  currents. 
These  sponges,  too,  as  well  as  others  which  are  attached,  present  a  concentnc 
arrangement  of  the  water  canals  ^vhich,  according  to  Rauff,  have  originated 
from  the  ramified  furrows  of  the  external  surface  and  have  eventually  become 
internal  hy  progressive  over-growth  of  the  sponge  in  concentric  layers. 
Such  concentric  canals  parallel  to  the  periphery  of  the  sponge,  are  simply 
large  conduits  receiving  the  external  water  through  a  great  number  of 
minute,  converging  epirrhysa  and  debouching  in  the  small  paragaster  near 
the  summit  of  the  sponge. 


FiouRE  1.  Diagrammatic  figure  sbowlng  the  structure  of  a  gimple 
sponge  In  vertical  section.  The  sponge  substance  {fi)\s  lightly  shaded. 
The  canal  system  is  black,  the  arrows  showing  the  direction  of  the 
currents,  pi),  are  the  pores  or  openings  of  the  inhalent  canals 
(epirrhysa)  which  lead  to  the  flagellate  chambers,  ch.  From  the  latter 
depart  the  exhalent  canals  (aporrh.vsa),  ic,  which  open  Into  the 
general  central  cavity  or  paragaster.  o,  Is  the  osculum  or  aperture 
through  which  the  water  Is  discharged.    (Haxckel.  ) 


16  DicriYOsi'ONr.iDA. 

Form.  Six)nges  assume  an  endless  variety  of  expressions  of  a  sac- 
shaped  body.  Aside  from  iiiherent  specific  traits,  Avhich  express  themselves 
most  forcibly  in  such  variations,  there  are  numerous  external  conditions  which 
superinduce  modifications  in  form,  even  within  the  limitations  of  a  species. 
Irre<?ulai-ities  in  growth  may  arise  from  the  sponge  receiving  more  nutriment 
on  one  side  than  on  the  other,  and  differences  in  the  aspect  of  individuals 
belonging  to  the  same  species  are  found  to  be  due  to  the  fact  that  some 
have  grown  in  quiet  waters  and  others  where  the  water  was  in  inotion. 
The  existence  of  such  variability,  which  is  corroborated  by  living  sponges, 
can  not  fail  to  embarrass  the  discrimination  of  specific  values  among  fossil 
forms.  The  Dictyosponges  present  some  interesting  examples  of  such 
poJytrope  species.  The  Jlydnoceras  tuherosti'm,  Conrad,  is  a  vase-shaped, 
sub-prismatic  sponge,  Avith  several  horizontal  rows  of  strongly  tufted  and 
pointed  nodes,  and  all  Dictyosponges  having  these  characters  might 
very  conveniently  be  referred  to  this  specific  type.  We  know,  how- 
ever, the  location  of  several  plantations  or  colonies  of  these  sponges 
on  the  sea-bottom  of  the  Chemung  period,  and  have  endeavored  to 
make  it  apparent  in  the  ensuing  descriptions,  that  in  each  of  these  colonies 
the  individuals  have  an  expression  which  characterizes  the  members  of  that 
colony  alone.  These  are  distinctly  local  expressions,  due  to  restricted 
variability,  and  they  serve  to  indicate  in  several  instances,  the  precise  spot 
where  they  have  grown. 

Individmdity.  Sponges  may  assume  either  a  simple  or  a  compound 
form  ;  growing  as  a  single  sac  which  never  branches  or  divides,  or  as  a  bush- 
like stock  increasing  its  branches  by  the  production  of  buds.  The  compound 
mode  of  growth  is  rare  among  the  siliceous  sponges  and  there  is  no  satisfactory 
e\ndence  that  in  the  family  DicTYospoNGiDiE  there  was  ever  a  tendency  to 
duplication  of  the  sponge  by  budding.  The  species  Madodictya  oscidata,  of 
the  Keokuk  shales,  requires  further  investigation  in  this  respect,  as  it  is  not 
■well  understood.  But  as  a  whole  these  ancient  sponges  have  rigorously 
maintained  their  individuality,  so  far  as  that  term  is  permissible  in  application 
to  a  single  skeletal  structure. 

Mode  of  attachment.  The  vast  majority  of  all  sponges,  recent  and  extinct, 
have  a  portion  of  the  body  more  or  less  distinctly  specialized  for  fixation. 
This  attachment  may  be  to  solid  extraneous  bodies,  rocks,  pebbles,  dead  or 
living  shells,  or  may  be  effected  by  the  simple  penetration  of  muddy  or  sandy 
sediments  by  the  hnig,  stout  fibers  of  the  skeleton. 


Ini'koductoby.  17 

Recent  species  and  the  later  fossil  forms  show  that  fixation  may  be  effected 
either  by  the  base  {l>asiempliytic)  or  by  one  side  {pleuremphytiv)  of  the  sponge. 
The  ancdent  sponges  were  largely,  indeed  almost  exclusively  basiem2)hytic. 
This  term,  however,  is  quite  general,  as  basal  attachment  is  effected  by  several 
special  appliances  ijopliopliytes).  8ome  basiemphytic  82)onge8  spread  their 
substance  over  the  opposing  surface  so  that  the  Ijasal  part  may  be  of  greater 
diameter  than  the  superior  portions  of  the  sponge  and  show  no  difference 
therefrom  in  spicular  composition.  In  others,  such  as  Polylopuus,  tufts  of 
spicular  threads  from  the  lateral  lobes  curve  downward,  and  are  interwoven 
into  a  short  basal  rope ;  while  in  Phkronema  attachment  is  attained  by 
means  of  several  distinct  basal  tufts.  In  Euplectklla  the  strong  vertical 
threads  of  the  skeleton  terminate  below  in  a  tuft  of  fibres  which  constitutes 
the  tip  of  the  elongate  obconical  body.  IIyalonema  possesses  a  long  twisted 
rope  of  thick  strands,  Holascus  and  Dictyospongia  have  shorter,  untwisted 
basal  bundles.  In  Tiiysanodictya  and  Phuagmodictya  attachment  was  effected 
by  means  of  a  frill  or  collar  (^perilAipli)  encircling  the  truncated  base,  and 
in  the  latter  genixs  appearing  to  have  been  aided  by  a  small  tuft  extended 
from  the  basal  diaphragm.  The  Devonian  Dictyosponges,  Hallodictta, 
AcTixoDicTYA  and  Cp^yptodictya,  ^vere  either  spreading  forms  encrusting  the 
surface  upon  which  they  grew,  or  pleuremphytic  sponges  attached  by  special- 
ized processes  yet  iinknown.  There  are  many  forms  among  the  Dictyosponges 
whose  mode  of  life  must  have  been  like  that  of  Euplkctella,  buried  in 
the  mud  or  sand  for  a  considerable  portion  of  their  length.  Such  are 
Dictyospongia  and  the  long,  slender,  annulated  horns  of  Ceratodictya. 

The  composition  of  the  sponge-skeleton  is  generally  regarded  as  of  funda- 
mental significance  in  the  classification  of  the  entire  division,  and  the  combi- 
nation of  the  skeletal  elements,  the  shape  of  their  component  parts  and  their 
mode  of  uiuon,  are  highly  essential  taxonomic  characters.  Tlie  proper  under- 
standing of  the  fossil  s^ionges  is,  therefore,  gi-eatly  embarrassed  by  the  usual 
nuitilation  and  destruction  of  the  skeletal  parts,  or  their  modification  by  the 
process  of  fossilization.  This  knowledge  can  be  acquired  only  by  most  patient 
investigation  of  the  best  preserved  material. 

Since  this  ^vork  does  not  deal  -with  the  ceratose  or  calcareous  sponges  they 
may  be  left  out  of  present  consideration. 

The  Siliceous  Sponges  —  Sheletal  Stvuctxire.  The  structure  of  the 
siliceous  sponges  is  simple,  though  often  difficult  to  elucidate  on  accoimt  of 
the  isolation  of  a  part  or  all  of  the  spicular  elements.  The  siliceous  sponges 
are  either  thick  or  thin-walled.  In  tlie  former  the  considerable  substance  of 
the  cup  permits  a  f  idl  development  and  differentiation  of  the  external,  medial 


18 


Dun'YOSPONOIDiE. 


and  internal  parts  of  the  skeleton,  while  in  the  thin-walled  sponges  these 
different  parts  are  necessarily  condensed,  and  probably  abbreviated  or  only 
partially  developed.  Thick-walled  siliceous  sponges  are  not  of  infrequent 
occurrence  in  the  earlier  faunas,  but  all  the  Dictyospongid^  and  their  most 
closely  allied  groups,  possessed  a  very  thin,  almost  lamellar  skeleton  in  which 
a  distinction  between  the  dermal  and  gastral  surface  is  not  always  clearly 
perceptible.  The  accompanying  figure  which  is  adapted  from  one  by  F.  E. 
StHiri.zK,  will  convey  a  conception  of  the  disposition  of  the  parts  of  the 
skeleton  in  the  thick- walled   siliceous  sponges.      It  may  be  presumed  that  a 


D/cfyonalia 


Dfrms/ii.- 


FlGURK  3.    Diagram  of  arrangement  of  the  spicules  in  the  HexactinelUda.     (Schulze.) 

corresponding  relation  of  parts  exists  in  the  thin-walled  sponges  but  it  is  not 
often  possible  among  the  fossil  species  to  distinguish  the  precise  position  of 
the  isolated  spicules  in  the  skeleton.  The  most  conspicuous  and  strongest 
parts  of  the  skeleton  are  the  long  rods  or  smooth  spicules  projecting  from  the 
surface  of  the  cup.  These  are  known  as  the  lyroHtalm.  They  may  form  a  root- 
tuft  at  the  l)ase  of  the  cup  {hasalia)  as  in  many  species  both  of  fossil  and  living 
foiTus,  and  in  such  cases  they  are  sometimes  furnished  ^vith  a  means  of  attach- 


iNTRODtlOTOBY.  1 9 

ment  to  the  sediment  in  the  form  of  a  minute  anchor  at  the  extremity  of  each 
rod.  Such  root-spicules  are  not  always  present  but  it  would  seem  that 
in  most  of  the  Dictyospongid^  they  have  been  more  or  less  highly 
developed.  About  the  aperture  or  osculum  there  may  also  be  a  series 
of  prostalia,  known  as  the  marginalia,  which  may  be  present  or  absent  in  the 
thin-walled  sponges.  At  intervals  over  the  lateral  surface  of  the  sponge  there 
may  be  tufts  of  lirostalla,  such  as  frequently  occur  in  the  Dictyospongid^  ; 
these  have  been  termed  plewalia.  The  Dictyospongid^  in  common  with 
the  closely  allied  family  Euplectellid^,  also  possess  horizontal  and  vertical 
series  of  smooth  rods  in  bundles  of  greater  or  less  size.  These  cross  each 
other  at  right  angles  and  produce  a  reticulated  surface  Avhich  is  a  leading 
characteristic  of  the  entire  group.  The  regularity  and  uniformly  descending 
order  of  this  reticulation  is  a  marked  character  of  these  sponges.  The  regu- 
larity of  the  meshes  is,  indeed,  a  feature  shared  with  the  Pkotospongid^,  but 
in  that  family  there  is  no  such  nmltiplication  of  the  meshes  by  subdivision  as 
occurs  in  the  Dictyosponges.  Though  this  gradually  descending  subordinate 
reticulation  is  seen  in  all  true  members  of  this  family,  it  has  not  been  deter- 
mined that  the  descending  series  of  intersecting  spicular  bundles  which  produce 
the  mesh-work  are  always  of  a  definite  number.  If,  in  non-prismatic  species, 
the  unifonnly  largest  prevailing  quadrule  be  fixed  upon  as  that  produced  by 
the  primary  spicular  bands  (Thysanodictya,  Clathrospongia)  or,  in  prismatic 
fomis,  by  the  bands  lying  on  the  prism  angles  and  intersecting  Avith  equi- 
distant horizontal  bands  (Prismodictya,  Hydjioceras),  it  is  found  that 
there  is  a  sub-division  of  the  primary  quadrule,  or  Tnere,  to  the  fifth 
degree;  that  is,  there  is  a  descending  series  of  six  sets  of  reticulating 
bands,  forming  quadndes  which,  in  terms  of  the  primary  (piadrule,  or 
mere,  may  be  designated  as  dimeres,  trimeres,  Utrameres,  pentameres  and 
hexanieres.  Oidy  in  rather  unusual  instances  of  delicate  preservation  are 
the  ultimate  quadrules,  or  hexaraeres,  retained.  Such  reticulation  is  a  quite 
uniform  cliaracter  of  the  majority  of  the  genera  of  these  sponges,  but 
there  are  instances,  seen  in  such  genera  as  Piiysospongia,  Roemerispongia 
and  Hypiiant^nia,  in  which  the  uniformity  of  this  gradation  is 
lost ;  and  Avhere  there  is  a  single  or  double  series  of  major  quadrules 
whose  surface  is  covered  by  an  exceedingly  minute  reticidation 
(Piiysospongia)  or,  possibly,  by  none  at  all  (Roeiierispongia,  Hypiiant^nia).* 

'The  fine  reticulation  of  tlie  major  quadrules  in  Piiysospongia.  seems,  as  elsewhere  observed,  to  bo 
due  not  to  intersecting  sincular  threads  or  bundles,  but  largely  to  regularly  arranged  hexiictios  of  thf 
parenchyma;  while  there  are  good  reasons  for  regarding  the  apparently  uncovered  (luadrules  in  Roemer- 
I8P0NGIA  and  UvpiiANTiENiA  ;!8  due  to  the  imperfect  retention  of  this  delicate  layer. 


20 


Dun'YOSPONGID^. 


mi    iMi 

f  an  HI! 

iitl    /ill 

mi    iiHj 

III!       !■■«' 


I     nil     la 

fill    ml 

nil  iiii 

ill  I 
•"J  L- 

••■1    Hlli 

■  ■■  !!!! 
1 11  "SL 
nil  i»t 

nil  " 

IIII 


Mil 

■■■■  Ini  mi 
■■■■  iiii  ill, 
!.  ill 


The  fossil  species  show  that  these  fascicles  of  rod-like  spicules  lie  upon  the 
l)aragastnil  sui-face  and  a  given  spicule  is  often  continuous  from  one  extremity 
of  the  sponge  to  the  other.  It  seems  evident  that  these  spicules  have  no 
connection  with  the  basalia  and  marginalia,  though  the  latter  are  continued 
into  the  substance  of  the  sponge  and  may  extend  through  to  the  paragastral 
surface.  In  the  living  Euplectella,  the  ground  work  of  the  skeleton  is  of 
similar  chanicter,  and  students  are  generally  agreed  in  regarding  the  cylindrical 
rods  composing  the  horizontal  and  vertical  bundles  as  modifications  of 
cruciform  spicules,  due  to  extreme  prolongation  of  the  rays  of  one  axis  and 
the  complete  suppression  of  the  others.  In  the  Dictyosponges  these  fascicles 
sometimes  include  rods  terminating  in  anchors,  and  others  with  a  series  of 

barbs  along  the  sides  (denies) ;  a  fact  which 
may  tend  to  corroborate  the  derivation  above 
assumed.  Under  this  interpretation  these 
rod-like  8j)icules  are  of  different  nature  from 
the  prostalia  ;  as  they  are  evidently  gastral 
in  position  they  are  properly  termed  gaMra- 
lid  and  may  be  distinguished  from  less  modi- 
fied gastral  spicules  lying  in  the  interspaces 
of  the  network  by  the  term  gastralia  later- 
alia.  Such  rod-shaped  spicules  when  clearly 
preserved  in  pyrite  show  very  distinctly  a 
tubular  axial  cavity  about  which  the  sub- 
stance of  the  rod  is  arranged  in  concentric 
bands  or  envelopes.  Other  spicular  elements 
of  the  Dictyosponges  such  as  have  been 
clearly  made  out,  can  be  referred  to  the  outer 
surface  of  the  cup  {dermalid)  and  there  are 
a  few  which  it  may  be  safe  to  regard  as  rep- 
resenting the  medial  layer  or  pai'euchyma  of 
the  thick-walled  sponges  (parenclnjinalia). 
The  variations  in  the  spicular  composition  of 
the  skeleton  will  be  considered  at  greater 
length  in  following  pages. 
Taxonomy.  The  siliceous  sponges  constituting  the  Class  SILICEA, 
(Gkay,  1867),  or  those  whose  skeleton  is  essentially  composed  of  siliceous 
spicules*,    are  divided    into    three    orders,  based    iipon    the    form    of    the 

•Recent  writers  (cf.  Raupp,  op.  oil.)  iacliide  in  tho  Silicea,  the  ceratose  or  horny  sponges. 


n 


__   *'  •■■■ 

Hrii  >!■■;  nil 


s,'masiiai^ 


<  nil  ml  I 


\\ii  ml  ■■■■  nil  iis; 
\ii*  •••■  ■■■■  »n  '■-' 


uii  nil  ■■■■  IHiUSS 

ull  nil  ■■■■  III!  tmtm  , 


Figure  8.  Diagrammatic  representation  of  the 
arrangameDt  of  the  spicular  parts  iu  a  typical  Dlctyo- 
sponge. 


Intkoducj'oby.  21 

fundamental  spicule  in  each.  These  are  the  (1)  Hexactinellida,  (2) 
Tetractinellida,  (3)  Monactinellida ;  in  which  the  predominating  spicule, 
whether  its  form  l)e  unmodified  or  derived,  lias  respectively  six  rays, 
four  rays  and  but  a  single  ray. 

In  the  Hexactinellida.  (Oscar  Schmidt),  the  normal  form  of  the 
spicular  element  is  a  six -rayed  star  whose  rays  are  of  equal  length  and  lie  in 
three  planes  at  right  angles  to  one  another.  These  spicules  are  termed,  with 
reference  to  the  rays,  1iex(ieti)is,  or  referring  to  the  three  mutually  })erpea- 
dicular  axes,  triaxons  /  f njm  the  latter  term  the  name  Triaxoiiia  was  applied 
to  the  order  by  Sciiulze.  The  variations  in  the  form  of  the  normal  hexactins 
are  manifold,  spicules  Avith  five,  four,  three  and  two  rays  being  not  infrequent, 
but  it  is  believed  that  all  are  derivatives  from,  or  modifications  of,  the  triaxial 
type.*  These  spicules  are  either  isolated  and  independent,  lie  side  by  side 
forming  fascicles,  or  are  coalesced  at  their  extremities  forming  a  continuous  net- 
work. Two  inqiortant  subdivisions  of  this  order  have  Ijeen  generally  recog- 
nized, (1)  the  Lyssacina  and  (2)  the  Dictyonina.  The  fonuer  is  characterized 
by  the  isolation  of  the  spicules  \vhich  are  held  together  only  by  the  sarcode  of 
the  sponge  or,  in  i-are  instances,  by  a  loose  and  irregular  deposit  of  siliceous 
matter;  in  the  latter  every  arm  of  each  hexactin  is  united  with  the  corre- 
sponding arm  of  an  adjacent  spicule,  the  skeleton  thus  formed  consisting  of 
a  frame-work  Avitli  cubical  meshes. 

The  forms  here  embraced  under  the  designation  Dictyospoxoid.e  belong 
wholly  to  the  suborder  Lyssaciim.  So  far  as  their  spicular  skeletons  have 
been  acertained  they  afford  no  evidence  whatever  of  the  continuous  network 
which  characterizes  the  Dictyonina. 

Symbiosis.  Very  interesting  instances  of  commensalism  occur  in  some 
of  the  living  hexactinellids.  A  well-known  case  is  that  of  the  gorgonian 
coral,  Palyth.oa  fatvM,  which  encrusts  the  glass  rope  of  Hyalonema.  Another 
is  that  of  the  crab  inhabiting  the  paragastral  cavity  of  Euplectella,  in  which 
it  is  imprisoned  by  the  reticulum.  As  it  is  stated  that  the  terminal  sieve-plate  is 
not  developed  in  the  early  stages  of  Euplectella,  the  crab  evidently  spends 
only  a  portion  of  its  existence  thus  incarcerated.  In  regard  to  this  commen- 
sal, J.  E.  Gray  has  written  :  f  "  The  crab  is  generally  found  in  the  cavity  of 
the  sponges.  The  Spaniards  in  Manila  regard  them  as  formed  by  the  crabs 
for  their  protection  and  they  do  not  consider  a  specimen  perfect  unless  a  crab 
is  contained  therein.     I  have,  within  the  last  few  days  had  a  pair  offered  to 

'See  Rauff,  op.  cit.  p.  187  ;  THOMSO<f,  The  Atlantic,  voL  1,  p.  141;  aud  others. 
tAnnaU  and  Magaziuo  of  N  itiiril  History,  3.1  ser.  vol.  18,  p.  489,  1866. 


22  DurryospoNoiOif;. 

me  for  an  extravagant  sum  (£200),  because  they  contained  tlie  ci-ab  that 
formed  llieni."  Sir  \\'ya'illk  Thomson  relates  similar  facts  *  and  adds :  "  It  is 
singular  tliat  while  Pahjthoa  fatna  is  as  constantly  associated  with  examples  of 
Hyalonema  from  the  coast  of  Portugal  as  with  those  of  Japan,  no  commen- 
sal crustacean  has  been  found  in  any  of  the  Atlantic  specimens  of  Euplec- 
TELLA."  DoDERLEiN  has  described  f  a  Japanese  lithistid  sponge  having  a  surface 
covered  with  small  protuberances  -within  each  of  which  dwells  a  cirripede  of 
the  genus  Acasta.  The  little  crustacean  communicates  with  the  outer  world 
through  a  small  opening  in  the  reticulum  which  is  closed  at  its  death. 

Amon<r  the  Dictyosi'oxoid.e  we  have  observed  several  instances  of  vermi- 
form  tube-fillings  complicated  with  the  imj)ressious  of  the  reticulum  and  dis- 
tinctly pronounced  on  the  internal  casts  of  the  sponges.  These  tubes  are 
shown  in  the  figure  of  Hydnoceras  tuberosum,  var.  gloasema,  on  Plate  IX 
(fig,  2),  and  in  the  illustration  of  Prismodictya  teluni,  on  Plate  XVII 
(figs.  3,  9,  10).  In  all  probability  these  markings  indicate  the  presence  of 
an  annelid  symbiotic  with  these  sponges. 

Affinities  of  the  Dictyospongidae. 

In  seeking  a  proper  definition  of  the  family  Dictyospongid^,  it  is  necessary 
to  consider  its  relations  to,  and  differences  from  the  other  admitted  families  of 
the  Lyssacine  sponges.  On  account  of  the  insufliciency  of  knowledge  concern- 
ing the  spicular  structure  of  these  fossils,  but  few  serious  attempts  have  been 
made  to  establish  their  systematic  position.  Hinde,  in  1883,  regarded  them 
as  Dictyonina,  and  placed  the  genus  Dictyopiiyton  in  the  family  Stauuoder- 
MiD^;J  in  1887,§  he  included  that  genus  in  his  proposed  family  Proto- 
SPONGiDiE.  These  opinions  were,  however,  based  upon  internal  sandstone 
casts  in  which  all  traces  of  the  spicular  elements  had  been  obliterated. 

Raufp  has  defined  the  family  very  satisfactorily,^  placing  it  among  the 
Lyssacina.  His  characterization  of  it  is  in  the  following  terms :  "  Funnel- 
shaped,  cylindrical  or  prismatic  sponges  with  thin  walls,  often  raised  into 
nodes  or  ridges;  skeleton  very  regularly  reticulated,  forming,  as  in 
the  Protospongid^,  quadrate  or  sub-quadrate  meshes  of  different  series, 
enclosing  one  another.  The  frame-work  is  composed  of  bundles  of  fine 
spicules,  but  whether  these  spicules  are  stauractins  |  \vith  their  rays  lying 

•  The  Atlantic,  vol.  1,  p.  141,  1878. 

♦  Zeitschr.  fiir  wissensch.  Zoologie,  vol.  xl,  1840.     See  also  Raui'F,  Pal.ieospongiologie,  part  1,  p.  140. 
%  Catalogoe  of  the  Fossil  Sponges  of  tbe  British  Museum,  p.  130,  1883. 

$  Monograph  British  Fossil  Sponges,  p.  90. 
fl  Palaeospongiologie,  p.  189,  1893. 

II  The  terms  "stauractins"  and  "cruciform  spicules"  are  used  interchangeably  for  any  hexactine 
spicules  showing  evidence  of  branches. 


Intkoductory. 


23 


close  together,  or  are  diactine  rods  (with  triaxial  foundation),  or  a  combina- 
tion of  both,  is  not  yet  certain.  Very  small  but  distinct  stauractins  are 
present.  [Sometimes  with  tenninal  sieve-plate  (?)]."  Although  the  characters 
of  the  family  can  now  be  stated  with  greater  precision,  still  the  above 
definition  serves  to  indicate  the  structural  features  which  distinguish  this 
from  other  groups  of  Lyssacine  spcmges. 

The  family  Puotospongid/e,  founded  by  IIinde  and  somewhat  restncted 
by  K.AUFF,  includes  a  number  of  Cambrian  and  lower  Silurian  forms 
(Protospongia,  Salter,  Diagonella,  DawfWn,  Phormohella,  IIinde),  in 
which  the  very  thin  net-work  consists  almost  exclusively  of  a  single 
layer  of  cruciform  spicules  which  lie  at  the  angles  of  the  quadrate  meshes, 
their  rays  forming  the  sides  of  the  meshes,  but  according  to  Rattff,  not 
overlapping  to  form  fascicles.  In  these  species  all  the  subordinate  series  of 
meshes  are  similarly  formed.  In  addition  to  the  regular  hexactins  or  pentactins 
of  the  reticulmu  there  are  also  prostalia  in  the  fonn  of  a  few  strong  root-threads ; 
sometimes  there  is  a  row  of  short  rods  about  the  margin  of  the  osculum. 
Small  lateral  surface  rods  (pleuralia)  are  also  present  in  some  instances. 
These  structures  are  shown  in  the  species  from  the  Little  Metis  slates 
described  by  Sir  William  Dawson,*  Protospongia  mononema,  P.  tetranema, 
P.  coronata,  etc.,  and  it  appears  from  an  examination  of  some  of  this  very  inter- 
esting material  that  the  cruciform  spicules  may  extend  so  far  as  to  overlap 
one  another,  and  also  that  there  are  pi'esent  a  few  fllifonn  spicules  lying  along 
the  axes  of  the  stauractins.  In  some  of  these  sponges,  (Diagonella  coronata 
and  D.  cyathiformis,  Dawson)  the  meshes  are  not  bounded  by  vertical  and 
horizontal  sides  but  are  arranged  in  oblique  rows.  Recognizing  these  as 
characters  of  the  Protospongid^,  it  seems  somewhat  difficult  to  find  substan- 
tial belief  in  the  usefulness  of  the  proposed  family  Plectospongid^,  Rauff,f 
which  ajipeare  from  the  description  to  have  for  its  principal  characteristic  the 
formation  of  spicular  bundles  by  the  overlapping  of  the  filiform  rays  of  the 
stauractins,  together  with  other  rod-like  spicules.  Dr.  Raitj"f  distinguishes 
in  the  skeleton  a  dermal  layer  which  can  not  be  discovered  in  the  Proto- 
SPONGLD^,  but  is  always  clearly  defined  in  the  Dictyospongidve.  Cyatiiopiiycus, 
its  best  known  genus,  also  shows  evidence  of  large  circular  mural  openings,  as 


"New  Species  of  Fossil  Sponges  from  the  Siluro-Cambrian  at  Little  Metis  on  the  Lower  St.  Lawrence ; 
Trans.  Roy.  Soe.  Canada,  vol.  vii,  Sect.  Iv,  pp.  31-55,  figs.  1-27,  pi.  iii,  1889. 

tit  may  be  observed  that  this  name  has  not  been  formed  strictly  iu  accordance  with  the  rulesof  nomen- 
clature. There  is  no  genus  Plectosponoia  and  there  should  not,  therefore,  be  a  family  termed  Plec- 
rospoNGiD/E.  The  type  of  the  family  is  not  specified  but  it  is  evident  that  the  Cyathophycus  reticulatun, 
Walcott,  has  furnislied  the  conception  of  family  characters,  although  the  group  also  includes  the  genua 
Pl,ECTODERMA,  Hinde. 


^4  l)icrYosi»oNGibiE. 

seen  in  the  figures  of  the  species  given  on  Plate  I.  They  increase  the 
siiniiarity  of  these  sponges  to  the  living  Euplectella,  in  which  such  openings 
are  the  vacant  or  unobstructed  portions  of  quadrules  formed  by  the  primary 
pentactins  and  the  rectangular  intersection  of  the  spicular  bundles,  rounded  at 
the  angles  by  the  encroachment  of  irregularly  branching  filigree  work. 

The  beautiful  specimens  of  Cyatiiopiiycus  which  have  been  found  in 
great  abundance  in  the  Utica  slate  at  Holland  Patent,  N.  Y.,  have  been  the 
means  of  affoi-ding  a  pretty  clear  conception  of  the  structure  of  this  sponge, 
and  the  fine  figures  given  by  Rauff,  supplemented  by  those  here  introduced 
for  comparison  with  the  Dictyosponges,  show  the  details  of  its  composition. 
Ui>on  consulting  figure  10,  Plate  I,  which  represents  a  large  and  finely 
preserved  example  of  this  species,  somewhat  conventionally  but  judiciously 
restored  in  parts,  it  will  be  observed  that  the  mural  openings  may  be  crossed 
by  the  larger  cruciform  spicules  of  the  outer  layer  (which  is  not  true  of 
Euplectella),  but  they  are  principally  outlined  by  a  pyritized  film  which 
appears  to  be  spicular,  though  not  readily  resolvable.  These  charactei's  are 
better  shown  in  figure  12,  which  is  an  enlarged  view  of  a  portion  of  another 
specimen.  Cyathophycus*  possesses  a  tuft  of  short  root  spicules,  arid 
Cyatliospongia  Quehecensis,  Dawson,  has  been  i-epresented  as  having  free 
prostalia  both  about  the  aperture  and  over  the  outer  surface  of  the  Avall.f 

Plate  I  bears  a  number  of  figures  representing  another  interesting 
species  from  the  Utica  slate,  occurring  in  association  with  Cyathopliycus 
{Cyatlwdictyd)  retieulatus,  in  which  the  mural  openings  are  numerous  and 
very  clearly  defined,  lying  on  the  inner  wall  of  the  sponge  and  being  sur- 
rounded by  a  highly  irregular  network  of  cruciform  spicules.  The  form  and 
structure  of  this  sponge  suggests  little  relation  to  C.  reticniatus,  but  again 
illustrates  the  occurrence  of  mural  areoles  which  open  through  the  inner  Avail  of 
the  sponge,  but  are  crossed  or  obstructed  by  the  stauractins  of  the  outer  Avail.;}: 

•This  name  Las  been  hitherto  retained  although,  like  Dictyophyton,  it  tends  to  perpetiinte  the  old 
and  erroneous  conception  of  the  algous  nature  of  these  fossils.  Dawson  pmposed  to  modify  the  term  to 
Cyathospongia,  which  would  have  been  an  appropriate  substitute,  had  not  the  name  already  been  in  use 
for  a  genus  of  thick-walled  sponges.  Cyathodictya  would  be  in  better  harmony  with  the  terminology 
in  this  group. 

t  After  examination  of  specimens  of  this  species  from  the  Little  Metis  slates,  it  seems  that  the  sim- 
plicity of  the  reticulum  and  the  absence  of  mural  openings,  ally  it  more  closely  with  the  species  which 
Sir  William  has  referred  to  Protospongia,  rather  than  to  Cyathophycus  (Cyathodictta)  which 
it  resembles  only  in  form. 

tMr.  Walcott described  from  the  Utica  slate  (Trans.  Albany  Institute,  vol.  x,  p.  19,  pi.  2,  fig.  17, 
1879)  as  Cyathophycus  subsphwricus,  a  flattened,  subcircular  body,  showing  no  structure,  except 
a  small  area  in  the  center,  which  was  regarded  as  aa  aperture.  Dr.  Rauff  {op.  cit.  p.  256,  pi. 
iv,  flgs.  3,  4,  1894)  has  more  recently  studied  these  or  similar  bodies,  anil  figures  them  as  shallow  cups 
with  rather  thick  walls  about  the  osculum.  Only  traces  of  the  spicular  structure  were  discerned. 
Although  the  composition  of  these  sponges  is  extremely  obscure,  Rauff  h:is  proposed  for  the  species  the 
new  genus  Teganidm,  and  is  disposed  to  laclade  under  the  specific  name  Teganium  suhsijhwricum  the  form 


Intkoductorv.  Sf) 

The  presence  of  inura,!  (Openings  in  these  nienihers  of  the  I'i.kctosi'onoid/B, 
is  a  feature  of  similarity  with  Euplectklla  and  its  allies  (Eiii'lectklmd^:), 
but  in  none  of  the  Dictyospongid^  is  there  evidence  of  precisely  similar 
structures.  The  Dictyospongid^e  are  distinguished  from  all  of  these  earlier 
hexactinellids  by  the  presence  of  an  inner  wall,  composed  of  strong  vertical 
and  horizontal  fascicles  of  rods  reticulating  at  right  angles.  In  the  great 
majority  of  instances  these  reticulating  bundles  are  present  in  a  gradational 
series,  the  size  of  the  bundle  depending  upon  the  number  of  rods  it  contains, 
rather  than  upon  the  size  of  these  rods.  In  normal  forms  the  series  (^f 
bundles  and,  consequently,  the  series  of  quadrules  made  by  them,  numbei-s  as 
high  as  five  or  even  six,  the  finer  reticidation  of  the  sixth  series  being  observ- 
able only  in  instances  of  the  most  favorable  preservation.  There  are  some 
genera,  however,  included  within  the  limits  of  this  family,  in  which  thei-e 
exists  but  a  single  set  of  reticulating  bundles.     Thus  in  Piiysospongia,  there 

oeacribed  l)y  Dawson  as  J'rotosponyia  delicatula,  from  the  Little  Metis  slates.  In  tlie  very  extensive 
material  representing  the  Utica  slate  sponges  which  has  been  under  examination  in  the  preparation  of  this 
volume,  similar  obscnre  bodies  have  frequently  been  seen.  There  are  however,  associated  with  them, 
othei  spherical  sponges,  showing  very  distinctly  the  composition  of  the  skeleton  and  presenting  some 
highly  interesting  characters.  These  bodies  in  their  adult  condition  and  as  usually  preserved,  appear  as 
illustrated  in  figure  18  (Plate  I),  that  is  to  say,  they  are  circular  disks  undoubtedly  representing  orgiually 
spherical  sponges  under  compression.  In  no  instance  bus  any  trace  of  aperture  been  obscrvjd.  These 
bodies  are  so  broken  by  the  parting  of  the  shaly  laminae  that  almost  invariably  a  portion  of  the  outer  wall 
oi  the  sphere  adheres  to  one  side  of  the  matrix,  so  that  ligure  18,  which  represents  the  usual  condition  of 
the  fossil,  shows,  over  the  circular  central  portion,  the  inner  wall  of  the  sponge,  the  j)cripheral  ring  being 
the  thickness  of  the  cup  to  the  outer  wall.  A  good  exposure  of  the  outer  surface  is  very  rarely  seen.  The 
inner  surface  of  the  sponge  bears  a  great  number  of  clearly  delined  areoles  or  mural  openings,  which  are 
much  more  closely  arranged  than  in  Cj/athodicli/a  reticulata,  and  the  network  by  which  they  are  surrounded 
is  highly  irregular,  devoid  of  quadrate  meshes,  though  clearly  made  np  of  simple  stauractine  spicules  which 
are  involved  in  not  fully  resolvable  patches  of  i>yriie  (see  figures  19  and  20).  The  thickness  of  the  wall  as 
shown  in  figures  18  and  22  is  considerable,  the  structure  being  composeil  of  short  rod-like  8[iieules  which 
may  all  be  stauractine  and  the  entering  arm  of  the  dermal  peiitactins,  but  the  specimens  do  not  show  this 
conclusively.  An  entire  layer  is  formed  of  a  single  thin  lamina  of  stauractine  spicules  making  a  regular 
quadrate  mesh  in  which  the  horizontal  rays  of  the  spicules  may  slightly  overlap,  but  no  distinct  spiculnr 
bundles  are  formed  and  no  filiform  rods  are  present.  A  portion  of  this  outer  layer  is  shown  in 
figure  20,  traces  of  the  mural  openings  being  seen  on  the  matrix  beneath  it.  The  entire  outer  surface  is 
armed  with  short,  erect  spicules.  Bodies  of  this  character  occur  in  clusters.  One  slab  measuring  18x8 
inches  bears  twenty  seven  of  them,  all  of  normal  size. 

The  same  rock  iilso  contains  clusters  of  much  smaller  circular  bodies  varying  somewhat  in  size, 
as  shown  iu  figure  IG.  These  also  seldom  show  evidence  of  an  aperture  except  sueh  as  is  seen  in  one  of  the 
larger  examples  in  figure  14,  and  this  appearance  may  be  wholly  due  to  the  exfoliation  of  a  fragment  of 
the  slate.  The  substance  of  the  smaller  of  these  spherical  bodies  or  disks  is  an  unresolvable  massofpyrite, 
of  similar  character  to  that  described  by  Uaui'K  for  Teganium  aubsphcericum,  except  that  about  the  periph- 
ery are  erect  spicules  of  proportionally  much  greater  length  than  the  surface  spicules  in  the  large  spheres 
just  described  (figure  17).  With  slight  increase  in  size,  however,  the  spicular  skeleton  with  its  mural 
areoles  becomes  grailually  defined,  leading  to  the  conclusion  that  these  Teganium-WkB  bodies  are  young  of 
the  larger  forms.  There  is  still  doubt  as  to  the  relation  of  all  of  these  forms  to  the  Cyathophyciia  auhspha;- 
ricus  of  Walcott  and  the  Teganium  aiibsphwricum  of  RAtTFF."  There  is  little  iu  their  structure  to  suggest 
the  characters  of  that  species  given  in  the  analysis  by  either  of  these  writers,  save  the  spherical  form  ascribed 
to  it  by  the  former  and  the  obscure  spiculization  recorded  by  the  latter,  notable  only  in  the  very  young  of 
our  specimens. 

The  generic  characters  ascribed  to  Teganium  fail  to  apply  to  these  spheroidal  sponges,  save  in  part  to 
their  youngest  stages,  and  while  we  can  not  be  certain  that  Dr.  Kaufk  may  not  have  had  similar  bodies 
before  him  in  the  description  of  Teganium,  the  uncertainty  attending  the  description  and  illustration  given 


20  DlCTYOSPONGID^ 

18  good  reason  to  believe  that  the  quadrilles  made  l)y  the  large  primary 
fascicles  ai-e  not  subdivided  by  similar  bundles  of  less  size,  although  an 
extremely  fine  reticulatitm  of  these  areas  may  be  seen,  which  may  be  due 
to  regularly  juxtaposed  stauractins  not  on  the  inner  wall,  but  over  the  outer 
surface  of  the  sponge. 

In  the  great  saucer-shaped  fronds  termed  Hypiiant^nia  (male 
Uphant^nia),  the  broad  radial  and  concentric  bands  form  quadrate  areoles 
which  may  have  been  largely  or  wholly  free  of  sj)icular  structures  of  any 
kind.  The  sandstone  matrix  in  which  the  few  known  specimens  of  this 
sjx)nge  have  been  found  is  not  favorable  for  the  i-etention  of  tenuous  spicules, 
though  a  similar  rock  shows  at  times,  the  continuity  of  very  fine  skeletal 
films  in  some  other  of  these  sponges.  But  had'  there  been,  over  these  open 
quadrules,  a  coating  of  fine,  isolated  dermal  spicules,  their  disappearance 
rather  than  their  retention  would  be  a  reasonable  expectation.  A  single  one 
of  the  five  specimens  of  Hyphant^nia  at  present  known,  shows  traces  of  such 
a  film  at  the  angles  of  the  great  quadrules. 

by  liim,  warrants  tbe  application  of  the  term  SpHiERODiCTYA  to  the  interesting  generic  type  of  strnctare 
here  analyzed. 

*Tbe  original  specimen  of  Cyathophycus  subaphcericu)  is  preserved  in  the  Miisetim  of  Comparative 
Zoology  at  Cambridge.  Lilte  the  speoimons  described  by  RaukI'' it  shows  no  evidence  ot  reticulation  or 
siiicnlar  strncture,  nor  are  there  any  traces  of  peripheral  spicules.  The  central  area,  supposed  to  be  the 
oscnlum,  is  irregnlar  and  there  are  no  means  of  determining  whether  it  is  actually  such  or  is  only  a  casual 
feature.*  Upon  application  to  the  curator  of  the  United  States  National  Museum  at  Washington,  for 
antbentic  specimens  of  Teganium  tubsphwricum  for  comparison  with  tbe  material  above  described,  wo  have 
been  kindly  allowed  to  examine  a  portion  of  the  material  studied  by  Dr.  Raukf  and  identified  by  Mr. 
Walcott  as  his  species.  Of  the  specimens  received,  one  is  a  minute  cup-sbaped  budy  similar  in  form  to 
those  described  by  Raufp  and  with  a  similarly  obscure  skeleton  ;  three  others  are  cups  of  like  form  but 
larger  size  and  with  a  quadrate  network  already  defined  ;  these  resemble  young  individuals  ot  Cyathodietya 
re<ic«Ja/a  and  all  the  restare  thesmall  circular  bodies  witli  pseudopodia  like  perijiherals-picules.  A  portion 
of  a  single  cluster  of  these  is  represented  in  figure  14,  and  the  enlargement  of  one  of  the  individuals  (tig.  15) 
shows  the  clearly  defined  spicular  layer  coexistent  with  the  spiny  outer  wall,  while  in  all  the  smaller  examples 
the  spicules  of  the  skeleton,  except  those  of  the  outer  wall,  are  not  evident.  As  it  was  still  difBcnlt  to  appre- 
hend the  evidence  upon  which  the  determinations  made  by  Raui'F  were  based,  Mr.  Charles  ScirucHKRT, 
curator  of  invertebrate  palaeontology  in  the  National  Museum,  was  requested  to  examine  the  specimens 
whith  bad  been  figured  by  that  writer,  with  especial  reference  to  the  following  points:  (IJ,  Whether 
the  specimens  figured  by  Ra0fk  iu  his  fig.  4,  pi.  iv,  showed  distinct  evidence  of  possessing  an  aperture,  or 
whether  the  areas  represented  as  apertures  might  be  exfoli.atious  of  the  thin  pyritized  films  ;  (2),  whether 
these  three  cups  are  all  on  the  same  block  and  are  convex  as  represented  ;  (3),  whether  any  of  these  small 
specimens,  when  moistened,  show  a  series  of  peripheral  spinules;  and  (4),  whether  tbe  larger  specimen 
represented  in  op.  cil.  fig.  3,  has  the  unresolvable  pyrite  film  and  the  great  aperture,  as  represented.  To 
these  inquiries  Mr.  SciiucHERT  has,  after  careful  inspection  of  the  material,  answered  very  explicitly: 
( 1 ),  "There  is  no  evidence  of  an  aperture  in  any  of  the  three  small  specimens  studied  by  Raukk.  Where  these 
arc  indicated  in  the  figures  there  has  been  a  partial  or  complete  removal  of  tbe  ])yritizcd  film  in  the  speci- 
men"; (2),  "  The  three  figures  given  by  Rauff  are  taken  from  three  separate  and  distinct  specimens  so 
that  fig.  4  is  ideal.  The  apertures  are  drawn  in  as  well  as  the  convexity."  (3),  "No  peripheral  spicules 
can  be  detected  in  these  specimens."  (4),  "There  is  a  thin  pyritized  film  which  has  been  varnished  over, 
aud  in  none  do  I  distinguish  spicules.     The  aperture  in  fig.  3  is  also  drawn  in." 

*We  have  had  no  opportunity  of  examining  this  specimen,  but  the  above  statements  are  based  upon  an 
czaminatiou  kindly  made  by  Prof.  Ali>hf.us  Hyatt. 


Intkoductokv.  27 

The  rods  of  the  vertical  and  horizontal  bundles  are  for  the  most  part 
smooth  throughout  their  entire  length;  they  afford  no  evidence  of  being 
stauractine,  but,  as  suggested  in  tlie  diagnosis  of  the  family  given  by  Rauff, 
are  apparently  extremely  elongated  diactins.  The  outer  wall  of  these  sponges 
usually  carries  a  regularly  (juadrate  mesh  produced  by  large  pentactins 
standing  at  each  angle  of  the  quadrules.  These  outer  quadrules  usually, 
though  not  necessarily,  correspond  with  the  varying  quadrules  of  the  inner 
reticulum,  and  it  is  not  at  all  certain  that  the  outer  meshes  are  greatly  sul)- 
divided.  This  outer  or  dermal  layer  may  also  carry  projecting  spicules  oi 
various  forms,  spinules,  umbrella-shaped  clavules  and  echinated  stauractins, 
which  will  be  more  fully  described  in  tlie  following  pages.  The  surface  may 
also  bear  projecting  ornamental  lamellae  and  ridges  like  those  in  Euplectella, 
and  strongly  developed  tufts  of  stout  rods,  as  in  Physospongia  and  Hydnocebas. 
To  the  thin  space  intervening  between  the  outer  and  inner  surfaces  of  the 
skeleton  are  ascribed  some  normal  hexactinellid  and  minor  modified  forms 
of  spicules. 

The  variations  in  form  independent  of  essential  alteration  in  structure  are 
most  noteworthy.  Some  of  the  Dictyosponges  are  obtusely  or  acutely  obconical 
with  a  simple  exterior  ;  other  obconical  forms  have  a  highly  nodose  and  tufted 
surface  with,  sometimes,  clusters  of  pendulous  pouches;  many  are  distinctly 
prismatic,  some  bear  strong  concentric  annulations  and  vertical  ridges,  a  few 
have  the  form  of  large  saucers,  others  of  graceful  vases  Avith  slender  neck  and 
expanded  bowl,  and  in  two  genera  the  form  is  almost  cylindrical  and  has  a 
broad  fringe  or  frill  about  its  flattened  base. 

One  should  pause  to  reflect  that  these  ancient  organisms  with  their 
manifold  variety  of  graceful  and  striking  shapes,  were  not  the  sombre  bodies 
we  find  them  to  be  in  the  rocks,  but  in  life  formed  the  most  delicately 
woven  fabrics  of  glass,  latticed  vases,  urns  and  cups  of  the  rarest  delicacy 
and  beauty.  They  must  have  been,  Avhen  denuded  of  their  sarcode,  among 
most  exquisite  structures  of  the  past,  as  their  descendants  are  of  tlie 
the  present.  Nothing  for  example,  could  have  surpassed  the  graceful 
filigreed  chalice  of  Botryodictya  rarnom,  with  its  slender,  tufted  pedicel 
expanding  above  into  a  cup  ornamented  with  pendant  pouches. 

Forms  like  those  just  mentioned  as  having  a  flattened  transverse  extremity, 
raise  an  important  question  as  to  the  existence  in  these  sponges  of  a  terminal 
sieve-plate  like  that  in  Euplectella.  The  determination  of  this  point  is 
essential   not  only  to  a  proper  conception  of  the  limitations  of   the   family 


28  DiCTYOSPONGIDiB. 

DicrrYosPONOiD.fi,  but  also  to  a  correct  orientation  of  certain  of  its  members. 
The  "'enera  PuKAfnionirTYA  and  Tiiy'sanodictya,  which  are  here  fully  described 
and  illustrated,  are  subcylindrical  sponges  having  the  margin  at  one  extremity 
ret'ular,  unmodified  and  free ;  the  other  extremity  is  an  almost  transverse 
plate,  septum  or  diaphragm  which  completely  closes  the  cup,  and  about  its 
edges  where  in  contact  with  the  walls  of  the  cylinder,  is  a  free  expansion  or 
frill.  In  EuPLECTELLA,  which  is  throughout  the  nearest  living  representative 
of  the  Dictyosponges,  there  is,  at  the  v/pper  extremity  a  tei-mlnal  plate,  the 
so-called  *'  sieve-plate,"  surrounded  by  a  free  marginal  frill  of  spicular  net- work.* 
This  terminal  portion  of  the  skeleton  is,  however,  less  a  perforated  plate  than 
a  coml)ination  of  stout,  inosculating  rods  or  trabecidae.  It  is  described  by 
SciiULZE  in  the  following  terms  :  "  The  watch-glass-like,  arched,  terminal  sieve- 
plate  consists  of  a  lattice-work  of  laterally  compressed  ridges  oi  various 
thickness,  which,  though  exhibiting  no  very  regular  arrangement,  yet  suggests 
a  wheel-like  reticular  structure.  One  can  distinguish  at  least  three  or  four 
main  beams  which  are  approximately  cii-cular  and  several  which  extend 
radially.  These  form  the  primary  meshes,  ^vhich  are  again  divided  by  narrower 
and  less  prominent  ridges.  Here  and  there  a  broader  plate  is  formed  in  the 
net-work,  as  if  by  the  confluence  of  the  stronger  beams."f 

This  structure  is  in  no  respect  similar  in  composition  to  the  net-work 
constituting  the  rest  of  the  sponge  and  its  beams  are  not  continuous  exten- 
sions of  the  ground-work  strands  of  the  latter.  It  appears  to  be  ^vholly  a 
protective  apparatus  for  the  oscuhun,  cemented  to  the  upper  mai'gin  of  the 
reticulum.  The  marginal  frill  in  Euplectella  is  simply  one  of  the  many 
surface  ridges  covering  the  outer  wall  of  the  sponge. 

Ill  Piiragmodici'ya  and  Thysanodictya  the  structure  of  the  terminal 
plate  is  wholly  different.  It  is,  pi-imarily,  a  low  cone,  and  is  constituted  of 
radial  spicular  strands  which  are  an  immediate  continuation  of  the  vertical 
fascicles  of  the  reticulum,  those  being  crossed  by  concentric  strands,  corre- 
sponding to  the  horizontal  bundles  of  the  rest  of  the  skeleton.  The  radial 
strands  converge  at  or  near  the  center  of  the  cone,  and  when  the  outer  part  of 
the  cone  is  preserved  there  is  evidence  of  the  existence  of  a  spicular  tuft 
about  the  apex.  Not  all  the  rods  of  the  vertical  bundles  extend  from  the  l)ody 
of  the  sponge  into  this  plate ;  a  portion  of  them  is  continued  into  the  broad 
thin  peripheral  frill,  the  surface  of  which  is  crenidated,  while  its  distal  margin 
is  irregularly  fringed  by  long  radiciform  extensions.     The  radial  spicules  of 

•OwKN  at  first  regarded  this  sieve-ptate  as  at  the  bottom  of  Eui'LKCTKLla  (Sec  Sciiui.zk,  op.cit.  y.  54). 
t  Challenger  Expedition,  vo).  xxi ;   Hexactiaellida,  p.  66. 


DuTYOSPONGIDiK 


2i» 


the  terminal  plate  in  these  genera,  and  their  continuity  with  the  rest  of  the 
skeleton,  of  itself  demonstrates  the  fact  that  this  is  the  base  of  the  sponge, 
and  is  not  in  the  remotest  sense  homologous  with  the  sieve-plate  in  Eupli-xj- 
tella;  this  evidence  is  corroborated  by  the  structure  of  the  peripheral  frill 
and  its  evident  adaptation  to  the  purposes  of  attachment  as  seen  in  the 
character  of  its  distal  margin. 

Prom  this  consideration  of  the  characters  of  the  Dictyosponges  and  the 
groups  allied  to  them,  may  be  deduced  the  following  definition  of  the  family : 

Family  DICTYOSPONGIDiE,  Hall. 

Obcanical,  prisirmtic,  cylindric,  explanate,  vase,  funnel  and  saucer-sfiaped, 
ihin-ioalhd  sponges,  with  the  inner  layer  of  the  shileton  composed  of  series  of  reticu- 
lating vertical  and  horizontal  bundles  rnosthj  of  smooth  diactine  rods  ;  the  outer 
layer  composed  of  regidarly  au^gianged  stauractins  forming  quadrate  meshes 
with  more  or  less  irregularly  disposed  spicules  scattered  over  the  quadrules  thus 
produced  (^pinulu^y  clavules,  etc.)-  Regular  liexactins  in  the  parenchyma. 
Single  lophophyte  generally  present.  Periloph,  marginalia  and  pleuralia 
present  or  absent.     No  mural  areoles  /  no  term,inal  sieve-plate. 

Structure  of  the  Skeleton. 
In  contrast  to  the  very  considerable  geneiic  variation  and  abundant 
specific  representation  now  recognized  in  this  family,  our  knowledge  of  the 
spiciilar  elements  of  the  skeleton  is,  and  must  remain,  very  incomplete.  This  is 
the  more  to  be  regretted,  since  the  structure  of  this  frame-work  is  the  essential 
basis  of  taxonomic  arrangement.  Among  the  abundant  species  of  the 
Devonian,  this  skeleton,  almost  without  exception,  is  lost,  however  clear  and 
undisturbed  the  original  form  of  the  sponge  and  the  impression  of  its  surface 
characters.  In  the  sands  and  sandy  shales  of  the  Chemung  group  from  which 
the  great  maj(jrity  of  these  sponges  has  been  obtained,  the  original  hydrated 
silica  of  the  skeleton  has  undergone  change,  probably  through  the  agency  of 
organic  acids  resulting  from  decomposition,  and  has  been  replaced  first  by 
pyrite,  and  by  oxidation  this  has  passed  into  limonite  and  been  quickly 
i-emoved  from  the  porous  rock.  In  rare  instances  portions  of  the  spicular 
rods  in  the  condition  of  limonite  have  been  seen  in  specimens  from  the  upper 
Devonian  and  more  often  in  those  from  the  sandy  layers  of  the  lower 
Carboniferous  (Waverly  group)  of  Ohio.  The  calcareous  shales  in  which  the 
sponges  of  the  Keokuk  group  at  Crawfordsville  and  Indian  Creek,  Indiana, 
are   embedded,   have,   however,  retained   a  greater   or   less   portion   of   the 


so 


DlCTYOSPONOIDiK. 


skeletons  in  a  pyritized  condition,  and  it  is  from  this  source  that  all  our 
knowledge  of  their  ultimate  composition  is  derived.  As  already  observed,  the 
frame  work  of  these  sponges  consists  of  bundles  of  rods  crossing  at  right 
angles.  The  majority  of  these  are  smooth  and  taper  very  gradually,  not 
necessarily  extending  for  the  entire  length  of  the  sponge.  Close  observation 
has  not  yet  shown  them  to  have  more  than  a  single  axis,  that  is,  they  are  not 


FiouBE  4.    Spicules  of  Dictjogpongidae. 

1.    Part  of  a  primary  pentactin  of  Cleodictya  Mohri,  witli  umbels  in  various  positions, 

and  fragments  of  ottier  spicular  rods,    x  60, 
8.    Two  angles  of  a  quadrule  in  Phragmodictya  catillifomiis,  the  primary  hexactins  or 

pentactins   being   in   place,  with  fragments  of   the  vertical  and    horizontal 

bundltis.    X  50 
S.    Part  of  a  quadrule,  showing  the  vertical  and  horizontal  bundles  with  parenchymal 

spicules  scattered  over  the  interspace,    x  40.    From  an  undetermined  fragment 

at  Crawfordsville,  Indiana. 
4.    Some  of  the  major  and  minor  hexactins  of  Dictyoapongia  f  Marcetlia.    x  30. 

crossed  by  other  rays.  Such  rods  may,  therefore,  be  regarded  as  extreme 
modifications  of  the  typical  triaxial  or  hexactinellid  form,  produced  by  the 
abbreviation  of  two  of  the  axes.  The  terminal  portion  of  a  very  small  rod 
{Mastodictya  osculatd)  enlarged  400  diameters,'is  represented  in  the  accompany- 
ing figure  5.  In  some  genera,  particularly  in  Piiysospongia,  vertical  bundles 
of  these  rods,  which  lie  on  the  inner  or  paragastral  surface  of  the  sponge, 
are  accompanied  by  comparatively  short  rods  which  terminate  in  an  anchor 
with  two  sharp  flukes  (figure  5).     These  had  nothing  to  do  with  the  fixation  of 


StIM'CTI'KK     <>I'     SKKMVroN. 


»l 


the  sponge  as,  in  PhywspongUi  Da^moni,  they  are  scattered  over  the  entire 
length  of  the  pi'iucipal  vertical  paragastrul  strands.     The  relation  of  their  form 
to  the  noimal  hexactins  may  ])e  uTiderstood  by  observing  that  the  sides  of  the 
iiead  of   the  anchor  not    extended  into 
spines  are  much    swollen,    forming   ab- 
breviated  representatives    of    the  rays 
of  the  third  axis.      With  these  anclioi-- 
ate  rods  are  others,  occurring  in  Phys- 
ospoNGiA,    Clatiirospongia    iiud    Lvno- 
DiCTYA,    which   may  be  termed  detnes; 
formed  of   flattened  rods  with  alternat- 
ing  broad  triangular    processes  at   the 
sides   (figure   5).       The  anchorate    and 
clemate  I'ods  are  found  together,  but  only 
in  the  vertical  fascicles.    The  spicules  of 
the  outer  or  dermal  Hvall  of  the  reticulum 
are  seldom  retained  in   place,  and   the 
precise  position  of  some  of  the  forms  is 
to  a  considei-able  degree  a  matter  of  con- 
jecture.     Here  there  are  no  smooth  rods, 
anchorate  spicules  or  clemes,  or  if   the 
first  are  ever  present,  they  are  small  and 
of  rare    occurrence.      The  mesh  of   the 
outer  wall  is,  as  already  observed,  com- 
posed of  stauractins,  principally  pentac- 
tins,  which    lie  at   the   angles  of   small 
quadrate   areas    (see   figure   4).     These 
pentactins  are  usually  smooth-rayed  but 
not  infrequently  they  bear  thorns  or  long 
spinules  (figure  6  (9)).      In  these  pen- 
tactins a  single  low  node  near  the  intersection  of  the  axes  is  frequently  present 
and  represents  a  sixth  ray.     The  surface  of  the  «piadrules  set  off  by  the  prin- 
cipal pentactins  is   covered  Avith,  or   filled   by,  isolated   spicules  of   various 
forms.      It  is  not   easy  to   conclude,  except   from   comparison   -with   living 
hexactinellids,  which  of  these  spicules  are  referable  to  the  dermal  layer  and 
what    portion   of   them   to    the    interstitial    or    parenchymal    layer   of    the 
reticidum.     On  account  of  the  thinness  of  the  entire  sponge  wall,  the  dis 
tinction  of  such  an  intermediate  layer  is  perhaps   a  purely  conventional  one. 
It  is   usually    extremely   difficult   to    discriminate    between   the   outer   and 
inner   layers   of   the   skeleton   especially  when    under  compression,  but  the 


FiouRK  5.    Spicules  of  Dictyospongidip. 

1.  Terminal  portion  of  a  small  vertical    rod  in  Motto- 

dictya  osc\daUt.    x  400. 

2.  Cleme  of  Physospon(/ia  Dawtoni.    x  60. 

3.  Part  of  cleme  of  AcUeodictya  marsipus.    x  400. 

4.  Group  of  anchorate  spicules,  Pkysospongia  Daumoni. 

x60. 


32 


l)lOTyt)SrONGIl)/K. 


presence  of  certain  spiculcis  Hirailar  to  those  which,  in  recent  siliceous  sponges, 

are  found  in   the  sarcode  of   the  inner  layer,  is  a  reason  for  ascribing  such 

layers  to  the  Dictyosi'ongid-*:. 

To  the  dermal  layer  may  be  ascribed  the  umbrella-shaped  clavules,  or 

ambels,  occuringin  Cleodictya,  Puysospongia  and  GiupiionicTYA  (figures  6,  7)  ; 

short,  thick  I'ods  ^^ith  an  arched, 
toothed  disk  at  the  end.  To  the 
same  layers,  also,  various  forms  of 
the  pinnies,  some  clearly  pentac- 
tin,  others  compound,  among 
them  the  peculiar  tn'-p/'nuh/s 
Avhich  consists  of  three  tufted 
rays  rising  from  a  smooth  base 
with  accessory  processes  (fig- 
ixre  6).  These  pinnies  have  sel- 
dom been  found  uiiliroken  with 
the  exception  of  the  tri-pinulus, 
a  form  occiirring  in  Cleodicty'a 
and  Lebedictya.  To  the  paren- 
chyma (or  perhaps  to  the  quad 
rules  of  the  inner  wall)  may  be- 
long certain  peculiar  pentactins 
with  curved  rays,  found  only  in 
Plirmjmodktya  catiUiformis  (fig- 
ure 6  (1-4))  and  the  pod-shaped 
diactins  ^vhich  occur  in  Phraomo- 

LeBEDICTYA,         PlIYSO- 


FioDRE  6.    Spicules  of  Dictyospongidae 

1,  2,  8,  4.    Modified  iiexactins  of  Phragmodictya  catiUiformis.    x  400. 
6, 8, 7.    Different  forms  of  the   pinuiug    from    Cleodictya  Mohri. 
X  400  ;  6  is  a  compiete  tri-pinulus. 
8,  9.    Parts  of  ecliinate  hexactins,  also  from  Cleo.  Mohri.    x  400. 

10.  Amuelimodifiedspicule  fromCripAodicfyacjjfp/taiies.   x  400. 

11.  hn\imti»l\  PhyiospongiaColletti.    x400. 


DICTYA, 


spongia,  Gbiphodictya,  etc.,  and 
perhaps  also,  the  highly  branched 
oxyhexasters  characteristic  of  the  genus  Gripiiodictya  (figiire  7). 

The  prostalia  or  tufts  of  spicular  rods  occurring  at  the  l)ase,  aperture  and 
at  various  nodes  upon  the  surface  of  some  species,  seem  to  penetrate  the  entire 
wall  and,  in  some  instances,  as  in  Hydnoceras  and  Piiysospongia,  to  bear  a 
definite  relation  to  the  intersections  of  the  vertical  and  horizontal  facicles  of 
the  inner  wall. 

Note.  The  terminology  of  the  spicular  elements  of  the  s[)onges  has  in 
recent  years  Ijecome  very  complicated  through  an  exuberant  growth  of  terms 
Avhich  do  not  explain  themselves.  The  present  profusion  of  names  is  quite 
bewildering  while  the  fundamental  c<mception  of  the  spicule  form  is 
perfectly  simple  and  the  mode  of  its  modification  not  of   itself  neccessaiily 


Form. 


83 


complex.  To  serve  the  purposes  of  this  work  we  have  "brouglit  together  in 
figure  8  a  group  of  the  more  characteristic  derivative  forms  of  the  hexactin 
spicule,  which  has  been  compiled  from  the  illustrations  given  by  Schulze 
{Report  Cltalletujer  Expedition)  and  Eauff  {Pala<iO>^(rmjiolo<jie).  These  will 
aid  in  elucidating  the  spicular  forms  occurring  in  the  Dictyspongid^ 


FiocRE  7.    Spicule!  of  Dictyospongidae. 

The  upper  figure  is  a  group  of  stout  bean-shaped  diactins  from  an  undetermined 
fragment  at  Crawfordsville,  Indiana.  The  two  lower  spicules  are  from  Griphodictya 
epi'phanes;  that  at  the  left  is  an  hexactin  much  branched  at  the  eztremityof  the  rays  ; 
that  at  the  right  an  umbel  with  deeply  cleft  top.    All  x  400. 


Vakiations  and  Genetic  Relations  of  Form. 

A  glance  through  the  plates  of  this  volume  will  convey  an  idea  of  the 
multiplicity  of  forms  under  which  the  Dictyosponges  manifested  themselves. 
Among  these  smooth,  nodose,  prismatic  and  annulated  cups  and  ^•ases,  the 
smooth  slender  obcone  as  expressed  in  the  typical  genus  Dicitospongia,  is 
doubtless  to  be  regarded  as  the  fundamental  form.  The  primary  relation 
of  this  form  to  all  the  abundant  expressions  occurring  in  the  Devonian 
faunas  is  evident  from  an  inspection  of  any  well-preserved  example  of  a 
nodose,  prismatic,  annulated  or  other  form,  in  all  of  which  the  early  stages  are 


34 


DiOTYOSPONO  IT)  JR. 


regularly  olK-onical.  In  IIydnocebas  this  primitive,  smooth  condition  was  of 
brief  duration  and  was  directly  followed  by  the  development  of  prism-faces, 
at  first  four  and  directly  thereafter  eight,  the  nonnal  number.  These  pi-ism- 
faces  are  well  developed   before  the  appearance  of  the   nodes   on  the  prism- 


FiouKE  8.    Primary  and  modified  forms  of  the  Hexactio. 

a.  Oxyhexactin,  x  100  ;  6.  Ecliinhexaotin,  x  300  ;  c.  StronBylhexactln  with  prickly  extremitiea,  x  450 ; 
d.  Tylhexaster,  I  300  ;  e.  Discuhexaster,  x  450 ; /.  Plumicome,  x  300 ;  g.  Amphidisk,  with  remnants  of  the 
abbreviated  arms,  x  100  ;    h.    Pinule,  x  300 ;  t.    Orthotetraen,  x  100;    j.  Clamle,   x  450  ;  k.  Qiaen,  x  150. 

(Rauff.) 

angles.  The  nodes  themselves  do  not  appear  abruptly  but  with  gradual 
increase  in  size,  and  after  the  passage  of  the  adult  period  their  size  again 
decreases,  and  eventually  they  disappear;  the  prismatic  faces  also  become 
obsolete,  the  apertural  portions  of  old  individuals  being  smooth  and  funnel- 
shaped. 

The  remarkable  genus  Botbyodictya  retains  the  smooth,  primitive, 
subcylindrical  form  of  the  cup  for  fully  one-half  its  length  and  then  abruptly 
swells  into  a  vase  with  obscure  prism-faces  and  an  extravagant  development 
of  compound  pouch-shaj^ed  nodes.  In  the  great  majority  of  all  of  the  Devonian 
species  of  these  sponges  the  prism-faces  are   more  or  less  clearly  discernible, 


FoKM.  85 

even  among  forms  belonging  to  the  genus  Dictyospongia,  so  that  tins  pris- 
matic form  seems  in  a  certain  sense  normal  to  the  group,  though  always  a 
progressed  condition.  In  some  of  the  annulated  species,  as  Riiabuosiponcha, 
the  annulations  are  clearly  of  subsecpient  appearance  to  the  prism-faces, 
though  the  latter  are  represented  throughout  the  adult  condition  of  the  sponge 
as  vertical  ridges;  while  in  Ceuatodici'YA  the  annulations  appear  to  follow 
directly  upon  the  smooth  conical  condition,  the  prismatic  condition  not  being 
manifested. 

It  is  a  significant  fact  that  even  with  our  present  impei-fect  knowledge 
of  this  group,  the  earliest  known  Dictyosponge,  Dictijosjxyrujia  Danhyi, 
McCoy  (sp.),  from  the  Ludlow  beds  of  England,  is  such  a  regularly  obconical 
species  with,  possibly,  an  obscure  development  of  the  prism-faces  toward  its 
upper  part.  This  simple  form  may  l)e  regarded  as  directly  derived  from  that 
which  characterizes  such  earlier  Silurian  forms  as  Cyatiiouictya.  The 
secondary  modifications  of  this  form  (prism-faces,  nodes,  annulations,  etc.),  are 
not  known  in  any  such  early  species  nor  outside  of  the  DicTYospoNdiD.E. 

Among  the  many  prismatic  or  banana-shaped  species  which  constitute  the 
genus  Prismodictya,  the  early  smooth  conical  stage  is  extremely  abbrev- 
iated, but  when  specimens  are  well  preserved  to  the  basal  apex  it  is  found 
that  such  a  condition  exists. 

With  the  disappearance  of  the  Devonian  faunas,  the  simple  prismatic 
sponges  quickly  disappear.  The  introduction  of  the  Waverly  fauna,  immed- 
iately succeeding  the  upper  Devonian  in  Pennsylvania  and  Ohio,  brought  in 
8e\'eral  new  generic  types,  some  of  which  retain  the  prismatic  form  under 
considerable  modification  (Calathospongia),  but  in  the  preponderating 
species  the  prismatic  structure  is  essentially  lost.  Tiiamnodictya  and  Cleo- 
DicTYA  are  sponges  with  a  smooth  surface ;  the  simple,  obconical  form  of  the 
cup  is  modified  in  both,  in  the  former  the  base  of  the  cup  being  broad  and  its 
apertural  portion  expanded  into  a  %vide  vase,  and  in  the  latter,  the  bowl-like 
lower  part  of  the  sponge  bearing  a  row  of  nodes  which  seem  to  have  no  rela- 
tion to  prismatic  faces  as  in  all  of  the  nodose  species  of  the  Devonian. 

The  still  later  fauna  of  the  Keokuk  group  as  represented  in  eastern 
Indiana,  brings  in  the  remarkable  genus  Piiysospongia,  a  broadly  conical  sponge 
with  wide  base,  and  having  the  inner  wall  of  its  skeleton  constructed  on  a  very 
simple  plan  ;  a  modification  of  the  normal  type  of  structure  which  is  evidently 
foreshadowed  in  the  Roemei'ispongia  Gerohtehiensis,  F.  Roemer,  of  the  middle 
Devonian  of  Germany.  In  this  fauna  is,  also,  the  genus  Piiragmodictya,  a  non- 
prismatic  sponge  characterized  by  its  depressed  conical  base  surrounded  by  a 


3g  DlOTYOSPONOID^ 

peripheral  frill;  a  form  anticipated  in  the  Tiiysanodictya  of  the  npper 
Devonian,  in  which  evidence  of  the  prism-faces  is  still  retained.  In  the  greater 
nnml)er  of  all  the  other  known  species  represented  by  the  genera  Cleodictya, 
Lkbedictya,  Calathospongia  and  Lykodictya,  the  form  is  some  modification 
of  a  rapidly  expanding  obcone  or  vase,  like  the  ultimate  expanded  apertural 
portions  of  the  Devonian  nodose  sponges.  With  all  these  generic  modifica- 
tions so  unlike  those  prevailing  in  the  Devonian,  are  still  to  be  found  a  few 
representatives  of  the  genera  Pkismodictya  and  Dictyospongia. 

Though  the  future  \n\l  doubtless  afford  much  evidence  helping  to  estab. 
lish  the  mutual  relations  of  form  among  the  Dictyosponges,  the  evidence  of  the 
present  clearly  points  to  the  simple,  slender,  smooth  obcone  as  primitive, 
probably  derived  from  the  antecedent  Plectodekmatid^  and  Protospongid^, 
and  attaining  its  maxinmm  in  the  upper  Devonian ;  to  the  prismatic  form  as  a 
primary  derivation  therefrom,  a  group  also  attaining  its  highest  development 
in  the  upper  Devonian ;  to  the  nodose  and  annulated  forms  as  a  secondary 
derivation,  and  the  explanate,  non-prismatic  vase-shaped  species  with  their 
various  modifications,  such  as  prevail  in  the  lower  Carboniferous  faunas,  as  a 
still  later  derivation  from  the  fundamental  type. 

Modes  of  Presekvation. 

The  skeletal  substance  of  the  siliceous  sponges  is  a  hydrated  or  chalcedonic 
silica.  In  these  fossils,  as  preserved  in  the  older  rocks,  this  skeleton  has  been 
wholly  removed  and  replaced  by  some  other  substance.*  In  a  single  instance 
among  all  the  examples  here  described,  Clathrospangia  Hamiltoneiisis,  from  the 
calcareous  shales  of  the  Hamilton  group,  the  skeleton  has  been  substituted  by 
a  deposit  of  carbonate  of  lime,  which  retains  the  general  aspect  of  the  net- work 
though  the  form  and  structure  of  the  individual  spicular  bodies  is  wholly  lost. 
In  all  other  cases  that  have  come  under  our  observation  the  replacement  of 
the  siliceous  skeleton  has  been  a  complicated  one  resulting  in  the  substitution 
of  iron  pyrites  for  the  silica,  and  with  such  delicacy  and  so  gradually  has  this 
process  been  carried  forward,  that  the  minutest  details  of  form  in  spicules  of 
highly  intricate  structure  have  been  retained. 

NotAvithstanding  the  fact  that,  almost  without  exception,  in  all  the 
numerous  sjionges  of  the  upper  Devonian  sandstones,  the  spicular  skeleton  is 
entirely  gone,  it  is  evident  that  its  absence  is  not  due  to  a  lack  of  replacement 
^y  pyrites   but  to  a  removal  of   the  pyrites  itself.     This   appears  from  the 

•ZiTTEL  observes  that  in  the  older  lithistid  siliceong  sponges  the  orisinal  amorphous,  sintrly  refractive 
silica  has  become  doubly  refractive,  and  almost  iusoluble  in  caustic  alkalies  {AnnaU  and  Magazine  oj 
Natural  Hitlory,  vol.  20,  1877,  pp.  260  et  seq.). 


Habitat  and  Occurrence. 


87 


occasional  presence  in  examples  from  these  and  the  similar  sandstones  of  the 
Waverly  group,  of  a  few  root  spicules  which  are  in  the  condition  of  hydrated 
oxide  of  iron  (limonite),  as  well  as  from  the  iron  stains  whicih  frequently 
accompany  much  of  the  material  from  all  such  rocks.  The  porous  sandy 
matrix  of  both  horizons  has  been  an  efficient  aid  to  the  removal  of  the 
skeleton.  Where  replacement  has  gone  on  in  the  presence  of  abundant 
decomposing  organic  matter,  as  in  the  bituminous  rocks  of  the  Quebec 
group,  Utica  slates  and  Marcellus  shales,  the  deposition  of  pyrite  is 
complete  and  sometimes  excessive ;  so  that  the  form  of  the  spicules  may 
become  obscured  by  pyritous  enlargement. 

In  the  calcareous  shales  of  the  Keokuk  group  in  Indiana,  the  pyritous 
i-eplacement  is  often  retained.  The  fact  that  it  is  not  always  present  but  that 
in  the  majority  of  instances  it  has  been  lost,  leads  to  the  inference  that  the 
removal  of  the  pyritized  skeleton  is  constantly  in  process.  It  is,  however,  to 
the  sponges  of  this  horizon  that  is  due  such  knowledge  as  Ave  now  possess  of 
the  ultimate  spicular  structure  in  the  DicTTOspoNGiDiE. 

Although  the  sandstones  of  the  Devonian  and  lower  Carboniferous  have 
proven  unsatisfactory  media  for  the  retention  of  the  skeleton  they  have  well 
preserved  the  form  of  the  sponges.  They  have  been,  at  times,  subjected  to 
slight  distortion  from  compression,  but  this  disfiguration  is  always  much  less 
in  these  rocks  than  in  those  more  highly  laminated  and  of  finer  texture.  The 
impressions  of  the  interior  and  exterior  of  the  skeleton  are  often  so  sharply 
defined  as  to  afford  a  very  clear  conception  of  the  structure  of  the  network 
and  the  ornamentation  of  the  surface,  the  slight  discoloration  of  superficial 
ridges  and  laminae  by  iron  stains  frequently  aiding  such  conceptions.  While 
specific  and  generic  characters  in  this  family  are  so  largely  dependent  upon 
variations  of  form,  this  mode  of  preservation  is  by  no  means  unsatisfactory, 
as  the  shape  of  the  sponge  is  usually  better  preserved  than  in  more 
schistose  sediments. 

Habitat  and  Mode  of  Occijrbence. 

The  prevalence  of  these  Dictyosponges  in  the  sandy  sediments  of  the 
Chemung  group  is  in  strong  contrast  to  the  distribution  of  existing  hexactin- 
ellids  over  the  present  sea-bottom.  The  Chemung  formation  of  the  central- 
western  part  of  the  State  of  New  York  is  essentially  a  succession  of  arenaceous 
beds  in  the  condition  of  compact  sandstones  and  sandy  shales,  with  occasional 
layers  of  argillaceous  shale  and,  toward  the  upper  part  of  the  series,  with 
gravelly  conglomerates.  In  all  of  these  sponge  remains  have  been  found,  but 
the  occurrence  of  isolated  specimens  alone  can  not  be  relied  upon  to  determine 


38  BlCTYOSPONGID^ 

the  nature  of  their  habitat ;  for  in  many  instances  there  is  collateral  evidence 
of  the  sponge  having  been  drifted  from  its  proper  position. 

The  known  colonies  of  these  sponges  afford  more  direct  evidence.  The 
colony  of  Hydnoceras  Bailiense,  at  Bath,  lies  overwhelmed  between  banks  of 
sandstone  and  involved  in  a  few  inches  of  a  somewhat  arenaceous  clay-shale. 
Its  existence  was  terminated  abruptly,  as  shown  by  the  fact  that  the  great 
majority  of  the  individuals  are  confined  to  a  very  thin  layer  at  the  surface,  of 
the  sandstone  and  by  the  additional  fact  that  all  have  been  greatly  compressed. 
Now  it  is  found  that  while  the  exposed  surface  of  these  overw^helmed  sponges 
may  be  covered  with  the  slialy  rock,  their  under  surfaces  are  involved  in  the 
sand  and,  further,  occasional  specimens  are  found  quite  buried  in  the  sandstone 
stratum  itself.  It  is  possible  that  this  extensive  sponge  plantation  grew  upon 
the  muddy  bottom  represented  by  the  thin  layer  of  shale,  though  the  data 
more  clearly  indicate  that  it  was  rooted  on  the  sandy  bottom  and  was  over- 
\vhelmed  by  the  influx  of  mud.  A  great  number  of  specimens  of  Hydtioceras 
hotroedenm  have  been  taken  from  the  soil  on  Irish  hill,  near  Bath,  and  though 
the  rock  beds  from  which  they  come  have  not  }^et  been  opened,  the  specimens  are 
uncompressed  internal  casts  in  sandstone,  both  their  form  and  matrix  indicating 
the  absence  of  a  muddy  sediment.  Tie  colony  of  Hydnoceras  kibei'osum  at 
Brown  hill,  near  Cohocton,  that  on  Hamlin's  farm,  near  Naples,  and  the  one  at 
Cotton  hill,  near  Avoca,  in  the  last  of  Avhich  is  a  number  of  diverse  species,  are 
all  in  a  soft  sandy  sediment,  which  is  highly  laminated  and  schistose  but  is  not 
shaly  nor  does  it  show  other  evidence  of  muddy  bottom  except  so  far  as  the  mud 
is  commingled  with  the  sand.  At  Cuba,  Allegany  county,  is  a  small  colony  of 
Dictyospongia  meptrwin  which  occurs  in  a  heavy  sandstone  stratum  and  the 
sponges  have  been  prostrated  not  by  any  change  in  the  nature  of  the  deposit, 
but  rather  by  the  impact  of  the  waves  or  water  currents.  This  is  indicated 
in  the  accompanying  cut  which  shows  three  overthrown  individuals  lying 
side  by  side  and  parallel,  having  evidently  fallen  where  they  grew. 

In  other  cases  there  is  evidence  of  considerable  admixture  of  muddy 
matter  in  the  sediment,  as  in  the  colony  of  Thysanodictya  Edwin-Halli  at 
Wellsville,  where  the  rock  is  soft  and  the  specimens  in  large  part  compressed ; 
but  even  here  the  matrix  is  still  arenaceous  and  many  specimens  are  without 
compression.  In  this  plantation  from  which  several  hiindred  specimens  have 
been  taken,  the  sponges  are  found  at  times  to  retain  their  upright  position, 
thus  indicating  how  gently  the  sediments  have  been  deposited  about  them. 
Fifty  miles  away  from  Wellsville,  which  may  be  regaj-ded  as  the  geographic 
center  of  the  most  prolific  sponge  region  in  the  upper  Chemung  beds,  occurs 


Habitat  and  Occuurknck 


89 


FiauR«  9.    Three  prostrate  iadlTiduals  of  DUtyospongia  tceptrum,  which  were  rooted  Id  a  Bandj  gediment. 
Chemung  sandstone,  Cuba,  N.  Y. 


40  DICTTOSPONGID.E. 

Boiryoilictya  ramom   in  a  soft  sandy  shale,  near  Mansfield,  Tioga   county, 
Pennsylvania,  the  only  locality  at  which  it  has  been  observed. 

Many  of  the  sandstones  of  the  Chemung  series  are  rich  in  molluscan  and 
other  fossils,  constituting  a  profuse  and  diverse  assemblage  of  organic  life,  but 
the  diffusion  of  this  fauna  is  not  uniform  throughout  the  series.  The  fossils 
predominate  in  rather  thin  layers  separated  by  intervals  of  barren  sands.  Tlie 
sponges,  ho\ve\er,  do  not  usually  occur  in  these  highly  fossiliferous  bands. 
In  the  colonies  at  Bath  and  near  Cohocton,  the  species  of  HYDNOCERAg  have 
almost  no  other  organic  associates  and  their  prolific  growth  seems  to  have 
crowded  out  other  forms  of  life.  Occasionally  the  cup  of  a  Taxocrinus  has 
been  found  among  the  sponges  at  the  latter  locality  but  no  other  organisms 
have  been  observed  in  the  colonies  at  Bath.  On  the  other  hand,  very  near  the  base 
of  the  Chemung  group,  on  Hamlin's  farm  in  the  township  of  Naples,  and  not 
over  150  feet  above  the  last  Portage  fauna,  is  a  colony  of  Hydnocekas  where 
the  specimens,  not  well  preserved  on  account  of  the  laminated  condition  of 
the  sediments,  are  associated  with  these  species :  Avihoccelia  iimbonata,  var. 
gregaria,  Oyrtina  Hamilton ensis,  Stropheoclonta  Cayuta,  8.  vai'iahiUs,  S. 
arcuata,  Leptostrophia  perplana  var.  nervosa,  Aviculopecten  cancellatus, 
SpJienotus  sp.,  Arthracantlia  sp.  Three  miles  away,  at  Deyo  basin,  in  the  same 
town,  Hydnocekas  occurs  sparingly,  with  Ceratodictya,  IIyduiodictya, 
SpiHfer  mesastrialis,  Ati'ypa  hystrix  and  Productella  speciosa. 

In  the  upper  part  of  the  formation  throughout  the  region  in  Allegany 
county  which  has  proved  so  prolific  in  Dictyosponges,  these  sjiecies  ai-e  fre- 
quently associated  with  Spirifer  disjunctus,  Prodtwtella,  Oamarotoxhia  and 
other  brachiopods. 

Such  a  habitat  for  these  sponges  as  is  indicated  by  the  foregoing  evidence 
is  dissimilar  to  that  of  the  recent  hexactinellids.  F.  E.  Sciiulze  reports  the 
number  of  species  dredged  by  the  Challenger  Expedition,  from  the  bottom 
of  mud,  clay  or  globigerina-ooze,  as  nearly  twice  that  found  on  all  other  kinds 
of  bottom.  Of  about  one  hundred  species,  five  were  trawled  from  sand  and 
but  two  from  gravel  and  stones.  The  Euplectellid^  and  other  forms  pro- 
vided with  a  root  tuft  were  mostly  found  on  soft  muddy  ground.  Euplectella 
aspergillnm  was  taken  in  mud  at  a  depth  of  95  fathoms,  while  K  crassistellata 
was  dredged  in  radiolarian-ooze  at  the  immense  depth  of  2750  fathoms,  the 
two  species  representing  the  extreme  of  depth  at  which  the  Hexactinellida 
were  found. 

These  ancient  allies  and  precursors  of  Euplectella  seem  to  have  been 
adapted  to  different  conditions  of  growth.     They  flourished  and  attained  their 


BiBMooRAPinr.  41 

highest  development  upon  the  sands  and  arenaceous  muds  of  the  upper 
Devonian,  and  among  them  are  those  most  frequently  provided  with  a  root 
tuft.  The  species  of  the  Keokuk  group  grew  on  a  soft  muddy  l)ottom  like 
many  of  their  modern  representatives,  accompanied  by  a  profuse  growth  of 
crinoids  and  Avith  many  mollusca  and  brachiopoda ;  and  wliile  some  of  tlie 
species  of  the  latter  fauna  are  provided  with  root-tufts,  the  majority  have 
broad  bases  indicating  a  sessile  mode  of  growth  on  solid  objects  such  as 
stones  or  dead  shells, 

HISTORICAL. 

Bibliography  of  the  Dictyospongid^ 
1842. 

1.  T.  A.  Conrad.     Observations  on  tlie  Silurmn  and  Devonian  Systems  of 

the  United  States,  with  descriptions  of  New  Organic  Jienmins.    (Proc. 
Acad.  Nat.  Sci.  Philadelphia,  vol.  8,  pp.  267-8,  pi.  xvi,  fig.  1.) 

Describes  the  genus  PIydnockras  and  the  species  H.  tuherosuni,  stated  to 
be  from  the  Chemung  sandstones  near  Addison,  Steuben  county,  N.  Y.  The 
author's  account  of  this  sponge  was  based  on  an  internal  cast  which  he  con- 
ceived to  represent  an  extravagant  form  of  cephalopoda. 

2.  L.  Vanuxem.     Geology  of  New  York;   lieport  on  the  Third  District, 

pp.  182-3,  fig.  60.  •  ' 

A  brief  description  is  given,  with  a  figure  of  an  incomplete  specimen 
of  Uphantcenia  (Ilyphantcenia)  CJiemungensis,  obtained  from  the  Chemung 
sandstones  "  near  the  south  end  of  the  Ithaca  and  Oswego  railroad."  The 
author  expresses  his  belief  that  the  fossil  was  a  marine  plant.  A  figure 
of  the  original  specimen  will  be  found  among  the  plates  of  this  voliune. 

1852. 

3.  Frederick  McCoy.     A  systematic  Description  of  the  British  Palaeozoic 

Fossils  in   the    Geological  Museum  of  the    University  of  Cambridge, 
p.  62,  pi.  1.  D,figs.  7,8. 

Under  the  name  Tetragonis  Danbyi,  the  author  describes  a  species  of 
DiCTYOSPONGiA  from  the  upper  Ludlow  rocks  of  AVestmoreland,  evidently 
regarding  the  fossil  a  cystidean.  It  is  placed  among  the  EchinalermMa 
in  Morris's  Catalogue  of  British  Fossils,  p.  90,  1854,  Murchison's 
Siluria,  third  edition  p.  536,  1859,  and  fourth  edition,  p.  509,  1872.  In 
Salter's  Catalogue  of  tlie  Cambrian  and  Silurian  Fossils   contained  in   the 


42  DlCTYOSPONGID^. 

Geological  Mmeum  of  the  University  of  Oambridge,  p.  176,  1873,  the  species 

is  placed  among  the  Sponges. 

1862. 

4.  J.  W.  Dawson.     Oti  tlw  Flora  of  tJie  Devonian  Period  in  Northeastern 

America.      (Quarterly  Journal  Geological  Society,  vol.  xviii,  p.    325, 

pi.  xvii.) 
Under  the  name  TJplumkenia  Gliermingensis,  the  author  figures  a  specimen 
of  Hydimceras  taberosam  from  the  Chemung  group  of  New  York  and  sug- 
gests that  such  forms  were  the  cylindrical  stems  of  the  flabellate  fronds 
described  by  Vanuxem  under  the  former  term.  They  are  regarded  "  as  Algce 
with  funnel-shaped  fi'onds,  sometimes  prolonged  into  cylinders,  and  when 
adult,  bearing  fructification  in  tubercles  on  the  sides  of  the  cylinders." 

1863. 

5.  James  Hall.    Observations  upon  the  Genera  Uphantcenia  and  Dictyophytmi, 

with  notices  of  some  Species  from  the  Chemung  group  of  New  YwTc  and 
tlie  Wdverly  sandstone  of  Ohio.  (Sixteenth  Ann.  Rep.  N.  Y.  State 
Cab.  Nat.  Hist.  pp.  84-91,  pis.  iii-v  a.) 

A  description  is  given  of  TIphantmnia  Chemungensis  with  reprint  from 
the  original  cut,  but  without  additional  specimens.  The  genus  Dictyophyton 
is  introduced,  based  upon  the  species  I).  Neivberryi,  and,  at  the  request  of 
Mr.  Conrad  (as  stated  in  a  foot-note  to  page  87),  it  is  allowed  to  replace  the 
term  Hydnoceras.     The  following   species  of  Dictyopiiyton  are  described : 

D.  Newberryi,  nov.  D.  fenestratum,  nov. 

D.  filitextile,  nov.  D.  annulatum,  nov. 

D.  Redfieldi,  nov.  D.  tuberosum,,  Conrad. 

D.  Conradi,  nov.  D.  nodosum,  nov. 

D.  rude,  nov. 
All  of  these  fossils  are  regarded  as  marine  Algae. 

1874. 

6.  E.  P.  Larkin.    Dictyophyton.    (The  Alfred  Student,  May,  p.  53  ;  A  paper 

published  under  the  auspices  of  Alfred  University,  at  Alfred,  Allegany 
county,  N.  Y.) 

Describes  a  number  of  supposed  variations  of  D.  tuberosum  and  new 
species,  viz. :  D.  serratiim,  D.  angulatum.,  D.  lobatum,  D.  cylindricum,  D. 
Imstatmn.  The  descriptions  are  very  incomplete  and,  in  the  absence  of  illus- 
trations, the  forms  intended  can  not  be  recognized. 


BlHLKXlRAJ'IIY.  43 

1879. 

7.  W.  Pii.  ScHiMPER.  ZitteVs  Handbuch  der  Palaeontologie  ;  Part  II.  Pdlaeo- 

phytolologie,  Lief.  l,p.  69. 

The  genera  Dictyophyton  and  Upiiant^nia  are  considered  under  a 
group  of  algae  termed  Dictyophyte^.  In  regard  to  the  genus  Dictvopiiyton, 
however,  the  author  observed :  "  This  regular  lattice-work  is  so  sharply  defined 
and  the  elevated  lines  so  uniformly  straight  that  such  impressions  could  be 
left  in  the  rock  only  by  bodies  composed  of  very  firm  parts.  Among  living 
algae  there  are  none  w^hich  have  such  a  composition.  There  are  indeed  very 
delicate  algae  which  become  hardened  by  calcareous  incrustation,  but  such  an 
ornamentation  as  this  can  not  be  the  result  of  incrustation. 

"  The  question  arises  whether  we  have  not  here  the  siliceous  skeletons  of 
sponges  rather  than  highly  abnormal  fomis  of  algae.  We  know  how  regularly 
the  skeletons  of  the  sponges  are  often  constructed,  and  that  a  quadrate  net- 
work is  not  infrequent  among  them.  It  is  to  be  farther  remarked  that,  as 
stated  by  Hall,  no  trace  of  coal  on  these  remains  has  been  observed." 

Similar  doubt  is  expressed  as  to  the  algous  nature  of  UpHANT.iENLA.. 

1880, 

8.  Ferdinand  Roemer.     Lethm,  Pakfozoica,  Erste  Lieferun^,  pp.  127,  128, 

fig.  3,  ami  p.  304,  fi^-  56. 

Describes  the  genera  Upiiant^nia  and  Dictyopiiyton,  giving  a  figure  of 
Dictyopliytoii  tuberosum  from  Steuben  coimt} .  Both  are  discussed  among  the 
fossil  algae  but  are  regarded  as  doubtfully  of  this  nature.  On  pp.  303,  304, 
the  genus  Tetragonis  and  the  species  T.  MnrcMsonil,  Eichwald,  T.  Danbyi, 
McCoy  and  T.  Eifeliensis,  n.  sp.  are  discussed,  the  last  being  illustrated  by  a 
fragment  of  the  surface  from  the  calcareous  middle  Devonian  shales  of  Gerol- 
stein.     The  genus  is  included  in  the  family  Receptaculitidce. 

1881. 

9.  R.  P.  Whitfield.  Observations  on  tlie  Structure  of  Dictyopiiyton,  and  its 
affinities  with  certain  sponges.  (Amer.  Journ.  Science,  vol.  xxii,  pp. 
53,  54.) 

Upon  comparison  of  specimens  of  Dictyophyton  which  had  been  described 
fix)m  the  sandstones  of  the  Chemung  and  Waverly  groups,  with  the  skeleton 
of    Eupleotella,  the  author   concludes  that    these  bodies   are   sponges,  at 


44  DlCTYOSPONOID^.. 

the  same   time   setting  forth  the  a  priori  evidence   against   their   supposed 
vegetable  nature. 

10.  K.  1*.  WiiiTFiKLD.       On   ilie  Nature  of  Dictyophytan.     (Amer.  Jouru. 

Science,  vol.  xxii,  p.  132.) 

Briefly  refers  to  additional  evidence  of  the  spongous  nature  of  these  fos- 
sils, based  upon  a  specimen  from  the  Keokuk  beds  at  Crawfordsville,  Indiana, 
for  which  the  name  of  Uphantwnia  Dawsoni  is  introduced.  The  specimen 
was  referred  to  Dr.  J.  W.  Dawson  for  examination. 

11.  J.  W.  Dawson.     Note  on  tli€  Structure  of  a  specimen  of  UphantcBnia, 

from  tlie  Collection  of  the  American  Mnsetim  of  Natural  History,  New 
Yorh  City.     (Amer.  Journ.  Science,  vol.  xxii,  pp.  132,  133.) 

This  examination  of  a  pyritized  specimen  from  Crawfordsville,  leads  the 
author  to  conclude  that  it  is  probably  a  sponge 

1882. 

12.  Ward's  Natural  Science  Bulletin,  vol  1,  No.  l,p.  5. 

Gives  a  notice  of  the  finding  of  \he  colony  oi  J) ictyophyton  {Hydnoceras) 
tuherosum,  Conrad,  in  the  town  of  Cohocton,  Steuben  county,  N.  Y. 

13  James  Hall.  Notes  on  the  Family  Dictyospongi[(r\w,  plates  17—20, 
with  explanations.  (Issued  in  advance  of  the  Thirty-fifth  Annual 
Report  N.  Y.  State  Museum  of  Natural  History,  1884.) 

The  explanations  accompanying  these  plates  differ  in  some  respects  from 
those  in  the  final  form  of.  the  paper  in  the  matter  of  generic  identifications. 
The  following  species  take  date  from  this  publication :  Dictyophyton 
[^Ectenodictya^  implexum.,  Clathrospongia  abacus,  Dictyophyton  \Tylodictya7\ 
tenue,  Plirafjmodictya  [Aclceodictya^  excentrica,  P.  scyphus  {=P.  catilliformis, 
Whitfield),  Physospongia  alternata,  Phragmodictya  [  lliysanodictya']  expansa^ 
P.  Grebristriata,  P.  lineata. 

14.  R.  P.  Whitfield.  Pemarhs  on  Dictyophyton  and  Descriptions  of  New 
Species  of  allied  forms  from  ilis  Keokuk  beds  of  Crawfordsville,  Indiana. 
CBuU.  No.  I,  Amer.  Mus.  Nat.  Hist.  pp.  10-20,  pis.  iii,  iv.) 

Reproduces  the  substance  of  observations  made  by  the  author  and  Princ. 
Dawson  (Nos.  9,  10,  11),  redescribes  TJphantoenia  \Physospongici\  Dawsoni 
and  gives  the  first  descriptions  of  Dictyophyton  [^Phragmodictya]  catilliform^ 
and  D.  \Dictyospongia~\  cylindricum,  all  from  the  Keokuk  group  of  Indiana 


BlBLIOOBAPflY,  4r 

[5.  K.  A.  VON  ZiTTKL.  Notizen  uber  fomile  Spongieii  {ProtoHpongia,  Dio 
tyophjtou.,  Corymlla,  StellispmujiM,  VerticilUtes).  (Neues  Jahrb.  fftr 
Mineral,  vol.  2,  pp.  203,  204.) 

From  the  examination  of  a  "  Dictyojjliyton  "  [tuherofrtim,  Conrad]  sent  to 
him  by  Professor  Wakd,  of  Rochester,  the  author  confirms  the  opinion  of 
WiiiTFiKLD  and  IIiNDE,  that  the  fossil  belongs  to  the  sponges  and  regards  it  as 
forming,  with  Pbotosponoia,  a  distinct  family  allied  to  Eupleotella. 

1883. 

16.  James  Hall.     On  the  Relations  of  Dictyophyton,  Phraymodictyum  and 

similar  Forms  with  Uplmntoetiia.     (Proc.  Amer.  Assoc.  Adv.  Sci.,  vol. 
31  (for  1882),  p.  419.     Title  only.) 

17.  Charles  Babkois.     Swr  les  lyictyospongUloB  des  Psammites  dii  Condroz. 

(Annales  de  la  Soc.  Geol.  du  Nord,  vol.  xi,  pp.  80-86,  pi.  1.) 

The  author  records  the  occurrence  of  these  sponges  in  the  upper 
Devonian  sandstones  of  jeuraont,  Brittany,  and  identifies  certain  of  them  with 
the  Dictyophyton  tuberosimi,  Conrad  [^Ilydiioceras  Barroisi\  Another  species 
is  described  as  D.  [^Dictyospon^ia^  Morini.  It  is  suggested  that  these  8i>onge8 
belong  to  the  suborder  Dictyonina,  Zittel. 

18.  George  J.  Hinde.     Catalogue  of  ihe  Fossil  Sponges  in  the  Geological 

Department  of  the  British  Museum,,  pp.  130-132,  pi.  xxviii,  fig.  3. 

The  author  places  the  genus  Dictyophyton  among  the  Dictyonine 
sponges,  under  the  family  Staurodermid^,  and  in  immediate  association  with 
Pkotospongia,  Salter.  The  view  is  expressed  that  the  nine  species  of 
Dictyophyton  described  by  Hall  (1863),  probably  all  belong  io  the  species 
D.  tuberosum,  Conrad,  and  that  the  Tetragonis  [^Primnodictya]  Fifeliensis, 
F.  Roemer,  represents  the  same  genus.  A  figure  is  given  of  a  specimen  from 
Cohocton,  N.  Y.,  probably  the  normal  form  of  the  species  from  the  Brown 
hill  colony.  McCoy's  species,  Tetrar/onis  Danhyi,  is  also  described  as 
Dictyophyton  Danbyi,  and  it  is  observed  that  an  examination  of  the  original 
species  of  Tetragonis  (73  Murchisoni,  Eichwald)  shows  it  to  be  closely 
allied  to  Ischadites. 

19.  Ferdinand  Roemer.     Notiz  ilber  die  Gattwng  Dictyophyton.     (Zeitschr. 

der  Deutsch.  Geolog.  Gesellsch.,  vol.  xxxv,  pp.  704-708,  figs,  a,  b.) 

Describes  and  figures  a  Dictyosponge  from  the  middle  Devonian  of 
Gerolstein,  under  the  name  Dictyophyton   Gerolsteinense.     It  is  shown  in  the 


46  BirrYospoNoiD^ 

following  pages  that  this  species  represents  a  distinct  type  of  generic  structure 
and  the  name  Roemerispon^ua  is  here  introduced  therefor.  The  author 
considere  neither  the  algous  nor  the  spongous  nature  of  these  fossils  established. 

1884. 

20  Ja>[es  Hall.  Descrlptimm  of  tlie  Specie^  of  B^ohsiI Reticulate  Spoiujes^con- 
stitittliuj  tlui  Family  DictyospomjidcB.  (Thirty-fifth  Ann.  Kept.  N;  Y. 
State  Mus.  Nat.  Hist.  pp.  465-481,  pis.  17  (18)-20  (21).) 

This  paper  defines  the  new  genera  Ectenodictya,  Lyrodictya,  Thamno- 
DiCTYA,  PiiRAGAioDicTYA  (1882),  Cleodictya  and  PiiYsospoNGiA  (1882),  and 
gives  brief  descriptions  of  all  the  species  known  at  that  time.  The  following 
species  take  date  from  this  publication :  Dictyophytoti  [Olathrosponyia^ 
Hamiltonense,  D.  \IIydriodictya^  patulum,  D.  [Prismodictya^  prismaticiim, 
D.  [P.]  telam,  D.  [P.]  hacidum,  D.  [P.]  paratkhim,  D.  \^C lathrospongia^ 
irregalare,  D.  [^Geratodictya^  cinctuni,  1).  \CalathoHpoiujia?\  sacculus, 
Ecteiwdictya  {Lyrodictya  ?)  Biuiingtonemis,  Lyrodictya  Romingeri,  Phrag- 
modictya  patellifonnis,  Oleodictya  gloriosa,  G.  Mohri  Pliysospongia 
Colhtti. 

21.  L.  LESQUEREtix.     Description  of  the  Coal  Flora  of  the  Carhoniferous 

Formation  in  Pennsylvania  ami  throughout  the  United  States.  (Second 
Geol.  Surv.  Penn.  Kept.  P,  vol.  iii,  p.  827.) 

The  author,  regarding  the  fossils  known  as  Dictyophyton  as  marine  algae, 
describes  a  new  form  under  the  name  Dictyophyton  ramosnm.  The  species  is 
considered  at  length  in  the  following  pages  uuder  the  name,  Botryodictya 
variwsa. 

1886. 

22.  R.  P.  AVniTFiELD.     Notice  of  a  New   Fossil  Body,  prohahly  a  Sponge 

related  to  Dictyophyton.     (Bull.  No.  8,  Amer.  Mus.  Nat.  Hist.  pp.  346- 
.348,  pi.  XXXV,  figs.  1-9.) 

Under  the  generic  name  Riiombodictyon,  are  described  two  species  from 
the  Hudson  River  slates  at  Kenwood,  near  Albany,  N.  Y.,  of  somewhat  irreg- 
ularly globose  outline  and  having  a  surface  marked  by  parallel  raised  lines 
crossed  by  flexuous  obli(pie  striae.  The  general  aspect  of  these  fossils 
suggests  their  spongous  nature,  but  there  is  no  satisfactory  evidence  of  rela- 
tionship to  the  Dictyosponges. 


BrBLIOORAPHY. 

1887. 


47 


23.  George  J.  Hinde.     A  Monograph  of  the  Bntinh  FohhU  Spongm,  Part  1, 

pp.  1-92,  ph.  i^iii. 

The  author  erects  the  family  Pi{oto8ponoiij.e  as  a  member  of  the  order 
Dictyoiiina.  It  is  made  to  inchide  Pkotosponoia,  Salter,  Pjiokmohella  and 
Plectodebma,  Hinde,  "  and  possibly  also  some  of  the  fonns  included  under 
Dictyophyton,  Hall "  (p.  90).  It  is  also  observed  that  "  by  some  authors  the 
spicules  are  believed  to  be  free  from  each  other  and  thus  of  a  Lyssacine 
character."  On  plate  ii,  are  given  figures  (4,  a-c)  of  Divtuophijton  [Dktyo- 
spongm^  Danhyl,  McCoy  (sp.),  from  the  Ludlow  beds  of  Westmoreland. 

24.  James  Hall.     iVbfe  on  the  Oanvrrence  of  the  Dictyospfmgidm  in  the  State 

of  New  Yorh.      (Sixth  Ann.  Rept.  N.  Y.    State  Geologist,  pp.  36-38, 
map.) 
Gives  a  table  showing  the  distribution  of   the   American   species   and'  a 
map  exhibiting  the   geographical  occurrence  of   specimens  in  western  New 

York  as  known  at  that  date. 

1887. 

25.  F.  E.  ScHULZE.      Report  on  tlie  Heooactinellida  collected  hy  H.  M.  S. 

Cftallenger  during  the  years  1873-1876,  p.  63. 

Reference  is  made  to  the  determinations  and  suggestions  by  Whh'field 
and  Dawson  as  to  the  hexactinellid  structure  of  some  of  the  DictyospongidvE, 
and  the  announcement  Ijy  Walcott  of  the  spongous  character  of  Cyathopiiy- 
cus.     These  forms  are  considered  as  "  fossil  Euplectellid^." 


1888. 

26.  J.  W.  Dawson.       New  Species  of    Fossil  Sporujes  from  Little  Metis, 

Province  of  Quebec,  Canada.      (Canadian   Record  of   Science,  April, 
pp.  49-59.) 

The  substance  of  this  paper  is  included  and  expanded  in  No.  30,  with 
similar  observations  upon  Dictyophyton. 

27.  George  J.  Hinde.     Notes  on  Sponges  from  tJie  Quebec  group  at  Metis 

and  from  tlie    Utica    slude.       (Canadian  Record  of    Science,   April, 
pp.  59-68.) 
The  author   remarks  that  both  names    Cyatiiophycus  and   Dictyophy- 
ton, are  unsuitable  for  sponges  and  that  it  might  prove  advisable  to  reinstate 
the  name  Hydnocekas,  Conrad. 


48  DlCTYOSPONGID^. 

28      George  J.  Hinde.     A  Monograph  of  tlie  British  Fossil  Sporujes,  Part 
II,  jyp.  93-188,  pi  ix. 

A  description  is  given  of  the  genus  Dictyophyton  (p.  126),  and  of  D. 
Danbiji,  McCoy  (sp.)  (p.  128).  It  is  observed  "that  the  genns  Tp;tragonis, 
Eichvvald,  is  merely  a  synonym  of  Ischadites,  Miirch.,  and  its  characters 
markedly  differ  from  those  of  Dictyophyton." 

1889. 

29.  Charles  E.  Beecher.      Brachiospongidce ;    a  Memoir  on  a  group  of 

Silurian  Spoiujes.  (Mem.  Peabody  Mus.  Yale  Univ.  vol.  2,  pt.  1,  pp. 
1-28,  pis.  i-vi.) 

Under  the  family  Dictyospongid^,  Hall,  are  included  (p.  16)  the  genera 
Protospongia  and  Cyaihophycus,  together  with  Hydnoceras  and  its  allies. 

30.  J.  W..  Dawson.     New  Species  of  Fossil  Sponges  from  the  Siluro-Camr 

hrian  at  Little  Metis,  on  ilie  Lower  St.  Laxorence  /  with  notes  hy  Dr. 
G.  J.  Hinde.  (Trans.  Roy.  Soc.  Canada,  vol.  VII,  Sect,  iv,  jip.  81-55, 
pi.  iii.) 

In  this  work  a  number  of  remarkable  forms  of  the  PROTOspoNGiDiE  are 
described,  together  with  other  sponges  whose  structure  is  too  imperfectly 
known  to  ensure  their  position  among  recognized  families.  A  distinction  is 
pointed  out  between  the  structure  of  Cyathophycus  and  the  Uplmntwnia 
(Phijsospomjia')  Dawsoni,  but  that  genus  is  ascribed  to  the  Dictyospongid.e, 
and  referred  to  the  sub-order  iJictijonina.  In  a  note  upon  page  39,  Dr.  IIinde 
suggests  that  there  is  good  reason  for  regarding  Protosponcjia  and  allied 
palaeozoic  genera  as  Lyssacine,  rather  than  Dictyonine  hexactinellids. 

31.  M.  Neumayr.     Lie  Stamms  des  Thierreichs,  p.  228. 

The  author  places  Dictyophyton  among  the  Dictyonine  hexactinellids 
but  suggests  that  it  may  be  Lyssacine. 

32.  E.  O.  Ulrich.     Preliminary  Lescription  of  New  Lower  Silurian  Sponges. 

(American  Geologist,  vol.  3,  p.  233.) 

The  species  Pauffella  filosa,  Ulrich,  of  the  Trenton  group,  which  the 
author  here  suggests  may  be  a  Dictyosponge,  has  not  the  characters  of  this 
family. 


BiBLKXiKAPiry.  49 

1890, 

33.  James   Hall.      Fossil  DictyospmujixhB  of  the  Devonian   and  Carhf/ni- 

ferous  Formations :  New  Forms  of  DiHi/ospon^/itlfP,  front  tJie  rocks  of 
tJie  Cliemmuj  group.  (Ninth  Ann.  Kept.  N.  Y.  State  Geol.  pp.  5H-60; 
Also  pu1)lished  in  Bull.  Geol.  Soc.  America,  vol.  1,  p.  22;  and  in 
Forty-third  Ann.  Kept.  N.  Y.  State  Mus.  pp.  258-2(32.) 

In  this  paper  two  new  genera  are  introduced:  A(Ti\(»i)icrYA  and 
Cryptodictya.     The  following  species  are  described  for  the  first  time : 

Dictijopliyton  sceptrum.  D.  tomaculmn. 

D.  vascellum.  I)  ?  {PhragmxxUclija)  JIalli. 

D.  Rttndalli.  Actinodictija  placenta. 

D.  scitum.  Cryptodictya  Alleni. 

D.  Amalthea. 

1892. 

34.  Clemens   Schluter.      Protospongia   rlienana.      (Zeitsch.  der    Deutsch. 

Geolog.  Gesellsch.,  pp.  615-618,  figure.) 

Under  the  above  name  is  described  a  species  from  the  lower  Devonian 
Hunsriick  slates,  near  Geniiinden,  which  is  represented  by  a  large,  (£uite 
incomplete  specimen,  showing  quadrate  reticulations  in  a  diminishing  series. 
The  quadrules  appear  to  be  formed  by  bundles  of  spicular  filaments  associated 
with  staiiractins.  The  species  is  undoubtedly  a  Dictyosponge  but  its  generic 
relations  are  uncertain. 

1893. 

35.  N.  H.  WiifciiELL  and  C.  Sciiuciiert.     Sponges,   Oraptolites  and  Corals 

from,  the  Ixyiver  Silurian  of  Minnesota.  (Extract  from  vol.  iii.  Final 
Report  of  the  Minnesota  Geol.  Surv.  p.  75,  pi.  F,  figs.  16-20.) 

Raujfella  filosa,  Ulrich,  is  redescribed  and  placed  under  the  family 
DicTYSPONGiDyE.     It  IS  asscrtcd  by  Rauff  that  this  is  not  an  organism. 

36.  Hermann  Rauff.      Palaeospongiologie,  Part  /,  Lieferungen  1-4,  pp. 

1-232.     (Palaeontographica,  vol.  xl.) 

In  this  compi'ehensive  undertaking,  which  plans  an  exhaustive  treatise  of 
all  fossil  sponges,  the  author  opens  with  an  elaborate  bibliography  of  the 
subject,  embracing  591  references  up  to  the  year  1892,  following  this  with  an 
analysis  of  the  work  of  those  writers  whose  investigations  have  been  of 
especial  influence  in   the  study  of  these  organisms.     A  long  chapter  on  the 


50  DlCTYOSPONOID^ 

morpliology  of  the  sponges,  both  of  their  soft  and  skeletal  parts,  with  abun- 
dant illustration,  is  constructed  with  especial  reference  to  the  fossil  forms.  In 
the  systematic  treatment  of  these  bodies  the  suborder  Lyssacina  is  made  to 
include  eight  families,  as  follows :     Protospongid^,  Hinde,  DictyospongiduE, 

Hall,    PLKCTOSPOXCilDyE,  Rauff,  EuPLECTELLIDyE,  Gray,  ASCONEMATID^E,    SchulzC, 

RossELLiD.E,  Schulze,  HYALONEMATiDyE,  Scliulze,  Bbachiospongidye,  Beeclier. 

The  author  is  the  fli'st  to  establish  distinctive  characters  between  the 
Protospongid^  and  the  Dictyospongid^  and  to  suggest  the  probability  that 
Cyathophycus,  Plectoderma  and  Piiormosella  are  not  safely  placed  with 
either  of  these  families ;  hence  the  erection  of  the  family  Plectospongid^. 
The  final  chapter  in  this  portion  of  the  work  relates  to  the  mode  of  preserva 
tion  of  the  fossil  sponges. 

1894. 

37.     Hermann  Rauff.     Palaeospongiohgie,  Part  1,  Liefermigen  5, 6, pp.  233— 
.  3^.6,  ph.  i-xvii.     (Palaeontographica,  vol.  xl.) 

This  instalment  of  the  same  work  is  devoted  to  descriptions  of  palaeozoic 
species.  The  various  forms  of  the  Protospongid^  described  by  Dawson,  and 
the  Plectospongid^e  (Cyatiiopiixcus,  Teganium),  are  described  and  in  many 
instances  illustrated  with  new  figures.  On  page  249,  under  the  family 
DicTYOspoNGiD^,  is  described  as  Didyopliytra  (.^)  Walcotti,  Rauff,  a  new 
species  from  the  Utica  slate  at  Holland  Patent,  N.  Y.  Dictyopuytka  is  evi- 
dently a  proposed  modification  of  the  term  Dictyophyton  ;  as  a  substitute  for 
the  latter  it  is  imobjectionable  etymologically  and  in  its  application  to  these 
fossils,  but  it  is  objectionable  historically  if  xised  in  the  broad  sense  in  which 
the  term  Dictyophyton  has  heretofore  been  applied.  The  definition  of  the 
genus  is  deferred  to  a  later  part  of  the  work  and  its  value,  therefore,  can 
not  be  determined  xuitil  the  author  makes  known  upon  what  forms  of 
"  Dictyophyton  "  the  genus  is  based. 

Dictyophytra  (?)  Walcotti,  (pi.  iv,  figs.  1,  2)  is  founded  upon  the  frag- 
ment of  a  large,  coarse-meshed  species,  quite  distinct  in  size  and  texture  from 
its  associate,  Gyatli&phyaus  feticulatus.  That  it  is  a  true  Dictyosponge,  how- 
ever, and  not  a  Cyathophycus,  can  not  be  regarded  as  established. 

The  7th  and  8th  Lieferungen  of  this  work,  dated  1895,  contain  nothing 
especially  relating  to  the  Hexactinellida. 

We  gi'eatly  regret  that  Dr.  Rauff's  observations  upon  the  Devonian  and 
Carboniferous  Dictyosponges  have  not  been  received  as  this  volume  goes  to 
press. 


Bibliography.  51 

38  John  M.  Clarkk.  JVew  ov  Rare  Species  of  FoHsiUfi-om  the  Jwrimns  of 
the  Livonia  Salt  Shaft.  (Thirteenth  Ann.  Kept.  N.  Y.  State  Geohtgist, 
vol.  1,  p.  177,  pi.  IV,  figs.  19-22.  Also  in  Forty-seventh  Annual 
Report  N.  Y.  State  Museum,  p.  371.) 

Descriljes  as  Hyahsteliu?  Marcellia,  a  species  from  the  hituminous 
shales  of  the  Marcellus  epoch,  the  generic  characters  of  which  are  (piite 
obscure,  though  the  presence  of  broad  unresolvable  spicules  of  great  size  and 
having  the  form  of  blades  of  grass,  indicates  a  structure  unlike  any  heretofore 
described.     The  original  fonn  of  the  species  is  unknown. 

1895. 

39.  K.  A.  v.  Zi'iTEL.     Grundzilge  der  Palaeontf}lo(/ie,  jyp.  52  et  seq. 

The  author  here  recognizing  the  two  suborders  of  the  Hexactinellida, 
Lyssacina  and  Dictyonina,  places  under  the  former  the  three  families,  Pro- 
TospoNGiD^,  DicTYospoNGiD^  and  Plectospong iD^.  The  definition  of  each  of 
these  families  is  essentially  that  given  by  Rauff. 

1896. 

40.  K.  A.  VON  ZiTTEL.     Text-hooTc  of  Palaeontology ;  Translated  and  edited  hy 

Charles  P.  Eastman,  pp.  54  et  seq. 

The  subject  is  here  treated  as  in  No.  39,  of  which  this  is  a  translation 
and  enlargement. 

41.  George  II.  Girty.      A  Pevision  of  the  Sponges  and   Coelenterates  of 

the  Laiver  Helderberg  Fauna  of  New  Yorh.  (Fourteenth  Ann.  Kept. 
N.  Y.  State  Geologist,  pp.  258  et  seq.  Also  Forty-eighth  Annual 
Report  N.  Y.  State  Museum,  p.  261.) 

For  certain  hexactinellid  spicules  from  the  Lower  Helderberg  rocks  the 

author  introduces   the  generic  term  Lysactinella  and  makes  use  of    two 

specific  names,  one  applied  to  smooth  spicules,  the  other  to  those  having 

thorny  rays. 

1897. 

42.  Ferdinand  Roemer  and  Fritz  Frech.      Lethcea  Geognostica,  1  TJieil ; 

Lethcea  Palceozoica,  1  Band 

This  is  a  continuation  of  Roemer's  work,  the  plates  of  which  appeared  in 
1876  and  the  first  instalment  of  the  text  in  1880.  So  far  as  the  Dictyosponges 
are  concerned  no  additions  or  changes  are  made  from  the  treatment  of  the 
genera  as  cited  under  No.  8. 


52  DiCTYOSPONGID.E. 

CLASSIFICATION  OF  THE  DICTYOSPONGID^. 

The  manifold  vuiiations  in  form  expressed  by  these  sponges  are  accom- 
panied by  certain  structural  diversities  ^vhich,  however,  do  not  lead  far  a-field 
from  the  comprehensive  definition  elsewhere  given  of  the  indicial  characters 
of  the  Family.  It  has  been,  in  our  judgment,  wise  not  to  endow  any  of  these 
structural  characters  Avith  so  high  value  as  to  necessitate  the  erection  of 
additional  Family  terms,  and  experience  indicates  that  few  thin-walled  siliceous 
sponges  of  the  Devonian  and  Carboniferous  faunas  will  fail  to  find  a  place 
within  the  limits  of  this  Family,  Dictyospongid^.  The  structural  variations  of 
chief  importance  are  herewith  summarized  and  utilized  as  the  basis  of  a 
subordinate  classification. 

Subfamily   I. 

DlCTYOSPONGIIN^. 

Simple,  vase-shaped  or  elongate,  smooth,  prismatic,  nodose  or  annulated 
forms  with  obcouical  base  ;  with  basal  prostalia ;  Avith  or  without  marginalia ; 
no  basal  disk ;  primitive  form  a  regulai'  obcone. 

Dictyospongia,  Helicodictya, 

(?)  Lysactinella,  lihabdosispongia, 

Hydriodictya,  Ceratodictya, 

Prismodictya,  Clathrospongia, 

Gongylospongia,  Lebedictya, 

Hydnoceras,  Lyrodictya, 

(?)  Tylodictya,  (?)  Mastodictya. 

BoTBYODicTYA,  Avliich  appertains  to  this  group  in  its  principal  characters, 
is  aberrant  in  the  possession  of  a  more  or  less  complete  diaphragm,  not  at  the 
base  but  near  the  middle  of  the  sponge,  just  Ijelow  the  nodlferous  body. 

Subfamily  II. 

Thysanodictyin^. 

Cylindrical  or  expanding  cups,  with  basal  disk  and  periloph. 
Thysanodictya,  Phragmodictya, 

Arystidictya,  (?)  Griphodictya. 


Classification.  53 

Subfamily  III. 

Calathosponoiin^. 

Expanding  forms  with  smooth,  tufted  or  nodose  surface  and  obtuse  basal 
extremity  withoiit  diaphragm  or  periloph. 

Calathospongia,  Clepsydrospongia, 

Thamnodictya,  Cleodictya. 

The  genus  Clepsydeospongia  presents,  in  its  tufted  nodose  surface, 
the  aspect  of  Hydnoceuas,  but  is  as  broad  at  the  basal,  as  at  the  apertural 
extremity. 

Cleodictya  is  tentatively  placed  in  this  division  on  account  of  its  appar- 
ently broad  base. 

Subfamily  IV. 
Physospongiin^. 

Stout  cups  with  a  single  or  double  series  of  strong  horizontal  and  vertical 
spicular  bundles,  tufted  at  their  intersections ;  meshes  covered  by  thin  bul- 
late  expansions  or  depressions  of  the  reticulum  (parenchymalia) ;  no  Ijasal 
disk  (?). 

Physospongia,  Roemerispongia. 

Subfamily  V. 

HYPIIANT^ENIINiE. 

Large  saucer-shaped  fronds  composed  of  a  single  series  of  ligidate  con- 
centric and  radial  reticulating  bands ;  meshes  partially  closed  by  parenchymal 
reticulum;  attached  by  basal  apex  (basalia?). 

Hyphantsenia. 

Subfamily  VI. 

Hallodictyin^. 

Irregularly  growing,  explanate  or  frondiform  sponges,  with  smooth  or 

tufted  nodate  surface. 

Hallodictya,  Cryptodictya. 

Actinodictya, 

Subfamily   VII. 

AOLII'HODICTYIN^. 

Discoid  sponges,  rapidly  expanding  and  contracting ;  attached  by  apex 
(basalia  ?)  ;  osculum  large. 

Aglithodictya. 


Synoptic  List  of  Genera  and  Species  of  the  Dictyospongidae. 


Subkingdom 
SPONGIyE. 

Class 
SILICEA,  Gray. 


Order 
HEXACTINELLIDA,  O.  Schmiot. 

Suborder 
LYSSACINA,  ZiTTEL. 

Family 
DICTYOSPONGID^,  Hall. 

Subfamily    DICTYOSPONGIINiE,  subfam.  nov. 
Genus  Dictyospongia,  gen.  nov. 

D.    Danbyi,  McCoy  (sp.).  D.  siraea,  sp.  nov. 

D.?  Marcellia,  Clarke  (sp.).  D.  Almondensis,  sp.  nov. 

D.    haplea,  sp.  nov.  D.  bacteria,  sp.  nov. 

D.    sceptrum,  Hall  (sp.).  D.  Morini,  Barrois  (sp.). 

D.    lophura,  sp.  nov.  D.  cylindrica,  Whitfield  (sp.). 

D.    charita,  sp.  nov.  D.  stylina,  sp.  nov. 

D.    eumorpha,  sp.  nov. 

Genus  Lysactinella,  Girty. 

L.  '  Gebhardi,  Girty.  L.    perelegans,  Girty. 

Genus  Hydriodictygi,  gen.  nov. 
H.   patula,  Hall  (sp.).  H.   nephelia,  sp.  nov. 

H.    cylix,  sp.  nov. 

Genus  Prismodictya,  gen.  nov. 

P.    palsea,  sp.  nov.  P.    banano,  sp.  nov. 

P.    telum,  Hall  (sp.).  P.    amicitiae,  sp.  nov. 

u 


List  ok  Gknkua  and  SpKcres. 


Ui) 


P.    baculum,  Hall  (sp.). 
spectabilis,  sp.  nov. 
parallela,  Hall  (sp.). 
prismatica.  Hall  (sp.). 
Allegania,  sp.  nov. 
choanea,  sp.  nov. 
corynia,  sp.  nov. 
filitextilis,  Hall  (sp.). 


P. 
P. 
P. 
P. 
P. 
P. 
P. 


P.  aulophia,  sp.  nov. 

P.  ptionia,  sp.  nov. 

P,  cercidea,  sp.  nov. 

P.  cithara,  sp.  nov. 

P.  narthecia,  sp.  nov. 

P.  CJonradi,  Hall  (sp.). 

P.  polyhedra,  sp.  nov. 


Gr.     Marshi,  sp.  nov, 


Genus  Gongylospongia,  gen.  nov. 


H.  legatum,  sp.  nov. 

H.  gracile,  sp.  nov. 

H.  tuberosum,  Conrad. 

H.  tuberosum, 

var.  glossema,  var.  nov 

H.  Bathense,  sp.  nov. 

H.  botroedema,  sp.  nov. 

H.  Avoca,  sp.  nov. 

H.  phymatodes,  sp.  nov. 

H.  hypastrum,  sp.  nov. 


Genus  Hydnoceras,  Conrad. 

H.  eutlieles,  sp.  nov. 

H.  raultinodosura,  sp.  nov. 

H.  Lutheri,  sp.  nov. 

H.  rhopalura,  sp.  nov. 

H.  variabile,  sp.  nov. 

H.  nodosum.  Hall  (sp.). 

H.  anthracis,  sp.  nov. 

H.  eumeces,  sp.  nov. 

H.  Barroisi,  nom.  nov. 

H.  Jeumontense,  sp.  nov. 


Genus  Botryodictya,  gen.  nov. 

B.  ramosa,  Lesquereux  (sp.).  B.    (or  Tylodictya)  sp.  nov. 

Genus  Tylodictya,  gen.  nov. 
T.    Warrenensis,  sp.  nov.  T.?  tenuis,  Hall  (sp.). 

Genus  Helicodictya,  gen.  nov. 
H.   tryjiania,  sp.  nov.  H.?  Scio,  sp.  nov. 

H.?  Concordia,  sp.  nov. 

Genus  Rhabdosispongia,  gen.  nov. 
R.    Amalthea,  Hall  (sp.).  R.    Condroziana,  sp.  nov. 

Genus  Ceratodictya,  gen.  nov. 

C.  Carpenteriana,  sp.  nov.  C.    annulata.  Hall  (sp.). 
C.    cincta.  Hall  (sp.).  C.    centeta,  sp.  nov. 

C.    zonata,  sp.  nov. 


56  DlCTYOSPONOID^. 

Genus  Clathrospongia,  Hall. 

C?  Hamiltonensis,  Hall  (sp.).  C.    irregularis,  Hall. 

C.    fenestrata,  Hall.  C?  desmia,  sp.  nov. 

C.    vascellura,  Hall.  C.    abacus,  Hall. 

C?  tomaculuui,  Hall.  C.    caprodonta,  sp.  nov. 


L.    crinita,  sp.  nov. 


Genus  Lebedictya,  gen.  nov. 

Genus  Lyrodictya,  Hall. 
L.    Romingeri,  Hall.  L.?  Burlingtonensis,  Hall  (sp.). 

Genus  Mastodictya,  gen.  nov. 
M.    osculata,  sp.  nov. 

Subfamily  THYSANODICTYINiE,  subfam.  nov. 
Genus  Thysanodictya,  gen.  nov. 
T.    Edwin-Halli,  Hall  (sp.).  T.    quasilluin,  sp.  nov. 

T.    pcecilus,  sp.  nov.  T.    scypliina,  sp.  nov. 

T.    rudis,  Hall  (sp.).  T.    hermenia,  sp.  nov. 

T.    Randalli,  Hall  (sp.).  T.    apleta,  sp.  nov. 

T.    saccus,  sp.  nov.  T.    turricnla,  sp.  nov. 

T.    Johnstoni,  sp.  nov.  T.    expansa,  Hall  (sp.). 

Genus  Phragmodictya,  Hall. 
P.    catilliforinis,  Whitfield  (sp.).  P.?  lineata,  Hall. 

P.?  crebristriata,  Hall. 

Genus  Arystidictya,  gen.  nov. 

A.    elegans,  sp.  nov.  A.    nodifera,  sp.  nov. 

Genus  Acloeodictya,  gen.  nov. 
A.    marsipus,  sp.  nov.  A.?  excentrica,  Hall  (sp.). 

Genus  Griphodictya,  gen.  nov. 
G.   epiphanes,  sp.  nov. 

Subfamily  CALATHOSPONGIINiE,  subfam.  nov. 
Genus  Calathospongia,  gen.  nov. 
C.    Redfieldi,  Hall  (sp.).  C?   sacculus,  Hall  (sp.). 

C.    carceralis,  sp.  nov.  C.    amphorina,  sp.  nov. 

C.    Carlli,  sp.  nov.  C?  magnifica,  sp.  nov. 

C.    Tiffanyi,  sp.  nov. 


List  of  Genkra  axd  Si'fx;rES. 

Genus  Thamnodictya,  Hall. 
T.    Newberryi,  Hall.  T.    Oi-toni,  sp.  nov. 

Genus  Clepsydrospongia,  gen.  nov. 

C.    matutina,  sp.  nov. 

Genus  Cleodictya,  Hall. 

C.    Claypolei,  sp.  nov.  C.    Mohri,  Hall. 

C.    gloriosaj  Hall. 

Subfamily  PHYSOSPONGIINiE,  subfain.  nov. 
Genus  Physospongia,  Hall. 
P.    Dawsoni,  Whitfield  (sp,).  P.     alternata,  Hall. 

P.    Colletti,  Hall.  P.    multibursaria,  sp.  nov. 

Genus  Roemerispongia,  gen.  nov. 
R.    Gerolsteinensis,  F.  Roemer  (sp.). 

Subfamily  HYPHANTiCNIINie,  subfara.  nov. 

Genus  Hyphantenia,  Vanuxem. 
H.    Chemungensis,  Vanuxem. 

Subfamily  HALLODICTYINiE,  subfam.  nov. 

Genus  Hallodictya,  gen.   nov. 
H.    Sciensis,  sp.  nov.  H.    Cottoniana,  sp.  nov. 

Genus  Actinodictya,  Hall. 

A.    placenta,  Hall. 

Genus  Cryptodictya,  Hall. 
C.    Alleni,  Hall. 

Subfamily  AGLITHODICTYINiE,  subfam.  nov. 

Genus  Aglithodictya,  gen.    nov. 
A.    numulina,  sp.  nov. 


DubitandcB  sedis. 
?  Genus  Ectenodictya,  Hall. 


?  E.    iinplexa,  Hall. 


58 


DiCTYOSl'ONGIDJE. 


Generic 

Distribution  of 

the  Dictyospongic 

ae. 

1 

1 

1 
4 

/famt/fon 

r 
2 

1 

^ 

1 

GENERA 

1 
1 

1 

Dictyospongia, 

? 

fc. 

Lysaetinella, 

Hydriodictya, 

Prismodictya, 

^- 

Hydnoceras, 

>■ 

Botryodictya, 

IJelicodictya, 

Rhabdosispongia, 

Ceratodictya, 

Clatbrospongia, 

— 

m^ 

Lebedictya, 

-«-. 

Lyrodictya, 

Mastodictya, 

Thysanodictya, 

^ 

— 

Generic  Distribution. 


'   59 


Generic  Distribution  of   the  DicTYOspoNorOif:  (Conchided). 


GENERA 


I 


1 


tf<aml/k 


'on 


"I 

I 


«5 

1 


C5 


i 
1 


Arystidictya, 

Phraginodictya, 

Griphodictya, 

Calathospongia, 

Tliainnodictya, 

Clepaydrospongia, 

Cleodictya, 

Roeinerispongia, 

Physospongia, 

Hyphantaenia, 

Hallodictya, 

Actinoflictya, 

Cryptodictya, 

Aglithodictya. 


^ifti— 


00 


DlOlTOSPONOin^E. 


Table    Showing     the    Geological    Distribution    of    the 

Dictyospongidai. 


2 
3 
4 
5 
G 
7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 
23 
24 
25 
2G 
27 
2.8 
29 
30 
31 
32 
33 
34 
35 
30 
37 
38 
39 
40 


NAME  OF  SPECIES. 


Dictyospongia    Danbyi,   McCov 

(SP.).     ••; 

DJ  Marcellia,  Clarke  (sp.). . . 

D.  haplea,  sp.  nov 

D.  sceptrum,  Hall  (sp.) 

D.  lophura,  sp.  nov 

D.  cliarita,  sp.  nov 

D.  eumorpha,  sp.  nov 

I),  sirfea,  sp.  nov 

I).  Almondensis,  sp.  nov.  . . . 

D.?  bacteria,  sp.  nov 

D.  stylina,  sp.  nov 

D.  cylindrica,  Whitfield  (sp.) 
D.  Morini,  Barrois  (sp.) .... 
Lysactinella  Gebhardi,  Girty 

L.  perelegans,  Girty   

Ilydriodictya  patula,  Hall  (sp.) 

H.  cylix,  sp.  nov 

H.  nepbelia,  sp.  nov 

Prismodictya  paltea,  sp.  nov. 

P.  telum,  Hall  (sp.) 

P.  baculum,  Hall  (sp.) 

P.  speetabilis,  sp.  nov 

P.  parallela,  Hall  (sp.) 

P.  prismatica,  Hall  (sp.) 

P.  Allegania,  sp.  nov 

P.  choanea,  sp.  nov 

P.  corynia,  sp.  nov 

P.  lilitextilis,  Hall 

P.  banano,  sp.  nov 

P.  amicitiiB,  sp.  nov 

P.  auiopbia,  sp.  nov 

P.  ptionia,  sp.  nov 

P.  cercidea,  sp.  nov .  , 

P.  cithara,  sp.  nov 

P.  narthecia,  sp.  nov 

P.  Conradi,  Hall  (sp.) 

P.  polybedra,  sp.  nov 

Gongylospongia  Marsbi,  sp.  nov 
Hydnoceras  legatum,  sp.  nov . 
H.  gracile,  sp.  nov 

•From  th'<  PMiurnites  da  Cindroz,  Fraoce. 


» 

5 


a. 


Hamilton 
Group. 


X 
X 
X 
X 
X 
X 
X 


X 
X 
X 

X 
X 

X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 


GkoUXUCAI,    DiHTHIItlTrON,  61 

Table  Showing  the  Geological  DiHtrihutioii  of  the  Dictvosponoid.e  (Cont'd). 


41 

42 

43 
44 
45 
40 

47 
48 
49 
50 
51 
52 
53 
54 
55 
56 
57 
58 

59 
CO 
61 
G2 
G3 
G4 
65 

66 
67 

68 
69 

70 
71 
72 

73 

74 
75 
76 
77 
78 
79 
80 
81 
82 
83 
84 

85 
86 
87 


H.  tuberosuu),  Conrad 

H.  tuberosuui,     var.     glossetiia, 

var.  iiov 

II.  Bathense,   sp.  nov 

II.  botrcedeina,  sp.  iiov 

11.  Avoca,  sp.  nov 

H.  phymatodes,  ep.  nov 

II.  hypastmm,  sp.  nov 

11.  eutheles,  sp.  nov 

II.  multinodosuiii,  sp.  nov... 

H.  Lutheri,  sp.  nuv 

II.  rhopaluin,  sp.  nov    

II.  variabile,  sp.  nov 

11.  nodosum,  llall  (sp.) 

II.  antbi-acis,  sp.  nov 

II.  eameces,  sp.  nov 

H.  Barroisi,  noni.  nov 

II.  Jeumontense,  sp.  nov 

Botryodictya     raniosa,     Lesque- 

reux  (sp.)    

B.  (or  Tylodictya)  sp.  nov 

Tylodictya  Warrenensis,  sp.  nov 

TJ  tenuis,  Hall  (sp.) 

Ilelicodictyatrypania, sp.  nov.  - 

II.?  Concordia,  sp.  nov 

II.?  Scio,  sp.  nov 

Rhabdosispongia  Amaltliea,  Hall 

^(«P-) 

K.  Condroziana,  sp.  nov 

Ceratodictya    Carpenteriana,    sp 

nov 

C.  cincta,  Hall  (sp.) 

C.  zonata,  sp.  nov 

C.  annulata.  Hall  (sp.) 

C.  centeta,  sp.  nov 

Clatbrospongia  (?)  Ilamiltonensis, 

Hall  (sp.) 

C.  fenestrata,  Hall 

C.  vascellum,  Hall 

C  ?  tomaculum,  Hall 

C.  irregularis,  Hall 

C?  desmia,  sp.  nov 

C.  abacus.  Hall 

C.  caprodonta,  sp.  nov 

Lebedictya  crinita,  sp.  nov 

Lyrodictya  Romingeri,  Hall  .... 
L  ?  Burlingtonensis,  Hall  (sp.). . . 
Mastodictya  osculata,  sp.  nov.  .  . 
Thysauod'ictya  Edwin-llalli,  Hal! 

T.  poecilus,  sp.  nov ■ . 

T.  riidis.  Hall  (sp) 

T.  Eandalli,  Hall  (sp.) 

•  From  the  Psammltes  du  Condroi,  France. 


7 
X 

X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 


X 
X 
X 
X 

X 


X 
X 
X 
X 


X 
X 


10 


X 
X 


62 


DlCTTOSPONGID,«. 


Table  Showing  the  Geological  Distribution 

of  the  DicTYospoNGiD^  {CoiwX 

d). 

88 
89 
90 
91 
92 
93 
94 
95 
96 

97 

98 
99 
100 
101 
102 
103 
104 

105 
106 
107 
108 
109 
110 
111 
112 
113 

114 
115 
116 
117 

118 

119 

120 
121 
122 

123 

124 
125 
126 
127 
128 

T.  saccus,  sp.  nov 

T.  Johnstoni,  sp.  nov 

T.  quasillum,  sp.  nov 

T.  scyphina,  sp.  nov 

T.  hermenia,  sp.   nov 

T.  apleta,  sp.   uov 

T.  turricula,  sp.  nov 

T.  expansa.  Hall  (sp.) 

Phragmodictya         catilliformis, 

Whitfield  (sp.) 

P.?  crebristriata,  Hall 

P.?  lineata.  Hall 

AclcBodictya  marsipus,  sp.  nov.  . 

A.?  cxcentrica.  Hall  (sp.) 

Arystidictya  elegans, sp.  nov. . . . 

A.  nodifera,  sp.  nov 

Griphodictya  epiplianes,  sp.  nov. 
Calathospongia    lledfieldi.    Hall 

(sp-)   

C.  cai-ceralis,  sp.  nov 

C.  Carlli,  sp.  nov 

C.  Tiffany i,  sp.  nov 

C?  sacculus,  Hall  (sp.) 

C.  amphorina,  sp.  nov 

C?  magnifica,  sp.  nov 

Thamnodictya  Newberryi,  Hall. 

T.  Ortoni,  sp.  nov 

Clepsydrospongia    matutina,    sp. 

nov 

Cleodictya  Claypolei,sp.  nov.. . . 

C.  gloriosa.  Hall 

C.  Mohri,Hall 

Roemerispongia    Gerolsteinensis, 

Roemer  (sp.) 

Pliysospongia  Dawsoni,  W  liitfield 

(sp-)  •    

P.  Colletti,  Hall 

P.  alternata.  Hall 

P.  multibursaria,  sp.  nov 

Hypliantaenia         Chemungensis, 

Vanuxcm 

Hallodictya  Sciensis,  sp.  nov. . . . 

II.  Cottoniana,  sp.  nov 

Actinodictya  placenta.  Hall 

Cryptodictya  AUeni,  Hall 

Aglithodictya  numulina,  s  ).  nov. 
Ectenodictya  implexa,  Hal 

1 

2 

3 

4 

s 

X* 

6 
X 

7 
X 
X 
X 
X 
X 
X 
X 

X 
X 

X 
X 
X 
X 
X 
X 

•  • 

8 
X 

X 
X 
X 
X 
X 

X 
X 

X 
X 

0 

10 

X 
X 
X 
X 

X 

iC 

X 

X 

.  . 
X 
X 

X 
X 

X 
X 

*  From  the  middle  Deronlan  of  Qerolstein  in  the  EifeL 


DESCRIPTIONS 

OF  Genera  and  Species. 

Of  the  considerable  number  of  genera  liere  discuHsed,  several  are  repre- 
sented by  many  species  whose  differentials  may  seem  obscure  or,  to  some 
students,  overestimated.  In  fixing  specific  values  within  a  given  genus  of  such 
variable  organisms,  it  has  been  found  necessary  to  depend  very  largely  upon 
variations  in  form  as  the  paramount  specific  character.  If,  in  this  practise, 
what  may  seem,  in  such  prolific  genera  as  Prismodicita,  Hydnoceras  and 
Thysanodictya,  to  be  traits  of  lesser  value,  are  dignified  with  specific  distinc- 
tion, it  is  because  such  distinctions  are,  in  the  existing  state  of  our  knowledge, 
the  truest  expression  of  these  relations,  as  well  as  the  most  convenient 
denomination.  To  future  investigations  with  the  aid  of  more  extensive 
material,  must  be  assigned  the  detennination  of  fundamental  type  forms,  in 
terms  of  which  all  of  these  manifold  variations  may  ha  expressed. 


SPECIES  OK  THE  SILURIAN. 

DICTYOSPONGIA,  gen.  nov. 
(For  description  see  page  72.) 

DicTYOsPONGiA  Danbyi,  McCoy  (sp.). 

1855.     Tetragonis  Danbyi,  McCoy.     British  Palaeozoic  Fossils,  p.  62,  pL  1  D, 

figs.  7,  8. 
1880.     Tetragonis  Danhyi,  F.  Roemer.     Lethaea  Palaeozoica,  Part  1,  p.  304. 
1883.     Dlclyophijton  Danbyi,  Ilinde.     Cat.  Fossil  Sponges  British  Museum, 

p.  131. 

1887.  Dictyophyton  Danbyi,    Hinde.     Monogr.  British  Fossil  Sponges,  Part 

1,  pi.  ii,  figs.  4  a-c. 

1888.  Dictyophyton  Danbyi,  Ilinde.     Monogr,  British  Fossil  Sponges,  Part 

2,  p.  128. 

"  Sponges  subovate  or  sub-conical  in  form,  growing  from  an  obtuse  basal 
point,  without  stem,  root  or  point  of  attachment;  the  base  is  flattened  or 
convex,  the  greatest  width  in  some  specimens  is  just  above  the  base,  in  others 
about  half  the  height  of  the  sponge,  from  this  it  gi-adually  tapers  to    the 


64  DicTTospoNorD.?;. 

suiniait  which  appears  to  have  been  t)peii.     The  specimens  vary  between  23 
and  30  mm.  in  height,  and  from  14  to  17  mm.  in  Avidth. 

"  The  sponge -wall  appears  to  have  been  smooth  and  even ;  the  lai'ger 
areas  of  the  rectangular  mesh-work  are  marked  out  by  more  prominent,  vertical 
and  transverse  raised  lines,  and  vary  from  1.5  to  3  mm.  in  length,  and  these 
are  subdivided  by  finer  lines  into  smaller  squares,  the  sides  of  which  are 
about  .75  mm.  in  length.  In  some  cases  the  stronger  lines  forming  the  larger 
squares  are  nearly  parallel  from  the  base  to  the  summit,  whilst  in  others  they 
converije  to  each  other  towards  the  base. 

"  No  spicular  structure  whatever  has  been  preserved  in  any  of  the  speci- 
mens yet  discovered,  which  are  merely  casts  in  a  matrix  of  micaceous  sandstone. 

No  structure  can  be    detected  in  the  interior  of    the 
specimens."     (Hinde,  op.  cit.  1888.) 

The  foregoing  description  was  based  upon  McCoy's 

original  sjjecimens  and  his  figures  are  here  reproduced. 

These  show   an  obscurely  piismatic  form  which  is  not 

.„  „  ^   .  «        .   reproduced    in    the   figures   of    the    sijecies    given  by 

FiouRE  10.    McCoy's  Sgurc:i  of  £  O  1  O  J 

TetragonisDanbyi.  IIiNDE  uor  mcutloned  lu  hls  description.     This  species, 

a   fossil    of    rare    occurrence,  is    the    oldest    known    representative  of    the 

DlCTYOSPONGID^. 

Localities.  From  the  upper  Ludlow  rocks  at  Brigsteer,  Benson,  Under- 
barrow  and  Kendal,  Westmoreland,  England. 

SPECIES     OK    THE     DEVONIAN 

SPECIES  OF  THE  LOWER  HELDERBERG  GROUP. 

LYSACTINELLA,  Girty. 

1896.     LysactiTiella,  Girty.     Fourteenth  Ann.  Report  N.  Y.  State  Geologist, 
pp.  258-261,  pi.  1. 

This  genus  has  been  recently  described  from  free  spicules  obtained  by 
etching  the  silica-bearing  limestones  of  the  Lower  Helderberg  grouj)  in 
Albany  county,  N.  Y.  Very  little  is  yet  known  of  the  form  of  the  sponge 
body  in  either  of  the  two  species  established.  Mr.  Girty  speaks  of  one 
species,  L.  Gehhardi,  as  being  flattened  spherical,  but  of  this  as  probably  a 
quite  accidental  shape  due  to  compression  and  distortion.  The  spicules  of 
this  species,  so  far  as  isolated,  are  smooth  forms,  mostly  hexactine  and  pentac- 
tine.     Lysactinella  perelegans  is  based  on  a  series  of  spicules  Avith  echlnate 


Speoiks   ok   thk  Hamilton  Group. 


65 


arms.  There  is,  from  analogy,  probably  no  reason  why  these  smooth  and 
ornate  spicules  may  not  have  belonged  to  the  same  8i)ecies,  but  the  present 
distmction  will  serve  a  useful  purpose  until  more  is  learned  of  these  fossils. 

SPECIES  OF  THE  HAMILTON  GROUP. 

DICTYOSPONGIA,  gen.  nov. 
(For  description  see  page  72.) 

DioTYOspoNGiA  (?)  Marcellia,   Clarke  (sp.). 


Plate  xli,  Figs.  8,  9. 


1894. 


Hyahstelia  (?)  Marcellia,  Clarke.     Thirteenth  Ann.  Rept.  N.  Y.  State 
Geologist,  vol.  1,  p.  177,  pi.  iv,  figs.  19-22. 

The  specimens  upon  which  this  species  was  based  were  found  at  the 
horizon  of  the  Marcellus  shales,  in  the  excavations  made  at  the  Livonia  salt 
shaft  in  Livingston  county,  N.  Y.,*  at  a  depth  extending  from  812  to  823  feet 
from  the  surface.  The  fossils  occur  as  broad,  thin,  film  like  expansions  which, 
being  pyritized,  make  irregular  splotches  of 
discoloration,  tingeing  the  surface  of  the 
black  bituminous  shales  a  yellowish  or 
greenish  hue.  Under  enlargement  these 
patches  are  seen  to  be  composed  of  masses 
of  rod-like  and  cruciform  spicules.  The 
latter  are  hexactins ;  many  of  them,  which 
exhibit  four  branches  on  the  same  plane, 
show  the  abbreviated  or  nodifonn  remnants 
of  the  other  two  rising  in  the  plane  at  right 
angles ;  in  some,  however,  there  seems  to  be 
no  evidence  of  more  than  four  rays.  With 
these  spicules  are  found,  similarly  pyritized, 
lar^e  and  coai'se  spicular  bands,  flat,  straight, 

^  '■  I  I  a       J  FIOURKII.    Major  and  minor  stauractlnes  ot  Diclyo- 

thickened  on  the  edges  or  in  the  middle,    'po^i/toyiforceuta,  wim  portion  or  iigaiat«  «ptcuio. 

X  30. 

and  similar  in  size  and  appearance  to  small 

blades  of  grass.     These  end  abruptly  at  one  end  but  taper  to  the  other,  though 

showing  no  evidence  of  terminating  in  any  appliance  for  attachment. 

There   is   no    evidence    that    these    broad    ribbon-like    spicular   bands 
are   composed  of   separate  spicular   rods ;  on  the  contrary  they  are   clearly 

*  A  detailed  account  of  these  excavationa  and  the  succession  of  the  faunas  through  a  vertical  section  of 
nearly  fifteen  hundred  feet,  is  given  in  the  worlt  cited. 


gg  DlCTYOSPONGID^E. 

distinct  and  simple  bodies.  Their  association  with  the  cruciform  spicules 
is  so  close  and  invariable  as  to  enhance  the  probability  of  the  organic 
connection  of  the  two,  and  yet  we  do  not  know  a  similar  structure  among  the 
hexactinellids.  It  has  been  impossible  to  gain  a  conception  of  the  original 
form  of  the  sponges  to  which  these  masses  of  spicules  belong.  The  generic 
reference  now  made  of  the  species  is,  therefore,  quite  uncertain. 

Locality.     In  the  Marcellus  shales  of  the  Hamilton  group,  Livonia,  N.  Y. 

CLATHROSPONGIA,   Hall. 
(For  description  see  page  121.) 

Cl^throspongia  (?)  Hamiltonensis,  Hall  (sp.). 

Plate  xiv,  Fig.  1. 

1884.     Dictijophyton  Hamiltomnse,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State 
Museum  Natural  History,  p.  468. 

Sponge  broadly  cyathif orm,  regularly  and  rapidly  expanding  from  the  base. 
Surface  apparently  regularly  obconical,  though  a  sub-prismatic  appearance  is 
given  to  it  by  the  very  strong  longitudinal  spicular  bands. 

Reticulum  characterized  by  the  sharply  elevated  bands  of  the  first  order 
which  produce  large  quadrules.  Near  the  middle  of  the  cup  these  primary 
meshes  measure  about  12  mm.  on  each  side,  but  become  smaller  toward  the  base 
of  the  spOnge  where  the  primary  bands  seem  to  have  been  more  or  less  free,  but 
it  is  not  clear  whether  these  were  developed  into  bundles  of  anchoring  spicules 
independent  of  the  reticulum.  The  intersections  of  the  primaiy  spicular  bands 
are  thickened  and  distinctly  nodate.  A  second  and  third  series  of  bands  are 
more  or  less  clearly  defined,  and  the  interspaces  are  very  closely  reticulated. 

The  preservation  of  the  single  known  specimen  of  this  species  is  peculiar 
and  unique.  The  entire  reticulum  is  compactly  calcified,  lying  in  an  argil- 
laceous shale.  This  calcareous  matter  seems  to  have  been  deposited  over  the 
inner  surface  of  the  net-work  or  to  have  replaced  it  from  within,  as  traces  of 
the  finest  spicular  lines  may  be  seen  over  the  surfaces  of  the  coarsest  bands ; 
this  would  indicate  the  posterior  position  of  the  finer  net-work  to  the  primary 
longitudinal  and  transverse  bands.  The  elevation  and  prominence  of  the 
latter  convey  the  impression  that  in  their  original  condition  they  were  extended 
into  vertical  tufts  of  similar  character  to  the  mural  lamellae  in  the  typical 
Chemxmg  forms  of  Clathrospongia  ;  the  species  is  therefore  referred  to  this 
genus  until  it  shall  have  become  more  fully  known.  The  lesser  spicular  bands 
abound  in  irregular  nodose  swellings  which  may  be  partially  due  to  imperfect 


Species  ok  tiik  Hamilton  Gkoup. 


67 


presei'vation  or  to  adventitious  deposits  of  calcareous  matter  ujKin  projecting 
points. 

Dimensions.     The  length  of  the  specimen  is  85  mm.,  but  neither  l^ase  nor 
aperture  is  retained.     Its  width  at  the  upper  extremity  is  100  mm. 

Locality.     In"  the  Hamilton  shales,  shore  of  Canandaigua  lake,  N.  Y. 


ROEMERLSPONGIA,  gen.   nov. 
1883.     Z>icti/oj)hi/to)i,  FGvd.  Roemer.     Zeitschr.  der  Deutsch.  Geolog.  Gesellsch. 
vol.  35,  p.  704,  figs,  a,  b  (p.  706). 
Obconical  rapidly  expanding  cups  apparently  arising  from  an  acute  base. 
Siirface  divided  by  strong,  sube(pial  vertical  spicular  bundles,  crossed  by  nar- 
rower horizontal  bands.     The   cpiadrules  thus   fonned   are    all    convex    and 
appear  not  to  be  subdivided  by  subordinate  series  of  spicular  bands.     Toward 
the  base  of  the  sponge  the  meshes  are  less  regular  than  elsewhere. 
Type,     Dictyophytoii  (rerolsteinensis,  F.  Roemer. 

RoEMEKispoNGiA  Geuolsteinensis,  F.  Roeuiei',  (sp.). 
1883.     Dictyophytoii  Gerolsteinense,  F.  Roemer,  loc.  cit. 

In  a  "  Notiz  uher  die  Gattuiuj  Dictyophytoii!\  the  late  Ferdinand  Roemer 
described,  as  above  cited,  an  interesting  Dictyosponge  from  the  middle  Devo- 
nian of  the  Eifel.  This  descrip- 
tion is  as  follows :  "It  shows,  like 
the  former  fossil  (Tetraganis 
[^Prismodictya]  Eifeliensis,  F. 
Roemer)  a  surface  divided  into 
rectangular  areas,  but  the  sub- 
division is  less  regular  and  more 
simple.  The  vertical  rods  are 
much  stronger  than  the  horizontal 
rods  and,  inasmuch  as  new  ones 
appear  toward  the  aperture,  are 
not  quite  straight.  Moreover  the 
transverse  i-ods  do  not  always 
make  right  angles  with  the  verti- 

°  "  nouBE  12.    RoemtritponoUi  Qtroiaeintntii.    (After  P.  Roemer).    The 

Cal    rods,  but    are,    especially  over    conTexlty9fthequadrulesl8notwell8iiown  in  these  copies. 

the   l()\\er   part,    directed   obliquely    against   them.     *     *     *     farther   sub- 
division of  the  rectangular  areas  by  liner  spicules  is  wanting." 


68  DlOTYOSPONGID^. 

The  two  wood-cuts  of  this  fossil  given  by  Roemek  and  reproduced  here, 
show  very  clearly  the  structure  described.  The  original  cuts  also  indicate 
distinctly  an  important  character  which  the  author  does  not  mention,  namely, 
the  convexity  of  each  of  the  non-reticulate  rectangular  areas,  which  we 
presume  to  be  of  a  similar  character  to  the  surface  nodes  in  Piiysospongia. 
The  relation  of  the  fossil  to  the  latter  genus  is  also  seen  in  the  relatively 
great  size  and  mode  of  reticulation  of  the  spicular  bundles,  while  a  generic 
difference  in  the  two  is  found  in  the  single  vertical  rows  of  quadrules, 
their  increase  by  intercalation  and  in  the  convexity  of  all  the  quadrules 
in  RoEMEKispoNGiA,  even  in  those  of  irregular  shape  over  the  early  portion 
of  the  cup. 

Geological  horizon.  From  the  Calceola-shales,  Gerolstein  (middle 
Devonian). 

Professor  Clemens  Sciilutee  has  described  under  the  name  Protospoiujia 
rJienana,  a  Dictyosponge  from  the  Hunsriick  slates  (lower  Devonian),  of 
Gemiinden,  Germany.*  The  original  specimen  is  a  mere  fragment  of  a  veiy 
coarse-meshed  sponge  whose  generic  characters  are  entirely  uncertain. 

SPECIES  OF  THE  PORTAGE  GROUP. 

DICTYOSPONGIA,  gen.  nov. 
(For  description  see  page  72.) 

DiCTYOSPONGIA  HAPLEA,  Sp.  nOV. 
Plate  xxxiii,  Fig.  1. 

A  small  species  of  this  genus  is  indicated  by  two  specimens  showing  a 
smooth,  regularly  expanding,  subcylindrical  cup  with  a  very  fine  reticulum. 
One  of  these  specimens  exhibits  only  the  apical  portion ;  the  other  represents 
nearly  the  entire  length  of  the  cup,  has  a  diameter  rather  more  than  one-third 
of  the  length  and  is  slightly  contracted  about  the  aperture.  No  trace  of  pris- 
matic faces  is  preserved.  This  entire  specimen  measures  40  mm.  in  length,  14 
mm.  in  greatest  width  and  11  mm.  in  width  at  the  aperture. 

Locality.  In  the  sandstones  of  the  Portage  group ;  second  falls  of  the 
Tannery  gully,  Naples,  N.  Y.     (Collected  by  D.  D.  Luther.) 


Zeitscbr.  der  Deutsch.  Geolog.  Gesellsoh.  vol.  4*,  p.  615,  1892, 


Species  of  tjie  Portaok  Group.  ^^^) 

PRISMODICTYA,  gen.  nov. 

(For  description  see  page  79.) 

Prismodictya  paljja,  sp.  nov. 

Platk  XXXIII,  Fio.  2. 

The  original  specimen  of  this  species  indicates  a  small,  elongate  cup, 
expanding  rather  rapidly  and  attaining  full  width  at  al)Out  one-third  of  its 
length  from  the  base.  Its  median  width  is  about  one-fourth  of  its  entire 
length  and  its  aperture  is  somewhat  contracted. 

The  surface  shows  that  the  prism  faces  are  distinctly  developed  and  these 
are  crossed  by  two  or  three  low  horizontal  ridges  making  vertically  elongate 
rectangles  with  the  prism-angles  and  at  their  intersections  are  low,  very  small 
nodes. 

Reticulum  very  fine. 

The  single  example  observed  has  a  length  of  50  mm.,  a  median  width  of 
14  mm.  and  an  apertural  width  of  10  mm. 

Locality.  In  the  Portage  sandstones  ;  second  falls  of  the  Tannery  gully, 
Naples,  N.  Y.     (Collected  by  D.  D.  Luther.) 

HYDNOCERAS,    Conrad. 
(For  description  see  page  95.) 

Hydnocebas  legatum,  sp.  nov. 

PLA.TE  XXXIII,  Fios.  5,  6. 

Recent  explorations  have  shown  the  presence  in  the  Portage  beds  of  a 
species  which  appears  to  be  closely  allied  to  typical  forms  of  II.  tnberomnn, 
Conrad,  of  the  Chemung  sandstones.  None  of  the  material  thus  far  collected, 
though  from  two  localities,  is  of  good  quality,  but  enough  is  retained  by  the 
fragments  gathered  to  show  the  presence  of  from  four  to  six  transverse  rows 
of  nodes  which,  over  the  lower  portion  of  the  sponge,  are  moderately  strong 
and  are  separated  by  deep  constrictions  which  bear  little,  if  any,  evidence  of 
prismatic  division.  A  fragment  of  a  large  individual  has  these  nodes  large, 
robust  and  blunt,  similar  to  those  in  the  gigantic  fonu,  II.  Avoca,  of  the 
Chemung.     (See  Plate  xi). 

The  reticulum  is  finely  divided,  rather  more  so  than  is  usual  in  II.  tube- 
rosum. Probably  additional  differentiating  features  will  appear  when  this 
Portage  fossil  becomes  better  known.  For  the  present  it  seems  wise  to  recog- 
nize  the  fossil   by  a  distinct  appellation,  its  specific  value  being  tentative. 


70  DlCTTOSPONOID^E. 

DimeTisions.  One  of  the  specimens,  somewhat  distorted,  has  a  length  of 
100  mm.  The  fragment  of  a  large  cup  indicates  an  apertural  width  of 
150  mm,  and  its  original  length  was  probably  not  less  than  250  mm. 

Localities.  The  first  specimen  found  was  taken  from  the  Portage  sand- 
stone at  the  second  falls  of  the  Tannery  gully,  Naples ;  subsequently,  the 
species  was  obtained  at  Varysburg,  Wyoming  county,  in  the  ravine  entering 
the  village  from  the  west.  The  latter  horizon  is  considerably  lower  in  the 
rock  series  than  the  former.     (Collected  by  D.  D.  Luther.) 

Hydnoceras   eutiieles  (?). 

Plate  xxxiii,  Fig.  7. 

(See  page  105  and  Plate  viii.  Figs.  5,  6.) 

This  species,  characterized  by  its  short  form  and  long,  pendant  nodes,  is 
described  among  the  Chemung  forms  obtained  at  Cotton  hill,  near  Avoca.  A 
specimen  from  the  Portage  beds  found  in  association  with  the  preceding 
species,  has  an  aspect  so  similar  to  H.  euthehs,  as  to  make  probable  the 
specific  identity  of  the  two.  It  is,  therefore,  for  the  present  referred  to  that 
form.     (Collected  by  D.  D.  Luther.) 

HyDNOCERAS  GRACILE,  Sp.  nOV. 
Plate  xxxiii.  Figs.  3,  4. 

Sponge  short,  rapidly  expanding,  with  three  rows  of  nodes  exposed, 
possibly  a  fourth  near  the  base  which  is  fractured.  In  two  of  these  horizontal 
rows  the  number  of  nodes  is  normal,  eight ;  but  in  the  outer  or  apertural  row 
some  of  the  nodes  are  absent  or  but  faintly  developed.  The  apertural  margin 
is  retained  in  its  entirety  and  while  the  absence  of  some  of  the  nodes  near  this 
edge  is  noticeable  and  in  accordance  with  the  structure  observed  in  other 
species,  yet  the  nodes  of  this  outer  row  are  close  to  the  margin,  and  not  sepa- 
rated from  it  by  any  considerable  area.  The  nodes  of  the  upper  row  are  low 
but  those  of  the  lower  rows  long  and  slender. 

The  reticulum  is  very  fine,  the  primary  subdivision  being  largely  obscured 
and  evidence  of  prismatic  faces  wanting.  The  two  specimens  which  represent 
this  species  are  similarly  preserved,  having  been  buried  in  the  sediment  with- 
out lateral  compression.  In  their  dimensions  and  individual  characters  they 
are  very  similar,  though  one  is  more  favorably  preserved  than  the  other.  Each 
has  an  apertural  diameter  of  50  mm.,  and  their  height  is  about  30  mm.  Both 
of  these  dimensions  have  been  somewhat  modified  by  the  slight  compression 


Spk(!Iks  of  the  Portage  Group.  71 

to  which  they  have  been  subjected,  i)ro])a])ly  the  former  wlightly  increjised  and 
the  latter  as  much  diminished. 

Locality.  In  the  flags  beh)\v  the  Portage  sandstones  in  the  east-side  ravine 
at  Varysburg,  Wyoming  county.     (Collected  by  D.  I).  Luther.) 

CLEPSYDROSPONGIA,  gen.  nov. 

Sponge  subcylindrical ;  subequally  expanded  at  both  base  and  aperture ; 
contracted  medially.  Apertural  margin  simple,  not  explanate.  Mode  of 
attachment  at  the  base  not  known  (possibly  by  a  transverse  disk  or  diji- 
phragm  as  in  Thysanodiotya  and  Phbagmodictya).  Surface  with  transverse 
rows  of  low,  pointed  nodes,  as  in  Hydnoceras. 

Type,  Clepsydrospongia  matutina,  sp.  nov. 

Clepsydeospongia  matutina,  sp.  nov. 

Plate  xxxiv,  Figs.  1,  2. 

Sponge  of  moderately  large  size ;  originally  subcylindrical.  Diameter  of 
basal  and  apertural  extremities  subequal;  constricted  medially  so  that  the 
median  width  is  about  two-thirds  of  that  at  the  ends.  Both  basal  and  aper- 
tural extremities  appear  to  end  abruptly,  the  latter  showing  a  regular  and  but 
slightly  expanded  margin.  The  basal  edge  is  not  defined  with  precision  but 
to  all  appearances  it  terminates  simply  as  in  the  case  of  the  aperture,  although 
a  transverse  basal  disk  may  have  existed. 

The  surface  shows  evidence  of  five  horizontal  rows  of  nodes  of  the  same 
character  as  those  of  Hydnoceras.  The  lowest  of  these  is  highly  obscure ; 
the  second  is  somewhat  more  pronounced,  as  shown  along  the  edges  of  the 
specimen.  These  lower  rows  are  separated  by  a  furrow  so  shallow  that  the 
surface  in  this  region  was  virtually  cylindrical.  The  third  or  median  row  of 
nodes  is  the  most  conspicuous  of  all.  It  is  separated  from  adjacent  rows  by 
broad  and  moderately  deep  constrictions.  The  nodes  also,  are  more  strongly 
developed  than  elsewhere  on  the  surface,  are  elongate  and  rather  low.  Both 
in  this  row  and  that  following  above,  there  are  ten  nodes,  varying  among 
themselves  in  size ;  those  of  the  fourth  row  are  not  as  strongly  pronounced 
as  those  in  the  median  row.  The  nodes  of  the  fifth  or  apertural  row 
are  hardly  discernible,  and  thus  the  apertural  one-third  of  the  cup  and  the 
basal  one-third  have  a  very  similar  aspect.  These  nodes  are  all  conical  and 
tufted. 


72  DlCTYOSPONGID^. 

Heticvlum.  The  net  work  is  very  fine  and  sharp.  The  obscuration  of  the 
fundamental  prism-faces  makes  the  prevailing  meshes  of  small  size,  about 
1.5  mm.  on  each  side,  and  though  these  may  be  again  divided,  combinations  of 
them  into  larger  quadrules  are  not  emphasized.  Thus  the  reticulation  appears 
very  uniform  throughout. 

Dimensions.  The  single  specimen  which  has  been  observed  of  this 
species,  has  a  length  of  170  mm.  Its  width  at  the  base  is  65  mm.;  at  the  aper- 
ture 75  mm.;  on  the  median  row  of  nodes,  60  mm.;  across  the  median 
constrictions,  42  mm. 

Locality.  In  the  sandstone  of  the  Portage  group ;  second  falls  of  the 
Tannery  gully,  Naples,  N.  Y,     (Collected  by  D,  D.  Luther.) 

SPECIES  OF  THE  CHEMUNG  GROUP. 

DICTYOSPONGIA,  gen.  nov. 
1855.     Tetragonis,  McCoy.     British  Palaeozoic  Fossils,  p.  62. 
1880.     Tetragonis,  F.  Roemer.     Lethsea  Geognostica,  1  Theil,  p.  304. 

Not  Tetragonis,  Eichwald,  1842.     Urwelt  Russlands,  Heft  2,  p.  81. 
1883.     Dietyophyton,  Barrois.      Sur  les  Dictyospongidse  des  Psammites  du 

Condroz,  p.  85. 
1890.     Dictyophyton,  Hall.     Ninth  Ann.  Kept.  N.  Y.  State  Geol.,  p.  56. 

Very  gradually  expanding,  smooth,  obconical  or  subcylindrical  sponges 
with  the  outer  surface  devoid  of  nodes,  tufts,  ridges,  annulations  or  other  orna- 
mentation; prism-faces  sometimes  very  obscurely  developed  toward  the 
summit ;  base  furnished  with  a  tuft  of  long,  straight  anchoring  spicules. 

Type,  Dictyophyton  sceptrum,  Hall. 

The  fossil  described  by  McCoy,  (op.  ait.)  as  Tetragonis  Danhyi,*  from  the 
upper  Ludlow  rocks  of  Westmoreland,  appears,  from  the  description  and  figure 
given  by  the  author  and  subsequently  by  Dr.  Hinde,  to  be  a  reticulate  sponge  of 
this  genus.  In  regard  to  the  nature  of  the  oi'iginal  species  of  Eichwald's  genus 
Tetragonis  (T.  Murcliisoni,  Eichwald),  we  may  accept  the  opinion  of  Hinde, 
who  says :  "  From  an  examination  of  undoubted  forms  of  Tetragonis  from  the 
Silurian  of  Gotland,  I  believe  that  it  is  a  sponge,  and  that  it  is  very  closely 
allied  to  Ischadites,  Murch." 

— —  "      —  ■--  ■       ■■  .1    ■  ...,.- '       T ' \ ■ 

*  Diciyospougia  Danhyi^  see  page  63  of  this  volume. 


Species  of  the  Chkmuno  Gkoup.  73 

DiOTYOSPONOIA  8CKPTRITM,  Hall  (sp.). 
Plate  xiv,  Fios.  3-6;  Plate  xv,  Fios.  8,  9;  Platk  xvi,  Fioh.  3,  4 ;  Platf.  xxvil,  Pio.  3; 

PlJkTE  XXXVI,  Fio.   7. 

1890.  Dictyophyton  sceptnim,  Hall.  Ninth  Ann.  Rept.  N.  Y.  State  Geolo- 
gist,  p.  56 ;  and  Forty-third  Ann.  Rept.  N.  Y.  State  Museum  of 
Natural  History,  p.  258. 

Sponge  elongate,  conical  or  subcylindrical ;  circular  or  elliptical  in  trans- 
verse  section.  Usually  abnormally  compressed,  sometimes  curved.  Very 
gradually  expanding  from  the  base  upward  for  one-half  to  two  thirds  its  length, 
narro-wing  more  abruptly  toward  the  aperture. 

Surface  smooth  or  without  prism-faces  or  any  iiregularities  except  those 
arising  from  distortion  in  fossilization. 

Reticulum  characterized  by  coarse,  subequidistant  transveree  spicular 
bands,  from  10  to  15  mm.  apart,  crossed  by  vertical  bands  of  subequal  size  not 
more  than  one-half  of  this  distance  from  each  other.  The  impressions  left  by 
the  .vertical  bands  are  not  as  deep  and  conspicuous  as  those  of  the  horizontal, 
hence  the  first  effect  produced  by  the  appearance  of  the  specimens  is  that  of  a 
strong  transverse  lineation,  and  a  secondary  and  very  characteristic  impression 
that  of  a  division  of  the  surface  into  vertical,  elongate  rectangles.  Some  of  the 
specimens  Avhich  have  been  enveloped  in  a  soft  clay-shale,  retain  the  spicular 
bands  themselves,  changed  through  pyritization  and  o.xidation  into  limonite,  a 
very  unusual  occurrence  in  species  from  the  Chemung  sandstones.  This  pres- 
ervation is  insufficient  to  demonstrate  the  character  of  the  individual  spicules, 
though  it  shows  the  very  fine  reticulation  of  the  primary  quadrules. 

Dimensions.  The  specimens  at  hand  show  considerable  variation  in  fonu 
and  size.  The  most  robust  of  these,  in  which  the  aperture  is  retained  Avhile 
a  small  part  of  the  basal  portion  is  lost,  has  a  length  of  200  mm.,  a  width 
at  the  base  of  40  mm.,  greatest  width  at  two-thirds  its  length  from  the  base,  of 
70  mm.  and  a  width  at  the  aperture  of  50  mm.  This  specimen  has  unquestion- 
ably been  much  flattened,  which  accounts  for  its  great  proportional  width. 
A  more  slender  cup,  nearly  complete  from  base  to  apertui-e  and  but  slightly 
compressed,  measures  252  ram.  in  length  and  38  mm.  in  greatest  width.  In 
one  example  the  cup  is  greatly  curved,  but  instances  of  this  kind  seem  to  be 
due  to  the  easy  distortion  of  such  long  and  slender  bodies. 

LocaUties.  In  the  shaly  sandstones  of  the  upper  part  of  the  Chemung 
group  at  Alfred,  Scio,  Friendship,  Nile,  Andover,  Cuba  and  Wellsville,  Allegany 
county ;  Olean,  Cattaraugus  county,  and  Jamestown,  Chautauqua  county,  N.  Y. 


74  DlCTTOSPONGID^:. 

This  species  is  of  very  frequent  occurrence  in  the  upper  beds  of  this 
foiTuation,  to  which  it  appears  to  be  restricted.  Most  of  the  specimens 
found  have  been  obtained  from  isolated  loose  blocks,  though  Mr.  E.  B.  Hall, 
of  Wellsville,  has  located  a  considerable  colony  of  them  near  that  village. 
The  species  is  Avell  characterized  and  always  readily  recognizable  by  the 
vertically  elongated  quadrules  of  the  reticulum. 

DiCTYOSPONGIA     LOPIIFRA,    Sp.   UOV. 
Plate  xv,  Fig.  5. 

Sponge  small,  slender,  long,  circular  in  cross-section ;  gradually  expanding 
for  one-third  of  its  length,  thence  upward  nearly  cylindrical.     Surface  smooth. 

Jietmdum  composed  of  fine  spicular  bands,  apparently  divisible  into  but 
two  clearly  defined  series.  The  transverse  bands  are  the  more  conspicuous, 
and  their  division  into  primary  and  secondary  series  is  clearly  evident,  but 
among  the  vertical  bands  this  difference  is  more  obscure.  The  general  aspect 
of  the  surface  is  that  of  series  of  vertical  parallelograms  whose  upper  and 
lower  boundaries  are  the  primary  horizontal  bands.  A  subordinate  reticula- 
tion is  visible  within  the  larger  quadrules. 

The  single  examj)le  observed  of  this  graceful  species  has  a  length  of 
88  mm.  to  the  base  of  the  cup.  So  far  it  is  an  internal  cast,  but  beyond  this 
point  is  an  external  cast  of  a  spreading  brush  of  anchoring  spicules  or  basalia, 
which  may  be  traced  for  a  distance  of  33  mm.,  giving  the  specimen,  which  is 
essentially  entire,  a  total  length  of  121  mm.  The  width  of  the  cup  at  its 
upper  end  is  7  mm.  and  about  the  same  at  the  middle  of  its  length. 

Locality.  In  the  upj)er  beds  of  the  Chemung  group.  Found  loose  at 
Wellsville,  N.  Y.     (Collection  of  E.  B.  Hall.) 

DiCTYOSPONGIA    CHARITA,    Sp,    UOV. 
Plate  xv,  Fig.  4. 

Sponge  very  small,  elongate,  slender  and  graceful ;  expanding  rather 
rapidly  from  the  base  for  about  one-half  its  length,  slightly  contracting 
toward  or  near  the  aperture.  Transverse  section  circular  or  elliptical.  Surface 
smooth. 

Meticulum  fine,  with  a  predominance  of  the  horizontal  bands  which,  with 
the  vertical  lines,  produce  a  division  of  the  surface  into  rectangular  quadrules 
about  3  mm.  in  length  and  2.5  mm.  in  width ;  these  are  subdivided  into 
approximate  squares  by  two  subordinate  horizonal  bands  and    one  vertical 


Spkcies  op  thk  CiiKMuxa  Group.  75 

W.     At  the  apex  of  the  best  preserved  specimen  is  the  impression  of  a 
short  basal  tuft  of  spicules. 

Dirmmwm.  One  example  measures  in  length,  exclusive  of  the  basal 
tuft,  36  mm. ;  its  median  diameter  is  4.5  mm.  and  its  apertural  diameter,  3  ram. 
In  a  second  specimen  two  individuals  lie  side  by  side,  one  of  them  measuring 
37  mm.  in  length,  with  a  median  diameter  of  7  mm.  and  an  apertural  diameter 
of  4  mm.     This  specimen  has  been  somewhat  flattened. 

Locality.  In  the  sandstones  of  the  upper  portion  of  the  Chemung  group 
afWellsviUe,  N,  Y.     (Collection  of  E.  B.  Hall.) 

DiCTYOSPONGIA  ETTMORPIIA,  Sp.  nOV. 
Plate  xv,  Pigs.  1-3. 
Sponge  small,  slender,  elongate-subfusiform,  expanding  for  about  one- 
fourth  its  length  from  the  apex,  thence  upward  with  parallel  sides,  contracting, 
somewhat  toward  the  aperture.  Surface  smooth ;  reticulation  very  fine,  com- 
posed of  minute  subequal  rectangular  meshes,  whose  ordinal  arrangement  is 
not  clearly  defined. 

Two  specimens  of  this  species  upon  the  same  slab  of  sandstone,  measure 
80  and  84  mm.  in  length,  each  having  a  maximum  width  of  15  mm.  In  the 
smaller  individual  the  aperture  is  retained,  but  in  neither  is  the  basal 
extremity  entire.  The  species  is  distinguished  from  the  other  forms  here 
described  by  its  more  rapidly  expanding  and  broader  cup. 

Locality.     In  the  upper  beds  of  the  Chemung  group  at  Wellsville  N  Y 
(Collection  of  E.  B.  Hall.) 

DiCTYOSPONGIA    SIR^A,  Sp.  nOV. 
Plate  xlii,  Figs.  4,  5. 

SpoifGE  rather  small,  expanding  gradually,  but  with  greater  rapidity  on 
one  side  than  on  the  other,  thus  giving  the  cup  a  semi-crescentic  outline,  a 
feature  persistent  in  all  specimens.  The  shape  is  thus  long  and  slender,  the 
aperture  being  somewhat  narrower  than  the  parts  immediately  belovv  it. 
The  lower  portion  terminates  in  a  long,  narrow  rope  of  anchoring  rods  whose 
length  is  nearly  if  not  fully  as  great  as  the  body  of  the  sponge. 

The  reticulum  is  composed  of  a  net- work  of  small  quadrules  of  generally 
uniform  size,  not  often  distinctly  retained.  The  surface  shows  no  nodes  or 
other  irregularities. 

Ditnemioiis.  A  specimen  entire  except  for  a  part  of  the  anchoring  rope, 
measures  in  length,  from  base  to  aperture,  72  mm.,  the  ^vidth  of  the  aperture 


76 


DlOTYOSPONGID^ 


being  17  mm.,  and  the  greate&t  width  of  the  cup  22  mm.  Another  specimen 
with  a  length  of  80  mm.  and  not  quite  entire  at  the  aperture,  has  an  anchoring 
rope  of  60  mm.  in  length  and  this  is  incomplete  at  its  lower  extremity. 
The  pereistent  curvature  of  this  species  and  its  relatively  rapid  expansion, 
together  with  its  remarkably  long  basal  tuft,  \\all  serve  as  distinguishing 
characters. 

Locality.  Several  specimens  have  been  found  in  the  lower  beds  of  the 
Chemung  group  at  Cotton  hill,  one  mile  north  of  Avoca,  Steuben  county,  in 
association  with  Hydnoceras  tuberosum,  II.  eutlieles,  H.  multinodosiim,  Arysti- 
dictya  elegans,  A.  nodifera  and  Ilallodictya  Cottoniana. 


DiCTYOspoNGiA    Almondensis,  sp.  nov. 

Plate  xliv,  Fig.  1. 
The  grey  sandstone  beds  at  Almond,  Allegany  county,  are,    in   pluces, 
crowded  with  fragments  of  a  species  of  very  large  size,  but  for  the  most  part 

these  fail  to  afford  a  definite 
conception  of  the  aspect  of 
the  entire  sponge.  Some  of 
these  large  expansions  like 
that  illustrated  on  Plate 
xliv,  show  under  compres- 
sion an  obscure  prismatic 
surface,  but  usually  the  sur- 
face is  smooth  and  the  gen- 
eral appearance  of  the  ac- 
cumulated fragments  indi- 
cates that  they  belong  to 
a  single  species  of  Dictyo- 
SPONGIA.  Among  the  large 
number  of  specimens  col- 
lected from  this  locality  no 

FionBKlS.    The  basal  portion  of  Biofi/osponBia  X!mo?i<i«jw<«.  siusrle  OUe    clearlv    shoWS    all 

the  specific  characters.  The  large  example  given  presents  the  maximum  size 
observed  while  the  accompanying  figui'e  represents  the  basal  portion  of  a 
specimen,  showing  its  rapid  expansion  and  curving  sides,  retaining,  as  well, 
the  elongate  parallelograms  of  the  reticulum  Avhich  are  more  or  less  noticable 
in  other  species  of  Dictyospongia. 


SpKCIES    ok    TlIK    CllKMlIXO    GitOUP.  77 

This  is  our  largest  species  of  the  genus,  and  doubtless  its  gi-eat  size  is  in 
itself  the  principal  occasion  of  so  delicate  a  structure  1)eiug  retained  only  in 
this  highly  fragmentary  state. 

Locality.     Almond,  N.  Y.     (Collection  of  E.  B.  Hall.) 

DiCTYOSPONGIA  (?)    BACTERIA,  Sp.  HOV. 
Platk  xxxviii,  Figs.  7,  8. 

This  is  a  slender,  nearly  cylindrical  sponge,  which,  in  the  entire  length, 
shows  but  slight  indication  of  tapering  in  either  direction.  Oi-iginally  its  sur- 
face" was  doubtless  smooth  and  this  aspect  is  modified  in  the  fossil  only  by 
some  low  casual  corrugations  and  indentations.  The  substance  of  the  skeleton 
was  extremely  thin  as  shown  in  transverse  section  and  is  i-epresented  by  a 
tenuous  ferruginous  film.  Under  favorable  light  there  is  evident  an  extremely 
fine  transverse  lineation  of  the  siu"face  but  vertical  striation  is  extremely 
obscure.  The  fossil  has  a  quite  different  aspect  from  the  other  species  of 
DiCTYOSPONGiA  here  described  and  while  there  is  good  reason  to  regard 
it  as  of  this  genus,  it  may  prove  to  be  of  a  somewhat  different  nature. 
Small  patches  of  iron  oxide  on  the  surface  of  the  specimen  show  actinic  spots 
which,  though  failing  to  resolve  under  high  magnification,  suggest  that  these 
may  be  parenchymal  or  dermal  spicules. 

The  length  of  the  specimen  illustrated  is  110  mm.  and  is  incomplete. 
Its  width  is  12  mm.  at  each  extremity  and  15  mm.  in  the  middle. 

Locality.  From  a  calcareous  sandstone  in  the  upper  Chemung  fonnation, 
associated  with  numerous  brachiopods :  Spirifar  disjiiiictas,  OrtJiotheten  Clie- 
mmigemis,  Stroplieodonta  demima  var.,  Frodtbctella  lachrynwsa.  The  precise 
locality  of  the  specimen  is  not  known,  but  it  is  probably  from  Steuben  county. 

HYDRIODICTYA,  gen.  nov. 

1884.     Dictyophytoii,  Hall.     Thirty-fifth  Ann.  Eept,  N.  Y.  State  Mus.  Nat. 
Hist.,  p.  469. 

Simple,   broadly   expanded,   vase  shaped  cups  without  nodes,  ridges   or 
other  conspicuous  surface  characters,  and  with  no  apertural  spiculai*  fiinge. 
Type,  Biciyophytoii patalwm,  Hall.    ■ 


78  DlCTYOSPONGID^.. 

Hydriodictya  patula,  Hall. 

Plate  iv,  Fig.  1. 

1884.     Diciyaphytoti  patidum,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State  Mus. 
Nat.  Hist.,  p.  469. 

Sponge,  of  which  the  upper  portion  only  is  known,  broadly  funnel-shaijed. 
Diameter  of  the  aperture  more  than  twice  that  at  60  mm.  below  the  aperture. 
Surface  marked  by  very  faintly  subprismatic  and  rapidly  expanding  faces. 

Reticulum  coarse-meshed,  the  primary  quadrules  being  l)ounded  by  ver- 
tical spicular  bands  which  traverse  the  edges  and  middle  of  the  subprismatic 
faces  and  are  intersected  by  horizontal  bands  of  the  same  strength.  These 
quadrules  are  trapezoidal  on  account  of  the  expansion  of  the  cup ;  an  average 
example  of  them  measures  11  mm.  at  the  base,  15  mm.  at  the  top  and  17  mm. 
in  height.  These  primary  quadrules  decrease  greatly  in  size  toward  the  aper- 
ture and  are  all  distinctly  subdivided  by  the  impressions  of  three  or  four 
subordinate  series  of  spicular  bands. 

The  original  specimen  has  a  length  of  63  mm. ;  an  apertural  diameter  of 
130  mm.,  and  a  diameter  at  the  lower  broken  end,  of  60  mm. 

Locality.  From  the  middle  beds  of  the  Chemung  group,  in  the  Hydno- 
ceras  tuberosum  colony,  at  the  Brown  hill  school-house,  near  Cohocton, 
Steuben  county,  N.  Y. 

Hydkiodictya  cylix,  sp.  nov. 

Plate  v,  Figs.  1,  2;  Plate  xvi,  Fu;.  2;  Plate  lV;  Fig.  7. 

Sponge  regularly  conical,  rather  rapidly  expanding  ;  surface  destitute  of 
nodes  or  other  ornamental  characters.  A  small  specimen  bears  a  few  irregular 
nodiform  elevations  which  ai'e,  however,  wholly  due  to  the  accidents  <^f 
fossilization. 

Ileticulum  regular,  the  prevailing  meshes  measuring  from  2.5  to  3  mm. 
on  each  side,  and  being  subdivided  by  one  or  two  subordinate  series  of  spicu- 
lar bands.  There  is  no  grouping  of  these  into  well-defined  quadrules  of 
large  size,  and  no  evidence  of  the  projection  of  the  spicular  bands  into  free 
lamellae. 

The  largest  specimen  observed  shows  a  patch  on  its  surface  where  the 
original  skeleton  had  evidently  been  broken  and  was  subsequently  I'epaired 
by  the  formation  of  a  highly  irregular  net,  totally  lacking  the  regular  c^uad- 
rulation  of  the  adjoining  parts  of  the  cup. 


Speciks  of  the  Ciikmuno  Group.  79 

Birnensions.  A  small,  appai-eiitly  entire  individual  has  a  length  of  60 
ram.  and  an  apertural  diameter  of  50  mm.  A  larger  e.vample,  which  is  nearly 
entire,  is  110  mm.  in  length  and  of  about  the  same  width  at  the  ui)per 
extremity. 

Locality.  In  the  lower  beds  of  the  Chemung  group  ;  from  a  shaly  sand- 
stone at  Deyo  basin,  Naples,  Ontario  county,  N.  Y.,  in  association  with  Hydiu>- 
ceras  variaUle,  Ceratodictya  annulata,  various  In-achiopods  of  the  Chemung 
fauna  and  abundant  remains  of  terrestrial  plants, 

Hydriodictya  nephelia,  sp.  nov. 

Plate  xxi,  Fig.  10. 

This  species  is  represented  by  the  upper  portion  of  a  flattened  cup, 
characterized  by  its  broadly  expanded  and  regular  aperture  and  perfectly 
smooth  surface.  The  median  or  lower  part  of  the  cup  tapers  much  less 
rapidly  than  the  apertural  portion. 

The  reticulmn  is  composed  of  extremely  fine  spicular  bands,  those  of  the 
horizontal  series  being  rather  the  more  clearly  defined.  They  occur  at  inter- 
vals of  about  4  mm.  and  form,  with  the  divergent  vertical  lines,  somewhat 
trapezoidal  quadrules.  These  spaces  become  obsolete  toward  the  aperture 
and  the  entire  surface  is  minutely  subdivided  by  the  subordinate  spicular 
bands. 

The  dimensions  of  the  single  example  observed  are  as  follows :  length 
108  mm. ;  width  at  the  lower  end  65  mm. ;  at  the  aperture  125  mm. 

Locality.  In  the  upper  beds  of  the  Chemung  group  at  Wellsville,  N.  Y. 
(Collection  of  E.  B.  Hall.) 

PRISMODICTYA,  gen.  nov. 

1863.     DictyophyUyn,  Hall.     Sixteenth  Ann,  Rept,  N.   Y.   State  Cab.  -Nat. 

Hist.,  pp,  88,  89, 
1882.     Blctyophyton  (partim).  Hall.     Notes  on  the  Family  Dictyospongidae. 
1884.     Dlctyophytoii    (partim),  Hall.     Thirty-fifth  Ann.  Rept,  N,  Y.   State 
Mus.  Nat.  Hist.,  pp.  466,  469-471. 

Elongate,  usually  slender  sponges,  with  acute,  gradually  expanding  base 
and  sharply  prismatic  surface,  the  prism-faces  numbering  eight ;  aperture 
slightly  contracted ;  surface  without  tufts  or  lamellar  ridges ;  rarely  with 
inconspicuous  nodes  on  or  between  the  interfacial  angles, 

Ty])e,  Dictyophyton  telum,  Hall. 


go  DlCTYOSPONGID^. 

Prisjiodictya  telum,  Hall  (sp.). 

Platk  xvii,  Figs.  »-10;  Plate  xxxv,  Figs.  7,  8;  Plate  xlii,  Fig.  2. 

1884.     DicUjophyton,  telum,   Hall.      Thirty-fifth    Ann.    Kept.   N.   Y.   State 
Mus.  Nat.  Hist.,  p.  470. 

Sponge  small,  sharply  prismatic,  octagonal  in  section  ;  prism-faces  flat  or 
slightly  concave.  The  base  is  broad  but  acute,  all  the  prism-faces  being  well 
defined  ;  the  sponge  expands  rather  rapidly  attaining  its  greatest  diameter  at 
about  one-third  of  its  length,  thence  it  contracts  very  gradually  upward 
becoming  naiTOAvest  at  about  one-fifth  of  its  length  from  the  upper  end, 
again  it  expands  slightly,  and  at  the  aperture  the  margin  is  somewhat  abruptly 
naiTowed. 

Reticwlwm  very  fine,  without  prominent  vertical  bands  upon  the  prism- 
faces,  but  showing,  in  a  few  instances,  rather  coarse  hoi'izontal  bands.  The 
prevailing  large  quadrule  appears  to  be  that  which  is  bounded  laterally  by  the 
prism-angles,  and  it  therefore  varies  in  Avidth  with  the  expansion  of  these 
faces.  These  major  quadrules  are,  however,  very  largely  obscured  by  the 
finer  reticulation  of  the  surface. 

Dimensions.  Entire  examples  of  this  species  are  not  uncommon.  The 
original  specimen,  Avhich  is  an  entire  internal  cast,  has  a  length  of  64  mm.  ;  its 
greatest  diameter  is  20  mm.  from  the  basal  point  and  measures  18  mm.;  the 
diameter  just  below  the  aperture  is  11  mm.,  and  the  aperture  itself  is  7  mm.  in 
major,  and  5  mm.  in  minor  diameter.  A  larger  specimen  is  87  mm.  in  length, 
21  mm.  in  greatest  diameter,  and  15  ram.  in  diameter  at  the  aperture. 

This  very  neat  and  pretty  species  is  distinguished  by  the  peculiar  cigar- 
shaped  form  of  the  cup  and  its  persistently  small  size.  It  appears  to  be  sub- 
ject to  but  slight  variation  in  these  respects  except  such  as  may  result  from 
compression  or  other  disturbances  in  fossilization.  Occasionally  an  abnormal 
tendency  to  spiral  growth  is  manifested,  and  one  of  the  exaniples  figured 
shows  how  completely  this  tendency  may  affect  the  entire  cup.  Such  a  result 
could  hardly  be  the  result  of  casual  cause. 

A  block  of  sandstone  from  Wells ville  is  crowded  with  these  sponges,  and 
the  exposed  surfaces  of  nearly  all  the  cups,  especially  of  the  larger  individuals, 
bear  numerous  irregularly  oblique  ridges  and  furrows,  which  traverse  the 
cup  more  or  less  completely  in  a  spiral  direction.  In  some  instances  their 
continuity  may  be  ti'aced  more  than  once  around  the  sponge.  In  one  example 
these  mai'kings  are  ridges  on  the  surface  and  grooves  on  the  opposite  side,  the 
two  being  to  a  certain  degree  continuous ;  the  difi"erence  in  their  chai-acter 


SpEOIES    ok    the    ClIKMIJ.Vii    flKOirP.  81 

IS  undoubtedly  a  matter  of  difference  in  retention.  Over  the  surface  of  these 
markings,  whether  ridges  or  grooves,  the  meshes  of  the  reticulum  are  distinctly 
seen.  There  is  so  close  a  resemblance  in  these  features  to  tubes  of  aiuielids 
or  those  left  by  certain  boring  sponges  as  to  justify  the  suggesticju  that  we 
have  here  an  indication  of  parasitism  or  commensalisni  of  such  boring  or 
tubicolous  animals  with  the  Dictyosponges.  On  Plate  ix  (fig.  2)  is  given  a 
figure  of  Hydmceras  tuberosum  var.  glosmna  which  very  distinctly  shows 
similar  vermiform  markings. 

Localities.  In  the  upper  beds  of  the  Chemung  group;  the  original 
specimens  were  derived  from  a  locality  now  unknown,  in  Cattaraugus  county. 
The  species  has  been  found  in  some  abundance  at  Wellsville  and  Fnendship, 

Allegany  county,  N.  Y. 

Prismodictya  baculum,  Hall  (sp.). 

Plate  xviii,  Figs.  1-3. 

1884.     Dictyophyton  baculum,   Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.   State 
Mus.  Nat.  Hist.,  p.  471. 

Sponge  prismatic,  distinctly  octagonal,  the  prism-faces  being  flat  or  very 
slightly  concave;  form  very  gradually  expanding  from  the  base  for  about 
one-half  its  length,  thence  enlarging  rapidly  and  again  contracting  toward  the 
aperture. 

Heticulum.  The  primary  quadrules  measure  about  7  mm.  on  the  edge, 
but  distinctly  enlaige  in  size  from  below  upward.  Each  quadrule  is  bounded 
vertically  by  the  band  upon  the  prism-angle  and  by  one  medially  dividing  the 
prism-face.  The  horizontal  bands  have  left  sharply  defined  impressions  upon 
the  surface  of  the  cast.  The  character  of  the  reticulation  is,  throughout, 
essentially  the  same  as  that  of  Prismodictya  Oonradi,  but  the  species  differs 
from  the  latter  in  its  more  gradual  expansion  from  the  base  upwai-d  and  the 
absence  of  the  peculiar  nodes  of  that  species;  it  differs  from  P.  parallela 
in  its  distal  expansion  and  the  absence  of  elevated  horizontal  ridges  with 
nodiform  intersections. 

Dimensions.  The  length  of  the  best  preserved  example  is  60  nun.;  this 
specimen  is  probably  but  slightly  imperfect  at  the  aperture,  though  somewhat 
more  at  the  basal  extremity.  Its  diameter  at  the  lower  end  is  26  ram.;  at  the 
greatest  width,  43  mm.,  and  at  the  upper  end,  39  mm. 

Locality.     In  the  Chemung  group  at  Wellsville,  Allegany  county,  N.  Y. 


g2  DiCTYOSPONGIDJS. 

Prismodictya  spkctabilis,  sp.  nov. 

Plate  xx,  Fig.  5. 

Sponge  large,  long,  rather  slender;  octagonal  cross-section  well-defined. 
Prism-faces  flat  or,  at  times,  slightly  convex  along  the  median  line ;  diameter 
gradually  increasing  from  the  base  to  a  point  distant  about  one-fifth  of  the 
entire  length  of  the  cup ;  from  this  point  upward  the  width  does  not  vary 
except  for  an  abrupt  contraction  at  the  aperture.  For  a  short  distance  near 
the  base  the  surface  of  the  sponge  is  smooth  and  all  evidence  of  the  prismatic 
faces  obscured;  the  latter,  however,  rapidly  become  defined  and  are  well 
developed  at  about  an  inch  from  the  basal  extremity.  The  impression  of  the 
reticulum  upon  the  original  internal  cast  shows  a  strong  development  of  the 
subordinate  spicular  bands,  the  coarser  meshes  being  pretty  generally 
obscured.  There  are  traces  of  elevated  horizontal  bands  at  intervals  of  about 
10  mna.  and  these  make  nearly  square  quadrules  with  the  prism-angles,  and 
subordinate  squares  or  rectangles  with  the  median  line  on  each  face.  Toward 
the  lower  end  of  the  sponge  the  vertical  bands  are  more  prominent  than  else- 
where. 

Dimensions.  The  specimen  which  has  served  as  the  basis  of  this  descrip- 
tion consists  of  two  fragments,  one  representing  the  basal  and  the  other  the 
upper  part  of  the  sponge,  a  portion  of  the  median  part  being  lost.  Judging 
from  the  slope  of  the  sides  the  oiiginal  length  of  the  sponge  must  have  been 
about  275  mm.  The  full  diameter,  which  is  attained  at  75  mm.  fi-om  the  base, 
is  37  mm.  and  this  is  maintained  nearly  to  the  aperture.  The  sponge  shows  a 
slight  curvature,  but  this  is  probably  casual. 

Locality.  In  the  upper  beds  of  the  Chemung  group  at  Wellsville,  Alle- 
gany county,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Prismodictya  parallela,  Hall  (sp.). 

Plate  xix,  Fig.  4 ;  Plate  xx,  Fig.  6 ;  Plate  xxi.  Figs.  8,  9 ;  Plate  xxxv,  Fig.  9. 

1884.     Dictyophytoii  parallelmn,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State 
Mus.  Nat.  Hist.,  p.  471.  \ 

Sponge  prismatic,  elongate,  slender,  very  gradually  enlarging;  prism- 
faces  slightly  convex,  divided  by  a  median  vertical  ridge  which  is  somewhat 
elevated.  Surface  crossed  by  elevated  horizontal  bands  at  intervals  varying 
from  8  to  10  mm.;  making  low  nodes  at  their  intersection  with  the  prism- 
angles.     The  subordinate  division  of  the  reticulum  is  normal. 


Species  of  tiik  Chemuno  Group.  88 

This  species  is  not  completely  understood.  The  original  speciine!i,  which 
is  an  external  cast  in  sandstone  ineasurin<'  130  mm.  in  length,  has  a  diameter 
of  28  mm.  and  evinces  little,  if  any  evidence  of  ex[)ansion  in  the  cuj).  This, 
of  itself,  is  the  character  which  will  best  serve  to  distinguish  the  form  from 
related  species.  The  horizontal  reticulation  of  the  cup  is  strong,  hut  it  is 
probable  that  its  relative  prominence  is  due  wholly  to  mode  of  pre.servation. 
A  second  specimen  referable  to  the  species  retains  a  portion  of  the  sponge 
without  compression  and  in  this  the  prominence  of  the  median  vertical  band 
on  each  prism-face  obscures  the  sharpness  of  the  primary  octagonal  form  of 
the  cross-section,  duplicating  in  some  degree  the  number  of  prism-fjices. 

A  third  specimen,  though  incomplete  at  both  extremities,  has  a  length  of 
200  mm.  showing  that  the  species  attained  considerable  length  without  notable 
variation  in  caliber. 

Localities.  The  original  specimen  is  from  the  Chemung  group  of  western 
New  York,  its  precise  locality  being  unknown.  Other  examples  of  the  sjiecies 
have  been  found  in  the  upper  sandstones  at  Wellsville,  Allegany  county,  and 
at  Olean  and  Hinsdale,  Cattaraugus  county,  N.  Y.,  by  Mr.  E.  B.  Hall. 

Prismodiotya  PRisMATicA,  Hall  (sp.). 

Plate  xvii,  Fig.  1 ;  Plate  xxi,  Figs.  1-6 ;  Plate  xlii,  Fia.  6. 

1882.     Dktyophyton  fiUtextile,  Hall.     Notes  on  the  Family  Dictyospongidae 

pi.  17,  figs.  2,  2a. 
(Not  D.jiUtextile,  Hall,  1863  and  1884.) 
1884.     Bictijophyton  prismaticum,  Hall.     Thirty -fifth  Ann.  Kept.  N.  Y.  State 

Mus.  Nat.  Hist.,  p.  469,  pi.  (17)  18,  fig.  2. 
1889.     Dictyophyton  prismaticum,  Lesley.     Dictionary  of  Fossils,  p.  200. 

Sponge  regularly  enlarging  from  base  to  aperture,  without  rapid  increase 
in  diameter  at  any  part ;  somewhat  variable  in  shape,  often  without  percepti- 
ble expansion  for  a  considerable  portion  of  the  cup,  and  with  a  slightly  con- 
tracted aperture,  but  usually  expanding  to  an  aperture  whose  diameter  is  the 
greatest  width  of  the  sponge ;  sometimes  curved  and,  not  infrequently,  some- 
what twisted ;  prism-faces  flat. 

Reticulum  as  in  the  allied  species  here  described,  there  being  a  median 
spicular  band  on  each  prism-face,  with  two  or  more  accessory  bands  which 
may  be  more  or  less  distinctly  retained.  The  expression  of  the  surface  varies 
with  the  condition  of  preservation  of  the  individual  specimens,  sometimes  the 


84  DicnrospoNGiD.E. 

major  quadrilles  predominating,  at  others  these  being  obscured  by  the  lesser 
meshes.  At  the  intersections  of  the  prism-angles  with  the  horizontal  spicular 
bauds,  the  internal  casts  show  small  pits  whicli  probably  indicate  the  presence 
at  those  points  of  spicular  tufts.  All  the  principal  bands,  and  perhaps,  also 
the  subsidiary  bands,  were  extended  laterally  from  the  body  of  the  cup  into 
narrow,  erect  lamellae,  from  .5  to  1  mm.  in  width. 

Dimenshm.  An  average  specimen  which  is  nearly  entire  has  a  length  of 
i)0  mm. ;  its  diameter  at  the  base  is  8  mm.  and  at  the  top  27  mm.  Incomplete 
specimens  of  larger  individuals  having  similar  dimensions  abound.  In  one 
specimen  there  is  no  increase  in  diameter  in  a  length  of  80  mm.,  examples  of 
this  character  being  of  somewhat  less  frequent  occurrence  than  the  others. 

This  sponge  is  not  widely  different  from  those  described  as  Prisniodictya 
parallekt,,  and  when  both  are  better  known  it  may  prove  that  the  more  regu- 
larly prismatic  forms  here  descnbed  are  identical  ^vith  that  species,  but  at 
present  there  seems  no  sufficient  ground  for  separating  the  various  forms  of 
P.  pristnatica,  all  of  which  occur  in  association. 

Localities.  In  the  upper  beds  of  the  Chemung  group ;  in  great  numbers  in 
a  light  grey  sandstone  at  Concord,  Erie  county,  Pennsylvania ;  also  at  Wells- 
ville,  Allegany  county,  Olean,  Cattaraugus  county,  N.  Y.,  and  at  "Warren, 
Pennsylvania. 

Prismodictya  Allegania,  sp.  nov. 

Plate  xx,  Fig.  4. 

Sponge  large,  octagonal  in  cross-section,  prism-faces  clearly  but  not 
sharply  defined ;  each  divided  by  a  more  or  less  prominent  median  vertical 
ridge  and  rendered  thereby  somewhat  convex.  Basal  portion  rounded,  the 
prismatic  surface  being  undefined  for  about  one-fifth  of  the  length  of  the 
sponge.  The  expansion  of  the  cup  is  gradual  for  nearly  two-thirds  of  its 
length  where  the  greatest  diameter  is  attained ;  the  upper  part  is  considerably 
expanded  and  again  contracted  towards  and  about  the  aperture. 

Meticulum.  The  prevalent  major  quadrules  are  formed  by  the  intersec- 
tion of  low  horizontal  bands  with  those  on  the  prism-angles  and  on  the  middle 
of  the  prism- faces.  These  average  about  9  mm.  across  the  base  and  11  mm. 
in  height.  The  subordinate  reticulation  is,  as  a  rule,  quite  clearly  retained  to 
the  fifth  or  sixth  degree. 

Dimensions.  The  largest  specimen  observed,  which  is  incomplete  at  the 
basal  extremity  and  for  a  short  distance  near  the  aperture,  is  210  mm.  in 
length,  53  mm.  wide  at  the  lower  end  and  75  rnna.  in  diameter  at  the  greatest 


Species  of  the  CriEMiiNo  Gnoirp.  85 

expansion,  which  is  70  mm.  from  the  upjw  end.  A  somewhat  smaller  but 
more  complete  example  has  a  length  of  185  mm. ;  an  apical  diameter  of  15  mm. ; 
a  greatest  width  of  60  mm.  and  a  subapertural  diameter  of  50  ram. 

This  species  bears  some  similarity  to  Prismodlctya  haculum  in  the  expan- 
sion of  the  cup  above  the  middle,  but  differs  in  its  proportionally  less  slender 
form  and  its  persistently  greater  size.  A  number  of  specimens  from  different 
localities  all  show  the  specific  characters  described. 

Localities.  In  the  upper  beds  of  the  Chemung  group,  at  Scio  and 
Almond,  Allegany  county,  and  Ischua  and  Olean,  Cattaraugus  county,  N.  Y. 
(Collection  of  E.  B.  Hall.) 

Pbismodictya  choanea,  sp.  nov. 

Plate  xix,  Figs.  1-3;  Platb  xxvii,  Fia.  5  ;  Plate  xxxv,  Fig.  6. 

Sponge  large,  elongate,  rapidly  expanding  from  the  base,  attaining  a 
greatest  dianieter  above  the  middle,  thence  somewhat  abruptly  contracting  to 
the  aperture.  Median  transverse  section  rounded  octagonal.  Surface  smooth 
about  the  base,  thence  upward  becoming  faintly  prismatic,  these  prism-faces 
again  becoming  obscure  and  finally  obsolete  toward  the  summit. 

Reticulum  fine ;  composed  of  quadrules  the  largest  of  which,  over  the 
basal  portions  of  the  cup,  are  bounded  vertically  by  the  edges  of  the  prism- 
faces,  crossed  by  horizontal  bands  at  subequal  distances.  This  makes  of  the 
main  quadrules,  squares  measuring  about  30  mm.  on  each  side.  These  large 
areas  are  distinctly  subdivided  by  one  inferior  series  of  bands,  and  each  of 
these,  in  its  ultimate  division,  contains  approximately  256  meshes. 

Toward  the  median  portion  of  the  sponge  the  prismatic  faces  become 
somewhat  more  obscure  and  the  division  into  the  quadrules  of  the  first  and 
second  order  more  sharp. 

Dimensions.  This  description  is  based  upon  specimens  which  represent 
only  the  lower  portion  of  the  sponge  and  we  are  not  justified  in  inferiing  the 
character  of  the  upper  part  except  that  while  the  species  may  assume  the 
proportions  of  Prismodictya  corynia  in  its  later  growth,  it  probably  does  not 
become  so  strongly  prismatic  as  does  that  species.  On  the  other  hand,  it  is 
easy,  from  the  obscurity  of  their  prism-faces,  to  confound  such  specimens  as 
those  figured,  with  the  Dictyospongia  Almondensis,  the  fragments  of  which, 
occuring  in  great  numbers  in  the  grey  sandstone  at  Almond,  sometimes  show 
faint  prismatic  divisions  which  are  wholly  the  result  of  compression. 

Localities.  In  the  upper  beds  of  the  Chemung  group  at  Ischua  and 
Olean,  Cattaraugus  county,  N.  Y.     (Collection  of  E.  B.  Hall.) 


86  DlCTTOSPONGID^. 

Prismodictya  cortnia,  sp.  nov. 

Plate  xvii,  Fig.  11. 

Sponge  large,  the  incomplete  specimen  on  which  the  species  is  founded 
indicating  a  greater  size  than  is  attained  by  any  other  known  representative 
of  the  genus;  expanding  with  some  rapidity  from  the  base  and  contracting 
toward  the  aperture.  Prism-faces  clearly  defined,  some  of  them  with  a 
low  elevation  in  the  median  line,  their  edges  scarcely  elevated  into  ridges. 
Surface  reticulated  by  spicular  bands  of  subequal  size,  which,  by  intersection, 
form  quadrilles  about  6  mm.  square ;  a  size  which  is  uniformly  maintained 
over  the  entire  surface,  except  for  a  slight  diminution  toward  the  base  and 
summit.  This  is  the  quadrule  which  impresses  one  as  being  the  most  promi- 
nent, though  a  combination  of  sixteen  of  these,  four  on  each  side,  covers  one- 
half  the  width  of  each  prism-face.  The  quadrules  are  subdivided  by  three 
sets  of  spicular  bands. 

This  sponge  attained  a  size  equal  to  that  of  P.  clinanea,  but  is  readily  dis- 
tinguished from  that  species  by  its  more  sharply  prismatic  surface.  Prismo- 
dictya clioanea  is  very  obscurely  prismatic  and  in  this  respect  approximates  the 
normal  condition  of  Dictyospongia. 

Dimensions.  Length  of  the  specimen  described,  140  mm.;  probable 
original  length  not  less  than  240  mm. ;  diameter  at  the  lower  end  90  mm. ;  at 
the  middle  110  mm. ;  at  the  top  90  mm. 

Locality.  In  the  upper  beds  of  the  Chemung  group.  Alma,  Allegany 
county,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Prismodictya  filitextilis.  Hall  (sp.). 

Plate  xxi,  Fio.  7 ;  Plate  xxvii,  Pig'  4. 

1873.     Dictyophyton  JiUtextile,  Hall.     Sixteenth  Ann.  Kept.  N.  Y.  State  Cab. 

Nat.  Hist.,  p.  88,  pi.  iv,  fig.  5. 
1884.      Dictyophytan  fiUtextile,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State 

Mus.  Nat.  Hist.,  p.  470. 

Sponge  small,  slender,  very  gradually  enlarging.  Surface  obscurely 
prismatic,  the  prism-faces  being  convex. 

Reticulum,  extremely  fine ;  a  very  sharp  external  cast  shows  that  the 
major  quadrules  are  formed  by  the  intersection  of  horizontal  bands  with  the 
prism-angles,  slight  nodes  being  made  at  these  points.  These  quadrules  measure 
about  8  by  9  mm.,  being  somewhat  higher  than  wide.  The  sub(.)rdiuate 
reticulation,  however,  usually  obscures  the  larger  meshes. 


SpECIKS    of    TliE    ClIKMUNO    GuOUP.  87 

Dimensions.  The  original  specimen  has  a  length  of  60  mm.,  a  width  at 
the  lower  end  of  11  mm.,  and  at  the  upper  end  of  15  mm.  Eswentially  the 
same  proportions  are  exhibited  by  other  specimens.  The  species  is  distin- 
guished by  the  obsolescence  of  the  prism-faces,  its  fine  reticulation  and  usually 
small  size. 

Localities.  In  the  upper  beds  of  the  Chemung  group.  Tlie  original 
specimen  is  from  some  locality,  now  unknown,  in  Steuben  county;  others 
have  been  found  at  Wellsville,  Allegany  county,  N.  Y.,  by  Mr.  E.  B.  Hall. 

Pbismodictya  banano,  sp.  nov. 

Plate  xvii,  Fig.  2. 

Sponge  elongate  and  quite  slender;  regularly  expanding  for  nearly  one- 
half  its  length,  thence  very  gradually  decreasing  in  width  to  the  aj^erture ', 
distinctly  prismatic ;  prism-faces  slightly  concave. 

Heticulum  fine,  the  principal  quadrules  apparently  being  bounded  by  the 
prism-angles  and  subequidistant  horizontal  bands. 

Dimensions.  The  original  specimen,  which  is  complete  at  the  apei'ture 
but  slightly  broken  at  the  basal  extremity,  has  a  length  of  120  ram.  \vith  a 
probable  original  length  of  135  mm.  The  diameter  at  the  lower  end  is 
11  mm.;  just  below  the  middle,  20  mm.,  and  at  the  aperture  13  mm. 

This  species  is  much  more  slender  than  any  other  of  the  forms  of 
Prismodictya  except  P.  aniicitice,  and  as  the  latter  is  quite  distinct  in  size 
and  contour,  the  form  will  be  readily  distinguished. 

Locality.  In  the  upper  part  of  the  Chemung  group  at  Scio,  Allegany 
county,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Prismodictya  amiciti^,  sp,  nov. 

Plate  xxxvi,  Figs.  2-4. 

Sponge  small,  slender,  graceful,  and  cigar-shaped,  being  somewhat 
broadest  toward  the  lower  end  just  above  the  basal  point,  thence  iipward 
tapeiing  very  gradually  and  regularly  to  the  aperture.  In  respect  to  this 
expansion  of  the  lower  part,  the  species  resembles  P.  telnm,  but  is  a  much 
more  slender  sponge  than  that. 

The  prism-faces  are  not  very  sharply  defined,  so  that  without  the  actual 
obliteration  of  the  interfacial  angles,  these  are  rendered  obscure  by  a  general 
convexity  of  the  surface,  which  characterizes  the  entire  length  of  the  sponge. 

Beticulum  as  in  other  forms  of  this  genus. 


gg  DlCTYOSPONGIUvE. 

This  species  will  be  readily  recognized  by  its  persistently  slender  and 
graceful  form  and  subprismatic  surface.  Among  the  examples  figured  one  is 
given  which  again  shows  that  a  spiral  mode  of  growth,  such  as  that  ali-eady 
shown  to  occur  in  specimens  of  P.  telum,  occasionally  manifests  itself  here. 

Bimensions.  The  best  preserved  specimen  has  a  length  of  90  mm.  Its 
greatest  width,  one- third  of  its  length  from  the  base,  is  13  mm,;  at  the 
aperture  the  width  is  9  mm. 

Locality.  Several  specimens  of  this  form  have  been  found  in  the  upper 
sandstones  of  the  Chemung  group  at  Friendship,  Allegany  county.  (Collec- 
tion of  E.  B.  Hall.) 

Prismodictya  aulophia,   sp.  nov. 

Pij^TE  XX,  Figs.  7,  8 ;  Plate  xxxiv,  Figs.  4,  5 ;  Plate  xli,  Fig.  7. 

Sponge  of  moderate  size,  gradually  expanding  to  its  full  width  at  about 
two-thirds  of  its  length,  again  contracting  towards  the  aperture.  Surface 
with  deeply  concave  prism-faces  and  elevated  prism-angles.  Each  prism-face 
is  divided  medially  by  a  vertical  spicular  band,  which  is  clearly  defined  on  all 
specimens.  These  and  the  longitudinal  bands  upon  the  prism-angles,  bound 
opposite  sides  of  major  quadndes  about  6  mm.  square,  the  reticulum  over  each 
of  these  areas  being  extremely  fine.  A  small  specimen  which  appears  to  be 
complete,  is  50  mm.  in  length  and  27  mm.  in  diameter  across  the  upper  ex- 
tremity. A  fragment  of  a  larger  cup  measures  85  mrii.  in  length,  and  the 
entire  length,  judging  from  the  inclination  of  the  sides,  was  upward  of 
100  mm.  The  concave,  fluted  surface  of  this  sponge  renders  it  unlike  any 
other  here  described. 

Localities.  This  species  has  been  found  in  numbers  in  some  blocks  of 
compact  and  finegrained  red  sandstone,  belonging  to  the  upper  beds  of  the 
Chemung  group,  at  Wellsville ;  less  abundantly  in  greenish  and  more  argilla- 
ceous layers  at  the  same  locality,  and,  also,  in  pome  abundance  in  a  greenish 
siliceous  sandstone  at  Scio,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Pkismodictya  ptionia,  sp.  nov. 

Plate  xxxiv,  Fig.  3. 

This  sponge  is  of  rather  large  size  and  of  somewhat  similar  aspect  to  the 
smaller  species,  P.  cithara  and  P.  cercidea,  the  cup  expanding  rapidly 
from  a  narrow,  doubtless  somewhat  extended  stalk  into  a  broad  and  inflated 
body  whose  prism-faces  are  gently  concave.  At,  or  just  above,  the  greatest 
width  of  the  sponge,  low  nodes  are  formed  on  the  prism-angles,  as   in  the 


Species  of  the  CnEMirNo  (troup.  89 

species  cited,  but  are  much  less  conspicuous  than  in  those.  Though  the 
specimen  described  is  incomplete,  it  is  evident  that  the  sponge  rapidly 
narrowed  from  the  line  of  these  nodes  upward. 

The  reticulum  is  as  in  other  prismatic  species.  Along  the  middle  of  each 
prism-face  is  a  conspicvious  spicular  band  and  the  major  quadrules  which  this 
forms  with  the  prism-angles  are  clearly  divided  by  four  series  of  lines,  all  of 
them  distinct  without  any  one  series  obscuring  the  rest. 

Di'mensions.  The  single  specimen  of  this  form  observed  has  a  length, 
covering  only  the  body  of  the  cup,  of  125  mm. ;  its  width  at  the  fractured 
base  is  20  mm.  and  across  the  swollen  body,  110  mm. 

Locality.  From  the  upper  Chemung  sandstone  at  Scio,  N.  Y.  (Collec- 
tion of  E.  B.  Hall.) 

Pbismodictya  cercidea,  sp.  nov. 

Plate  xx,  Fig.  1-3 ;  Plate  xlii,  Fig.  3. 

Sponge  small,  fusiform,  with  well  defined  prism-faces  which  are  flat  or 
somewhat  concave ;  rather  gradually  expanding  for  somewhat  more  than  one- 
half  its  length,  where  its  greatest  diameter  is  attained,  thence  tapering  ^\•ith 
greater  rapidity  to  the  aperture  which  is  contracted  and  very  small.  At  the 
apex  the  eight  prismatic  faces  are  distinctly  defined  and,  at  the  summit,  these 
faces  so  nearly  meet  that  in  some  instances  it  is  impossible  to  determine  which 
is  the  basal  and  which  the  apertural  extremity.  At  the  greatest  expansion  of 
the  sponge,  near  the  middle,  there  are  thickenings  or  elongated  nodes  upon 
the  prism-angles. 

Meticulam  fine ;  prism-faces  devoid  of  prominent  vertical  spicular  bands, 
while  the  horizontal  bands  are  numerous  but  not  conspicuous.  The  largest 
quadrule,  measuring  4  mm.  on  each  side,  is  frequently  repeated,  but  the  minor 
subdivisions  of  the  net-work  prevail  on  all  the  specimens  observed. 

Dimenffiom.  Of  three  entire  individuals  one  measures  62  mm.,  and  two 
55  mm.  in  length.  All  are  somewhat  compressed,  but  agree  in  having  a 
greatest  diameter  of  22  mm.  The  apertural  diameter  is  not  more  than  2  or 
3  mm. 

The  presence  of  low  nodes  on  this  species  shows  its  relationship  to 
P.  ptionia  and  P.  dtlmra.  To  the  latter  P.  cercidea  is  very  closely  allied 
but  does  not  possess  its  long  slender  base. 

Locality.  In  the  upper  part  of  the  Chemung  group  at  Wellsville,  Alle- 
gany county,  N.  Y.,  associated  with  the  lamellibranch,  Sphenotais  contraA;tu8. 
(Collected  by  E.  B.  Hall.) 


90  DieryospoNGiDA 

PRisMODicryA  ciTHARA,  sp,  nov. 

Plate  xxxv,  Fiob.  1-3;  Plate  xxxvi,  Fig.  5. 

Sponge  of  small  size,  with  a  very  gently  expanding,  stalk-like  basal 
portion  which,  at  about  mid-length  of  the  sponge  rather  abrui)tly  widens, 
very  gradually  tapering  toward  the  contracted  aperture. 

The  lower  part  of  the  sponge  is  more  or  less  distinctly  prismatic  and  the 
faces  and  angles  are  retained  over  the  expanded  portion.  Where  the  gi-eatest 
width  of  the  cup  is  attained,  low,  elongate  nodes  are  developed  upon  the 
prism-angles.  These  are  of  the  same  character  as  those  already  described  for 
P.  cei'ciJsa  and  P.  ptionia,  being  very  much  elongated  vertically  and  scarcely 
thickened.  As  in  the  species  cited,  also,  they  appear  in  but  a  single  trans- 
verse row.  By  distortion,  the  nodes  in  one  of  the  specimens  figured  are  made 
to  appear  of  exaggerated  size ;  usually  they  are  low  and  obscure. 

The  reticulum  is  fine,  as  in  P.  cercidea,  and  the  chief  difference  in  these 
species  is  in  the  extended  basal  portion  of  that  under  consideration. 

Dimensions.  A  large,  slightly  distorted  but  nearly  entire  specimen  is 
73  mm.  long,  7  mm.  wide  near  the  base,  and  22  mm.  wide  where  most  expanded. 
This  seems  to  be  exceptionally  large  size  and  a  considerable  number  of  others 
indicate  prevailing  less  size,  a  normal  small  specimen  having  a  length  of 
39  mm.,  a  width  at  the  base  of  4  mm.  and  a  greatest  diameter  of  15  mm. 

Locality.  In  a  compact  siliceous  sandstone  of  the  upper  part  of  the 
Chemung  group  at  Scio,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Prismodiotya  narthecia,  sp.  nov. 

Platk  XXXV,   FiQS.  4,  5. 

This  small  species  is  allied  to  P.  cithara  and  P.  cercidea;  more  closely 
to  the  former,  from  which  it  differs  in  its  more  slender  form,  and  though  it  is 
not  without  the  median  expansion  of  that  species,  yet  this  is  very  much  less 
in  degree  and  bears  only  the  faintest  suggestions  of  nodes.  The  slender 
base  which  it  possesses  in  common  with  P.  cithara,  will  distinguish  it  from 
P.  cercidea. 

But  a  single  specimen  of  this  little  sponge  has  been  observed.  This 
measures  44  mm.  in  length,  3  mm.  in  width  at  the  base,  9  mm.  in  width  Avhere 
broadest,  and  5  mm.  across  the  aperture. 

Locality.  In  the  upper  beds  of  the  Chemung  group  at  Scio,  N.  Y.,  asso- 
ciated with  the  preceding  species.     (Collection  of  E.  B.  Hall.) 


Spkciks  of  the  Cukmuno  Group.  91 

Prismodictya  Conbadi,  Ilall  (sp.). 

Plate  xviii,  Fio.  4 ;    Platk  xix,  Kio8.  5,  6. 

1863.     Dkiyophyton   Conradi,  Hall.     Sixteenth  Ann.  Kept.  N.  Y.  State  Cab. 

Nat.  Hist.,  p.  89,  pi.  v,  fig.  2,  pi.  v  a,  fig.  2. 
1884.     T)icty<q)hyton   Conradi,  Hall.      Thirty-fifth  Ann.  Kept.  N.  Y.  State 

Mas.  Nat.  Hist.,  p.  471,  pi.  (17)  18,  figs.  8,  4,  4a. 
1882.     Dictyophyton  Conradi,  Hall.     Notes  on  the  Family  Dictyospongidse, 

pi.  17,  figs.  3,  4,  4a. 

Sponge  obconical,  slightly  arcuate  toward  the  base;  the  original  specimen 
is  somewhat  compressed  laterally,  sharply  prismatic,  the  eight  j)ri8ra-face8 
being  without  evidence  of  median  longitudinal  ridges,  with  obscurely  concave 
surfaces  and  brondly  elevated  margins.  Toward  the  upper  extremity  are  a 
few  nodiform  protuberances  produced  by  the  depression  and  elevation  of  the 
surfaces  between  the  prism-angles.  These  nodes  are  low,  evenly  rounded  and 
alternate  in  position  on  adjoining  faces,  presenting  thus  to  some  degree 
the  aspect  of  the  nodiform  sui-face  in  the  genus  Physospongia.  The  speci- 
men unfortunately  terminates  soon  after  the  appearance  of  these  nodes. 
The  prism-angles  are  not  modified  by  this  configuratif)n  of  the  faces.  On 
one  side  of  the  specimen,  however,  at  something  less  than  one-half  its  length 
from  the  apex,  are  two  or  three  lower  elevations  in  a  transverse  line  on  the 
prism-angles. 

lidic/fdum.  Each  prism-angle  bears  the  impression  of  a  strong  vertical 
strand  of  spicules,  which  is  crossed  at  subecpial  intervals  by  similarly  strong 
horizontal  bands.  This  interval  is  about  5  mm.  and  is  maintained  from  base 
to  top  without  material  increase  or  diminution.  Where  these  intersect  the 
primary  bands,  there  are  slight  cavities  or  minute  pits  such  as  are  more  dis- 
tinctly seen  in  specimens  of  P.prismatica,  and  which  probably  indicate  the 
existence  of  small  spicular  tufts  at  these  points  of  intersection.  The  absence 
of  prominent  vertical  bands  upon  the  surface  of  the  prism-faces,  gives  to  the 
primary  division  of  the  reticulum  the  aspect  of  a  series  of  transverse  and  flat 
annular  bands.  Considering  the  primary  quadrules  to  be  about  5  mm.  square, 
it  is  possible  to  trace  on  some  parts  of  the  specimen  a  division  of  the  reticulum 
to  the  sixth  degree. 

Dimendons.  The  length  of  the  original  specimen  is  1 30  mm. ;  its  diam. 
eter  at  the  apical  end,  5  mm. ;  at  one-half  its  length  its  major  and  minor  diam- 
eters are  38  and  23  mm.,  while  at  the  upper  extremity  these  are  44  and  22  mm. 


92  DicTYospoKoiD.?;. 

Locality.  In  the  upper  beds  of  tlie  Chemung  group,  at  Randolph, 
Cattaraugus  county,  N.  Y. 

GONGYLOSPONCtIA,  gen.  nov. 
Elongate,  subprismatic    sponges,    with  irregular  protuberances  or  nodes 
developed  on  the  prism-faces  in  single  alternate  or  opposite  vertical  rows. 
Type,  Gongylospongia  Mwrsld,  sp.  nov. 

GoNGYLOSPONGIA   MaRSIII,    Sp.    nOV. 

Sponge  approaching  Prismodicty a  in  general  aspect,  the  prism-faces  being 
clearly  manifested  throughout  the  length  of  the  skeleton.  The  expansion 
from  the  base  upward  is  gradual  and  is  greatest  at  about  two-thirds  of  the 
length  from  the  point  of  attachment. 

Near  the  base,  protuberances  appear  on  each  prism-face  in  a  vertical  row. 
These  protuberances  are  at  first  low  and  irregular  lumps,  separated  from  each 
other  by  more  or  less  longitudinal  depressions.  With  increase  in  the  size  of 
the  sponge,  the  protuberances  become  more  prominent  and  may  show  several 
points  or  apices,  while  the  separating  depressions  are  much  deeper.  When  at 
full  size  the  surface  of  the  protuberances  may  carry  several  small  and  irregular 
concavities. 

On  adjacent  prism-faces  these  nodes  seem  to  correspond  for  the  most  part 
in  position,  but  there  are  noticeable  exceptions  in  which  a  depression  in  one  row 
lies  against  a  node  of  another.  Moreover  there  is  a  frequent  difference  in  the 
size  and  number  of  these  projections  in  the  various  rows.  Irregularity  in  all 
traits  is  their  leading  cliaracteiistic. 

The  prism-angles  are  pretty  sharply  developed  between  these  rows  of 
nodes  and  on  some  of  the  faces  the  vertical  angles  of  the  second  order  are 
seen  to  traverse  the  nodes. 

After  the  maturity  and  full  size  of  the  sponge  has  been  attained,  the 
node-like  irregularities  rapidly  disappear  and  all  the  later  portion  of  the 
sponge  is  free  of  them.  Over  the  apertural  region,  hence,  the  prismatic  and 
primitive  form  of  the  sponge  is  stiikingly  manifested. 

Keticidimi  as  in  Prismodictya. 

Dirnenskms.  The  single  specimen  found  of  this  interesting  species  is 
repi-esented  in  the  accompanying  illustrations.  Its  length  is  190  mm.;  greatest 
width  (under  some  compi-ession)  70  mm.,  and  apertural  width  55  mm. 

Locality.  In  the  Chemung  beds  at  AVellsville,  N.  Y.  (Collection  of 
E.  B.  Hall.)     Named  for  Professor  O.  C.  Marsh,  of  New  Haven,  Conn. 


f 


SpKOIKS    ok    TIIK    CllKMUNG    GliOUP. 


98 


f 


FiODUs  14,  IS.     Oongylotpongia  Manht. 


94 


DlCTYOSPONGlD.IC. 


^    t 


V. 


b  '' ^  i"  I 


>    J  -.  r    ♦ 


V/     c 


FiOHREs  16,  17.     O'otKrvIoaponiria  MarsM. 


Speciks  ok  tiik  Chemiino  Group.  95 

Ohservatlonn.  The  teiuleiicy  among  Diutyosponges  of  prismatic  founda- 
tion to  produce  nodes  upon  the  surface,  manifests  itself  in  two  ways;  these 
nodes  develop  either  along  the  prism-angles,  as  in  Hydnocebas  and  Botbyo- 
DicTYA,  or  on  the  prism-faces,  as  in  Prisnwdlctya  Conradi,  Gonoylospongia 
and  Tylodictya.  In  the  last  named  genus,  however,  the  prismatic  foundation 
is  almost  completely  suppressed,  although  the  nodes  show  very  much  the  same 
irregular  characters  as  in  Gongyl.  Marshi. 

The  characters  of  Prismodictya  Conradi  have  already  been  pointed  out. 
The  single  specimen  known  of  the  species  shows,  near  the  fractured  upper 
end,  an  obscure  nodation  on  several  of  the  sharp  pi-ism  faces.  These  are 
gentle  and  regular  convexities  and  concavities  lying  in  alternating  rows. 
How  irregular  they  may  have  become  in  later  growth  it  is,  of  course,  impos- 
sible to  say,  but  the  most  primitive  form  of  the  node  in  Gongyl.  Marshi  is 
of  more  complicated  shape  than  the  nodes  shown  in  Prism.  Conradi,  while  in 
the  latter  even  these  simple  nodes  appear  at  a  distance  from  the  basal  point 
which,  in  Gongyl.  Marshi,  covers  nearly  the  entire  growth  of  the  sponge  to 
maturity.  It  seems  quite  clear  that  Prism.  Conradi  exemplifies  a  primitive 
stadium  of  Gonoylospongia. 

HYDNOCERAS,  Gonrad. 
1842.     Hydnoceras,    Conrad.        Journal   Philadelphia  Academy    of  Natural 

Sciences,  vol.  viii,  p.  267. 
1863.     Dictyophyton,  Hall.     Sixteenth  Ann.  Rept.  N.  Y.  State  Cab.  Nat.  Hist., 

pp.  84,  87,  90,  91. 
1874.     Dictyophyton,  Larkin.     The  Alfred  Student  (May),  p.  53. 

1879.  Dictyophyton,    Schimper.       Zittel's    Handbuch    der    Palaeontologie ; 

Palaeophytologie,  p.  69. 

1880.  Dictyophyton,  F.  Roemer.     Lethaea  Palaeozoica,  p.  128. 

1881.  Dictyophyton,   Whitfield.     American   Journal   of    Science,  vol.    xxii, 

pp.  53,  132. 

1882.  Dictyophyton,  Whitfield.     Bull.  No.  1,  Amer.  Mus.  Nat.  Hist.,  p.  10. 

1882.  Dictyophytmi,  Zittel.     Neues  Jahrb.  fiir  Mineral.,  etc.,  vol.  ii,  p.  203. 

1883.  Dictyophyton,   Barrois.      Ann.    de   la   Soc.    Geol.  du    Nord,  vol.  xi, 

pp.  80-86. 

1883.  Dictyophytmi,  Hinde.     Cat.  Fossil  Sponges  British  Museum,  p.  130. 

1884.  Dictyophyton.,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State  Mus.  Nat. 

Hist.,  pp.  466,  472,  473. 


96  DiCTYOSPONOIDiE. 

1887.     DlcUjophyton,    Hall.      Sixth    Ann.    Kept.    N.    Y.    State    Geologist, 

p[).  30-38. 
1889.     Jlydnocems,  Beecher.     Mem.  Peabody  Museum,  vol.  ii,  pt.  1,  p.  14. 
1889.     Dictyophyton,  Neumayr.     Die  Stamme  des  Thierreichs,  p.  228. 
1895.     Dictyophyton,  Zittel.     Grundziige  der  Palaeontologie,  p.  52. 

Obconical,  more  or  less  rapidly  expanding  sponges,  with  an  acute  or 
subacute  base  furnished  with  a  short,  often  obscure  tuft  of  anchoring  spicules. 
The  surface,  in  elementary  stages  smooth,  soon  develops  eight  prism-faces  and 
thereafter  prominent  nodes  arranged  in  horizontal  and  vertical  rows,  each 
node  normally  lying  at  the  intersection  of  adjoining  prism-faces.  These 
nodes,  therefore,  correspond  in  number  with  the  prism-angles,  and  are 
normally  eight  in  any  horizontal  row ;  the  number  of  horizontal  rows  in  the 
adult  condition  varying  from  four  to  eight.  Abnormalities  in  place  of 
appearance  of  the  nodes  are  not  infrequent.  The  nodes  may  be  small  and 
acutely  triangular  or  very  broad  and  semi-ovoid  in  outline.  The  extremity  of 
each  is  furnished  with  a  small  tuft  of  spicular  rods.  By  the  development  of 
the  nodes  the  prism-faces  of  the  body  of  the  sponge  are  greatly  obscured. 
The  surface  also  bears  short  erect  lamellae  connecting  the  nodes  of  any  ver- 
tical row,  and  it  is  probable  that  there  were  similar  erect  lamellae,  both 
.vertical  and  horizontal,  which  correspond  to  the  principal  reticulating  bands. 
About  the  aperture  the  surface  is  slightly  expanded  and  the  margin  is  smooth 
and  free  of  prostalia. 

Type,  Hydnoceras  tuheroBum,  Conrad. 

Notwithstanding  the  fact  that  Conrad's  term  Hydnoceuas,  based  upon 
the  now  well  known  s).ecies,  //.  tuberosum,  was  introduced  under  the  con- 
ception that  the  fossil  represented  an  aberrant  or  extravagant  type  of  orthoceran 
cephalopod,  there  is  nothing  in  the  etymological  construction  of  the  name  that 
of  itself  conveys  any  erroneous  notion  of  the  structure.  Consequently  it  seems 
proper  to  rehabilitate  this  finely  characterized  genus  of  sponges  ^vith  its  first 
appellation,  and  restore  to  Mr.  Conrad  the  credit  for  its  discovery  and  first 
description.  It  was,  in  fact,  at  Mr.  Conrad's  own  suggestion,  that  his  name 
gave  \vay  to  the  later  term  Dictyophyton,*  and,  as  else\vhere  observed, 
although  this  term  has  been  generally  adopted  in  the  literature  of  these 
fossils,  it  seems  the  wisest  plan  to  discontinue  it,  not  because  it  was  founded 
on  a  misconception  but  because  it  perpetuates  one.  The  specified  type-species 
of  Dictyophyton  were  three  in  number,  representing  according  to  our  present 

*  See  Sixteenth  Ann.  Kept.  N.  Y.  State  Cab.,  p.  87. 


SpKCraS    OF    THK    CHKMITN(J    GrOITP,  07 

understaudiiig  tliree  distinct  genera ;  none  of  thene,  however,  are  nodc.se  species 
of  the  Hydnoceuas  type. 

The  genus  is  well  characterized,  definitely  restricted  and  most  abundantly 
developed  in  the  Chemung  sandstones  of  New  York. 

Hydnocekas   tubeuosum,  Conrad. 

PLATE  ",  Fios   5-7;  PLAT.  i„,  Fio-s.  1,  2,4;  Platk  iv,    F.os.  2-.;    P,.^tk  v,  F.oh.  3.  4;   Pi.^rr.  v., 
Fig.  3;  Plate  vh,  Figs.  2,  3;  Plate  vin,  Figs.  1,  2;  Platk  xxxvm.  Fio«.  1,  2. 

1842.  Hydnocerm  tuberosum,  Conrad.  Journal  Philadeli.hia  Academy  of 
Natural  Sciences,  vol.  viii,  p.  267,  pi.  xvi,  fig.  1. 

1863.  Dictyophyton  tuberosum,  Hall.  Sixteenth  Ann.  Kept.  N.  Y.  State  Cab. 
Nat.  Hist.,  p.  90,  pi.  iii,  fig.  i. 

1880.     Dictyophyton   tuberosum,   F.    Roemer.      Lethiea   Pala3ozoica,    part    1 
p.  128. 

1883.  Duityophyton   tuberosum,    Hinde.     Catalogue  Fossil  Sponges  British 

Museum,  p.  130,  pi.  xxviii,  fig.  3. 

1884.  Dictyophyton  tuberosum.  Hall.      Thirty-fifth  Ann.  Kept.  N.   Y.  State 

Mus.  Nat.  Hist.,  p.  473,  pi.  17  (18),  figs.  7,  8. 
Not  Dictyophyton  tuberosum,  Barrois.     Annales  de  la  Soc.  Geol.  du  Nord,  vol. 
xi,  p.  82,  pi.  1,  figs.  la-e.     1883. 

Sponge  of  ample  dimensions,  turbinate  or  obconical,  expanding  with 
regularity  ami  moderate  rapidity;  aperture  not  contracted.  Transverse  sec- 
tion of  the  Ijody  of  the  cup  normally  subcircular  or  obscurely  octagonal,  but 
usually  subelliptical  from  compression.  Surface  bearing  strong,  elevated, 
subangular  nodes  which  are  distinctly  arranged  in  two  series,  one  horizontal! 
the  other  longitudinal.  The  horizontal  rows  do  not,  in  any  mature  individual 
obsei-ved,  exceed  four  or  five,  and  each  row  is  separated  from  the  next  adjacent 
by  a  regularly  and  broadly  concave  constriction.  The  vertical  i-ows  are,  nor- 
mally, eight  in  number,  and  these  usually  present  the  aspect  of  four  double 
rows,  the  concavity  of  the  surface  between  adjoining  double  rows  being  notice- 
ably deeper  than  that  between  nodes  of  the  same  double  row.  The  effect  of 
this  arrangement  into  double  rows  is  heightened  by  the  compression  of  the  cup. 
The  nodes  are  low.  subacute  or  blunt,  with  broad  bases  connected  vertically 
by  sharp,  slightly  elevated  ridges  which  cross  the  transverse  constrictions,  and 
honzontally  by  quite  shallow  depressions.  In  the  most  slender  examples 
the  height  of  these  nodes,  measured  from  the  greatest  depth  of  the  horizontal 
constrictions,  is  not  more  than  one-seventh  of  the  full  diameter  of  the  cup  • 
in  robust  individuals  and  young  sponges  it  is  very  much  less. 


98  DlCTYOSPONGID^. 

Tliese  nodes  lie  at  regular  intervals  along  the  margins  of  appressed 
})rismatic  longituditial  faces,  and,  presumably,  at  the  intersection  of  such  ridges 
Avith  horizontal  spicular  bundles  of  the  first  order.  Instances  of  irregularity 
in  the  development  of  these  nodes  are  very  frequent,  indeed  it  is  rarely  that  the 
normal  number  and  order  of  the  nodes  is  retained  in  full-grown  examples. 

Reticulum.  The  specimens  of  this  species  are  preserved  as  external  and 
internal  casts  in  sandstone.  The  very  extensive  material  which  has  been 
imder  examination  has  not  furnished  a  single  instance  in  which  the  parts  of 
the  spicular  skeleton  have  been  indicated  except  as  impressions  in  the  sand- 
stone. Such  impressions  are,  lu)weA'er,  very  sharp,  and  afford  a  complete 
conception  of  the  gross  structure  of  the  skeleton.  The  sui'face  is  very  I'egu- 
larly  i-eticulated  by  vertical  and  hoiizontal  series  of  spicular  bands.  The 
vertical  bands  of  the  first  order  are,  as  already  observed,  those  along  which 
the  vertical  rows  of  nodes  are  arranged  and  correspond  to  the  angles  of  the 
elementary  prism-faces ;  the  horizontal  bands  passing  over  the  nodes  do  not, 
however,  appear  to  be  perceptibly  larger  than  their  neighbors.  The  general 
appearance  presented  by  the  surface  is  that  of  regular  quadration  by  sub- 
equilateral  rectangles,  whose  size  varies  with  the  individual,  and  the  number  of 
which  between  any  two  nodes  in  a  vertical  row  also  appears  to  be  an  indi- 
vidual peculiarity.  In  a  slender  example,  one  of  these  quadrules  measures 
appi-oximately  5  mm.  on  each  side  and  there  are  nine  of  them  between  two 
nodes,  while  in  a  very  I'obust  individual  they  measure  about  10  mm.  on  each 
side  and  there  are  six  between  successive  nodes. 

When  the  surface  is  Avell  preserved,  four  distinct  series  of  rectangularly 
reticulating  bands  may  be  observed  within  each  of  these  quadrules.  External 
casts  of  the  cups,  broken  in  a  favorable  manner,  show  that  the  vertical  bands 
connecting  the  nodes  were  produced  beyond  the  surface,  and  this  we  may 
infer  to  have  been  the  condition  of  all  the  larger  bands ;  thus  producing  an 
exterior  somewhat  similar  to  that  of  Clatukospongia,  though  these  lamellae 
are  broader  in  the  latter  genus.  As  these  longitudinal  lamellae  rise  over  the 
bases  of  the  nodes  they  become  much  broader,  expanding  to  form  a  flattened 
triangular  tuft  at  the  apex  of  each  node,  often  greatly  extended,  and  consider- 
ably broader  than  the  depth  of  the  node.  From  the  summit  of  the  node,  at 
the  center  of  this  triangular  expansion,  extends  a  stout  bundle  of  spicules  to 
and  beyond  the  apex  of  the  lamella  (see  Plate  vi,  fig.  3).  Upon  internal  casts 
of  H.  tuberosum,  there  is  frequently  an  irregular  depi'ession  or  pit  at  the 
summit  of  each  node,  and  it  may  be  inferred  therefrom  that  the  proximal  ends 
of  the  central  tufts  have  left  these  cavities  (see  Plates  iv  and  v). 


Spfx)ies  of  the  Chkmung  Group.  99 

Mode  of  grointli.  The  young  HydiioceraH  tiiheroKum  is  an  anodose, 
narrowly  turbinate  cup,  in  the  pi-iniitive  stage  obcouical  and  thei-eafter  with 
eight  flat  or  slightly  convex  sides.  The  tendency  to  form  nodes  is  first 
indicated  by  low  swellings  a  little  distance  above  the  base.  At  a  short 
interval  there  is  another  row  slightly  more  prominent  than  the  firet,  but 
neither  of  these  is  sufficient  to  obscure  the  plasmatic  form  of  the  cup  (see 
Plate  vii,  figs.  2,  3 ;  Plate  viii,  figs.  1,  2). 

It  has  been  mentioned  above  that  four  or  five  transverse  rows  of  nodes 
characterize  the  adult  of  this  species.  This  statement  is  intended  to  refer  only 
to  the  fully  developed  nodes,  which  may  be  preceded,  over  tlie  ba.sal  and 
prismatic  parts  of  the  cup,  by  one  or  two  rows  of  incipient  nodes ;  further, 
individuals  will  occasionally  bear  a  subapertural  row  of  low  supernumerary 
nodes.  The  frequent  irregularity  in  the  development  of  the  nodes  is  due 
to  various  causes.  A  node  may  fail  to  appear  at  its  proper  place  without 
causing  any  irregularity  in  the  reticulum  (compare  Plate  ix,  fig.  2,  where  in 
the  two  lower  rows  there  is  one  node  wanting,  the  normal  ntimber  being  filled 
only  in  the  third  row  by  the  abrupt  appearance  of  a  full-sized  node).  Many 
of  these  irregularities  of  development  or  growth  are  due  to  injuries  to  the  cup, 
which  the  sarcode  of  the  animal  quickly  repaired,  but  the  net-work  once 
broken  was  rarely  fully  restored  to  its  proper  pattern.  Figure  2  on  Plate  vii 
shows  a  cup  which  suffered  a  rent  or  break  in  the  net-work  at  an  early  stage 
of  development,  and  this  has  left  its  traces  in  all  siibsequent  growth.  The 
irregulaiity  shown  in  figure  1,  Plate  iii,  apparently  produced  by  some 
accident  to  the  cup,  has  most  effectually  modified  the  shape  of  the  sponge. 
On  Plate  vii  (figure  3),  is  shown  a  cup  on  which,  as  a  result  of  injury,  a 
node  lies  in  one  of  the  horizontal  constrictions  and  produces  a  slight  deforma- 
tion, from  which  the  sponge  shortly  recovered.  The  disappearance  of 
the  nodes  towards  the  aperture  may  be  abrupt  or  gradual.  The  apertural 
surface  is  moderately  broad  and  smooth,  distinctly  expanded,  and,  as  shown  in 
several  instances,  not  fringed  or  tufted  on  the  margin  as  in  the  lower  Carbo- 
niferous genus  Lebedictta. 

The  characters  of  this  species  as  above  given  have  been  chiefly  derived  from 
specimens  obtained  at  the  locality  known  as  the  Brown  hill  school-house, 
four  miles  west  of  Wallace,  Steuben  county,  N.  Y.  A  remarkable  colony  of 
these  sponges  was  there  discovered  in  1877,  when,  by  the  grading  of  an  old 
road,  several  hundred  specimens  of  the  species  were  turned  out  of  the  super- 
ficial rock-beds,  and  since  that  time  systematic  operations  have  produced 
many  more,     It   ha§    been  found  that  this  colony  is  quite  restricted  in  its 


100  DlCl'YOSPONGID^. 

extent,  though  it  is  highly  probable  that  its  resources  are  not  yet  exhausted. 
Probably  not  less  than  two  thousand  specimens  have  already  been  collected 
at  this  spot  within  a  radius  of  five  hundred  feet  from  the  district  school-house 
situated  on  the  roadside.  The  species  occurs  here  almost  to  the  exclusion  of 
other  organic  remains.  We  have  a  record  of  three  other  sponges,  Hydnoceras 
rhopalwm,  Jlijdriodictya  patuln  and  Actinodictya  placenta,  represented  by 
single  specimens,  and  also  of  a  few  crinoids  and  an  occasional  brachiopod.  It 
is  especially  important  to  observe  that  in  this  prolific  development,  variations 
from  the  specific  type,  so  far  as  the  great  amount  of  material  studied  would 
indicate,  are  altogether  rare,  if  not  absent.  The  uniformity  of  specific  charac- 
ters is  maintained  throughout  all  the  variations  in  size  and  development.  In 
the  vicinity  of  this  locality  specimens  have  been  found  at  various  points  within 
a  mile  to  the  "west  and  south,  and  all  of  these  occurrences  appear  to  be  at 
about  the  same  geological  horizon ;  they  may  prove  to  be  offshoots  or 
extensions  of  the  Brown  hill  colony. 

Elsewhere  the  species  occurs  in  its  normal  proportions  at  somewhat 
widely  separated  localities  in  the  northern  part  of  Steuben  county  and  the 
southern  part  of  Ontario  coiinty.  A  short  distance  north  of  the  village  of 
Avoca,  on  the  land  of  the  late  Thomas  Co'itox,  specimens  occur  in  abun- 
dance but  in  an  inferior  condition  and  showing  a  constant  tendency  to 
assume  the  characters  normal  to  the  species  H.  Avoca  ^  they  have  recently 
been  found  in  great  numbers  by  Mr.  D.  D.  Luther,  about  two  miles  southeast 
of  the  village  of  Naples,  Ontario  county.  The  species  has,  iu  fact,  the  most 
general  distribution  in  the  Chemung  rocks  of  any  of  the  Dictyosponges,  but 
it  does  not  range  high  in  this  series  of  faunas;  although  the  specimen 
described  by  Cojntbad  was  stated  to  have  come,  from  Addison,  Steuben  county, 
no  examples  of  the  species  from  so  high  an  horizon  have  come  under  our 
observation. 

The  discovery  of  similar  nodose  sponges  in  the  Psaramites  du  Condroz,  of 
France,  the  e(|uivalent  of  the  Chemung  sandstones  of  New  York,  has  been 
recorded  by  Bakuois.*  The  fossil  which  this  author  has  referred  to  Dictyo- 
pliyton  titheroswm,  Conrad,  seems  to  represent  a  species  clearly  distinct  from 
any  of  the  known  American  nodose  forms,  as  is  indicated  by  its  deep  hori- 
zontal constrictions,  sharp  prismatic  faces  and  low  carinate  nodes  situated  on 
broad  and  stout  horizontal  annulations.  This  fossil  is  described  and  figured 
elsewhere   in   this    volume    under    the    name,  Hydnoceras   Barraisi,  and  in 

*  Sur  les  Dictyospongidee   des  Piiatnitiites  du  Condroz ;  (Aunales  de  la  Socictd  Gdologiquo  du  Nord, 
vol.  xi,  p.  80,  pi.  i,  1883), 


Species  op  the  Chemung  Group.  101 

connection    therewith    other  nodose   species  from  the  same  formation,  most 
kindly  submitted  by  Professor  Barrois. 

Dimensions.  The  material  stndied  shows  wide  differences  in  respect  to 
size.  Some  comparatively  small  individuals  indicate  full  growth,  while  occa- 
sionally much  larger  specimens  evince  immaturity  in  the  development  of  the 
nodes.  The  specimen  figured  on  Plate  v  (tig.  4),  is  a  very  good  example 
of  an  average  individual,  its  onginal  length  having  been  about  200  mm.  A 
large  but  slender  cup  has  a  length  of  280  mm.,  Avith  a  greatest  diameter 
of  95  mm.,  while  a  shorter  example,  220  ram.  in  length,  has  a  maximum 
width  of  135  mm. 

Localities.  In  the  lower  and  middle  portions  of  the  Chemung  sand- 
stones; in  the  town  of  Italy,  Yates  county,  and  in  the  southern  part  of 
Naples,  Ontario  county;  at  various  localities  in  Steuben  county  principally 
west  of  the  Cohocton  river.  Brown  hill  school-house.  Potter  hill,  scattered 
localities  in  the  towns  of  Howard  and  Avoca.  The  species  also  occurs, 
though  rarely,  in  the  Jenks  quarry  at  Bath. 

Hydnocebas  tuberosum,  Conrad, 
var.  GLOS8EMA,  var.  nov. 

Plate  vii,  Figs.  4,  5 ;  Plate  ix,  Fia.  2. 

This  name  is  introduced  for  a  form  distinguished  from  IT.  tuheros^im  by 
its  slender,  very  gradually  expanding  cup,  with  low,  horizontal  constric- 
tions and  depressed,  subacute  nodes  in  five  or  six  well-developed  rows. 
The  vaiiety  is  based  upon  several  specimens  whose  dissimilarity  from  forms 
of  IL  tuberosum'  is  very  apparent.  In  both  of  the  specimens  figured  the 
spicular  net  work  is  finer,  that  is  the  principal  quadrules  are  smaller  than 
in  II.  taberosuni.  The  specimen  on  Plate  ix,  fig.  2,  is  interesting  for  the  traces  it 
shows  of  annelidan  tubes,  suggesting  a  coramens;ilism,  similar  evidences  of 
which  have  been  observed  in  other  Dictyosponges.  (See  page  81,  and  also 
remarks  on  Symbiosis  in  the  Hexactinellida,  page  21.) 

Locality.  In  the  sandstones  of  the  Chemung  group,  in  New  York. 
Both  examples  are  from  unknown  localities,  but  probably  from  the  middle 
portion  of  the  group  in  northern  Steuben  county. 

Hydnoceras  Bathense,  sp.  nov. 

Plate  hi,  Fig.  3;  Plate  viii,  Fig.  4;   Plate  ix,  Fig.  1. 

This  species  has  the  same  general  form  of  cup  as  Hydnoceras  tuherosxim, 
but  is  distinguished  from  that  by  the  greater  number  of  horizontal  roAvs  of 


102  DlCTYOSPONGID^ 

nodes  in  the  mature  individual,  there  being  usually  five  or  six  well  developed 
rows,  and  also  by  the  size  of  the  nodes  themselves,  which  are  strongly  elevated, 
compressed  laterally  and  have  an  elongated  semielliptical  or  oval  outline. 
The  height  of  these  nodes  is  often  equal  to  one-third  and  sometimes  even  to 
one-half  the  diameter  of  the  body  of  the  cup.  Their  summits  are  almost 
invariably  wanting,  indicating  a  tufted  reticulum  at  these  points.  The  sub- 
prismatic  surface  of  the  cup  is  less  clearly  retained  than  in  II.  tuheroHum,  and 
the  rate  of  expansion  and  the  concavity  of  the  horizontal  constriction  is  less 
pronounced. 

Sponges  with  these  characters  well  defined  have  thus  far  been  found 
largely  localized  in  the  neighborhood  of  Bath,  N.  Y.,  where  the  specific  traits 
described  are  maintained  by  multitudes  of  individuals.  Away  from  this 
locality  there  seems  to  be  a  progressive  modification  in  the  form  of  the  nodes, 
which  lose  their  elongate  tapering  outline  and  gradually  approach  their  very 
distinct  expression  in  the  species,  II.  Avoca,  The  predominance  of  these  two 
forms  under  different  geographical  conditions  justifies  their  description  as 
distinct  species  notwithstanding  the  desultory  occurrence  of  passage  forms  in 
other  localities. 

Locality.  Ilydnoceras  Bathense  is  exceedingly  abundant  at  the  Jenks 
quarry,  one  mile  southeast  of  the  village  of  Bath,  Steuben  county,  where  it 
occurs  in  thin  shaly  layers  between  two  heavy  compact  beds  of  greenish 
sandstone,  whose  fine  grain  has  preserved  the  lines  of  the  reticulum  with  much 
distinctness.  In  association  with  it  at  this  locality  occurs  an  occasional  speci- 
men of  H.  hotrcedema.  Outside  of  the  Jen1c8  quarry  colony,  typical  specimens 
have  not  been  found. 

Hydnoceras  botrcedema,  sp.  nov. 

Plate  ii,  Fig.  4. 
Sponge  rather  small,  slender,  very  gently  expanding  ;  with  four  horizontal 
rows  of  very  strong  digitiform  nodes  separated  by  deep  horizontal  con- 
strictions. Over  these  concave  areas  the  surface  of  the  cup  is  totally  without 
trace  of  the  prismatic  surface  shown  in  allied  species,  and  when  uncom- 
pressed is  circular  in  section.  The  body  of  the  cup,  were  the  nodes  removed, 
would  be  very  narrow  and  quite  cylindrical.  The  nodes  are  arranged  in  eight 
vertical  rows,  which  are,  in  reality,  four  double  rows,  and  any  two  adjacent 
meml>ei-s  of  these  double  rows  arise  from  a  common  and  thickened  base. 
Nearly  one-half  the  height  of  these  nodes  from  the  bottom  of  the  horizontal 
constriction  upwards,  is  below  the  level  of  the  concavity  dividing  their  prox- 
imal surfaces.     Thus  each  pair  of  nodes  stands  upon  an  elevated  base.     The 


SpEcras  OF  TiiE  Chkmuno  Group.  103 

direction  of  the  nodes  of  any  pair  is  divergent  and  to  such  a  degree  as  to 
bring  the  summits  of  all  the  nodes  to  about  the  same  interval.  Each  node  is 
laterally  compressed  and  projected  downward,  so  that  each  horizontal  ro^v 
presents  a  circlet  of  eight  drooping,  almost  pendulous  pouches. 

Reticulum  rather  fine-meshed. 

Dimensions.  The  species  does  not  vary  much  in  size.  An  average 
example  measures  from  apex  to  aperture,  135  mm.;  its  greatest  width,  across 
the  top  row  of  nodes,  is  60  mm.;  the  diameter  of  the  body  of  the  cup  at  the 
subapertural  constriction  is  33  mm. 

The  characters  of  this  species  are  pronounced  and  readily  recognized.  Its 
small  size,  subcylindrical  body,  elevated  rows  of  double  nodes,  and  the  elon- 
gate, I'ecurved  form  of  the  latter  give  it  an  expression  not  otherwise  reproduced. 

Locality.  The  species  occurs  in  hundreds  at  a  locality  known  as  Irish 
hill,  near  the  village  of  Bath,  Steuben  county.  All  the  specimens  that  have 
been  examined  from  this  spot  (the  Ii'IhIi  hill  coUymj)  were  found  loose  in  the 
fields  and  no  systematic  excavation  has  been  made  for  them  into  the  underlying 
rocks.  It  also  occurs,  though  not  abundantly,  in  association  with  II.  Batlienm 
in  the  Jeiiks  quarry  cohny.  A^vay  from  the  immediate  vicinity  of  liath  the 
fossil  has  not  been  observed. 

Hydnoceras  Avoca,  sp.  nov. 

Plate  ii,  Figs.  1,  3;  Plate  ix,  Fig.  3;  Plate  x,  Pig.  1 ;  Plate  xi,  Fio.  1. 

This  form  of  Hydnoceras  is  rendered  very  distinct  by  its  short,  broadly 
expanded  nodes,  very  stout  and  wide  at  the  base  and  Avith  a  subsemioval  out- 
line. The  number  of  horizontal  rows  of  nodes  is  usually  four  or  five,  includ- 
ing the  incipient  row  near  the  base.  The  horizontal  constrictions  are  narrow 
and  deep,  and  the  prismatic  surface  is  obscurely  retained.  All  the  fonns  of 
this  species  observed  indicate  great  size  at  maturity,  as  shown  by  the  specimens 
represented  on  Plates  x  and  xi.  That  on  the  latter  plate  is  the  largest  nodose 
sponge  yet  seen  and  must  have  had  a  length  when  entire  of  not  less  than  625 
mm.,  or  about  2  feet.  Reference  has  been  made  above  to  the  occasional  occur- 
rence of  passage  forms  from  H.  Batliense  to  this  species,  but  the  two  represent 
sharp  extremes  of  variation  in  the  form  of  the  nodes. 

Locality.  Hydnoceras  Avoca  is  represented  in  the  material  under  study 
by  perhaps  a  dozen  specimens  from  the  vicinity  of  Avoca,  Steuben  county. 
Their  precise  position  in  the  rocks  has  not  yet  been  ascertained  as  the  spec- 
imens, most  <jf  them  collected  by  the  late  Thomas  Cotton  of  that  village, 
have  been  found  loose.      There  is  no  satisfactory  evidence  of  theii-  occurrence 


X04  DictyosponoidjE. 

in  the  Dictyosponge  colony  of  H.  tuherosum  upon  the  farm  of  Mr.  Cotton. 
The  lai'ge  example  referred  to  was  taken  from  the  foundation  wall  of  a 
ruined  mill  in  the  village  of  Avoca.  A  single  example  Avith  the  characteristic 
nodes  of  the  species  comes  from  the  Jenks  quarry,  Bath,  N.  Y. 

Htdnoceras  phymatodes,  sp.  nov. 

Plate  vi,  Pigs.  4,  5 ;  Plate  xxxvii,  Fig.  3. 

Sponge  distinguished  from  other  species  of  this  genus  by  its  broadly 
expanded  form  and  small,  abundant  nodes.  The  basal  portion  of  the 
cup  is  anodate  for  some  distance,  and  the  eight  prismatic  faces  are 
clearly  apparent  though  somewhat  obscured,  the  nodes  of  the  vertical  rows 
being  connected  by  longitudinal  ridges  which  are  also  clearly  defined  over  their 
summits.  To^vards  the  base  of  the  cup  these  nodes  are  in  eight  longi- 
tudinal rows,  but  near  the  middle  or  toward  the  aperture  this  number 
is  increased  by  intercalation  to  nine,  ten  or  eleven.  The  specimen 
figured  on  Plate  vi  shows  the  introduction  of  a  single  vertical  line  of  nodes 
at  the  fourth  horizontal  row  (counting  the  incipient  row  near  the  base),  and 
on  the  opposite  side  of  the  cup  the  abrupt  appearance  of  two  new  lines  in 
the  fifth  horizontal  row.  The  cup  also  shows  faint  evidence  of  an  eighth  sub- 
apertui-al  horizontal  row.  The  specimen  shown  on  Plate  xxxvii  has  similar 
irregularities  in  the  number  and  appearance  of  the  nodes.  All  of  these  nodes 
are  low,  with  rounded  margins  and  surfaces  sloping  nearly  equally  in  all 
directions,  the  concavities  between  the  members  of  each  horizontal  row  being 
nearly  as  deep  as  those  between  successive  rows.  The  horizontal  constrictions 
are  thus  very  shallow ;  they  are  also  closely  approximate  and  distinctly 
marked  by  the  elevated  edges  of  the  vertical  prismatic  faces. 

Reticulum  fine,  the  various  series  of  spicular  bands  not  being  very  dis- 
tinctly defined. 

Dimensions.  Specimens  of  this  species  are  usually  much  compressed 
laterally,  probably,  however,  without  much  exaggeration  of  the  actual  rate  of 
expansion  of  the  cup.  The  figured  example  has  the  following  dimensions : 
Length,  180  mm.;  greatest  width,  83  mm.;  height  of  nodes,  5  to  8  mm.;  dis- 
tance between  the  horizontal  rows  of  nodes,  about  10  mm.  A  larger  example 
has  somewhat  stouter  and  less  rapidly  multiplying  nodes. 

Localities.  In  the  schistose  Chemung  sandstones  near  Bath,  Steuben 
county,  N.  Y.  The  species  is  rare  at  this  })lace,  and  its  occurrence  can  not 
be  given  more  precisely.     A  specimen  of  much  smaller  size  but  of  similar 


Species  of  the  Chemung  Group.  105 

aspect,  has  been  found  by  D.  D.  Luthkr  in  the  lower  Chemung  beds  at  Italy 
hollow,  Yates  county. 

Hydnocekas  hypastrttm,  sp.  nov. 

Plate  tii,  Fig.  1. 

Among  the  specimens  obtained  from  the  late  Rev.  Jonathan  Allen, 
of  Alfred,  is  one  showing  in  cross-section  a  cup  with  a  cii'cular  or  subcylindrical 
body  and  eight  very  strong  nodes  in  four  pairs.  Each  pair  is  raised  on  an 
elevated  base  as  in  H.  hotroedema.  These  nodes  are  very  long  and  droop 
toward  the  base  of  the  cup.  Though  the  summits  of  all  are  broken,  sufficient 
is  retained  to  indicate  their  great  size  and  length.  The  greatest  diameter  of 
the  specimen  between  the  broken  extremities  of  the  nodes  is  1 30  mm. ;  the 
diameter  of  the  cylindrical  body,  57  mm.  Though  the  fragment  indicates  a 
form  similar  to  II.  hotrcedema,  the  nodes  are  proportionally  larger  and  the  cup 
itself  of  much  greater  size  than  in  that  species.  It  is  probable  that  when 
better  known  this  form  will  prove  to  be  distinct  from  the  others  here 
enumerated. 

Locality.  In  the  Chemung  group ;  from  a  greenish  sandstone  in  associa- 
tion with  Cryptodwtya  Alleni ;  probably  from  the  lower  part  of  the  series  in 
Steuben  county,  N.  Y, 

Hydnoceras   eutheles,   sp.  nov. 

Plate  viii,  Figs.  5,  6 ;  see  also  Plate  xxxiii,  Fig.  7. 

This  form,  which  is  represented  by  two  very  similar  fragments  from  the 
same  locality,  combines  in  a  certain  measure  the  form  of  II.  Batlien.^  with  the 
pendulous  nodes  of  II.  hotroedema.  The  sponges  are  short,  expand  rather 
rapidly  and  show  obscure  evidence  of  the  prismatic  faces ;  the  horizontal  rows 
of  nodes  were  probably  not  more  than  four  in  number ;  the  nodes  themselves 
are  extremely  long,  somewhat  compressed  laterally,  the  outline  being  incom- 
plete at  their  extremities,  thus  indicating  a  strong  spicular  tuft  at  these  points. 
In  the  elongate  form  of  the  nodes  the  species  is  even  more  extreme  than  H. 
hotroedema,  while  it  lacks  the  subcylindrical  body  and  slender  form  of  the 
latter. 

Birrmisions.  The  better  of  the  two  examples  has  a  length  of  80  mm. 
from  the  aperture  to  the  end  of  the  third  horizontal  row  of  nodes,  which  could 
not  have  been  far  from  the  base  of  the  cup.  Its  diameter  at  the  aperture  is  90 
mm. ;  between  the  two  lower  rows  of  nodes,  38  mm.  One  of  the  nodes  has  a 
length  of  25  ram. 


106  DlCTYOSPONGID^. 

Localities.  In  the  schistose  Chemung  sandstones,  on  the  farm  of  Thomas 
Cotton,  near  Avoca,  N.  Y.  Associated  with  this  species  here  are  great  numbers 
of  a  coarsely  meshed  sponge,  probably  H.  tuberosum,  which  is  always  poorly 
preserved ;  also,  //  muUinodosum,  Arystidictya  elegans,  A.  nodifera  and 
HaUodidya  Cottoiiiami.  Reference  has  already  been  made  to  the  probable 
presence  of  this  species  in  the  Portage  group  at  Varysburg,  Wyoming  county 
(page  70). 

Hydnocebas  multinodosum,  sp.  nov. 

Plate  xxxix,  Fig.  1. 

Among  the  specimens  from  the  Dictyosponge  colony  at  Cotton  hill,  the 
great  majority  of  which  are  very  large  though  usually  imperfect  individuals 
approaching  Hydnoceras  tuberosum,  are  a  few  fragments  indicating  a  species  of 
this  genus  noteworthy  in  its  size  and  the  remarkable  abundance  of  nodes.  The 
best  of  these  specimens  is  that  figured,  which  represents  the  upper  part  of  a 
large  cup  laterally  compressed,  as  is  the  case  with  all  the  sponges  at  this  local- 
ity. Three  transverse  rows  of  nodes  are  shown  on  the  exposed  side,  the  lower 
bearing  eight,  the  middle  one  ten,  and  the  upper,  which  appears  to  be  the 
apertui-al  row,  seven.  All  are  small  and  low.  The  adventitious  character  of 
many  of  these  nodes  is  evident  from  a  comparison  of  the  different  rows  ;  thus 
one  incipient  node  in  the  lower  row  is  not  represented  in  either  of  the  others ; 
similarly  one  node  of  the  second  row  has  no  equivalent  above  or  below, 
while  another  is  represented  in  the  top  row  but  not  in  the  lower.  The  total 
number  of  nodes  in  each  upper  row  must  have  been  not  less  than  from  14  to 
20,  a  feature  which  at  once  distinguishes  the  species.  The  more  conspicuous 
nodes  in  the  vertical  rows  are  connected  by  slightly  elevated  ridges  or  i-em- 
nants  of  the  prism-angles.  The  apertui'al  width  of  the  fragment  is  180  mm., 
its  length  110  mm. 

Locality.  In  the  lower  Chemung  group,  one  mile  north  of  Avoca, 
Steuben  county,  in  association  ^vith  II.  taberosiim,  Arystidictya  elegans,  A. 
nodifera,  Hallodictya  Cottoniana,  etc. 

Hydnoceras  Lutiieri,  sp.  nov. 

Plate  viii,  Figs.  7-10. 

Sponge  very  small  and  graceful,  very  gradually  exjaanding,  prism-faces 
greatly  obscured,  growth  usually  regular.  The  rate  of  expansion  from  the 
base  to  the  first  row  of  nodes  is  greater  than  elsewhere  except  about  the 
aperture.  The  nodes  are  well  developed,  with  rounded  extremities,  and 
are   arranged  in  three  or  four  horizontal  rows  which  are  separated  by  very 


Species  of  the  CiiEHnNG  Ghoitp.  107 

broad  depressions.  AVhen  there  are  but  three  rows  of  these  nodes,  the  first 
or  lowest  is  sharply  developed,  and  at  times  there  is  evidence  of  an  obscure 
row  in  the  apertural  region,  as  shown  in  one  of  the  specimens  figured.  When 
the  four  rows  are  nonnally  developed  those  of  the  first  row  ai-e  small  and 
incipient.  The  aperture  is  situated  at  somewhat  less  than  the  usual  interval 
al)ove  the  last  row  of  nodes,  and  is  expanded  to  nearly  the  full  diameter  of 
the  cup  at  the  last  row.  The  number  of  nodes  in  each  row,  while  normally 
eight,  is  sometimes  but  seven  on  the  earliest  rows. 

The  reticulam  is  fine,  usually  sharply  retained,  and  shows  primary 
sjjicular  bands  midway  between  the  nodes  both  longitudinally  and  horizon- 
tally. The  tufted  extremities  of  the  nodes  are  seldom  preserved  in  the  speci- 
mens examined. 

Dimensions.  Nearly  all  the  examples  of  this  diminutive  and  graceful 
species  are  entire,  a  most  unusual  occurrence  among  species  of  this 
genus.  One  such  individual,  mature  in  all  its  characters,  has  a  length  of 
40  mm.,  a  width  at  the  first  nodes  of  8  mm.,  greatest  width  of  13  mm.  and  an 
apertural  diameter  of  12  mm.  A  somewhat  stouter  specimen  has  the  same 
length  but  a  maximum  diameter  of  20  mm.  Individuals  much  smaller  than 
either  of  these  hav'e  also  been  observed. 

Locality.  In  the  sandstones  of  the  lower  part  of  the  Chemung  group. 
Nearly  fifty  individuals,  characterized  by  their  slender  form  and  persistently 
small  size,  have  been  found  by  D.  D.  Lutiikk,  in  a  single  loose  slab,  in 
Segur's  gully,  township  of  Italy,  Yates  county,  N.  Y.  Specimens  of  Hydiw- 
cei-as  tuhei-osum  occur  in  the  same  rock. 

Hydnoceras  bhopalum,  sp.  nov. 

Plate  xxxvii,  Figs.  1,  2. 

This  form,  found  only  in  the  Brown  hill  colony  of  H.  tuberosum,  is 
too  unlike  the  latter  to  justify  its  being  regai-ded  as  an  abnormal  individual 
expression  of  that  species. 

The  sponge  is  elongate  and  slender,  its  most  expanded  portion  at  two-thirds 
of  its  length  from  the  base,  having  about  thrice  the  Avidth  of  the  fractured 
lo^ver  end.  Nearly  the  entire  lower  half  of  the  cup  is  devoid  of  well  developed 
nodes,  though  showing  distinctly,  under  compression,  the  pnsm-faces. 

The  number  of  horizontal  rows  of  nodes  is  large,  but  the  nodes  them- 
selves are  seldom  conspicuously  developed.  These  rows,  counting  all  which 
show  incipient  nodes,  are  eight  or  nine  in  number,  and  appear  to  bear  normally 
ei'^ht  in  each  row,  but  in  several  instances  they  are  irregular  in  this  respect, 


o 


108  DiCTYOSPONGIDiE. 

nodes  failing  to  appear  at  their  proper  position.  The  upper  portion  of  the 
cup  is  somewhat  distorted  from  accident  and  the  mending  of  the  net-work. 

The  reticulum  is  similar  to  that  in  H.  tuberosum. 

Dimensions.  This  specimen  has  a  length  of  125  mm.,  which  is  nearly  the 
original  extent  of  the  sponge.  At  the  lower  extremity  its  width  is  13  mm. 
Where  broadest,  the  width  is  36  mm. 

Locality.  Lower  Chemung  sandstone.  Brown  hill  school  house,  near 
Wallace,  Steuben  county.     (Collection  of  E.  B.  Hall.) 

Hydnocebas  variabile,  sp.  nov. 

Plate  xxxvi,  Fig.  1 ;  Plate  xxxvii,  Figs.  4-6. 

Sponge  of  moderate  size ;  expanding  slowly  from  the  base  for  one-third  or 
nearly  one-half  of  its  length,  thence  upward  widening  more  abruptly  over  the 
node-bearing  surface.  The  form  is  thus  somewhat  more  slender  below  than 
in  other  species  and  is  further  characterized  by  an  apparent  absence  of  nodes 
over  the  basal  region,  or  if  not  altogether  absent  their  existence  is  only  sug- 
gested here  and  there.  Normally  there  appear  to  have  been  not  more  than 
three  horizontal  rows  of  nodes  over  the  expanding  portion  of  the  cup.  In 
all  the  internal  casts  representing  the  species  these  are  low  and  the  full  num- 
ber, eight,  is  seldom  developed.  Fragments  of  the  lower  parts  of  the  cup  are 
so  nearly  free  of  nodes  as  to  fail  to  even  suggest  the  genus  Hydnoceras.  Yet 
when  a  specimen  is  retained  in  its  matrix,  as  shown  in  figure  1,  on  Plate  xxxvi, 
this  apparently  anodate  basal  surface  is  found  to  have  borne  sharply  defined 
nodes  nearly  to  the  apex.  As  ordinarily  preserved  the  specimens  are  very 
variable  in  their  general  aspect,  not  only  on  account  of  irregularity  of  appear 
ance  in  the  nodes  but  also  because  of  the  irregular  expansion  of  the  reticulum. 
The  fundamental  prism-faces  are  almost  wholly  obliterated  even  on  the  basal 
portions  of  the  cup. 

Reticulum.  The  net-work  appears  on  internal  casts  to  be  coarse,  on 
account  of  the  predominance  of  the  secondary  spicular  bands  forming  meshes 
measuring  about  2  mm.  on  each  side.  The  general  aspect  of  the  reticulum  is 
a  composition  of  such  rather  coarse  meshes,  but  these  are  more  or  less  dis- 
tinctly subdivided. 

Dimensions.  A  specimen,  the  most  nearly  entire  of  any  observed  and 
having  the  nodes  well  developed,  is  125  mm.  in  length  and  probably  its  origi- 
nal length  was  not  less  than  140  mm.  At  the  lower  end  its  width  is  30  mm.; 
its  greatest  width  65  mm. 


Speciks  op  thk  Cmkmitnc}  Group.  109 

Locality.  From  the  lower  beds  of  the  Chemung  group,  not  more  than 
one  hundred  feet  above  the  Portage  sandstones,  at  Deyo  basin  in  the  southern 
part  of  the  town  of  Naples,  N.  Y. 

Associated  with  it  are  Ceratodietya  annnkita,  Spirifer  mesacoHtalis, 
Atrypa  hystricc,  ProdwtelUiU;liryitwm,  Amhocidm  umboiuiUt,  Ilystracanthua, 
sp.,  etc.     (Collected  by  D.  D.  Luthek.) 

HTDNOCJiRAS   NODOSUM,    Hall    (sp.). 
Platk  ii,  Fio.  2. 

1863.     Dictyophyton  nodosum,  Hall.     Sixteenth  Ann.  Kept.  N.  Y.  State  Cab. 

Nat.  Hist.,  p.  91,  pi.  iii,  fig.  2. 
1884.     Dictyophyton  nodosum,  Hall.    Thirty-fifth  Ann.   Rept.  N.  Y.  State 

Mus.  Nat.  Hist.,  p.  472,  pi.  17  (18),  fig.  6. 

"A  fragment  two  and  a  half  inches  in  length,  being  an  impression  in  shaly 
sandstone,  preserves  the  marks  of  longitudinal  ranges  of  rounded  nodes  there 
being  six  [five]  nodes  in  each  longitudinal  row  in  the  length  indicated. 

"This  species  differs  from  the  preceding  [^  tuberosum]  in  having  longi- 
tudinal rows  of  nodes  more  nearly  parallel,  smaller  and  more  clearly  arranged 
in  rows,  while  they  are  not  angular.  The  surface  is  finely  reticulated  by 
longitudinal  and  transverse  striae. 

"Geoloijical  formation  and  locality.  In  the  shaly  sandstones  of  the 
Chemung  group  in  Cattaraugus  county  (N.  Y.)."     (op.  cit.  1863.) 

This  is  a  small  species  which  appears  to  have  some  resemblance  to  H. 
hotrosdemxi  in  its  slender  cup  and  strong  nodes  though  possessing  one  more 
horizontal  row  of  nodes  than  any  examples  of  the  latter  species  have  yet  shown. 
Nothing  closely  comparable  to  the  original  specimen  has  been  observed,  and 
hence  our  knowledge  of  the  specific  form  is  still  quite  imperfect. 

Hydnoceras  anthraois,  sp.  nov. 

Plate  ii,  Fig.  8;  Plate  vi,  Figs.  1,  2;  Plate  xxxiii,  Figs.  8,  9. 

Sponoe  small,  narrow  and  gracefully  expanding  for  one-third  of  its  length, 
thence  upward  gently  contracting  to  a  well-defined  constriction  just  below  the 
aperture. 

Surface  sharply  prismatic  for  nearly  one-half  the  length  of  the  cup.  Not 
until  the  sponge  attains  its  full  diameter  do  nodes  appear  and  these  manifest 
themselves  with  some  irregularity.     In  the  single  entire  specimen  observed 


no  DlCTYOSPONGIDiF,. 

there  are  hut  three  transverse  rows  oi  nodes,  and  in  the  lowest  row  but  three 
of  the  normal  eight  nodes  are  distinctly  developed,  three  others  are  merely 
suggested  while  two  are  absent.  In  the  second  and  third  rows  all  nodes  are 
present  but  are  unetjually  developed.  The  fi'agment  of  a  second  and  some- 
what larger  specimen  shows  four  transverse  I'ows  of  nodes.  All  of  these 
nodes  are  small  and  are  elongated  by  the  prism-angles  on  \\hicli  they  lie. 
Their  presence  does  not  materially  obscure  the  fundamental  prismatic  form  of 
the  s2)onge. 

In  full  development  each  node  is  prominent,  its  height  ecpialling  about 
one-fourth  of  the  diameter  of  the  cup.  They  are  directed  gently  downward 
and  distinctly  tufted.  The  upper  or  apertural  portion  of  the  sponge  is  smooth 
for  a  considerable  distance  and  noticeably  expanded.     The  margin  is  regular. 

RetiGulmn  fine,  the  principal  spicular  bands  being  those  at  the  angles  of 
the  prism-faces,  with  a  corresponding  horizontal  series  dividing  the  constric- 
tions between  the  nodes.  The  remainder  of  the  net- work  is  subequally 
divided  into  quadrules  measuring  about  1  mm.  on  each  side,  and  these  are 
again  divided  by  a  single  series  of  bands. 

Dimensions.  An  entire  specimen  has  a  length  of  60  mm.,  a  greatest 
width  of  20  mm.  and  an  apertural  width  of  17  mm.  Other  specimens  indicate 
about  the  same  proportions. 

A  very  interesting  feature  of  this  species  is  the  retention  of  a  pronounced 
prismatic  surface,  the  late  appearance  of  the  nodes  and  their  feeble  develop- 
ment. Other  species  demonstrate  the  relationship  of  Hydnoceras  to  Prismo- 
DicTYA,  but  in  none  is  the  connection  so  clearly  shown  as  in  this  later  and 
probably  degenerate  form. 

Localities.  A  rare  species,  which  has  been  observed  only  in  the 
higher  beds  of  the  Chemung  group,  in  a  very  compact  green  sandstone  at  Scio 
and  Wellsville,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Hydnoceras  eumeces,  sp.  nov. 

Plate  xii,  Figs.  1-3. 

This  is  a  very  large  species  vnth  narrow,  elongate,  sharp  nodes  and 
narrow  prism-faces.  The  original  specimen  has  been  greatly  compressed 
vertically,  but  shows  features  Avhich  are  not  possessed  by  the  other  species  of 
the  genus.  This  is  indicated,  first  by  the  dimensions  of  the  fossil,  the  internal 
cast  having  a  diameter  of  180  mm.,  while  the  distance  across  the  cup  as  indi- 
cated by  the  external  impression,  was  about  225  mm.;  a  size  not  surpassed  in 
any  known  specimens   of   this   genus.     The   example  affords   no   conclusive 


Si'KOIKS   OF   THE   ChEMUNO    GroUP.  Ill 

evidence  of  the  number  of  horizontal  rows  of  nodes,  but  the  nodes  themselves 
are  seen  to  be  low  and  sharp,  [)r()l()nged  into  acuminate  tufts,  separated  from 
adjacent  meinljers  of  the  same  horizontal  row  hy  a  sludlow  depression,  while 
the  horizontal  constrictions  between  the  rows  are  very  deep.  Tiie  interval 
between  such  of  these  nodes  as  are  shown  on  the  external  and  internal  casts 
indicate  the  normal  number  of  eight  vertical  rows.  The  spicular  bands  have 
left  sharp  impressions,  the  prevailing  size  of  the  quadrule  being  G  nini.  on  a 
side,  this  being  divided  by  a  subordinate  series  of  bands. 

Locality.  From  the  upper  portion  of  the  Chemung  group  at  Alfred, 
Allegany  county,  N.  Y. 

Hydnoceeas,  sp. 

Plate  viii,  Fio.  3. 

A  much  distorted  and  broken  specimen  from  the  compact  green  sand- 
stones of  the  upper  part  of  the  Chemung  formation  at  Wellsville,  N.  Y., 
indicates  still  another  species  of  this  genus,  in  which  the  nodes  are  small  but 
prominent,  the  prismatic  faces  sharply  defined  and  concave,  and  the  reticulum 
very  fine.  Its  characters  are,  however,  too  obscure  to  ensure  a  correct 
diagnosis. 

BOTRYODICTYA,  gen.  nov. 

Dictyosponges  consisting  of  two  distinct  parts ;  the  lower  subcylindrical  or 
very  gradually  expanding  and  Avithout  sui-face  ornamentation ;  the  upper  part  a 
subturbinate  cup,  very  strongly  and  complexly  nodose.  These  two  {)arts  are 
divided  by  a  sort  of  diaphragm  evidently  produced  by  the  interweaving  and 
thickening  of  the  reticulum  internally  at  the  base  of  the  cup.  The  upper  part  of 
this  sponge  has  the  aspect  of  Hydnoceeas,  the  compound  pouch-shaped  nodes 
being  arranged  in  horizontal  rows;  the  lower  part  resembles  Dictyospongia. 

Type,  Dictyophyton  ramosuin,  Lesquereux. 

Botryobictya  eamosa,  Lesquereux  (sp.). 

Plate  xiu,  Figs.  1-*. 

1884.  Dictyophytum  ramosum,  Lesquereux.  Description  of  the  Coal  Flora 
of  the  Carboniferous  Formation  in  Pennsylvania  and  throughout 
the  United  States,  vol.  iii,  p.  827.  (Second  Geological  Survey  of 
Pennsylvania,  Kept.  P.) 

1889.  Dictyophytum  ramosum,  Lesley.  Dictionary  of  Fossils,  vol.  i,  p.  200. 
Original  Description :     "  Stem  nodose  or  tubercular  at  the  articulations, 

the  tubercles  projecting  outside  at  right  angles  to  the  stem,  as  short  branches, 

inflated  and  button-like  at  the  apex. 


112  DlCTYOSPONGID^. 

"This  species  maybe  a  form  of  D.  tuherosum,  Hall,  1.  c,  p.  90,  pi.  iii,  fig.  1, 
which  has  the  articulations  inflated  into  short,  obtuse  tubercles,  some  of 
them  broken,  but  none  prolonged  as  branches.  The  specimens  I  have  had  for 
examination  represent  the  tubercles  arranged  in  horizontal  rows,  as  in  D. 
tuheromirn,  but  instead  of  being  round  and  obtuse  they  are  prolonged  horizon- 
tally into  short  branches  inflated  into  half  round  knots  at  the  apex.  The 
ribbon-like  divisions  surrounding  the  stems  and  branches  are  parallel  and 
quite  distinct. 

"  Habitat.  Upper  Chemung,  head  of  Hill  creek,  Charleston  township, 
Tioga  county,  Penna.     A.  Sherwood." 

The  foregoing  description,  based  upon  a  conception  of  the  vegetable 
nature  of  this  organism,  was  derived  from  highly  impei-fect  specimens.  It 
might  in  fact,  be  difficult  to  recognize,  in  the  absence  of  illustrations,  the  form 
which  it  is  intended  to  portray,  were  it  not  that  we  have  specimens  collected 
from  the  same  locality  by  Mr.  Andrew  Sherwood,  and  that  these  represent  the 
only  nodose  Dictyosponge  known  to  occur  there.  The  fossil  has  proven  to 
be  a  remarkable  and  extremely  interesting  one,  and  by  careful  nianipulation 
of  the  very  badly  collected  material  representing  it,  it  has  been  possible  to 
elucidate  the  main  features  of  its  structure. 

Sponge,  in  general  form,  stout  and  subturbinate,  resting  upon  along,  sub- 
cylindrical,  hollow  base  or  pedicel.  The  latter  expands  very  gradually,  is  sub- 
circular  in  cross-section  and  smooth  on  the  exterior.  Toward  its  upper  portion 
it  broadens  where  passing  into  or  joining  the  vase-like  upper  part  of  the 
sponge.  Its  length  seems  to  be  variable,  irrespective  of  variations  in  the  size 
of  the  nodiferous  vase ;  ^vhere  best  preserved  it  is  considerably  longer  than 
the  vase,  and  even  here  its  basal  point  is  not  retained.  In  this  instance  its 
width  is  about  one-fifth  that  of  the  body  of  the  vase,  and  this  varies  but  little 
from  top  to  bottom  except  at  the  proximal  expansion. 

Form  and  general  aspect  of  the  upper  part  or  vase  as  in  IIydnoceras. 
Body  of  the  vase  subcylindrical,  somewhat  rapidly  expanding  beneath,  and 
slightly  contracting  at  or  near  the  aperture.  Surface  covered  with  a  series  of 
compound  iz-regular,  pendulous  lobes  or  pouches  which  are  arranged  in  eight 
vertical  and  four  horizontal  rows,  these  being  separated  by  low  vertical 
depressions  and  deeper  horizontal  constrictions.  The  specimens  are  so  pre- 
served that  most  of  these  nodes  are  greatly  flattened  and  somewhat  distorted 
from  their  noi-mal  shape,  though  those  lying  in  the  plane  of  lamination  of  the 
rock  have  generally  escaped  such  distortion.  The  structure  of  these  compound 
nodes  is  essentially  as  follows  :  Of  the  three  lower  horizontal  rows,  each  rises 


Species  of  tiik  Ciikmuno  Group.  113 

from  the  surface  of  the  vase  with  a  stout  body  which,  at  some  little  distance 
above  that  sui-face,  is  divided  by  a  vertical  and  a  horizontal  gro<jve,  usually  of 
unequal  depth.  These  grooves  make  a  division  of  the  surface  into  four  lobes 
of  irregular  form  and  distinctly  pendulous ;  upon  the  lowest  horizontal  row 
the  two  lobes  of  the  inferior  division  of  the  node  are  very  long  and  confluent 
for  a  considerable  distance,  producing  digitiform  pouches.  Those  of  the 
median  rows  appear  to  be  somewhat  less  elongated,  but  all  have  a  downward 
direction.  There  are  some  irregularities  in  the  division  of  the  nodes.  In  one 
instance  the  horizontal  groove  is  wanting,  leaving  but  two  nodes  on  the  lol)e ; 
in  another;  one  of  the  four  lobes  appears  to  be  divided  into  three,  and  the  other 
into  two  parts.  The  uppermost,  or  apertural,  series  of  nodes  differs  from  the 
rest  in  having  the  lobes  short  and  ari'anged  singly  or  in  simple  pairs  which 
seem  not  to  correspond  invai-iably  in  position  with  the  vertical  rows  in  which 
the  lower  nodes  are  arranged.  The  growth  of  the  nodes  in  this  row  was, 
however,  quite  irregular ;  they  are  short,  and  while  in  some  specimens  they 
are  erect  and  in  others  inclined  downward,  their  normal  position  was  probably 
directly  outward.  No  specimen  is  sufficiently  perfect  to  indicate  precisely  the 
number  of  nodes  in  this  row. 

A  slight  variation  from  the  type  of  structure  described  is  indicated  by  one 
example  which  shows  a  coalescence  of,  or  an  indistinctness  between  the  first 
and  second  horizontal  rows  of  nodes  at  their  base  and  also  a  partial  loss  of  the 
quadrilobed  surface  of  the  second  row.  Each  specimen,  however,  presents  its 
peculiar  modifications  in  the  number  and  form  of  the  nodes  and  lobes. 

Retwulum.  Over  the  pedicel  the  spicular  bands  are  extremely  fine,  but 
the  reticulation  is  distinct.  Toward  the  base  of  the  cup  the  longitudinal 
bands  become  prominent  while  the  horizontal  bands  are  obscured.  On 
the  vase,  the  net-work  is  fine  in  comparison  with  that  of  Hyonocekas,  and 
the  obscurely  subprismatic  form  prevailing  in  that  genus  is  here  totally  lost. 
The  quadrules  of  the  most  prominent  series  measure  about  2  mm.  on  each 
side,  and  these  are  crossed  by  one  or  two  finer  series.  Over  the  lobes  the 
spicular  bands  all  spread  apart  as  in  Hydnoceras.  There  is  no  satisfactory 
evidence  of  spicular  tufts  at  the  extremities  of  the  nodes,  though  presumably 
such  tufts  existed ;  nor  are  there  other  spicular  expansions  except  fine,  narrow, 
longitudinal  laminae  following  the  courses  of  the  larger  spicular  bands. 

Toward  the  base  of  the  cup  there  seems  to  have  been  a  thickening  of  the 
spicular  strands  which  on  the  inner  or  gastral  side  of  the  sponge  apparently 
converged  to  an  apex  near  the  union  of  the  cup  with  the  pedicel.  This  inter 
ual  convergence  is  exhibited  in  several  instances  where  the  sponge  has  been 


114  DlCTTOSPONOID^:. 

broken  transversely  at  this  point.  There  is  no  evidence  that  the  pedicel  is 
filled  up  with  spicular  deposits  and  hence  the  convergence  of  the  spicules  of 
the  cup  forms  a  sort  of  diajihragm  which  permits  a  ready  separation  of  the 
parts  at  this  place. 

Dimensions.  Of  two  of  the  best  preserved  examples,  one  measures  aa 
follows  :  Entire  length,  170  mm.,  of  which  93  mm.  appertain  to  the  incom- 
plete pedicel.  Width  of  pedicel,  10  mm.;  apertural  width  of  cup,  68  mm. 
The  other  specimen  has  a  length  of  160  mm.,  of  which  but  63  belong  to  the 
pedicel ;  the  median  diameter  of  this  cup  is  45  mm.,  and  tlie  apertural  width 
about  52  mm. 

Locality.  In  greenish  sandy  stales  belonging  to  the  upper  part  of  tlie 
Chemung  group,  near  Lawrenceville,  Tioga  county,  Pennsylvania. 

BOTRYODIOTYA    (or    TtLODICTYA),    Sp. 

From  the  upper  beds  of  the  Chemung  group  at  Scio,  N.  Y.,  have 
been  obtained  a  few  fragments  of  a  quite  small  species  which  show  com- 
pound and  pendulous  nodes.  These  doubtless  represent  a  species  either 
of  BoTBYopicTYA  or  Tylodictya.     (Collection  of  E.  B.  Hall.) 

IIELICODICTYA,  gen.  nov. 

Dictyosponges  with  funnel-shaped  or  explanate  skeletons,  bearing 
equidistant  spiral  surface  ridges. 

Type,  HeUcodictya  trypania,  sp.  nov. 

Helioodictya  trypania,  sp.  nov. 

Plate  xxii,  Figs.  I,  2. 

The  specimen  upon  which  this  species  is  founded  shows  an  elongate, 
funnel-shaped  cup  of  small  size,  subcylindrical  for  nearly  one-half  its  length 
from  the  base,  expanding  more  rapidly  above.  Of  the  basal  portion  both 
external  and  internal  casts  are  preserved,  but  of  the  upper  part  only  one  sur- 
face is  retained.  The  cup  is  characterized  by  a  series  of  strong  parallel, 
nearly  ecpiidistant  spiral  ridges,  from  4  to  5  mm.  apart,  ^vhich  ap})ear  to 
have  been  extended  into  short,  erect  lamellae.  The  continuity  of  these  ridges 
about  the  entire  surface  of  the  cup  is  clearly  shown  upon  the  internal  cast. 

Reticulum  composed  of  sharply  defined  and  minute  meshes,  the  spicular 
bands  lying  parallel  and  perpendicular  to  the  spiral  ridges.  The  latter  appear 
to  con-espond  to  the  vertical  ridges  in  Prismodictya.  The  texture  of  the 
reticulum  is  extremely  fine,  there  being  scarcely  any  visible  distinction  in  the 


Sl'KCrKS    OF    TKK    ClIKMUNO    GroITP.  1  1  O 

size  of  the  qiiadrules.  In  some  parts  of  the  specimen  tlie  prevailing  (piadrule 
appears  to  l)e  somewhat  less  than  1  mm.  stpiare  and  this  is  siilxlivided  by  one 
or  two  subordinate  paii-s  of  spictilar  bands.  The  entire  surface  shows  twelve 
of  the  spiral  ridges  and  the  last  two  interspaces  near  the  apertural  extremity 
are  divided  medially  ])y  a  much  fainter  ridge  and  elevated  line. 

Dimensions.  Entire  length  of  the  specimen,  which  seems  to  be  ap[)ro.\i- 
mately  the  original  length  of  the  cup,  70  mm.;  width  at  the  base,  9  mm.;  at 
30  mm.  above  the  base,  15  ram.;  at  its  greatest  diameter,  30  nun. 

LoeaUty.  In  the  upper  beds  of  the  Chemung  group  at  Wellsville,  N.  Y. 
(CoUection  of  E.  B.  Hall.) 

The  following  species  are  referred  to  this  genus  provisionally.  They 
indicate  a  spiral  mode  of  growth  but  their  structure  is  too  imperfectly  known 
to  render  the  definite  determination  of  their  generic  characters  possible. 

Helicodictya  (?)  Concordia,  sp.  nov. 

Plate  xxvii,  Fig.  2. 

The  single  specimen  of  this  fossil  is  subfusiform  in  outline,  with  a 
depressed  upi)er  surface ;  expanding  from  its  lower,  subacute  extremity 
through  somewhat  more  than  one-half  its  length,  thence  broadly  contracting 
without  revealing  any  definite  evidence  of  an  aperture  at  the  upper  end, 
although  the  margin  here  is  entire  for  a  portion  of  its  extent.  The  surface  is 
crossed  by  spiral  ridges  at  wide  and  enlarging  intervals.  These  ridges  are 
three  in  nimiber  and  on  the  left  margin  of  the  specimen  are  about  10  mm. 
apart,  this  interval  increasing  on  the  right  margin  to  15  or  18  mm.  The  two 
upper  intervals  are  each  divided  in  the  middle  by  a  finer  ridge  parallel  to  the 
others.  There  are  no  reticulating  ridges  of  corresponding  size.  The  net-work 
is  composed  of  fine  spicular  bands  among  which  none  is  consjiicuously  pre- 
dominant, though  the  prevailing  quadrule  measures  about  2  mm.  on  a  side 
and  is  repeatedly  subdivided.  The  length  of  this  specimen  is  60  mm.  and  its 
greatest  width  35  mm. 

While  this  specimen  presents  a  specific  aspect  rendering  it  distinct  from 
associated  forms,  there  is  no  very  good  reason  for  regarding  it  as  congeneric 
with  Helicodictya  trypania  or  as  a  genuine  spiral  Dictyosponge.  Upon  com- 
parison of  this  with  the  specimen  figured  on  Plate  xvii  (fig,  1),  the  similarity 
in  the  character  of  the  reticulum  in  both  may  be  a  reason  for  inferring  that 
the  spiral  direction  of  the  ridges  in  H.  (?)  Concordia  may  be  w  holly  casual,  as 
the  example  referred  to  evidently  represents  the  apei'tural  portion  of  a 
regularly  expanding  cup,  whose  divergent  ridges  are  vertical. 


116  DiCTYOSl'ONGID.E. 

LocaliUj.  From  the  upper  beds  of  the  Chemung  group  ;  in  the  Prismo- 
dictya  prismatica  colony,  at  Concord  station,  Erie  county,  Pennsylvania. 

Helicodictya  (?)  Scio,    sp.  nov. 

Plate  xxvii,  Fig.  1. 

Sponge  moderately  large  and  apparently  regularly  expanding.  Surface 
crossed  by  subequidistant  spiral  ridges  from  17  to  20  mm.  apart,  revolving 
from  left  upward  and  over  to  right.  These  are  parallel  to  one  series  of 
spicular  bands,  the  principal  reticulating  series  making  therewith  quadrules 
measuring  about  10  mm.  on  a  side.  The  length  of  the  original  fragment  is 
about  70  mm. ;  its  width  at  the  lower  extremity  33  mm.  and  at  the  upper 
extremity  55  mm.     Its  surface  bears  four  spiral  ridges. 

The  reason  for  placing  this  form  among  the  spiral  sponges  is  simply  the 
aspect  of  the  best  preserved  specimen.  While  the  specific  form  may  be 
distinct,  the  evidence  afforded  by  a  second  example  indicates  that  the  generic 
reference  here  provisionally  employed  may  be  of  doubtful  value. 

LocaUtij.  In  the  upper  beds  of  the  Chemung  group  at  Scio,  Allegany 
county,  N.  Y.     (Collection  of  E.  B.  Hall.) 

RHABDOSISPONGIA,  gen.  nov. 
Elongate,  broadly   and  strongly  aunulated   sponges,  fluted   by  vertical 
ridges. 

Type,  Dlctyophyton  AmaltJiea,  Hall. 

Rhabdosispongia  Amaltiiea,  Hall  (sp.). 

Plate  xvi,  Fig.  1. 

1890.     Dietyophyton  Amalthea,  Hall.     Ninth  Kept.  N.  Y.   State  Greologist, 
p.  58 ;  Forty-third  Ann.  Kept.  N.  Y.  State  Museum,  p.  260. 

Sponge  slender,  expanding  veiy  gradually,  gently  curved  near  the  base ; 
transverse  section  probably  circular.  In  the  original  specimen  the  base  is 
wanting,  but  the  remaindei-  of  the  surface  bears  five  broad  and  deep  trans- 
verse constrictions  which  increase  in  width  toward  the  aperture.  These  give 
to  the  cup  the  appearance  of  being  strongly  annulated,  the  annulaticms  increas- 
ing in  breadth  with  the  constiictions.  The  area  about  and  below  the  aperture 
is  without  constriction  at  the  usual  interval.  The  surface  is  ornamented  by  a 
series  of  distant,  longitudinal,  narrow  spicular  ridges,  diverging  toward  the 
aperture  with  the  expansion  of  the  cup.  Six  or  seven  of  these  are  visible 
upon  the  exposed  surface  but  no  other  trace  of  the  reticulum  is  apparent. 
The  interspaces  between  these  narrow  ridges  are  flat  and  broad. 


Speciks  of  tiik  Ciiemuno  Group.  117 

Dimensions.  The  only  known  example  has  a  length  of  280  mm.  from 
the  aperture,  which  appears  to  be  entire,  to  the  broken  basal  e.xtremity ;  a 
width  of  70  ram  at  the  aperture  and  20  mm.  at  the  lower  end. 

This  description  is  based  upon  a  plaster  cast  of  a  specimen  formerly  in  the 
possession  of  the  late  Rev.  A.  H.  Riley,  of  Montrose,  Pennsylvania,  but 
the  present  ownership  of  which  is  unknown. 

Locality.  In  the  upper  beds  of  the  Chemung  group  at  Great  Bend,  Sus- 
quehanna county,  Pennsylvania. 

CERATODICTYA,  gen.  nov. 

Elongate,  gently  expanding  cups,  transversely  annulated  and  without 
predominating  ridges  or  other  surface  ornamentation. 
Type,  Dictyophyton  annulatum,  Hall. 

Ceratodictya  Carpenteriana,*  sp.  nov. 

Plate  xlii,  Fig.  1 ;  Plate  xliii,  Fios.  1-3. 
Sponoe  extremely  elongate,  slender,  very  gradually  and  regularly  ex- 
panding. Sometimes  the  cup  assumes  a  gentle  crescentic  curve  which  appears 
to  be  not  abnormal  though  one  specimen  of  considerable  length  is  straight 
thi'oughout  its  extent.  The  surface  is  covered  by  strong,  regular,  continuous, 
duplex  annulations  separated  from  each  other  by  deep  grooves,  both  grooves 
and  annulations  having  about  the  same  width  over  the  median  portion  of  the 
sponge.  The  annulations  are  rendered  duplicate  by  a  transverse  groove 
on  each,  which  under  compression  and  especially  on  the  incipient  portions 
of  the  skeleton,  may  sometimes  approximate  in  breadth  the  interannular  grooves. 
Each  annulation  is  thus  divided  into  two  well-defined  ridges  rounded  at  their 
summits  and  with  their  deeper  slope  toward  the  interannular  grooves.  Some 
variation  is  apparent  in  the  form  of  these  annulations  during  the  process  of 
growth.  Near  the  basal  point  the  summit  grooves  are  rel.atively  nuich 
broader  than  at  any  later  growth-period,  and  although  all  observed  specimens 
of  the  species  are  noteworthy  for  the  stability  and  uniformity  of  the  annula- 
tions, yet  the  example  figured  on  Plate  xliii  (fig.  2),  shows  that  the  second 
annulation  which  is,  in  fact,  the  upper  member  of  the  divided  first  annulation, 
is  itself  faintly  duplicate.  As  growth  continues  beyond  maturity  the  inter- 
annular grooves  become  diminished  in  width,  and  apertural  fragments  of  full 
grown  and  old  specimens  show  that  both  grooves  and  annidations  are  much 

*  Thus  named  as  a  token  of  respect  and  esteem  for  Mrs.  Fannib  Hall  Carpenter,  daughter  of  E.  B. 
Hall,  Esq.,  of  Wellsville,  N.  Y. 


118  DiOTVOSPONOin  J5. 

diniinislied  in  prominence,  though  the  duplication  of  the  latter  is  still  more 
pronounced  than  in  Ceratodictya  cincta. 

The  reticulum  is  very  fine-meshed  throughout,  with  no  evidence  of  any 
especially  predominating  series  of  spicular  bands.  The  prevailing  mesh 
measures  about  1  mm.  on  each  side  and  is  subdivided  by  a  single  minor  series. 
There  is  no  evidence  of  any  ornamental  projections  from  the  surface  in  the 
form  of  tufts  or  spicular  ridges. 

Dimensions.  This  sponge  is  remarkable  for  the  great  length  which  it 
attained.  Plate  xlii  shows  one  excellently  preserved  sj)ecimen,  incomplete, 
however,  at  both  ends,  having  a  length,  measured  along  its  gently  curving 
axis,  of  317  mm.  In  its  entire  length  it  bears  thirteen  double  annulations, 
the  earliest  of  which,  measured  from  the  outer  edges  of  the  ridges,  is  15  mm. 
wide,  and  the  last  of  them  has  about  the  same  width.  A  second  fragment 
(Plate  xliii,  fig.  2)  is  nearly  complete  at  the  base  and  is  straight  for  a  distance 
of  365  mm.,  the  remainder  being  lost.  The  very  long  specimen  shown  on  the 
same  plate  (fig.  1)  measures  500  mm.  in  length,  with  slight  curvature,  and 
is  complete  at  neither  end.  Fragments  of  the  apertural  end  of  full  grown  or 
old  specimens  found  in  association  with  those  referred  to,  have  a  much  greater 
width.  One  of  these,  doubtless  of  this  species,  is  130  mm.  wide  and  to  have 
attained  this  width  at  the  rate  of  increase  shown  in  other  specimens,  the  entii-e 
sponge  must  have  been  not  far  from  three  feet  in  length.  Allowance  is,  in 
this  estimate,  made  from  the  very  evident  fact  that  the  sponge  expanded  much 
more  rapidly  dui-ing  the  first  one-third  of  its  length  than  in  later  gro\vth. 

Localities.  All  the  more  extended  specimens  have  been  obtained  in  the 
upper  beds  of  the  Chemung  group,  on  Milo  Cole's  fann,  near  Ischua,  Cattarau- 
gus county.  A  single  fragment  of  an  old  individual  is  from  Randolph,  N.  Y. 
Associated  with  it  in  the  soft  greenish  shales  at  the  former  locality  are  speci- 
mens of  Lingida,  Camarotcechia  and  Sfirifer.     (Collection  of  E.  B.  Hall.) 

Ceratodictya  cincta.  Hall  (sp.), 

Platk  XXII,  Pigs.  7-9. 

1884.  Dietyophyton  cinctnm.  Hall.  Thirty-fifth  Ann.  liept.  N.  Y.  State 
Mus.  Nat,  Hist.,  p.  472. 
Sponge  large,  subcylindrical,  showing  slight  evidence  of  tapering,  and 
though  much  compressed,  evincing  a  nonnally  circular  cross-section.  Surface 
bearing  closely-set,  subequidistant,  nan-ow  and  strong  annulations  separated  by 
furrows  equal  to  them  in  Avidth.  These  form  a  purely  ornamental  configura- 
tion of  the  surface  and  have  no  other  relation  to  the  reticulum.     The  reticular 


Spkciks  of  tiik  Ciikmuno  Group. 


119 


net-work  is  extremely  fine  and  homogeneous,  no  large  (juadrules  being  present. 
The  prevailing  mesh  is  about  1  ram.  square  and  is  subdivided  by  a  subsidiary 
pair  of  spicular  bands.  These  cover  the  entire  surface  both  of  the  annulations 
and  furrows,  without  the  intersection  of  any  coarse  bands. 

The  entire  dimensions  of  the  cup  are  unknown.  A  large  fragment  with 
a  length  of  120  mm.  bears  eight  annulations,  the  height  of  each  being  about 
5  mm.  The  diameter  of  this  specimen  is  120  mm.  On  one  side  of  the  cup  the 
growth  of  the  annulations  has  been  interrupted,  evidently  from  accident,  and 
in  the  repair  of  thereticulum  the  regularity  of  the  annulations  ha?  been  greatly 
disturbed.  This  species  will  be  distinguished  from  the  terminal  e.xpanded 
portions  of  O.  Carpenteriana  by  more  regularly  equidistant  annulations. 

Locality.  In  the  sandstones  of  the  Chemung  group ;  from  a  boulder  at 
Erie,  Pennsylvania.  Portions  of  several  intlividuals  occur  in  the  same  frag- 
ment of  rock  and  with  them  is  the  original  specimen  of  Ceratodictya  zonata. 
(Collection  of  the  Philadelphia  Academy  of  Natural  Sciences.) 


Ceratodictya  zonata,  sp.  nov. 

Plate  xu,  Fiq.  8. 

1884.     Dictyophyton  cinctum,  Hall  {partim).     Thirty-fifth  Ann.  Kept.  N.  Y. 
State  Mus.  Nat.  Hist.,  p.  473. 

Sponge  small,  slender,  gradually  expanding.  Surface  crossed  by  broad 
and  low  annulations  which  are  separated  by  narrower  and  shallow  constrictions 
toward  the  basal  extremity  which  is  wanting.  These  horizontal  rings  are  not 
sharply  defined ;  those  over  the  median  and  distal  parts  of  the  cup  are 
flattened  or  somewhat  depressed  on  their  summit.  There  are  eight  annulations 
in  a  length  of  45  mm. 

Reticulum  exceedingly  fine,  all  of  the  quadrules  being  minute.  The  most 
conspicuous  of  the  spicular  bands  are  represented  by  sharp  horizontal  lines  at 
the  bottom  of  each  constriction. 

Length  of  the  only  observed  specimen  53  mm. ;  diameter  at  the  lower 
end,  6  mm.;  greatest  diameter  14  mm. 

Locality.  In  the  arenaceous  beds  of  the  Chemung  group ;  from  a  boulder 
found  at  Erie,  Pennsylvania.  (Collection  of  the  Philadelphia  Academy  of 
Sciences.) 


120  DiCTYOSPONOIDvE. 

Ceratodictta  annulata,  Hall  (sp.). 

Plate  xxii,  Figs.  3-6. 
1863.     Dictyophyton  annulatam,  Ilall.     Sixteenth  Ann.   Rept.  N.  Y.  State 

Cab.  Nat.  Hist.,  p.  90,  pi.  iii,  fig.  3. 
1882.     Dictyophyton  annulatum,  Hall.     Notes  on  the  Family  Dictyospongidae, 

pi.  17,  fig.  5. 
1884.     Dictyophyton  annulatum,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.   State 

Mas.  Nat.  Hist.,  p.  472,  pi.  17  (18),  fig.  5. 

Sponge  slender,  very  gradually  expanding,  straight  or  slightly  curved ; 
cross-section  circular. 

Surface  smooth,  annulated  by  a  series  of  low,  horizontal  constrictions, 
varying  somewhat  in  depth,  and  much  narrower  than  the  successive  swellings 
of  the  cup.  Over  the  basal  portion  of  the  cup  the  constrictions  occur  at  much 
mder  intervals. 

Meticulum  fine,  without  the  strong  predominance  of  any  series  of  spicular 
bands.  The  prevailing  (piadrule  appears  to  measure  about  2x3  mm.,  and  is 
repeatedly  subdivided. 

Dimensions.  The  original  specimens  are  three  in  number  and  the  best 
of  these  has  a  length  of  80  mm.,  and  a  width  of  21  mm.  On  this  specimen 
there  are  five  annulations  in  a  length  of  63  mm.  A  somewhat  larger,  more 
completely  flattened  example  retains  nearly  the  entire  cup,  a  portion  of  the 
base  being  wanting.  The  original  length  of  the  cup  was  about  85  mm. ;  the 
diameter  of  its  aperture  34  mm.  This  specimen  bears  but  one  constriction 
and  hence  two  annulations,  the  lower  half  of  the  cup  being  regularly  conical. 

Localities.     The  orginal  specimens  are  from  a  sandy  shale  of   the  upper 

Chemung  group  in  western  New  York ;  their  precise  locality,  however,  is  not 

known.     With  them  are  associated  the  brachiopods  Amhocoelia  gregaria  and 

Wiynchonella  duplicata.    The  specimen  represented  on  Plate  xxii  (figs.  3,  4)  is 

from  the  lower  beds  of   the  Chemung  formation   at   Deyo   basin,    Naples, 

Ontario   county,   where   it   is    associated    with    Hydnoceras    variahile    and 

Hydriodictya  cylix. 

Ceratodictta  centeta,  sp.  nov. 

Plate  xli,  Figs.  1-5. 

A  small  colony  of  annulated  Dictyosponges  has  recently  been  found  in  a 

soft  sandy  shale  which  was  evidently  deposited  about  the  gro\ving  cups  so 

that  the  greater  number  of  the  specimens  cross  the  sedimentation-lines  of  the 

rock  at  large  angles,  and  have,  in  consequence,  been  variously  compressed  and 


Spkcips  of  tfik  Ciikmuno  Gkoup.  121 

distorted.  Some,  which  became  detached  or  hiy  more  nearly  parallel  to  the 
lines  of  deposition,  are  better  preserved,  though  in  none  of  the  examples  is 
the  entire  form  retained. 

These  sponges  vary  considerably  in  form  and  size,  the  young  individuals 
being  slender  with  broad  or  narrow  annulations  at  une([ual  distances.  The 
largei-  individuals  have  the  annulations  more  regular  and  equidistant.  All 
the  annulations  are,  as  a  rule,  low  and  rounded,  with  broad  and  shallow  inter- 
spaces. In  a  specimen  measuring  57  mm.  in  length  and  having  a  width  at  the 
upper  end  of  24  mm.,  there  are  six  annulations  at  subequal  intervals ;  a  young 
example  46  mm.  long,  with  a  width  at  the  top  of  12  mm.,  and  at  the  bottom 
of  7  mm.,  has  six  annulations  at  somewhat  irregular  intervals.  The  largest  frag- 
ment observed  has  a  diameter  of  40  mm.  The  reticulum  in  all  of  these 
specimens  is  fine  and  shows  no  predominant  spicular  striae,  the  prevailing 
mesh  being  uniformly  from  .5  to  1  mm.  square.  The  species  is  distinctly 
unlike  any  of  the  other  forms  referred  to  this  genus  and  although  having  an 
aspect  somewhat  similar  to  that  of  O.  zonata,  diifei's  from  this  in  its  less  regular 
and  unflattened  annulations  as  well  as  in  the  character  of  its  reticulation. 

LocaUtij.  In  the  upper  part  of  the  Chemung  group  at  Ischua,  Cattarau- 
gus county,  N.  Y.  About  fifty  individuals  were  found  in  place  distributed  over 
a  very  limited  area. 

CLATHROSPONGIA,  Hall. 

1863.     Dwtyophyton,  Hall.     Sixteenth  Ann.  Kept.  N.  Y.  State  Cab.   Nat. 

Hist.,  p.  90. 
1882.     Clailirospongia,  Hall.     Notes  on  the  Family  Dictyospongidse,  Expl. 

pi.  18,  figs.  3-4. 
1884.     Dictyophyton  {Clatliro&pongia) ,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y. 

State  Mus.  Nat.  Hist.,  pp.  468-470,474. 
1890.  Dictyophyton,  Hall.  Ninth  Ann.  Eept.  N.  Y.  State  Geol.,  pp.  57,  58. 
This  term  was  introduced  without  definition  and  was  intended  to  cover, 
in  a  subgeneric  sense,  the  species  Dictyophyton  abomis^  Hall,  which  presents  a 
strong  development  of  the  erect  reticulating  surface  lamellae  such  as  also 
occur  in  the  genus  Tiiysanodictya.  Unlike  the  species  of  the  latter  genus, 
however,  the  form  of  the  cup  is  regularly  obconical,  and  it  appears  to  terminate 
below  in  an  acute  point.  There  is  still  some  uncertainty  in  regard  to  this 
character,  but  nevertheless  the  differences  in  the  form  of  the  cups,  and  the 
characters  of  their  earlier  growth  are  evident.  It  has  been  found  necessary  in 
the  consideration  of  a  number  of  different  species  where  imperfect  preserva- 
tion has  precluded  definite  knowledge  of  the  basal  structure,  to  refer  to  this 


122  DlCTYOSPONGID^. 

genus  some  forms  which,  \vhen  better  known,  may  prove  to  belong  in  some  other 
association.  Indeed,  in  the  type-species  only  are  the  generic  features  clearly 
defined,  and  hence  the  group  must  for  the  time  being .  serve  in  a  measure  as  a 
convenient  receptacle  for  certain  imperfectly  known  forms. 

Type,  Cluth-ospongia  abacus,  Hall.  (See  description  of  this  form  among 
the  sjiecies  of  the  Waverly  group.) 

CliATIIROSPONGIA   FENESTBATA,  Hall    (sp.). 
Plate  xiv,  Fig.  2;  Plate  xxv,  Fig.  5. 

1863.     Bktyophyton  fenestratum,  Hall.    Sixteenth  Ann.   Rept.  N.  Y.  State 

Cab.  Nat.  Hist.,  p.  90,  pi.  iii,  fig.  4. 
1884.     Dictyophyton  femstraPum,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State 

Mus.  Nat.  Hist.,  p.  469. 

The  original  specimen  of  this  species  is  a  very  incomplete  fragment  of  a 
moderately  large  cup,  a  portion  of  whose  surface,  involved  in  soft  green  shale, 
retains  the  exterior  character  of  the  reticulation,  while  the  remainder  of  the 
specimen  is  only  an  internal  cast  of  the  skeleton.  This  specimen  gives  no 
accurate  idea  of  the  form  of  the  cup,  except  that  it  was  subcylindrical  with  an 
essentially  smooth  exterior,  but  an  internal  cast  of  a  more  complete  example 
from  the  same  locality  as  the  original  furthers  our  knowledge  of  the  species. 

Sponge  of  considerable  size,  elongate  subcylindrical  through  most  of  its 
length,  with  a  short,  rapidly  expanding  basal  portion.  Transverse  section 
subelliptical  or  subcirculai".  Surface  with  only  traces  of  low  nodes  at  the 
angles  of  the  principal  quadi'ules. 

ReticAtlmii  in  its  external  character,  composed  of  (juadrules  of  ^vhich  the 
principal  series  measures  about  12  mm.  on  the  side.  The  honzontal  and  ver- 
tical bands  of  the  first  series  are  of  about  equal  prominence  and  at  their  inter- 
section distinct  nodes  are  formed.  These  main  quadrules  are  regularly  sub- 
divided by  four  subordinate  series  of  spicular  bands  and  upon  the  internal 
cast  almost  all  distinction  between  these  series  is  lost,  the  ultimate  subdivision 
of  the  quadrules  being  nearest  the  internal  surface  of  the  cup. 

It  is  evident  that  the  principal  spicular  bands  were  produced  outwaj-d  in 
the  form  of  free  lamellae ;  at  one  spot  on  the  cast  these  show  a  width  of 
about  2  mm. 

Dimensions.  The  original  specimen  has  a  length  of  105  mm.  and  a  width 
of  45  mm.  A  second  example  measures  138  mm.  from  base  to  distal 
extremity,  and  for  five-sevenths  of  this  length  the  diameter  of  the  cup  is 
45  mm.     The  basal  cone  has  a  length  of  about  30  mm. 


Speciks  ok  the  OiiEMiJNn  Giiotri'.  123 

This  species  bears  some  resemblance  to  G.  cajprcxT/ynia  of  the  Waverly 
group,  but  differs  from  it  in  the  smaller  size  of  the  i)rincipal  quadrilles  of  the 
reticulum,  its  more  cylindrical  form  and  more  abruptly  tapei-ing  base. 

Locality.  In  the  upper  beds  of  the  Chemung  group,  Chemung  nar- 
rows, N.  Y. 

Clathkospongia  vascellum,  Hall  (sp.). 

Plate  xxi,  Fig.  12;  Plate  xxv,  Pio.  4. 

1890.     Dictyopliyton  vascellum,  Hall.     Ninth  Ann.   Rept.  N.  Y.  State  Geolo- 
gist, p.  57;  Forty-third  Ann.  Rept.  N.  Y.  State  Museum,  p.  259. 

The  original  specimen  of  this  species  is  a  large,  explanate  fragment  indi- 
cating a  flattened  cup,  one  of  whose  transverse  diameters  is  six  times  the  other. 
In  its  normal  condition  the  cup  must  have  been  large,  quite  rapidly  expanding 
and  probably  regularly  conical  in  shape.  The  extreme  length  of  the  fragment 
is  188  mm.  and  the  slope  of  the  lateral  margins  would  indicate  a  length  from 
the  upper  edge  of  the  specimen  to  the  base,  of  260  mm.,  making  no  allowance 
for  the  unknown  extent  of  the  cup  beyond  this  upper  limit  of  the  specimen. 
The  surface  bears  a  series  of  low,  broad  vertical  bands  or  obscure  ridges, 
which  are  crossed  by  somewhat  more  obsciire  transverse  bands,  making  a 
series  of  large  quadrules,  a  little  longer  than  broad,  and  measuring  on  an 
average  about  26x30  mm.  The  specimen  is  an  intei-nal  cast,  and  the  inter- 
mediate series  of  meshes  are  obscured  by  the  predominance  of  the  ultimate 
and  penultimate  quadrules. 

Several  fragmentary  examples  of  this  species  have  been  o])served,  all  of  them 
being  somewhat  water- worn.  That  here  figured  is  the  largest  and  most  complete, 
affording  specific  characters  not  easily  confounded  with  those  of  other  s[)ecies. 

Locality.     From   the   upper   portion  of   the   Chemung   group;  drift  at 
Alfred,  Allegany  county,  N.  Y. 

ClATHBOSPONGIA  (?)  TOMACITLUM,  Hall  (sj}.). 
Plate  xvm,  Figs.  5-6;  Plate  xxi,  Fkj.  11. 

1890.      Dictyophytmi   tomamlum,   Hall.      Ninth    Ann.    Rept.    N.    Y.    State 

Geologist,     p.    58;     Forty-third     Ann.     Rept.     N.    Y.    State 

Museum,  p.  260. 

Sponge  long,  slender,  probably  normally  subcylindrical  or  very  gradually 

expanding,  though  the  original  specimen  is  flattened  and  somewhat  arched 

laterally.     The  length  of  this  specimen  is  185  mm.  and  neither  the  apertural 

nor   basal  parts  of  the  cup  are  retained.     Its  diameter  at  the  lower  end  is 

50  mm.  and  at  the  upper  end  60  mm. 


124  DlCTYOSPONGID^. 

Retiaulum.  The  surface  is  covered  with  the  fine  reticulations  of  the 
ultimate  spicular  bands,  but  these  do  not  obscure  the  coarse  ridges  of  the 
primary  bands.  The  latter  form  quadrules  which  are  somewhat  broader  than 
long,  and  measure  on  the  average  17x15  mm.  The  vertical  bands  appear  to 
be  somewhat  stronger  than  the  primary  horizontal  series,  and  this  gives  a 
slight  subprismatic  aspect  to  the  surface. 

This  sponge  is  not  greatly  iinlike  Clathi'ospongia  fenestrata ;  a  more 
complete  knowledge  of  both  may  demonstrate  their  identity ;  with  our  present 
information  however,  O.  tomacidum  appears  to  have  been  a  rather  more 
slender  cup,  with  a  coarser  primary  reticulation. 

A  small  fragment  of  a  second  specimen  retains  the  definition  of  the 
surface  characters  much  more  sharply.  Here  each  of  the  principal  quadrules 
is  divided  by  a  subordinate  pair  of  spicular  bands,  which  are  nearly  as  promi- 
nent as  those  of  the  first  order,  and  the  finer  subdivision  is  clearly  shown  in  its 
ultimate  division,  each  principal  quadrule  containing  256  minimum  quadrules, 
produced  by  the  intersection  of  five  series  of  spiciilar  bands. 

Locality.  From  the  upper  part  of  the  Chemung  group ;  in  a  loose  block 
at  Alfred,  Allegany  county,  N.  Y. 

ClATHROSPONGIA    (?)    IRREGULARIS,  Hall    (sp.). 
Plate  xv,  Figs.  6,  7. 

1884.     Dictyophyton  irregulare,  Hall.     Thirty-fifth  Ann.  Kept.   N.  Y.  State 
Mus.  Nat.  Hist.,  p.  470. 

Sponge  small,  gently  tapering,  very  faintly  subprismatic.  In  the  original 
specimen,  which  is  an  internal  cast,  these  prism-faces  are  much  obscured, 
partly  by  compression  and  partly,  also,  by  the  surface  reticulation. 

The  surface  is  divided  into  major  quadrules  by  transverse  bands  equidis- 
tant with  the  edges  of  the  prism-faces,  thus  forming  squares  measuring  about 
9  mm.  on  a  side,  at  each  of  the  angles  there  being  a  low  node.  All  of  these 
characters  are,  for  the  most  part,  very  obscure,  but  sufiicient  is  preserved  to 
show  that,  in  its  original  condition,  the  sponge  was  divided  into  strong  quad- 
rules  as  described,  these  being  tufted  at  their  angles.  The  subordinate 
reticulation  of  the  surface  is  normal. 

Dimensions.  The  single  example  of  this  species  is  complete  at  neither 
extremity,  though  probably  little  is  wanting  at  the  upper  end,  and  its  basal 
extent  is  indicated  by  the  slope  of  the  sides.  The  length  of  the  part  retained 
is  fi5  mm. ;  the  original  length  was  probably  not  more  than  75  mm.  The 
greatest  diameter  of  the  cup  is  25  mm. 


SpKcrKs  OK  TiiE  Chkmuno  Ghoth'.  125 

Locality.  This  sponge  was  obtained  many  years  ago  at  Ithaca,  N.  Y.,  by 
Prof.  F.  W.  SiMONDS.  Its  precise  position  in  the  strata  thei-e,  whether  from 
the  beds  of  the  Portage  or  from  those  of  the  overlying  Chemung  rocks  of  the 
vicinity  is  not  known, 

ClATHROSPONGIA     ( ?  )     DESMIA,     Sp.     HOV, 
Plate  xxv,  Fig.  3. 

A  fragment  of  the  surface  of  a  flattened  cup  indicates,  in  the  absence  of 
decisive  generic  characters,  a  species  distinct  from  any  of  the  forms  elsewhere 
described.  The  primary  and  secondary  spicular  bands  are  so  nearly  of  the 
same  size  that  the  surface  appears  to  be  divided  into  vertical  rectangular 
parallelograms  which  measure  on  the  average,  3  mm.  on  the  base  and  6  ram. 
in  height ;  that  is,  the  horizontal  bands  over  the  entire  extent  of  the  exposed 
surface  are  about  twice  as  far  apart  as  the  vertical  bands,  and  all  leave  im- 
pressions of  about  the  same  strength  upon  the  external  cast.  The  predominant 
rectangles  are  divided  into  squares,  by  faint  transverse  bands.  All  of  the 
principal  spicular  bands  form  erect  lamellae  without  producing  no<les  at  their 
intersections.  The  form  of  the  quadrule  in  this  species  is  somewhat  similar  to 
that  prevailing  in  Dictyospongia  sceptrum,  but  we  are  without  evidence  of  the 
existence  of  spicular  lamellae  in  that  species  and  genus.  In  Thymrwdictya 
rvdis  and  T.  Maiidalli  the  quadrules  are  larger  and  more  nearly  square.  The 
length  of  the  fragment  is  90  mm.  and  its  greatest  width,  40  mm. 

Locality.  In  the  upper  beds  of  the  Chemung  group  near  Alfred  and  at 
Scio,  ADegany  county,  N.  Y. 

THYSANODICTYA,  gen.  nov. 

Subcylindrical  or  tapering  Dictyosponges  with  prominent,  projecting, 
rectangularly  reticulating  spicular  bands  or  lamellae  forming  series  of  fenes- 
trated quadrules  upon  the  surface,  as  in  Clatiirospongia.  Base  ending 
abruptly  in  a  transverse  disk  or  broad  and  shallow  reticulated  obcone 
or  basin,  which  may  be  encircled  by  an  expanded  bjisal  fiill,  as  in 
Phragmodictya. 

Type,  Dictyophyton  HalU,  Hall. 

Species  of  this  genus  combine  some  of  the  characters  of  the  two  genera 
last  mentioned,  and  while  the  form  of  the  sponge  is  similar  to  that  of 
Phragmodictya,  it  lacks  both  the  irregular  surface  ridges  of  that  genus 
and    also    the    regularly    radiated    reticulum    of   the    diaphragm    or    basal 


1 26  DlCTYOSPONGID^. 

surface.  In  Thysanodictya  the  reticulation  of  this  surface  is  rendered 
reguhuly  quadrate  by  the  crossing  of  the  strands  whicli  are  vertical  in  the 
wall  of  the  cup;  the  margins  of  the  base  are  frequently  puckered  into  low  and 
irregularly  disposed  nodes,  the  peripheral  fi-ill  being  produced  directly  from 
this  margin.  The  aspect  of  this  surface  is  somewhat  similar  to  that  in  the 
species  Oncosella  catinum,  Rauff,  from  the  Wenlock  limestone*,  although  the 
latter  is  a  thick-bodied  sponge,  and  in  this  respect  unlike  Tiiysajj^odictya, 
Rauff  is  disposed  to  believe  that  Oncosella  possessed  one  or  more  tufts  of 
anchoring  spicules,  though  they  have  not  yet  been  found,  and  that,  therefore, 
the  sj>onge  was  not  yet  attached  by  this  flattened  base.  Presumably,  in 
Thysanodictya,  the  basal  attachment  was  essentially  effected  by  the  encircling 
frill-like  tuft. 

It  may  be  added  that  all  the  species  of  this  genus  now  known  show  evi- 
dence of  a  sixth  series  of  reticulating  spicular  bands,  which  produces  an 
extremely  minute  ultimate  division  of  the  quadrules. 

Thysanodictya  Edwin-Halli,  nom.  nov. 

Platk  XXIV,  Figs.  1-9. 

1890.  Dkiyopliiitoti?  {Pliragmodictya)  HalU,  Hall.  Ninth  Ann.  Rept.  N.  Y. 
State  Geologist,  p.  59;  Forty-third  Ann.  Rept.  N.  Y.  State 
Museum,  p.  261. 

Sponge  generally  small,  obconical  or  subclavate,  narrowest  above  the 
base,  rapidly  expanding  to  the  aperture ;  very  variable  in  form,  sometimes  long 
and  extremely  slender,  at  other  times  short  and  stout.  Base  somewhat 
expanded,  nearly  flat,  in  some  instances  apparently  concave ;  nodose  on 
the  periphery  and  with  a  broad  marginal  spicular  frill  extended  downward  or 
over  the  surface  of  attachment. 

Meticulmn  of  the  lateral  walls  coarse,  the  main  spicular  bands  forming 
vertically  elongate  quadrules  measuring  on  an  average  7x11  mm.  The  size  of 
these  quadrules  varies  greatly  in  different  individuals  and  in  different  parts  of 
the  same  individual.  The  subordinate  reticulation  even  to  the  last  degree,  is 
often  retained,  but  does  not  in  any  way  obscure  the  predominance  of  the  main 
ridges.  These  vertical  and  transverse  ridges  of  the  first  and  second  series  are 
expanded  into  lamellae  which  lie  vertical  to  the  surface  of  the  cup  and  are 
free  for  a  width  of  from  2  to  4  mm.  At  the  base  of  the  cup  the  vertical 
ridges  thicken  and  form  a  single  series  of  sharp,  nodiform  periphei-al  ridges 
which  may  extend  for  a  short  distance  on  the  surface  of    the  flattened  or 

•8«e  Ra0ff;  Palaeospongiologie,  Part  1,  p.  264,  pi.  vii,  figs.  5-10  (especially  fig.  7).    1894. 


SpKcrKs    OF   TirK  CiiEMiJNO  Guotrp.  127 

depressed  basal  diaphragm.  The  basal  frill  (periloph,  Ilauff)  is  broad  and 
where  best  retained  appears  to  be  gently  undulated,  but  tliis  expansion  is  not 
so  preserved  as  to  show  clearly  the  character  of  the  reticulation.  The  l)aMa1 
diaphragm  Is  slightly  convex  or  even  concave,  and  bears  a  low  but  coarse 
rectangular  reticulation. 

Dimensions.  A  well  developed  individual  of  rather  large  size,  measures 
60  mm.  from  aperture  to  base.  The  aperture  is  48  mm.  and  the  base  26  mm. 
in  diameter;  the  width  of  the  cup  across  the  constriction  just  above  the  base  is 
20  mm.  A  very  slender  example  is  82  mm.  in  length,  30  mm.  in  diameter  at 
the  aperture,  11  mm.  at  its  narrowest  point  and  13  nmi.  across  the  ba.se  which 
is  not  entire.  In  still  another  specimen,  which  may  be  somewhat  compressed 
vertically,  the  length  is  but  20  mm.,  the  diameter  of  the  aperture  27  mm.,  and 
that  of  the  base  25  mm.  In  an  individual  about  50  mm.  in  length  and  having 
a  basal  diameter  of  25  mm.,  the  basal  expansion  or  frill  has  a  diameter  of  about 
17  ram.  A  single  example  surpasses  the  usual  dimensions  of  the  species, 
presenting  a  length  of  175  mm.  and  even  thus  incomplete. 

Localities.  In  the  upper  arenaceous  beds  of  the  Chemung  group  at  Wells- 
ville,  N.  Y.  This  usually  small  species  has  been  found  here  by  Mr.  E.  B.  Hall 
in  considerable  quantities.  About  fifty  specimens,  all  from  the  same  spot,  have 
been  examined  and  these  we  are  informed  represent  but  a  small  part  of 
the  number  there  obtained.  It  also  occurs  at  Scio,  Allegany  county,  Ischua, 
and  Olean,  Cattaraugus  county,  and  has  been  found  on  the  Cassadaga  creek  in 
Chautauqua  county. 

TlIYSANODICTYA  P(ECILU8,   Sp,  nOV. 
Plate  xxiv,  Figs.  10, 11. 

Sponge  somewhat  variable  in  form  ;  short  and  robust  or  long  and  slender ; 
subcylindrical,  contracting  above  the  base  for  less  than  one-half  its  length,  and 
again  expanding  to  the  aperture,  whose  diameter  is  about  equal  to  that  of  the 
base.  Basal  disk  very  depressed,  almost  flat;  margins  thickened  and  also 
extended  into  a  peripheral  frill. 

Reticidihm  having  the  general  expression  of  that  in  ThusaiUKlidya  rudis, 
though  the  elongate  rectangular  primary  quadrules  are  regularly  smaller,  each 
comprising  four  parallelograms  measuring  on  an  average  4x6  mm.  At  the 
base  and  near  the  aperture  these  areas  are  more  nearly  square.  The  pri- 
mary spicular  bands  are  sharply  elevated,  minutely  nodose  at  their  extremi- 
ties upon  the  external  cast,  and  both  the  vertical  and  horizontal  series 
are  carried  out  into  broad,  erect  lamellae  which  have  a  diameter  of  from 


128  DlCTYOSPONGID.15. 

3  to  5  ram.  At  the  intersection  of  these  lamellae,  somewhat  thickened  nodes 
or  tufts  are  formed  which  extend  beyond  the  edges  of  the  lamellae  themselves. 
The  aiwrture  appeai-s  to  be  without  evidence  of  a  marginal  tuft.  The  basal 
disk  is  somewhat  thickened  and  bears  ridges  at  the  margins,  but  the  character 
of  its  surface  and  that  of  the  peripheral  frill  is  not  well  shown. 

Dimensions.  Four  examples  of  this  species  are  preserved  upon  a  single 
slab  of  sandstone.  Of  the  best  two  of  these,  one  measures  100  mm.  in  length  ; 
diameter  at  the  base  (exclusive  of  marginal  expansion),  55  mm.;  at  the 
aperture,  53  mm. ;  at  the  narrowest  portion  of  the  cup,  37  mm.,  and,  with  the 
surface  lamellae,  46  mm. ;  the  other,  a  much  more  slender  cup,  has  a  length  of 
120  mm.  (incomplete),  an  apertural  diameter  of  30  mm.  and  a  narrowest  diam- 
eter of  22  mm. 

This  species  is  distinguished  in  its  stouter  form,  from  Thysarwdictya 
rudis,  and  in  its  more  slender  form,  from  T.  Mlwi/i-IIalli,  by  the  smaller 
size  of  its  primary  quadrules. 

Locality.  In  the  upper  beds  of  the  Chemung  sandstones,  associated  with 
Camarotcechia  and  Leptodesimi,  Scio,  Allegany  county,  N.  Y.  (Collection  of 
E.  B.  Hall.) 

Thysanodictya  rudis,  Hall  (sp.). 

Platk  XIX,  Fig.  7 ;  Platk  xxiii,  Figs.  1-6. 

1863.     Diet yophy ton  rude,  Hall.     Sixteenth  Ann.  Rept.  N,  Y.  State  Cab.  Nat. 
Hist.,  p.  90,  pi.  V,  fig.  3. 

1884.     Dictyopliyton  riide,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State  Mus. 

Nat.  Hist.,  p.  469. 
1890.     Dictyopliyton  scitum,  Hall.'    Ninth  Ann.  Rept.  N.  Y.  State  Geologist, 

p.    58;  Forty-third  Ann.  Rept.  N.  Y.  State  Museum,  p.  260. 

Sponge  moderately  large,  cylindrical.  Base  subtruncate,  closed  by  a 
broadly  obconical  diaphragm.  Surface  sharply  divided  by  a  series  of  vertically 
elongate  (piadrangles  which,  in  the  middle  portion  of  the  original  specimen, 
appear  to  be  all  of  approximately  the  same  size.  Each  of  the  quadrules  of  the 
first  order  eniljraces  four  of  these  rectangles  and  measures,  on  an  average, 
1 5x18  mm.  Toward  the  base  of  the  cup  these  primary  quadrules  are  more  nearly 
square.  The  bands  of  the  first  and  second  series  form  low  nodes  at  their  in- 
tersection and  both  the  vertical  and  horizontal  seiies  are  produced  into  erect 
lamellae  which  may  extend  outward  from  the  surface  for  a  distance  of  from 
3  to  5  mm.  The  third,  fourth  and  fifth  series  of  spicular  bands  are  seldom 
clearly  developed,  being  to  some  extent  obscured  by  the  ultimate  division  of 


Species  of  the  Chkmuno  Gkouk  129 

the  reticulum  produced  by  the  sixth  Hcries  of  bandH.  The  presence  of  this 
series  produces  1024  divisions  in  each  primary  (puidruh;.  Tlie  conspicuous 
ridges  of  the  lateral  surfaces  extend  over  the  broad,  sulxtonical  base,  the  ver- 
tical bands  converging  at  the  apex  and  the  horizontal  bands  l>eing  distinctly 
concentric.  The  basal  cone  is  shown  in  two  examples,  in  one  of  which  it  is 
attached  to  a  portion  of  the  cup  and  is  extremely  compressed  laterally.  In 
the  other,  such  distortion  as  it  has  undergone  is  in  the  opposite  direction, 
vertical  rather  than  lateral.  The  latter  specimen  represents  a  cup  with  a 
diameter  t)f  about  80  mm.,  bears  a  decidedly  eccentric  ajKix  and  shows  the 
characteristic  coaree  and  minute  reticulation  of  the  surface.  The  depth  of 
this  cup  is  17  mm. 

■  Dimensions.  All  specimens  of  this  species  observed  are  of  about  the 
same  size.  The  original  example,  which  is  nearly  entire  in  length  though  not 
retaining  the  basal  cone,  is  135  mm.  long  and  64  mm.  wide.  Another  entire 
specimen  has  a  length  of  120  mm.  and  a  width  at  the  aperture  of  70  mm. 

This  species  is  readily  recognized,  even  in  the  absence  of  its  basal 
diaphragm  by  its  stout,  cylindrical  shape  and  coarse  primary  reticulation.  It  is 
closely  allied  to  T.  Handalli  though  rather  less  stout.  The  characters  of  the 
latter  species  are  not,  however,  so  well  known. 

Localities.  In  the  upper  beds  of  the  Chemung  group,  at  Little  Genesee, 
Alma  and  Wellsville,  Allegany  county,  at  Ischua,  Hinsdale  and  Olean,  Catta- 
raugus county,  and  Chemung  narrows,  Chemimg  county. 

Thtsanodiotta  Randalli,  Hall  (sp.). 

Plate  xxiv.  Fig.  12. 

1890.     Dictyophyton  Randalli^  Hall.     Ninth  Ann.  Kept.  N.  Y.  State  GeoL, 
p.  57 ;  Forty-third  Ann.  Kept.  N.  Y.  State  Museum,  p.  259. 

The  specimen  upon  which  the  oiiginal  description  of  this  species  was 
based,  is  the  upper  portion  of  a  flattened,  once  cylindrical  cup  whose  surface 
is  covered  with  low  longitudinal  and  horizontal  ridges  which  form  quadrules 
measuring  about  17x'il  mm.,  each  of  these  being  subdivided  by  secondary 
ridges  of  nearly  the  same  pi-ominence  as  the  first.  The  subordinate  c'uicella- 
tion  of  the  surface  is  very  fine.  At  the  upper  extremity  of  the  specimen  where 
the  apertural  margin  is  retained  there  is  some  indication  of  a  continuation  of 
the  vertical  spicular  bands  beyond  the  edge  of  the  cup.  The  length  of  the 
specimen  is  90  mm.;  its  apertural  diameter  is  78  mm.  This  form  is  very 
closely  allied  to  T.  riidis,  though  present  evidence  indicates  that  the  cup  ^vas 


1 30  DlCTYOSPONOID^. 

broader  and  the  primary  surface  quadrules  more  nearly  equilateral  than  in 
that  species ;  yet  it  is  probable  that  further  knowledge  will  demonstrate  the 
superfluity  of  this  specific  name. 

•   Locality.     In  the  Chemung  group,  associated  with  Sjpirifer  disjunctus; 
Warren,  Pennsylvania. 

TlIYSANODICTYA  SACCU8,    Sp.  nOV. 
Plate  xxv,  Figs.  1, 2. 

Sponge  short,  stout,  broadest  at  the  base,  narrowing  upward,  with  gently 
incurving  margins.  Height  somewhat  less  than  the  basal  diameter.  Circular 
or  subelliptical  in  cross-section.  Basal  obcone  or  diaphragm  moderately  deep 
with  a  subcentral  apex. 

Meticulum.  Primary  quadrules  somewhat  variable  in  size,  smaller  and 
more  nearly  square  than  in  Thysanodictya  riidis.  Subordinate  divisions  of 
the  net-work  clearly  shown  in  j)laces.  The  basal  diaphragm  is  thickened 
or  nodose  about  the  edges  and,  in  the  specimens  observed,  has  a  some- 
what  irregular  reticulation,  though  the  meshes  of  the  lateral  surfaces  ai-e  here 
to  some  extent  retained.  This  species  may  be  recognized  by  its  short  and 
robust  form  and  its  tapering  rather  than  expanding  cup.  Two  individuals 
fi'om  different  localities  sho^v  the  same  characters.  In  the  better  of  these,  one 
surface  is  covered  ^vith  low,  apparently  concretionary  tubercles  which,  in  a 
measure,  obscure  the  character  of  the  reticulum ;  a  phenomenon  not  infre- 
quent among  these  fossils. 

Dimensions.  A  typical  example  has  a  length  of  74  mm.,  a  basal 
diameter  of  88  mm.,  and  an  apertural  diameter  of  63  mm.  A  second  specimen 
has  nearly  the  same  proportions. 

Localities.  In  the  upper  part  of  the  Chemung  group,  at  Wellsville  and 
Whitesville,  Allegany  county,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Thysanodictya  Johnstoxi,  sp.  nov. 

Plate   xxiii,  Fiqs.  7,  8. 

Sponge  robust,  subcylindrical  or  somewhat  inflated  medially,  and  sharply 
constiicted  above  the  base.  Basal  obcone  or  diaphragm  very  shallow,  nearly 
flat ;  sharply  and  irregularly  nodose  about  the  margin.  The  surface  of  this 
diaphragm  is  composed  of  fine  and  sharp  spicular  bauds  intersecting  at  right 
angles,  the  i)revailing  (piadrule  measuring  from  3  to  5  mm.  on  a  side.  There 
are  no  coarse  spicular  lidges  of  a  higher  order  distinguishable,  but  the  reticu- 
lum shows,  in  addition  to  the  regularly  intersecting  bands,  iri-egularly  arranged 
or  disordered  bands  piincipally  of  large  size.     The  nodes  of  this  basal  surface 


Speciks  of  thk  Chemcno  Group.  131 

are  most  prominently  developed  at  the  margin,  taking  somewhat  the  form  of 
elongated  ridges  directed  toward  a  common  center  and  hence,  here  and  there, 
they  are  to  be  found  extending  upon  the  inferior  surface  in  an  irregular  and 
desultory  way.  The  surface  of  the  cup,  in  the  only  example  in  which  it  is 
retained,  is  covered  with  low  nodes  or  coarse  papillae,  an-anged  with  some 
degree  of  regularity ;  they  show,  however,  no  trace  of  the  spicular  net-\vork. 
This  specimen  is  involved  in  a  soft  arenaceous  shale  and  these  rounded  bcxlies 
on  the  surface  appear  to  be  concretionary,  as  in  places  where  the  rock  is 
exfoliated  they  show  a  succession  of  thin  laminae.  Probably  they  have 
fonned  about  the  projecting  points  and  intersections  of  the  principal  spicular 
strands. 

Dimenfdons.  The  most  complete  example  observed  retains  a  large 
portion  of  the  cup  which  has  been  bent  over  so  that  it  lies  in  neai-ly  the  same 
plane  with  the  basal  diaphragm.  Its  entire  lengtli  is  135  mm. ;  its  diameter 
at  the  upper  end,  67  mm.;  at  the  middle,  78  mm.;  just  above  the  base,  53  mm. 
The  greatest  and  least  diameters  of  the  base  are  70  and  55  mm.  Another 
example  preserving  the  base  only,  has  the  major  and  minor  diameters  85  and 
55  mm. 

This  species  may  be  distinguished  from  T.  Edwin-Halli  and  T.  radis  liy 
its  different  form,  extremely  nodose  and  finely  reticulate  base. 

Locality.  In  the  upper  beds  of  the  Chemung  group,  Wellsville,  N.  Y. 
(Collection  of  E.  B.  Hall.  Named  for  Mr.  John  S.  Johnston,  of  Wellsville, 
in  recognition  of  his  intelligent  activity  in  collecting  the  sponges  of  the 
upper  Chemung  group.) 

THYSANODICryA   QTJASILLUM,  Sp.  HOV. 
Plate  xxxviii,  Figs.  3,  4. 

Sponge  large  and  stout ;  general  form  as  in  the  other  pjjecies  of  the  genus, 
with  a  subcylindrical  body  slightly  expanded  above  the  middle  and  at  the 
base.     The  aperture  is  not  well  retained. 

Surface  marked  by  large,  coarse  meshes  of  somewhat  variable  size, 
bounded  by  distinct  ridges  which  were  probably  bases  of  vertical  surface 
lamellae.  These  quadrules  are  approximately  square,  the  largest  of  them 
measuring  about  16x14  mm.  These  occur  about  the  more  expanded  portion 
of  the  sponge,  toward  the  base  the  prevailing  meshes  being  somcAvhat  smaller. 
At  the  intersection  of  the  primary  bands  there  is  a  noticeable  tendency  to 
elevation  in  the  form  of  low  nodes.     A  faint  subdivision   of  the  principal 


132  DicnrospoNGiD^. 

quatlniles  by  spicular  bands  of  the  second  order  is  visible  here  and  thei-e  over 
the  sui-face,  and  over  the  entire  reticulum  may  be  seen  the  extremely  fine 
ultimate  di\-ision  of  the  skeleton.  At  the  base  the  margins  of  the  diaphragm 
are  someAvhat  irregularly  thickened  into  nodes,  but.  the  structure  of  the 
reticulum  over  the  basal  surface  is  not  retained  although  the  form  and  outline 
of  this  part  are  distinctly  shown. 

The  species  is  distinguished  from  T.  rudis  and  T.  Randalli  by  its  greater 
size  and  the  coarseness  of  its  primary  meshes. 

The  specimen  we  have  used  for  the  illustration  of  this  species  shows  in  a 
very  interesting  manner  how  differently  the  reticulum  may  be  retained  on 
parts  of  the  same  individual.  Figures  3  and  4  (Plate  xxxviii)  represent 
opposite  sides  of  a  specimen,  on  one  the  primary  net-work  only  being  apparent 
while  on  the  other  the  minute  ultimate  division  of  the  net  is  i-etained.  This 
difference  appears  to  be  due  to  the  fact  that  the  specimen  is  a  somewhat  con- 
cretionary mass  of  sandstone  which  was  involved  on  one  side  in  a  film  of  fine 
mud  capable  of  retaining  the  impression  of  the  most  delicate  spicular  divisions. 

Dimensions.  Length  of  an  essentially  entire  specimen,  140  mm. ;  width 
above  the  middle  and  at  the  base  85  mm. 

Locality.  From  the  Chemung  group  at  Weston  mills,  Cattaraugus 
county,  and  Wellsville,  Allegany  county.     (Collection  of  E.  B.  PIall.) 

Thysanodictya  scyphina,  sp.  nov. 

Plate  xlii,  Fig.  8. 

Sponge  small,  subcylindrical,  the  original  specimens  being  somewhat  con- 
tracted at  or  near  the  aperture  and  having  the  prevailing  mesh  of  the  net-work 
of  small  size,  averaging  2  to  3  mm.  on  a  side.  The  species  is  distinguished  by 
these  characters  from  other  members  of  the  genus. 

The  basal  frill  is  clearly  retained  over  a  portion  of  the  lower  extremity 
and  shows  a  broadly  crenulated  or  furrowed  surface  and  an  irregular  margin. 

Dimensions.  The  length  of  the  specimen,  which  has  been  somewhat  com- 
pressed vertically,  is  36  mm.,  its  basal  diameter  30  mm.  and  the  width  of  the 
basal  expansion  7  mm. 

Localitij.  In  the  upper  part  of  the  Chemung  group,  Olean,  Cattaraugus 
county,  N.  Y.     (Collection  of  E.  B.  PIall.) 


Spkciks  of  the  Chemuno  Gkoitp.  188 

Thysanodictya  iiermenia,  sp.  nov. 

Plate  xxvii,  Fig.  7;  Plate  xxxix,  Via.  2  ;  Plate  xl,  Fios.  1-3. 

Sponge  attaining  large  size ;  subcylindrical,  expanded  both  at  base  and 
aperture.  Primary  quadrules  of  the  wall  of  the  cup  very  large,  varying  some- 
what in  size.  These  are  generally  longer  than  wide  and  the  largest  of  them 
may  measure  as  much  as  42x21  mm.  Such  large  quadrules  generally  lie  over 
the  median  part  of  the  sponge,  those  toward  the  extremities  being  shorter  and 
more  nearly  square.     The  finer  net-work  is  retained  in  places. 

The  length  of  a  large  specimen,  incomplete  at  the  aperture,  is  250  ram. 
Three  specimens,  two  of  which  were  evidently  overwhelmed  by  encroach- 
ing sediments  while  growing  in  their  normally  erect  j)Osition,  afford  a 
very  clear  conception  of  the  structure  of  the  basal  plate  or  diaphragm,  though 
the  vertical  walls  of  the  cup  have  been  so  compressed  as  to  wholly  destroy 
their  characters.  In  the  matter  of  size,  they  can  be  compared  oidy 
to  2]  quasillmn,  whose  basal  parts  have  not  been  precisely  made  out. 
Of  these  specimens  one  has  a  diameter  of  150  mm.,  the  other  measuring  to  the 
edge  of  the  peripheral  frill,  175  mm.,  the  base  of  the  cup  itself  being  130  mm. 
in  diameter. 

No  other  examples  of  this  genus  have  preserved  so  distinctly  the  structure 
of  these  basal  parts  and  of  these  three  specimens  two  retain  the  reticulum  A\-ith. 
out  distortion.  Both  show  that  the  coarse  meshes  with  strongly  elevated 
spicular  margins,  such  as  characterize  most  other  species  of  the  genus,  are  con- 
tinued from  the  lateral  walls  to  the  base,  and  while  they  present  the  same 
general  aspect  as  do  those  of  the  upright  surfaces  of  the  sponge,  their  composition 
is  distinctly  different.  There  is  no  radial  arrangement  evident  in  these  bands, 
no  center  of  departure  nor  apex  of  attachment.  The  vertical  bands  from  the 
lateral  walls,  bent  abruptly  into  the  diaphragm,  cross  each  other  from 
opposite  sides,  making,  over  the  central  portion  of  the  base,  large  square  meshes 
of  the  same  aspect  as  those  occuring  elsewhere,  but  here  formed  only  by  the 
intersection  of  the  vertical  spicular  bands. 

The  actual  crossing  of  these  vertical  bands  is  restricted  to  the  central 
area  of  the  base ;  away  from  this  area  and  in  the  peripheral  region  of  the  disk, 
the  bands  of  both  sets  gradually  put  on  the  aspect  of  the  horizontal 
bands  of  the  cup  and  actually  form  such  bands.  There  are  but  a  few  of 
these  concentric  ridges  on  the  disk,  not  more  than  three  or  four  of  the  first 
order,  and  the  inner  ones  are  somewhat  squared  to  the  quadrate  form  of  the 
central  area,  so  that  the  transition  from  the  concentric  to  the  vertical  ridges 


134  DiCTYOSPONGID/E. 

is  barely  noticeable  on  a  cursory  inspection.  The  character  of  this  peculiar 
reticulation  is  such  that  four  small  wedge-shaped  areas  remain,  one  in  each 
quadrant,  where  rectangular  reticulation  can  not  occur.  These  are  at  the 
sides  of  the  broad  bands  of  vertical  spicules  extending  over  from  the  lateral 
walls  of  the  cup,  and  here  the  net-work  is  not  only  irr-egular  but  seems  to 
have  formed  a  node  or  tuft. 

Beyond  the  edge  of  the  basal  plate  extends  a  frill  of  considerable  %vidth, 
very  sirailiar  in  character  to  that  occurring  in  Phragmodictya  catilUformis, 
but  not  as  wide  as  has  been  observed  in  that  species,  and  with  a  crenulated  or 
deeply  furrowed  surface.  At  the  union  of  the  frill  with  the  body  of  the 
sponge,  low,  irregular  and  elongate  nodes  are  formed,  as  observed  in  other 
species  of  the  genus.  The  more  prominent  of  these  extend  outward  from  the 
periphery  over  the  lower  surface  of  the  expansion  and  may  become  consider- 
ably elevated,  with  smooth,  concave  interspaces.  The  coarse  reticulation  of 
the  skeleton  is  here  visible  only  near  the  base  of  the  frill,  further  outward 
the  net-work  being  obscured.  In  the  specimens  under  consideration  the 
actual  extent  of  the  frill  is  pretty  clearly  defined  by  a  discoloration  of  the 
matrix. 

The  strong,  coarse  net-work  of  the  basal  disk  does  not  obscure  the 
exceedingly  fine  ultimate  reticulation  which  is  everywhere  seen.  In  one 
of  the  three  specimens  oblique  compression  has  disturbed  the  regularity  of  the 
basal  net-work,  but  the  strong  irregular  nodes  at  the  periphery  are  very 
pronounced.  Confirmatory  evidence  of  the  view  that  these  convex  plates 
are  basal  and  not  apertural  like  the  sieve-plate  in  Euplectella,  is  the  follow- 
ing fact ;  The  arenaceous  slabs  of  the  upper  Devonian  of  western  New 
York  frequently  have  their  surfaces  covered  with  various  forms  of  wave 
and  ripple  marks  and  furrows  left  by  the  dragging  of  pebbles,  tracks  of 
crustaceans,  etc.  Such  surface  markings  were  made  on  the  soft  muds 
which  subsequently  became  shales,  while  the  encroachment  of  more  sandy 
sediments  produced  relief  impressions  of  these  configurations  so  that  such 
reliefs  are  found  upon  the  lower  side  of  the  sandy  layers.  One  of 
these  basal  disks  of  Thysanodictya  liermenia  lies  at  the  upper  surface 
of  such  a  sandy  slab,  its  convex  face  downward,  the  compressed  cup  above, 
while  the  lower  surface  of  the  slab  is  covered  with  relief  impressions  of 
mechanical  markings  made  upon  the  sediments  immediately  beneath. 

Locality.     In  the  upper  beds  of  the  Chemung  group  at  Hinsdale,  Cat- 
taraugus county,  N.  Y.     (Collection  by  E.  B.  Hall.) 


Spkcies  op  thk  Ciiemuno  Group.  135 

TlIYSANODIOTYA   APLETA,   Sp.   nOV. 
Platk  XXVI,  Fio.  1 ;  Puitk  xxvii,  Fio.  8. 

This  name  is  applied  to  incomplete  examplew  indicating  cups  of  gi*eat  size 
with  a  very  coarsely  lamellose-reticulate  surface.  None  of  the  observed  speci- 
mens is  sufficiently  complete  to  give  a  clear  conception  of  the  form  of  the 
sponge  or  a  decisive  idea  of  the  generic  character  of  the  species. 

One  of  the  largest  individuals  seen  is  that  figured  on  Plate  xxvi ;  this  has 
a  length  of  about  275  mm.  and  a  greatest  width  of  220  mm,  The  drawing  has 
been  made  from  a  plaster-cast  of  an  external  impression,  which  shows  the 
curvature  of  the  surface  and  probably  gives  an  approximately  correct  idea  of 
the  diameter  of  the  cup,  but  no  accurate  indication  of  its  length.  So  far  as 
the  original  form  of  the  cup  is  here  suggested,  it  does  not  widely  differ  from 
that  of  the  known  species  of  TnYSANODiCTVA.  The  character  of  the  surface 
fortifies  this  resemblance.  The  principal  spicular  bands  form  strong,  erect 
lamellae,  nodose  at  their  intersections,  characters  which  indicate  the  surface 
structure  in  that  genus.  The  primary  quadrules  are  usually  square  and 
measure  about  14  mm.  on  a  side,  though  very  variable  both  in  fonn  and  size. 
The  secondary  bands  are  usually  clearly  defined  in  all  external  casts,  and  no 
internal  cast  has  been  observed  showing  the  finer  meshes  of  the  reticulum. 
The  surface  shows  no  other  ornamentation  than  that  described. 

Localities.  The  largest  of  the  specimens  of  this  species  was  found  loose 
among  the  sandstones  of  the  upper  Chemung  at  Alfred,  Allegany  county, 
associated  with  Spirifer  dlsjunctus.  Others  nearly  as  large  have  been 
obtained  at  Hinsdale,  Cattaraugus  county,  and  fragments  at  Scio  and  Ischuji. 
Smaller  fragments  have  been  received  from  Nile  and  Wellsville,  localities 
in  Allegany  county,  and  from  Olean,  Cattaraugus  county. 

Thysanodiotya  tukricula,  sp.  nov. 

Flats  xxxviii,  Figs.  6,  6. 

This  is  the  most  slender  and  graceful  of  all  the  species  of  the  genus.  The 
sponge  is  small  and  either  tapers  or  expands  upward  from  a  relatively  broad 
base.  The  surface  is  cancelled  into  narrow,  elongate  meshes  which  grow 
shorter  about  the  aperture.  While  these  quadrules  may  belong  to  a  minor 
series,  yet  they  are  predominant  upon  the  surface  to  the  obscuration  of  othera. 
The  periloph  is  quite  broad.  These  characters  are  well  shown  in  the  two 
entire  specimens  figured,  one  of  which  has  a  length  of  40  mm.,  a  basal  width 
of  1 2  mm.  and  an  apertural  mdth  of  7  mm.     The  width  of  the  basal  expansion 


136  DlClYOSPONOID.E. 

is  4-5  mm.     The  other  has  a  length  of  42  mm.,  a  basal  width  of  10  mm.,  across 
the  periloph,  18  mm.  and  across  the  aperture,  15  mm. 

The  species  is  well  characterized  by  its  elongate  form,  small  size  and 
the  nature  of  its  quadration. 

LiK-ality.  In  the  upper  Chemung  sandstones  at  Wellsville,  N.  Y.  (Col- 
lection of  E.  B.  Hall.) 

ARYSTIDICTYA,  gen.  nov. 

This  genus  comprises  sponges  which  are  cylindrical  toward  the  base 
but  expand  xipward  to  a  broad,  flaring  aperture,  the  general  form  of  the 
sponge  being  like  that  of  Lebedictya.  The  lower  extremity  is,  however, 
abruptly  truncated  by  a  circular,  transverse  or  slightly  convex  diaphragm  as 
in  TuYSANODicTYA  and  Phragmodictya.     No  basal  frill  has  been  observed. 

Surface  smooth  or  with  low,  irregularly  arranged  nodes;  more  or  less 
distinctly  reticidate. 

Apertural  region  with  fine,  branching  channels  running  to  the  margin. 

Type,  Arysiidictya  elegans,  sp.  nov. 

Arystidictya  elegans,  sp.  nov. 

Plate  xli,  Pig.  10;  Plate  xlii,  Fio.  7. 

Sponge  of  considerable  size.  The  single  specimen  has  been  compressed 
but  was  undoubtedly  circular  at  the  base  and  regularly  obconical  for  about 
one  third  of  its  length,  thence  expanding  abruptly  to  the  very  large  aperture. 
The  form  of  the  cup  is,  thus,  somewhat  broadly  vase-shaped. 

Surface  smooth,  without  nodes  or  other  irregularities.  Margin  regular 
and  entire. 

The  reticulnm  as  preserved  appears  to  be  composed,  over  the  lower  part  of 
the  sponge,  of  rather  sparse  and  loosely  arranged  spicular  bands  crossing  at 
right  angles  and  forming  square  meshes  about  4  mm.  on  a  sitle.  Whatever 
subdivision  of  these  meshes  existed  appears  to  have  been  quite  irregular  and 
the  principal  quadrules  themselves  seem  to  be  at  times  crossed  by  deviating 
threads  of  the  principal  strands.  Over  the  upper  expanded  surface  of  the  cup 
the  reticulation  is  not  distinctly  visible,  the  surface  being  smooth  or  covered 
with  irrregularly  distributed  depressions,  and  toward  the  aperture  these  are 
extended  into  grooves  or  branching  channels,  subparallel  to  one  another  and 
perpendicular  to  the  margin. 

The  base  of  the  sponge  is  truncated  by  a  diaphragm  the  compression  of 
which  has  obscured  the  reticulation  of  the  surface. 


Spkcies  ok  the  CiiKMUNfi  Gijour.  137 

Dimensions.  Length  145  mm.;  diameter  at  the  base  53  mm.,  at  the 
aperture,  160  ram. 

Locality.  In  the  lower  part  of  the  Chemung  group  at  Cotton  hill,  one 
mile  north  of  Avoca,  Steuben  county,  N.  Y.     (Collection  of  E.  B.  Hall.) 

Arystidictya  nodifeba,  sp.  nov. 

Plate  xli,   Fig.  11. 

In  this  species  the  ciip  expands  somewhat  more  regularly  than  in  A. 
elegans,  and  is  of  smaller  size.  Its  surface  is  covered  with  low,  vertically 
elongate  nodes  which  appear  to  be  without  definite  order;  these  are  larger 
toward  the  base  and  become  small  and  obscure  toward  the  aperture,  the 
region  about  the  apertural  margin  being  smooth.  This  margin  is  smooth,  unin- 
terrupted, thin  and  regidar. 

This  is  an  interesting  species,  but  is  imfortunately  known  only  from  a 
part  of  a  cup  representing  probably  about  one-half  the  original  wdth. 

Locality.  In  the  lower  part  of  the  Chemung  group  at  Cotton  hill,  near 
Avoca,  N.  Y.,  associated  with  A.  elegans,  Hydnocera-s  tuberosum,  H.  multino- 
dosrnn,  H.  eutJieles  and  Hallodictya  Cottoniana.     (Collection  of  E.  B.  Hall.) 

HYPHANT^NIA,  Vanuxem  (emendata). 

1842.      JJphantenia,  Vanuxem.     Geology  of  New  York;    Kept.  Third  Dist., 

p.  184,  fig.  50. 
1863.     UphantcBnia,  Hall.     Sixteenth  Ann.  Rept.  N.  Y.  State  Cab.  Nat.  Hist. 

pp.  84-87. 

The  remarkable  bodies  to  which  the  term  UpiiANTiENiA  has  been  long 
applied,  are  very  large,  circular  and  shallow  saucer-shaped  cups,  composed  of 
two  series  of  intersecting  spicular  straps,  one  radiating  from  a  common  center, 
the  other  concentric  about  the  same  point.  The  example  which  was  described 
by  Vanuxem  under  the  name  Uphantenia  Oliemungeiisis  was  a  flattened  quad- 
rant, the  original  circular  form  of  which  was  only  inferential ;  but  specimens 
of  the  species  since  found  show  the  aperture  of  the  cup  to  have  been  circidar. 
In  shape  it  appears  to  have  been  slightly  elevated  at  the  basal  point,  thence 
outward  to  have  curved  gently  downward  and  then  upward  again  toward  the 
apertural  margin.  This  at  least  is  the  foma  presented  by  what  seem  to  be 
internal  casts,  from  the  sandstones  of  the  Chemung  group.  The  peculiar 
feature  of  this  shape,  however,  is  the  upward  or  inward  projection  of  the  apex 
or  base  of  the  sponge.     In  the  two  specimens  in  which  this  part  of  the  cup  is 


138  DlCTYOSPONGID.*;. 

retained  there  is  a  small  mass  of  sandy  matrix  adhering  to  the  basal  portion, 
though  the  remainder  of  the  cup  so  far  as  retained  is  cleanly  separated  from 
the  rock.  From  this  it  is  perhaps  to  be  inferred  that  the  center  was  largely  a 
tufted  pompon  of  spicules  whose  structure  is  obscured  by  complication  with 
the  sediment. 

It  is  possible  that  the  nonnal  form  of  the  cup  was  the  reverse  of  that 
we  have  suggested ;  instead  of  being  saucer-shaped,  it  may  have  been  broadly 
umbrella  shaped.  This  is,  however,  less  in  accordance  with  the  form  usually 
assumed  by  the  palaeozoic  Dictyosponges  which  have  been  described,  but  it 
is  a  shape  occasionally  reproduced  among  living  hexactinellids.  The 
material,  however,  representing  the  genus  is  insufficient  for  the  detennination 
of  this  point. 

From  the  center  or  basal  point  radiates  a  series  of  flat,  gradually  widening 
ligulate  bands,  with  elevated  or  thickened  edges.  These  bands  are  approxi- 
mately of  equal  size  and  are  sube(juidistant,  but  a  slight  difference  in  the 
interval  between  them  at  their  origin  becomes  greatly  increased  toward  the 
margin.  There  is  no  evidence  of  more  than  a  single  series  of  these  vertical  or 
radial  bands,  all  of  them  meeting  at  the  center.  They  are  crossed  by  con- 
tinuous concentric  bands  of  similar  character,  though  with  a  less  palpable 
thickening  of  the  margins.  These  regularly  increase  in  width  as  they  recede 
from  the  center,  and  the  diameter  of  the  intervals  between  them  also  increases, 
but  less  rajiidly.  At  their  intersection  with  the  radial  bands  they  do  not  lose 
their  definition,  and  are  probably  interwoven  with  them  as  there  is  no  evidence 
of  their  displacement  at  these  surfaces,  as  would  in  all  probability  have 
occasionally  occurred  had  there  been  no  actual  connexion  or  interlacing  of  the 
spicules. 

Over  the  interspaces  between  the  radial  bands  it  is  observable  that  the 
position  or  angle  of  these  concentric  bands  is  slightly  changed ;  they  do  not 
there  confonn  to  the  general  curvature  of  the  surface  of  the  cup  but  are 
horizontal  or  approximately  so ;  the  effect  of  this  arrangement  is  to  give  to 
each  of  these  radial  interspaces  including  the  bounding  I'ays  a  suggestive 
resemblance  to  a  much  inclined  step-ladder. 

The  meshes  or  quadrides  produced  by  the  intersection  of  the  two  series  of 
bands  usually  show  only  clean  rock  surfaces ;  one  specimen  however  indicates 
that  these  quadrules  have  had  their  angles  filled  by  a  delicate  parenchymal 
film  which  encroaches  on  the  quadrules  to  such  an  extent  as  to  leave  a  sub- 
circular  opening.  We  have  no  palpable  evidence  of  the  nature  of  the  spicules 
composing  the  radial  or  vertical,  and    the    concentric    or   horizontal   bands 


Spkcies  op  the  CnKMUNO  Group. 


139 


Their  surface  is  smooth  and  usually  presents  such  an  aspect  as  might  he 
produced  by  the  pressure  of  fine  sand  grains  agaiirst  a  thin,  flat  b<Mly.  It  is 
so  extremely  seldom  that  any  traces  of  the  spicular  skeleton  are  retained  in 
the  sandstones  of  the  Chemung  group,  that  the  absence  of  such  evidence  here 
is  not  surprising.  It  is  possil)le  that  their  structure  is  similar  to  that  of  the 
ligulate  spicules  observed  in  I/ictysponr/ia  MarcelUa.  The  character  of  the 
apertural  margin  is  not  shown  in  any  of  the  specimens. 

Hypiiant^nia  seems  to  be  a  Dictyosponge  in  which  the  compacting  and 
condensation  of  the  vertical  and  horizontal  spicular  bands  has  been  earned  to 
an  extreme,  leaving  between  the  sharply  delimited  bundles  broad  areoles 
which  in  generic  types  like  Hydnocekas,  Dictyospongia,  etc.,  would  be  sub- 
divided by  the  minor  spicular  bands,  or  in  Piiysospongia,  be  filled  by  dermalia 
and  parenchynialia. 

The  only  known  species  of  Hypiiant^nia  is  the  II.  Chemvngensis, 
Vanuxeni,  and  of  this  species  but  five  specimens  have  come  to  our  knowledge, 
four  of  them  being  here  figured. 


Hypiiant^nia   Chemungensis,  Vanuxem. 

Plate  xxviil,  Fio.  1,  2;  Plate  xxix,   Fig.  1;   Plate  xlv,  Pio.  I, 

1842.      Uflmntenia  Chemungensis,  Vanuxem.     Geology  of  New  York;  Rept. 

Third  Dist.,  p.  184,  fig.  50. 
1863.      Uphantcenia  (Jlieimingensis,  Hall.     Sixteenth  Ann.  Rept.  N.  Y.  State 

Cab.  Nat.  Hist.,  pp.  84-87,  figure  (on  p.  86). 
1884.      Uphant(enia  OJiermmgensis,  Hall.    Thli-ty-fifth  Ann.  Rept.  N.  Y.  State 

Mus.  Nat.  Hist.,  p.  481. 
1890.      UplMnUBTnia  Chermcngensis,    Lesley.     Dictionary   of   Fossils,  vol.    3, 

p.  1246. 

Sponge  large,  explanate,  broadly  saucer-shaped,  or  umbrella-shaped  with 
a  short  base  of  attachment.  Apertural  outline  circular.  The  radiating  ligu- 
late spicular  bands,  all  of  which  depart  from  the  center  or  apex,  ninnber 
thirty-two  in  the  two  specimens  which  retain  this  part  of  the  sponge ;  the 
original  specimen  which  is  approximately  a  quadrant  of  the  cup  indicates  the 
existence  of  the  same  number  of  radial  bands.  These,  as  above  observed,  are 
not  always  equidistant,  but  the  irregularities  in  this  respect  are  not  con- 
spicuous. The  concentric  horizontal  bands  regularly  increase  in  width 
from  apex  to  margin.  What  their  number  was  in  the  completed  cup  it 
is  impossible   to   say ;   the  greatest  radius  of    the   largest   specimen   shows 


140  DlOTYOSPONGID^. 

nineteen  of  these  and  the  appearance  of  the  outeiTnost  one  indicates  the 
margin  of  the  cup. 

There  is  some  variation  in  the  relative  proportion  of  the  parts  shown  in 
the  material  at  hand.  The  greatest  radius  exposed  by  any  specimen  is  192 
mm. ;  upon  this  are  nineteen  concentric  bands  ;  upon  the  original  specimen  a 
radius  (almost  but  not  quite  complete)  of  147  mm.  shows  twenty  concentric 
bands.  In  another  example  there  are  fourteen  bands  in  a  radius  of  100  mm. 
There  is  no  regular  variation  in  size  in  either  series  of  bauds,  but  sometimes 
there  is  a  slight  change  in  the  width  of  a  concentric  band  ;  as,  for  example,  the 
15th  concentric  band  on  the  original  specimen  measures  5  mm.  at  one  place,  and 
8  mm.  at  another.  In  three  of  the  specimens  the  13th  concentric  band  is 
situated  at  a  radial  distance,  measuring  its  outer  edge,  of  93  ram.,  92  mm.  and 
118  mm.  In  the  first  of  these  the  width  of  this  band  is  4.5  mm. ;  in  the  second 
3.2  mm. ;  in  the  third  5  mm.  These  differences  are  undoubtedly  individual 
variations. 

The  largest  example  of  this  sponge  yet  observed  has  a  greatest  diameter  of 
347  mm.;  its  radius,  which  appeal's  to  be  complete,  is  192  mm.  so  that  its  full 
diameter  was  about  380  mm.  or  15  inches.  A  smaller  individual  not  complete 
at  the  margin  has  a  diameter  of  170  mm. 

Geological  horizon.  In  a  compact  and  fine-grained  greenish  or  rusty 
sandstone  from  the  upper  beds  of  the  Chemung  group,  in  the  vicinity  of 
Owego,  Tioga  county,  N.  Y.  Upon  one  of  the  specimens  are  crinoidal 
segments  and  abundant  examples  of  Tentacnlites  spiculus. 

HALLODICTYA*,  gen.  nov 
Sponge  growing  in  an  explanate  frond  without  definite  limitation  of  its 
margins ;  in  this  respect  similar  to  the  genera  Actinodictya  and  Crypto- 
DiCTYA.  Surface  highly  irregular,  moulded  into  circular  or  elongate  nodes 
irregularly  grouped,  and  broader  swellings  and  depressions.  Reticulation  of 
the  surface  fine,  regular  and  rectangularly  developed. 
Type,  Ilallodictya  Sdensis,  sp.  nov. 

Hallodictya  Sciensis,  sp.  nov. 

Plate  xxxvi.  Fig.  8. 

This  very  interesting  sponge  is  of  considerable  size,  showing  that  its 
explanate  mode  of  growth  was  maintained  in  all  stages.     Its  surface  is  of  so 

•  This  generic  term  is  intended  as  a  partial,  though  inadequate  recognition  of  the  important  contri- 
bntions  to  the  knowledge  of  the  Dictyospongea  ol  the  Chemung  group  made  by  E.  B.  Hall,  Esq.,  of 
WellsvUle,  N.  Y. 


Speciks  ok  thk  CuKMirNo  Gkoup.  141 

irregular  a  character  as  to  convey  the  impression  that  the  frond  miglit  have 
conformed  its  shape  to  the  irregidarities  of  the  surface  ui)on  which  it  grew. 
Yet  our  knowledge  of  the  mode  of  growth  in  living  siliceous  sponges  does  not 
favor  this  conception ;  rather,  that  the  frond  was  attached  at  one  extremity,  its 
expanded  surface  free  and,  consecj^uently,  its  irregularities  not  due  t<j  such 
conformation,  but  altogether  idiocratic.  These  irregularities  may  be  roughly 
classed  as  great  and  small.  The  former  are  the  main  depressions  and  eleva- 
tions of  the  frond ;  the  latter  are  groups  of  nodes  or  pits  upon  the  surface  of 
the  former,  which  are  probably  of  similar  nature  to  the  tufted  node  of 
Hydn^ocp:uas  and  other  genera.  On  the  cxi)osed  surfivce  of  the  sjHjcimen  in 
hand,  these  nodes  are  expressed  as  pits  generally  narrow  and  elongate  rather 
than  circular,  and  seldom  occurring  singly. 

The  reticulum  is  fine  and  sharp  but  shows  only  the  ultimate  and  penulti- 
mate quadrulation,  spicular  bands  of  higher  order  being  wholly  obscured. 
The  direction  of  the  spicular  lines  indicates  no  convergence  to  any  single 
point;  those  of  the  horizontal  series  curve  as  the  contour  of  the  surface 
requires,  and  seem  to  approximate  a  circular  curvature  in  both  the  upper  and 
lower  portion  of  the  specimen.  The  upper  area  is  the  better  defined,  and  near 
the  center  of  the  general  curvature  of  the  spicular  bands  is  the  best  defined 
group  of  small  nodes  upon  the  surface.  We  may  thus  infer  a  remote  pos- 
sibility of  the  attachment  of  the  frond  at  more  than  one  point,  whence  the 
growth  of  the  spicular  skeleton  progressed. 

This  species  has  been  observed  in  but  a  single  specimen  having  a  length 
of  190  mm.  and  a  greatest  width  near  the  middle  of  90  mm. 

The  interpretation  of  the  obscure  genus  Cryptodictya,  as  well  as  Aci'ino- 
DicTYA,  is  made  clearer  by  the  aid  of  Hallodictya.  In  Actixodictya,  the 
evidence  of  spicular  reticulation  is  restricted  to  coarse,  ridge-like  bands  which, 
though  seeming  to  make,  at  times,  quadrangular  meshes,  usually  appear  to 
be  without  much  order,  so  that,  by  itself  considered,  there  might  be  a  possi- 
bility of  misconstruction  in  regarding  them  as  representing  the  actual  mesh- 
work  of  the  sponge.  In  Cbyptodictya,  this  condition  is  the  more  obscured 
by  the  actual  obliteration  of  nearly  every  trace  of  the  spicular  skeleton  except 
the  tufts  at  the  tips  of  the  nodes.  In  this  genus,  however,  the  siu-face  nodes 
are  quite  as  strongly  developed  as  in  Hallodictya.  All  three  genera  agree  in 
their  explanate,  irregular  and  undefined  mode  of  growth. 

Locality.  Hallodictya  Sciensis  occurs  in  a  compact  siliceous  sandstone  at 
Scio,  Allegany  county.     (Collection  of  E.  B.  Hall.) 


142  Dictyospongid.t;. 

« 

Hallodictya  Cottoniana,  sp.  nov. 

Platk  xli,  Fig.  12. 

A  j)ortioii  of  a  frond,  evidently  of  the  mode  of  growth  characterizing 
this  genus,  is  less  irregular  than  the  foregoing  species  in  the  nodulation  of  its 
surface,  which  bears  only  low  and  inconspicuous  elevations.  The  reticulation 
is  quite  as  sharply  defined  as  in  IT.  Sciensis,  but  it  is  notably  less  regular  and 
lacks  even  the  suggestion  of  arrangement  about  any  point.  Notwithstanding 
this  irregularity,  the  impression  of  the  reticulum  clearly  pertains  to  one  and 
the  same  layer,  there  being  no  evidence  that  such  irregularity  is  due  in  any 
way  to  the  overlapping  and  compression  of  the  reticular  walls. 

The  single  specimen  observed  is  of  small  size,  measuring  85  by  50  mm., 
evidently  not  the  entire  extent  of  the  frond. 

Locality.  In  the  middle  beds  of  the  Chemung  group,  at  Cotton  hill,  one 
mile  north  of  Avoca,  associated  with  Hydnoceras  multinodosum,  Arystidictya 
elegans  and  A.  nodifera.     (Collected  by  Jacob  Van  Deloo.) 

ACTINODICTYA,  Hall. 

1890.  AcUiwduiya,  Hall.  Ninth  Ann.  llept.  N.  Y.  State  Geol.,  p.  59  ;  Forty- 
third  Ann.  Kept.  N.  Y.  State  Museum;  p.  261. 

This  genus  was  constituted  for  explanate,  somewhat  irregularly  growing 
sponges,  whose  surface  is  depressed  convex  and  whose  form  appears,  from  the 
aspect  of  the  majority  of  specimens,  to  have  been  broadly  oval.  The  fronds 
have  not  shown  the  mode  of  attachment,  but  the  absence  of  any  cicatrix,  or 
stem,  may  perhaps  indicate  that  the  sponge  was  fixed  by  a  marginal  foot-stalk. 
There  are  among  the  living  hexactinellids  species  having  this  explanate  or 
leaf-shaped  aspect  and  attached  by  such  a  stalk-like  basal  extension ;  and 
these  bodies  may  be  compared  with  the  Euryplegma  auriculare  as  figured  by 
SciiULZE  (Voyage  of  H.  M.  S.  "Challenger";  Hexactinellida,  pi.  cii,  fig.  1). 
The  surface  of  the  specimens  is  crossed  in  all  directions  by  ridges  and  lines 
of  various  sizes  representing  the  interlacing  spicular  bands,  and  yet,  notwith- 
standing the  conspicuous  irregularity  in  the  disposition  of  these  bands,  every 
specimen  shows  more  or  less  distinctly  a  certain  degree  of  rectangidar  reticu- 
lation among  both  the  coarser  and  finer  bands.  The  coarse  bands  are  com- 
paratively few  in  number,  and  the  differences  in  size  in  the  coarse  and  fine 
ridges  are  gradational  and  never  abrupt.  The  irregular  intersection  of  the 
various  spicular  bands  gives  the  reticulum  a  some^vhat  stellate  aspect. 

In  the  living  species,  Walteria  Flemiyvlngl^  Schulze  (see  work  cited,  pi.  ix)^ 


Spkcikh  ok  thk  Ciikmun*-,  Gkoijp.  143 

there  is  a  very,  similar  irregularity  in  the  sponge  meshes  produced  by  the 
intersection  of  spicular  bands  at  varif)us  angles,  while  the  rectangular  inter- 
section is  to  some  degree  retained.'  Though  Walteria  is  a  cup-shaped 
sponge,  the  structure  of  its  reticulum  indicates  that  these  fossil  bodies, 
AcTiNODicTYA,  are  correctly  construed  as  hexactinellid  sponges. 
Type,  Actinodictya  placenta,  Hall. 

AoTrNODICTYA   PLACErTTA,    Hall. 
Plate  xxx,  Figs.  1-3;  Platb  xxxi,  Figs.  1,  2. 

1890.     Actinodictya  placenta,  Hall.     Ninth  Ann.  Rept,  N.  Y.  State  Geologist, 
p.  60 ;  Forty-third  Ann.  Rept.  N.  Y.  State  Museum,  p.  262. 

Sponge  explanate,  broadly  depressed-convex,  most  of  the  specimens 
having  a  suboval  though  unfinished  outline,  occasionally  with  margins  deflected 
or  incui-ved.  Surface  without  nodes  but  covered  with  spicular  ridges  of 
various  sizes,  the  larger  usually  crossing  the  frond  through  its  entire  length, 
the  smaller  being  traceable  for  a  less  distance;  these  ridges  crossing  one 
another  in  all  directions  without  obscuring  a  certain  degree  of  rectangular 
reticulation.  The  margins  of  the  specimens  are  not  defined  by  any  of  the 
spicular  bands.  The  variation  in  size  from  the  coarsest  to  the  finest  bands 
is  very  considerable.  The  surface  of  these  fronds  is  usually  of  a  darker  color 
than  the  matrix,  and  this  fact  contributes  to  the  actual  appearance  of  the 
specimen  in  evincing  the  very  close  interweaving  of  the  spicular  bands. 

Localities.  This  species  has  proven  to  be  quite  abundant  in  some  of  the 
lower  or  middle  strata  of  the  Chemung  group  in  the  vicinity  of,  and  to  some 
extent  in  association  with  the  colony  of  Hydnoceras  tuberosum  at  the  Brown 
hill  school  house  near  Wallace,  Steuben  county.  A  few  specimens  have  come 
from  this  spot  itself,  and  about  a  half-mile  away,  upon  the  farm  of  Mr.  John 
Roth,  a  considerable  number  of  fine  specimens  has  been  collected  by  Mr.  Jacob 
Van  Deloo,  in  close  proximity  to  the  horizon  of  H.  tuberosum.  A  few  miles 
to  the  southeast,  in  the  town  of  Howard,  Mr.  D.  D.  Luther  has  found  some 
loose  arenaceous  slabs  undoubtedly  derived  from  the  immediate  neighbor- 
hood, and  which  are  nearly  composed  of  small  fronds  of  this  sponge. 
Among  the  earliest  specimens  of  the  species  found  are  some  from  the  upper 
Chemung  beds  near  Lawrenceville,  Tioga  county,  Pennsylvania,  collected  in 
1879,  by  A.  S.  Sherwood.  Associated  with  it  in  the  localities  in  Steuben 
county  are  Atrypa  hystrix,  Spirifer  niesacostalis,  Productella  lachrymosa,  and 
Poteriocrinvs  Glarhei. 


144  DlOTTOSPONGID^. 

CRYPTOBICTYA,  Hall. 

1890.     CryptoiUetija,  Hall.     Ninth  Anji.  Rept.  N.  Y.  State  Geologist,  p.  60; 
Forty-third  Ann.  Rept.  N.  Y.  State  Museum,  p.  202. 

Irregularly  growing,  tuberose,  coarsely  nodose,  and  probably  uninclosed 
sponges,  with  a  surface  showing  no  reticulating  spicular  bands;  the  skeleton 
being  presiunably  composed  of  extremely  fine  isolated  spicules,  except  at  the 
summits  of  the  nodes  which  show  evidence  of  short  tufts  of  prostalia. 

Athough  these  specimens  do  not,  of  themselves,  clearly  demonstrate  their 
title  to  a  position  among  the  hexactinellid  sponges,  yet  in  the  light  of  the  asso- 
ciated generic  structures,  Hallodictya  and  Actinodictya,  we  are  doubtless 
justified  in  regarding  them  as  Silicea,  having  a  matted  or  felted  reticulum  of 
very  minute  spicules,  and  nodes  with  distinct  pi'ostalia  developed  at  their 
summit. 

In  addition  to  this  structure,  the  surface  frequently  presents  a  sort  of 
linear  striation  or  brush  of  lines  departing  from  the  nodes  or  sweeping  irregu- 
larly over  the  concave  spaces. 

Type,  Cryptodictya  Alleni,  Hall. 

Cryptodictta  Alleni,  Hall. 

Platk  XXXI,  Figs.  3-6 ;  Platb  xxxii,  Figs.  1-4. 

1890.     Cryptodictya   Alleni,    Hall.     Ninth    Ann.    Rept.  N.  Y.  State  Geol., 
p.  60 ;  Forty-third  Ann.  Rept.  N.  Y.  State  Museum,  p.  262. 

Sponges  of  moderately  large  size,  with  a  general  tendency  to  a  suboval 
or  elongate  outline,  and  a  depressed  convex  sux'face  with  abruptly  deflected 
margins.  In  the  majority  of  the  best  preserved  examples  this  margin  is 
clearly  defined  for  more  or  less  of  its  extent,  showing  that  the  form  was  frond- 
shaped  or  scaphate,  though  highly  irregular  and  tubei'ose  in  contour  and  vari- 
able in  outline.  In  one  example  the  frond  has  been  so  enfolded  by  lateral 
compression  as  to  bring  the  edges  close  together  and  produce  the  appearance 
of  a  nearly  closed  cup.  The  attachment  of  these  bodies  must  have  been 
either  by  means  of  a  foot-stalk  or  by  the  margins  of  the  frond.  The  nodes 
upon  the  surface  are  of  various  shapes  and  irregular  in  disposition ;  usually  a 
number  of  small  nodes  are  situated  upon  the  summit  of  a  large  one,  these 
being  connected  by  a  more  or  less  sharply  defined  ridge,  an  arrangement  also 
seen  in  the  genus  Hallodictya.  The  entire  surface  of  the  rock  covered  by 
the  frond  is  invariably  tinted  of  a  deep  chocolate  or  reddish  bro\vn,  in 
strong  contrast  to  the  color  of  the  matrix.     An  obscure,  irregular  lineation  of 


Spkciks  of  thk  CiiEMrNo  Gitoiri'.  145 

the  surface  can  be  made  out  on  nearly  all  examples,  but  it  is  not  such  as  to 
give  a  clew  to  the  skeletal  structure,  except  as  we  may  infer  that  the  skeleton 
was  fine  and  dense,  and  we  know  that  the  nodes  were  tufted. 

A  symmetrical  individual  of  suboval  outline  has  a  length  of  155  mm.  and 
a  median  diameter  of  90  mm.;  this  is  about  the  average  size  of  most  of  the 
specimens. 

Geological  Iwrizon.  In  the  Chemung  sandstones  near  Cohocton  and 
Avoca,  Steuben  county,  and  at  Lyndon,  Cattaraugus  county,  N.  Y.  Named  for 
the  late  Reverend  Jonathan  Allen,  former  President  of  Alfred  University, 
Alfred,  N.  Y. 

AGLITHODICTYA,  gen.  nov. 

Small,  button-shaped  sponges,  greatly  expanded  horizontally  at  mid-length, 
abruptly  tapering  to  both  extremities.  Aperture  narrow ;  surface  finely  retic- 
ulate, without  nodes  or  other  ornament. 

Type,  Aglitliodictya  numuhna,  sp.  nov, 

Aglithodiotya  numulina,  sp.  nov. 

Plate  xxv,  Figs  6-9. 

Sponge  small,  subdiscoidal,  biconvex  with  reference  to  a  median  horizontal 
plane,  the  parts  below  and  above  this  plane  being  subconical,  that  above  being 
considerably  the  more  depressed.  Radical  portion  of  the  cup  not  fully  retained ; 
lateral  expansion  from  the  inferior  extremity  about  thrice  the  vertical  growth. 
The  sponge  attains  its  greatest  diameter  at  a  horizontal  plane  lying  about  three- 
fourths  of  the  vertical  distance  from  the  inferior  extremity  to  the  aperture, 
and  is  thence  bent  sharply  ujjwards.  Upon  the  matrix  there  is  some  trace  of 
an  erect  spicular  lamella  or  expansion  about  the  cup  at  this  plane  ;  otherwise 
the  surface  is  smooth.  The  margin  of  the  apertural  orifice  is  rounded  and 
somewhat  swollen  into  low  irregular  nodes.  Departing  from  the  orifice  and 
traversing  the  apertural  slope  to  the  peripheral  angulation  are  traces  of  diver- 
gent surface  grooves  or  canals  such  as  are  frequent  among  the  lithistid 
sponges  and  are  not  unknown  among  the  thick-walled  hexactinelliils. 

The  reticulum  is  exceedingly  fine,  but  distinguishable  over  all  of  the 
principal  exposures  of  the  surface.  There  is  no  noticeable  predominance  of 
any  spicular  bands,  and  the  quadrules  are  rather  too  indistinct  to  be  accurately 
measured.  The  species  is  based  on  a  single  fairly  preserved  specimen  in  a 
.  sandstone  matrix ;  the. form  is  well  retained  and  the  skeleton  in  the  usual  con- 
dition of  other  sponges  of  this  fauna.  This  specimen  has  a  vertical  length  of 
11  mm.,  a  greatest  diameter  of  24  mm.,  and  an  apertural  width  of  8  mm. 


146  DrCTYOSPONGID^. 

Observations.  It  may  be  possible  that  in  this  species  we  have  not  a  true 
thin-walled  Dictyosponge,  as  its  form,  relatively  small  aperture,  and  the,  radial 
canals  upon  its  surface  all  suggest  some  of  the  thicker-walled  sponges,  such 
for  example,  as  those  of  the  genus  Pachytrichisma,  one  of  the  dictyonine 
family,  Ventriculitid^.  (For  comparison  see  Zittel,  Handhuchder  Palaeon- 
tologie,  vol.  1,  p.  177,  P.  (Jarteri,  Zittel,  fig.  92.)  Nevertheless,  such  positive 
evidence  as  here  presented,  affords  no  good  reason  for  excluding  this  peculiar 
genus,  Aglithodictya,  from  the  Dictyospongid^. 

Locality.  In  the  sandstones  of  the  Chemung  group,  near  Lawrenceville, 
Tioga  county,  Pennsylvania. 


! 


SPECIES  FROM  THE  PSAMMITES  DU  CONDROZ  (FRANCE). 
Sabmitted  by  Professor  Charles  Uabrois,  UniTeriity  of  Lille. 

DicTYospoNGiA  MoBiNi,  Barpois  (sp.). 

Platk  xltz,  Fiob.  1,2. 

1883.     Dictt/ophf/ton  Mormi,'Ba,rrois.     Annalea  de  la   Soc.  G60I.   du    Nord, 
vol.  xi,  p.  83,  pi.  1,  figs.  2  a-c. 

This  species  is  one  of  the  smooth  conical  sponges  for  which  the  generic 
term  Dictyospongia  has  been  here  proposed.  The  specimen  which  we  have 
examined  represents  the  upper  portion  of  the  cup  including  the  a])erture, 
which  is  regular  and  apparently  unornamented.  The  impression  of  the  reticu- 
lum is  obscurely  retained  and  extremely  fine.  Professor  Barbois  has  figured 
not  only  the  specimen  here  illustrated,  but  also  one  which  retains  the  basal 
part  of  the  cup  and  shows  that  the  expansion  of  the  sponge  was  gradual  and 
uniform. 

Locality.     Jeumont  (Departement  du  Nord),  Brittany. 
Htdnoceras  Barroisi,  nom.  nov. 

Platb  xlti,  Figs.  3,  1. 

1883.     Dictyophyton  tuberosum,  Barrois.     Annales  de  la  Soc.  G60I.  du  Nord, 
vol.  xi,  p.  82,  pi.  1,  figs.  1  a-e. 

Attention  has  already  been  directed  to  the  fact  that  the  species  described 
as  Dictyophyton  tuberosum,  Conrad,  by  Professor  Barrois,  in  his  memoir  on 
the  DiCTYOspoNGiD^  of  the  Psammites  du  Condroz,  proves,  after  a  careful 
revision  and  comparison  of  the  nodose  sponges,  to  represent  a  specific  form 
which  is  not,  to  our  knowledge  reproduced  among  the  American  species  of 
this  genus.  The  distinctive  characters  of  the  French  fossil  lie  in  (1)  the  very 
deep  and  broad  horizontal  constrictions  which  render  the  node-bearing  ridges 
very  conspicuous;  (2)  the  sharply  defined  prism-faces,  their  margins  being 
distinctly  continuous  ridges  extending  from  the  apex  and  crossing  the  hori- 
zontal constrictions ;  (3)  a  faint  median  ridge  dividing  each  of  the  eight  prism- 
faces  ;  (4)  the  low,  elongate  and  sharp  nodes  forming  ridges  by  fusion  with  the 
vertical  prism-margins.  The  single  specimen  is  broken  across  the  third  node- 
bearing  ridge  and  there  is,  hence,  no  means  of  inferring  the  number  of  such 

147 


148  DlCTYOSPONGID^.. 

horizontal  ridges  in  the  total  length  of  the  sponge.  There  is  still  another 
feature  in  which  this  species  diifers  from  H.  tuberosum,  viz. ;  (5)  the  preva- 
lence of  a  very  fine  network,  the  meshes  being  of  subequal  size,  contracting 
over  the  constricted  areas.  Large  quadrules  such  as  are  usually  found  in 
H.  tuberosum  are  wanting.  Professor  Babrois's  excellent  figures  of  this  species 
show  the  curvature  of  the  sponge,  which  seems  to  be  probably  an  accidental 
feature  of  the  specimen. 

Locality.  In  compact  greenish  sandstone  of  the  Psammites  du  Condroz 
(=Chemung  group  of  New  York),  Jeumont,  Brittany. 

Hydnoceuas  Jeumontense,  sp.  nov. 

Plate  xlvi,  Figs.  5,  6. 

This  form  is  represented  by  several  more  or  less  compressed  speci- 
mens in  which  the  cup  is  broad  and  has  somewhat  the  aspect  of  the 
Chemung  species,  H.  pJiymatodes,  though  Avith  coarser,  less  abundant  and 
less  regular  nodes. 

The  form  of  sponge  is  rather  broadly  turbinate,  its  lower  portion  being 
distinctly  prismatic  for  a  considerable  distance  from  the  apex.  The  horizontal 
constrictions  appear  to  be  rather  narrow  and  shallow,  while  the  node-bearing 
ridges  are  decidedly  elevated  and  often  quite  broad.  The  nodes  themselves 
are  elongate  and  those  of  the  vertical  rows  are  more  or  less  distinctly  con- 
nected by  the  ridge  at  the  angle  of  the  prism-faces.  A  remarkable  character 
of  the  species  is  the  rapid  multiplication  of  nodes.  At  the  base  of  the  sponge 
the  number  of  the  prism-faces  appears  to  be,  as  normally,  eight,  liut  even  on 
the  first  node-bearing  row,  nodes  may  appear  on  the  interspaces  between  the 
angles  of  the  prism  and  as  many  as  twelve  nodes  may  be  counted  upon  the 
second  row  in  one  of  the  specimens.  The  increase  by  intercalation  is  not, 
however,  always  so  rapid,  as  another  example  shows  nine,  with  an  incipient 
tenth,  in  the  second  row.  The  nodes  also  multiply  in  another  manner,  namely 
by  the  transverse  division  of  a  single  node  into  two,  or  to  express  the  fact 
with  greater  accuracy,  the  appearance  of  two  nodes  one  above  the  other  in  the 
place  usually  occupied  by  one.  This  is  seen  on  the  specimens  figured,  Avhile 
another  example  not  here  illustrated  shows  the  duplication  of  all  the  nodes  of 
one  row,  thus  making  a  complete  double  row  of  nodes.  This  interesting 
mode  of  duplication  of  the  nodes  has  not  been  observed  in  any  American 
species,  but  seems  to  indicate  an  approximation  to  the  condition  attained  in 
the  genus  Botryodictya. 


Species  of  tiie  Psammttrs  dit  Condroz.  149 

The  reticulwm  is  very  fine,  as  in  //.  Barroini,  showing  none  of  the  coarner 
meshes  prevailing  in  the  Chemung  species  of  this  genus.  None  of  the  speci- 
mens are  so  preserved  as  to  sho\v  tufts  at  the  end  of  the  nodes. 

Dimensions.  The  most  complete  individual,  which  bears  three  horizontal 
ro\vs  of  nodes  and  a  considerable  portion  of  an  anodose  base,  has  a  length  of 
155  mm.  and  a  width  across  the  top  (exaggerated  by  compression)  of  100  mm. 

Locality.      In  the  Psammites  du  Condroz,  Jeumont,  Brittany. 

RlIABDOSISPONGIA   CoNDROZIANA,  Sp.  HOV. 
Plate  xlvii,  Fios.  1,  2. 

This  magnificient  species  shows  some  of  the  exterior  characters  of  the 
Chemung  form,  Dictyophyton  AmaUliea,  which  has  elsewhere  been  taken  as 
the  type  of  the  proposed  genus  Rhabdosisponoia.  These  are,  briefly,  the  suc- 
cessive annulations,  separated  by  deep  constrictions  and  crossed  by  vertical  or 
divergent  ridges. 

In  the  French  specimens  the  apical  portion  of  the  cup  is  regular  and  with- 
out annulations  for  a  considerable  distance,  the  surface  here  being  divided  into 
eight  prism-faces.  In  the  best  preserved  example  the  first  annulation  is,  on 
one  side,  an  enormous  bulging  of  the  cup,  which  covers  more  than  one-half 
the  length  of  the  specimen  and  makes  its  greatest  width.  On  the  other  side, 
this  large  swelling  is  divided  into  two,  by  a  broad  constriction  near  its  lower 
portion,  which  extends  rather  more  than  half-way  around  the  cup.  This  forms 
a  small  annulation  beneath  and  a  large  one  above.  The  constriction  above  this 
double  annulation  encircles  the  entire  cup,  but  the  succeeding  swelling  is  of 
less  size  and  is  apparently  of  equal  width  on  both  sides  of  the  sponge.  Above 
this  is  evidence  of  a  still  fainter  constriction  and  annulation,  making  in  all 
four  annulations  on  the  best  preserved  side.  There  are  three  specimens  of 
this  species  in  the  material  at  hand,  and  all  of  them  show  the  tendency  to 
develop  an  annulation  or  swelling  on  one  side  earlier  than  on  the  other.  To 
this  fact  are  due  the  slight  curvature  and  lateral  asymmetry  of  the  sponge. 
The  eight  vertical  ridges  near  the  base  of  the  cup  maintain  themselves  over 
the  entire  extent  of  the  sponge,  diverging  greatly  as  they  pass  over  the 
expanded  portions.  Their  number  is  increased  by  the  gradual  appearance  of 
new  ones  at  the  center  of  the  faces,  but  the  latter  are  never  so  strong  as  the 
primary  ridges. 

The  reticulum  is  fine-meshed  throughout,  as  in  the  other  species  from  this 
horizon. 


160  DlCTYOSPONGID^. 

Dimensions.  The  largest  and  best  preserved  of  the  three  specimens  had 
a  length  of  300  mm.,  which  must  be  considerably  less  than  its  original  length, 
as  parts  are  lacking  at  both  extremities.  Its  width  at  the  base  is  50  mm,, 
across  the  first  annulation  115  mm.,  and  at  the  top  100  mm. 

Ziocality.     In  the  Psammites  du  Condroz,  Jeumont,  Brittany. 


SF»ECIES    OK    THE    CARBONIFEROUS. 

SPECIES  OF  THE  WAVERLY  GROUP. 

TYLODICTYA,  gen.  nov. 

This  generic  form  is  as  yet  but  imperfectly  imderstood.  Some  recently 
discovered  sponges  from  Warren,  Pennsylvania,  present  the  api>earance  of 
erect,  reticulated  cups,  smooth  or  somewhat  irregularly  undulated  for  a  con- 
siderable part  of  their  length,  but  abruptly  breaking  out  into  one  or  possibly 
more  whorls  of  quite  unsymmetrical  and  irregular  simple  or  compound  nodes. 
Apparently  there  are  eight  nodes  in  each  whorl,  but  this  is  not  certain. 
The  nodes  are  j)endidous  when  large,  and  in  their  subdivision  have  foUoAved 
no  rule  or  order.  The  aspect  of  the  cylindrical  portions  of  the  sponge  is 
similar  to  that  of  the  forms  of  Calatiiosponqia  with  which  it  is  associated, 
being  fine-meshed  and  free  from  prismatic  faces. 

The  fi'agments  to  which  our  knowledge  of  this  genus  is  now  restricted, 
thoiigh  higlil\-  imperfect,  are  still  sufficient  to  distinguish  it  from  any  other. 

Type,  Tylodictya  Warrenensis,  sp.  nov. 

Tylodictya  Waerekensis,  sp.  nov. 
Erect,  subcylindrical  cups,  apparently  contracting  slightly  above  the  base ; 
surface   smooth   for  a   considerable    distance,   then    gently   expanding   and 
developing   a   horizontal   row  of   nodes.     In  the 
smaller  of   the  fragments  which   have  been  ob. 
served,  these    nodes   are  low,  simple,  somewhat 
elongate  vertically  and  divided  by  narrow  fiirrows 
which  reach  to  the  general  surface  of  the  sponge, 
displaying  no  tendency  to  subdivision  or  irregu- 
larity of  arrangement ;  in  the  larger  specimen  less 
of  the  inferior  surface  of  the  cup  is  retained,  but 
the  nodes  are  very  strongly  developed,  are  vertic- 
ally elongate,  and  were  apparently  somewhat  fan- 
shaped,    are   pendent   toward  their   rounded    ex- 
tremities and  separated  by  grooves   of    different 
depth,  so  that   each  pair  of   nodes  seems  to  be 
elevated  on  a  stout  base,  as  with  the  nodes  in  cer 
tain  species  of  HrDNOCERAS.     Upon  one  side  of 
this  specimen  the  nodes  are  much  less  regular  than 
on  the  other,  one  pair  having  appeared  below  the 
others.     This  pair  is  divided  very  unequally  by  a 
vertical  groove  and  again  horizontally,  so  that  the  effect  produced  is  some- 
what like  that  observed  among  the  nodes  of  Botryodiotya.     To  what  extent 

161 


Jjgb 


FiouKB  18.  Viilodietya  Warmeiut;  War 
ren,  Penus^lvanla.  A  youn«  fndlTldual  boar 
Ing  nodes  only  toward  the  nppor  part. 


152 


DlCTYOSPONGID^ 


this  condition  of  the  nodes  is  normal  and  in  how  far  the  sponge  is  affiliated 
with  the  last  named  genus  can  not  at  present  be  determined. 

The  reticulwm  is  very  fine-meshed,  bearing  no  conspicuous  major  divisions, 
and  resembles  that  of  Calatiiospongia. 

Of  the  two  specimens  referable  to  this  species,  one  has  a  length  of  75  ram., 
the  lower  portion  of  the  cup  to  the  base  of  the  nodes  measuring  50  mm.     At 


30 


Figures  19,  20.      Tuhtdictya  Warrenensis,  Waverly  sandstone,  Warren,  Pennsylvania. 
Figure  19  gives  a  side  view  of  the  type  specimen  In  which  thestronj;,  Irregular  nodes  are  much  foreshortened.    The  prominence 
of  these  nodes  Is  brought  out  in  flgure  20,  In  which  the  specimen  is  viewed  from  above. 

its  base  this  specimen  has  a  width  of  45  mm.,  and  its  diameter  at  mid-length 
is  35  mm.  The  other  fragment  is  65  mm.  in  length  and  is  broken  across  the 
expanded  nodiferous  portion,  measuring  in  diameter  to  the  exti-emities  of  the 
nodes,  90  mm. 

Locdlity.    In  the  Waverly  sandstone, Warren,  Pennsylv;mia.    (Collection  oi 
Prof.  C.  E.  Bekciij<:r.) 

TyLODICTTA  (?)  TEHTTIS,  Hall  (sp.). 
Platk  liii,  Figs.  4,  5. 

1882.      DwtyopTiyton   teniie,   Hall.      Notes  on    the  Family  Dictyospongida) ; 

Expl.  pi.  18,  fig.  5. 
1884.     Dictyophyton  temie,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State  Mus. 

Nat.  .Hist.,  p.  474,  pi.  18  (19),  fig.  5. 

The  only  known  specimen  of  this  species  is  a  small  fragment  of  a  cup 
bearing  t\vo  large,  compound  nodes  separated  ])y  a  deep  longitudinal  groove, 


Spkcxks  of  the  Wavkbly  Group. 


153 


and  eacli  divided  across  the  top  by  a  low  furrow.  Tlie  nodes  are  vertically 
elongated,  not  constricted  at  the  base,  and  the  lobes  are  low  and  obtuse. 

The  reticulum  is  composed  of  very  fine  spicular  bands  among  which  scarcely 
any  difference  in  size  is  apparent.  There  is  no  evidence  of  a  pnsmatic  division 
of  tlie  surface.  So  few  specific  characters  are  retained  by  this  single  frag. 
nient  that  it  is  referred  with  considerable  hesitation  to  the  genus  Tyixjdiotya  ; 
yet  its  nodose  surface  suggests  such  relationship  and  its  occurrence  in  the 
same  fauna  Avith  T.  Warrenensis  enforces  this  suggestion. 

Locality.     In  the  Waverly  sandstone,  at  Warren,  Pennsylvania. 


CLATIIROSPONGIA,  HalL 
(For  description  see  page  121,  part  1.) 

Clatiirospongia  abacus.  Hall. 

Plate  xlix,  Fiqs.  6-8. 

1882.     Clath'ospongia  ahacus,  Hall.     Notes  on  the  Family  Dictyospongidaa ; 

Expl.  pi.  18,  figs.  2-4. 
1884      Dictyopliyton  {Glatlii'ospongid)  ahacus,  Hall.     Thirty-fifth  Ann.  Kept. 

N.  Y.  State  Mus.  Nat.  Hist.,  p.  474,  pi.  18  (19),  figs.  2-4. 

Sponge  of  moderate  size,  regularly  turbinate  or  obeonical ;  actual  base 
not  preserved,  but  judging  from  the  slope  of  the  sides,  acute.  Expansion 
rather  rapid.     Aperture  unknown. 

Surface  bearing  coarse  jirimary  spicular  ridges,  which  enclose  quadrules 
measuring  on  the  average  10  ram.  in  width  and  9  mm.  in  length.  These  ai"e 
subdivided  into  four  squares  by  the  secondary  spicular  bands;  the  finer 
divisions  of  the  reticulum  may  also  be  preserved.  The  primary  bands  are 
extended  into  free  horizontal  and  vertical  expansions  measuring  about  8  mm. 
in  width  and  the  deep  fenestrations  thus  made  are  divided  into  smaller  areoles 
by  the  crossing  of  the  subordinate  reticulating  bands  in  each ;  this  inference 
is,  at  least,  to  be  made  from  the  appearance  of  finer  reticulating  lines  on  the 
surface  of  the  primary  expansions. 

Dimensions.  Length  of  the  original  specimen,  100  mm.-  greatest 
diameter  of  the  cup  (at  upper  extremity),  50  mm.;  median  diameter,  38  mm. 
without,  and  54  mm.  with  the  lamellar  expansions. 

Locality  In  the  sandstones  of  the  Waverly  group,  associated  with 
Ectenodictya  implexa,  and  an  undetermined  Spirifer  or  Syringothyris ; 
Warren,  Pennsylvania. 


154  DlCTYOSPONGID^ 

Clatiirospongia  caprodonta,  sp.  nov. 

Platk  I,,  Figs.  8,  9. 

Sponge  slender,  elongate  obconical,  gradually  enlarging  from  an  acnte 
}>ase  to  the  aperture,  the  rate  of  expansion  being  the  most  rapid  over  the 
earliest  one-third  of  the  length.  Surface  faintly  subprismatic,  with  low  nodea 
at  the  intersection  of  the  primary  spicular  bands. 

Upon  the  single  specimen  observed,  which  is  an  internal  cast,  the  reticulum 
shows  twelve  longitudinal  spicular  ridges  which,  with  the  intersecting  horizon- 
tal ridges  of  about  the  same  size,  form  large  primary  quadrules.  Over  the  body  of 
the  cup  these  measure  about  13  mm.  in  width  and  19  mm.  in  length,  making  a 
comparatively  large  qiiadriile  which  was  evidently  very  shai-ply  defined  over 
the  entire  cup,  except  at  the  base  and  about  the  aperture.  The  angles  of 
these  quadrules  bear  pronoiinced  nodes,  not  so  highly  developed  as  in  Hydno- 
ceras,  but  indicating  a  tufted  projection  of  the  spicules,  undoubtedly  accom- 
panied by  a  strong  lamellar  spicular  band  along  the  primary  vertical  and  hor- 
izontal ridges.  These  ridges  and  nodes  are  somewhat  obscured  by  the 
secondary  reticulation,  the  entire  surface  being  covered  with  meshes  measur- 
ing about  2  mm.  on  a  side,  the  intermediate  reticulation  being  virtually  lost. 

The  cup  has  an  entire  length  of  185  mm.  and  an  apertural  diameter  of 
63  mm.  Its  diameter  at  50  mm.  from  the  apex,  is  38  mm.,  and  at  125  mm. 
irom  the  base,  it  is  57  mm. 

Locality.  In  the  sandstone  of  the  Waverly  group,  at  Portsmouth,  Ohio. 
(Collection  of  the  School  of  Mines,  Columbia  University,  New  York.) 

THYSANODICTYA,  gen.  nov. 
(For  description  see  page  125.) 

Thysanodictya  expansa.  Hall  (sp.). 

Plate  liii,    Fig.  3. 

1882.     Phragmodictya  expansa,Yl.i\\\     Notes  on  the  Family  Dictyospongidse ; 

Expl.pl.  19,  fig.  10. 
1884.     Ectenodictya  expansa,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State  Mus. 

Nat.  Hist.,  p.  475,  pi.  (19)  20,  fig.  10. 

The  original  specimen  of  Ectenodictya  expansa  is  a  portion  of  a  very  large 
basal  disc  or  diaphragm,  representing  a  species  of  those  fonus  occurring 
abundantly  in  the  Chemung  sandstones  for  which  we  have  erected  the  genus 
TiiYSANODicTYA.     The  s^tccies  is  HomeAvhat  irregularly  nodose  over  the  surface 


Species  of  the  Waverly  Grottp.  155 

and  quite  strongly  so  at  the  margins,  the  elevations  there  being  elongated  and 
extended  for  some  distance  over  the  basal  surface  ;  jiresenting  an  asjject  simi- 
lar to  this  part  of  the  cup  in  Th.  Mwin-IIalU  and  Th.  rudis,  of  the  Chemung 
group. 

The  reticulum  is  irregular,  as  in  all  of  these  basal  disks  which  have  been 
subjected  to  compression,  but  at  certain  spots  a  strong  reticulation  is  retained, 
composed  of  major  quadrules  measuring  about  6  mm.  on  a  side,  subdivided  by 
two  or  three  series  of  subordinate  bands.  The  primary  ridges  are  elevated 
and  indicate  a  strong  and  coarse  net- work  over  the  lateral  walls  of  the  cup. 

The  diameter  of  this  disc  is  150  mm.,  a  size  which  is  not  attained  by 
specimens  of  any  other  species  of  the  genus. 

Locality.    In  the  sandstone  of  the  Waverly  gronp,  Warren,  Pennsylvania. 

CALATHOSPONGIA,  gen.  nov. 

1863.     Dictyophyton,   Hall.     Sixteenth    Ann.  Rept.  N.  Y.  State  Cab.  Nat. 

Hist.,  p.  88. 
1882.     Phragnwdictya  ? ,  Hall.     Notes  on  the  Family  Dictyospongidse ;  ExpL 

pi.  17,  fig.  9. 
1884.  .  ,Dictyophyton  f,'H.&\l.  Thirty-fifth  Ann.  Rept.  N.  Y.  State  Mus.  Nat. 
Hist.,  p.  474. 
Stout,  subcylindrical  cups,  vidth  truncated  bases,  without  basal  cone  or 
diaphragm  as  far  as  known,  but  probably  attached  by  the  basal  margins ; 
contracted  medially  and  more  or  less  expanded  at  the  aperture.  Surface 
without  nodes  or  other  ornament. 

Type,  Dictyophyton  Redjieldi,  Hall. 

Calatiiospongia  Redfieldi,  Hall  (sp.). 

Plate  xlviii,  Figs.  1,  2;  Plate  xlix,  Figs.  1-4. 

1863.     Dictyophyton  RedfieUi,  Hall.     Sixteenth  Ann.  Rept.  N.  Y.  State  Cab. 

Nat.  Hist.,  p.  88,  pi.  v,  fig.  1 ;  pi.  v  a,  fig.  1. 
1882.      Phragmodictya  f  Redpldi,  Hall.      Notes  on    the  Family  Dictyo- 

spongida; ;   Expl.  pi.  17,  fig.  9. 
1884.     Dictyophyton  f  JiedfieUi,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State 

Mus!  Nat.  Hist.,  p.  474,  pi.  (17)  18,  fig.  9. 
1889.     Dictxjophytum  RedfieUi,  Lesley.     Dictionary  of  Fossils,  p.  200. 

Sponge  large,  probably  subcircular  in  cross-section ;  base  bi-oad,  expanded 
at  the  margin ;  the  basal  exi)ansiou  continues  for  about  one-third  the  length  of 


156  DlOTYOSPONGID^. 

the  cup,  passing  upward  into  a  long,  shallow  constriction ;  the  apertural  part 
of  the  cup  is  again  gradually  expanded,  the  greatest  diameter  being  at  the 
aperture,  which  is  unconstricted.  The  normal  form  of  the  sponge  as  thus 
described,  is  best  represented  by  the  example  shown  upon  Plate  xlix, 
(figs.  2  and  3)  which  is  essentially  entire.  The  skeleton  of  this  species  was 
evidently  very  delicate  and  fragile,  as  larger  specimens  are  all  more  or  less 
broken  and  distorted.     Surface  quite  smooth. 

Meticulwm  composed  of  fine  and  minute  quadrules.  Very  faint  longitudi- 
nal ridges  are  visible  in  places,  especially  upon  one  side  of  the  largest  observed 
example,  but  these  can  be  discerned  only  over  the  apertural  part  of  the  cup. 
The  more  noticeable  spicular  bands  are  the  horizontal  ones  which  succeed 
each  other  at  pretty  regular  intervals  of  from  3-4  mm.  The  vertical  bands 
corresponding  with  these  are  rarely  well  preserved,  so  that  the  usual  aspect 
of  the  surface  is  a  succession  of  narrow  horizontal  bands.  The  quadrules 
formed  by  the  horizontal  bands  and  their  corresponding  vertical  bands  are 
subdivided  by  four  subordinate  series  of  spicules,  so  that  the  ultimate  division 
of  the  surface,  which  is  usually  sharply  retained,  is  very  minute.  Upon 
certain  portions  of  any  given  specimen,  the  coarser  reticulation  may  be  Avholly 
lost,  leaving  traces  of  only  the  finer  division  of  the  reticulum.  The  apertural 
margin  is  regular  and  entire. 

Dimensions.  A  small,  but  entire  and  slightly  compressed  example 
has  a  length  of  70  mm.;  it  measures  32  mm.  in  diameter  at  the  base ;  30  mm. 
just  above  the  base,  and  33  mm.  at  a  distance  of  10  mm.  above  the  base ;  at 
the  middle  of  the  cup  its  diameter  is  25  nam.,  and  at  the  aperture,  51  mm.  A 
large  example  has  been  somewhat  shortened  by  vertical  compression  which 
has  produced  an  abnormal  swelling  about  the  middle  of  the  cup.  Its  length 
is  183  mm.;  its  basal  diameter  about  96  mm.;  at  a  point  slightly  above  the 
middle,  72  mm.;  and  at  the  aperture,  135  mm.  The  original  specimen  has  a 
length  of  200  mm.;  a  basal  width  of  86  mm.;  a  median  width  of  38  mm., 
the  upper  part  of  the  cup,  which  is  somewhat  distorted,  being  105  mm.  in 
diameter. 

Localities.  The  specimen  upon  which  the  original  description  of  the 
species  was  based  was  collected  near  Harrisville,  Medina  county,  Ohio,  in  a 
yellowish  sandstone  of  the  Waverly  group,  by  the  late  W.  C.  Redfield.  It 
has  also  been  found  in  the  Cuyahoga  shale  of  the  Waverly  group,  at  Akron, 
and  at  Richfield,  Ohio,  and  also  in  the  Waverly  sandstone  on  Nelson's  farm 
near  Pleasantville,  Venango  county,  Pennsylvania. 


SptXIES   OF   THE    WaVEULY    (tROTTP.  157 

CALATIIOaPONGIA   CARCERALI8,    Sp.  nOV. 
Plate  u,  Fios.  2-4 ;  Plate  ui,  Fius.  2,  ;i ;  Platk  lx.  Figs.  1,  3. 

1863.     Bictyophyton  Newberryi,  Hall.      Sixteenth  Ann.  Rept.  N.  Y.  State 
Cab.  Nat.  Hist.,  pi.  iv,  fig.  3. 

Among  the  specimens  utilized  for  the  original  illustration  of  the  species 
DietyophyUm  Newherryi,  was  one  large  subcyliudrical  cup  which  was  doubt- 
fully regarded  as  representing  the  jiedicel  of  that  species.  A  re-examination 
of  this  specimen,  supplemented  by  other  material,  some  of  which  has  been  col- 
lected since  the  date  of  that  description,  shows  very  clearly  that  such  speci- 
mens represent  a  quite  distinct  form  of  sponge. 

Cup  elongate,  rather  stout,  probably  circular  in  cross-section,  though  all 
the  specimens  are  more  or  less  flattened.  Base  broad,  terminating  abruptly ; 
from  the  base  upward  the  body  contracts  slowly  and  then  very  gradually 
expands  to  the  aperture.  The  form  is  thus  somewhat  like  that  in  certain 
species  of  Thysanodictya  (e.  g.  Th.  poecUus)  but  the  truncated  base  affords 
no  evidence  of  a  diaphragm  or  any  other  method  of  closure. 

The  reticidum  is  characterized  ])y  strong  vertical  and  horizontal  primary 
ridges,  the  foi-mer  being  the  more  conspicuous  and  in  some  instances  so  highly 
developed  as  to  give  a  subprismatic  appearance  to  the  cup.  In  the  specimens 
from  the  Waverly  sandstones  these  vertical  ridges  are  twelve  in  number  and 
their  intersection  with  the  principal  horizontal  ridges  forms  quadrules  measur- 
ing about  8  mm.  in  width  and  1 2  mm.  in  height ;  the  development  of  the 
horizontal  spicidar  ridges  is,  however,  (piite  variable.  The  subordinate  reticu- 
lation is  sharply  developed.  The  apertural  margin  is  regular  and  the  primary 
reticulation  extends  to  it  without  material  loss  of  definition. 

Dimensions.  The  original  example  which  is  essentially  entire,  has  a 
length  of  145  mm. ;  a  width  at  the  base  of  33  mm. ;  at  40  mm.  above  the  base 
the  diameter  increases  to  46  mm. ;  decreases  above  to  43  mm. ;  and  expands 
to  the  aperture  which  is  65  mm.  in  width.  This  specimen  is  from  the  Wav- 
erly group.  An  entire  specimen  from  the  Keokuk  group  is  130  mm.  in  length ; 
48  mm.  in  diameter  at  the  base ;  75  mm.  at  the  aperture,  and  40  mm.  where 
narrowest. 

Localities.  In  the  shales  and  sandstones  of  the  "Waverly  group  at  Richfield 
Ohio ;  also  in  the  calcareous  shales  of  the  Keokuk  group  at  Indian  Creek, 
Indiana.     (The  latter  from  the  collection  of  A.  S.  Tiffany.) 


158  DictyospongiDjE. 

Calathospongia  Carlli,  sp.  nov. 

Plate  iii,  Figs.  4-7. 

Among  some  specimens  of  Dictyosponges  from  the  Waverly  group  of 
nortliern  Venango  county,  Pennsylvania,  kindly  loaned  for  study  by  Mr. 
JojiN  F.  Carll,  formerly  of  the  Second  Geological  Survey  of  that  state,  is  one 
rather  lai-ge  specimen  of  the  upper  portion  of  the  cup  which  in  the  aspect  of 
the  surface,  bears  no  little  similarity  to  G.  Redfiddl  but  it  has  a  proportion- 
ally much  broader  aperture  than  is  possessed  by  that  species.  The  specimen 
has  been  somewhat  compressed  obliquely,  but  apparently  without  serious  dis- 
tortion to  the  best  exposed  surface.  The  body  of  the  cup  is  moderately  nar- 
row, measuring  55  mm. ;  the  width  of  the  aperture  is  145  mm.,  but  this  is 
perhaps  somewhat  below  its  original  proportions  as  the  apertural  portion  of 
the  cup  has  been  slightly  enfolded  on  one  side. 

The  character  of  the  reticulum  is  essentially  the  same  as  that  of  C  Red- 
fieUU,  the  surface  being  crossed  transversely  by  numerous  strong  spicular 
bands,  while  the  vertical  bands  are  very  fine  and  all  trace  of  conspicuous  ver- 
tical ridges  is  wanted. 

Localities.  In  the  sandstones  of  the  Waverly  group.  "  Found  loose  on 
the  flats  of  Oil  Creek,  near  the  mouth  of  Pine  Creek.  The  stream  here  cuts 
down  about  15  feet  below  the  Berea  grit"  (Mr.  Carll's  letter)  ;  near  Pleasant- 
ville,  Venango  county,  Pennsylvania.  The  species  also  occurs  in  the  same 
rocks  at  Warren,  Pennsylvania,  and  some  of  the  specimens  from  there,  like 
that  shown  in  figure  21,  page  159,  indicate  the  probable  identity  of  G.  Carlli 
with  the  fragmentary  remains  described  as  Ecteuodictya  implexa. 

Calathospongia  Tiffanyi,  sp.  nov. 

Plate  li,  Figs.  5,  6. 

Sponge  subcylindrical  toward  the  base,  expanding  with  moderate  rapidity 
to  the  aperture,  producing  an  elongate,  subconical  vase-shape.  Surface  with 
obscure  traces  of  prismatic  faces  which  widen  upward.  Where  best  preserved 
these  faces  are  seen  to  be  marked  by  moderately  strong  spicular  ridges  crossed 
by  less  prominent  bands,  the  two  making  quadrules  measuring  approximately 
12x10  mm.  where  the  faces  are  least  expanded,  but  increasing  in  width, 
without  increase  of  length,  toward  the  aperture.  The  intersections  of  these 
primary  l)ands  are  obscurely  nodose.  The  subdivision  of  the  quadrules  to 
the  fifth  degree  is  very  clearly  seen  on  the  internal  cast.  The  surface  is 
otherwise  devoid  of  ornamentation. 


SpKOIKS    ok    TIIK    WaVKRLV    GkoUP, 


159 


ffiauBB  31.    Portion  of  a  large  ■peclmea  of  Calathoapongia  Carlli,  from  the  Warerly  taadstono  at  Warren.  Pennaylvaala. 


160  DlCTYORPONGID.E. 

The  tyi>e  specimen,  which  is  imperfect  toward  the  base,  has  somewhat 
the  aspect  of  G.  carceralls  and  evinces  a  probable  agreement  in  form  with  the 
other  species  of  the  genus  Calatiiospoxgia.  Its  condition  of  preservation  is 
not  such  as  to  show  with  certainty  specific  differences  from  C.  carceralls  except 
in  its  stouter  form  and  more  rapid  apertural  exjiansion 

Dimensions.  Length  (incomplete)  100  mm.;  apertui'al  diameter  (slightly 
flattened)  84  mm.;  diameter  at  lower  extremity,  39  mm. 

Locality.     Waverly  group,  Ohio.     (Loaned  by  A.  S.  Tiffah^y.) 

CALATnOSPONGIA  ?  SACCTJLUS,  Hall  (sp.). 
Plate  l,  Fig.  7. 

1863.     DictijopJiyton  Eedfieldi,  Hall.     Sixteenth  Ann.  Eept.  N.  Y.  State  Cab. 

Nat.  Hist.,  pi.  iv.,  fig.  6. 
1884.     Dictijophyton  sacGulum,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.   State 

Mus.  Nat.  Hist.,  p.  473. 

The  specimen  upon  which  this  species  (?)  was  founded  is  a  small,  short, 
apparently  subcylindrical  cup,  with  a  smooth  surface  and  very  fine  retic- 
iilum.  One  extremity  is  i-ather  irregular  and  apparently  incomplete,  while 
the  other  seems  to  be  closed  or  enfolded.  There  is  very  slight,  if  any,  increase 
in  diameter  from  one  end  to  the  other.  It  seems  probable  that  the  specimen 
is  incomplete  and  affords  no  precise  conception  of  its  original  form.  The 
reticulation  is  somewhat  similar  to  that  of  Calalliospoiuj ia  HeJfieldi,  but  lacks 
the  strong  horizontal  bands  of  that  sj)ecies.  Its  recognition  as  a  species  and 
its  reference  to  this  genus  are  only  provisional. 

The  length  of  the  specimen  is  33  mm. ;  its  width  about  18  mm. 

Locality.  In  the  shaly  sandstone  of  the  Waverly  group  at  Richfield, 
Ohio. 

TIIAMNODICTYA,  Hall. 

1863.     DictyopJiyton,  Hall.     Sixteenth  Ann.  llept.  N.    Y.    State    Cab.    Nat. 

Hist.,  p.  87. 
1882.     Phraginodictya,  Hall.     Note  on  the  Family  Dictyospongidaj ;   Expl. 

pi.  17,  figs.  10,  11. 
1884.     TJmmnodictya,  Hall.     Thir-ty-fifth  Ann.  Rept.  N.  Y.  State  Mus.  Nat. 

Hist.,  pp.  466,  477. 

Dictyosponges  with  a  narrow,  subcylindrical,  tubular  stem  below, 
abruptly  Avidening  above  into  a  broad  funnel-shaped  circular  cup.  Surface 
with  prominent  spicular  ridges,  but  without  riodes  or  well  defined  prism-faces. 


Species  of  the  Waverly  Gropp.  Ifil 

Type,  Thamnodictya  Newhen-yi,  Hall. 

In  the  observations  made  upon  the  genus  Hydnocekas,  reasons  liave  been 
given  for  discontinuing  the  generic  term  Dictyopiiyton.  It  is  not  necessary 
to  recount  these  further  than  to  recall  that  the  tenn  was  introduced,  not 
primarily  to  replace  Conrad's  name  Hydnoceras,  but  as  a  designation  for 
other,  and  anodose  species,  the  first  among  the  specified  types  being 
BicUjophjton  Newherryi,  the  form  subsecpiently  adopted  as  the  type  of  Tham- 
nodictya. Dictyopiiyton  has  proved  to  be  a  misleading  term  among  the 
sponges,  and  its  place  is  better  filled  by  the  various  generic  terms  which  the 
recjuirements  of  our  present  knowledge  have  originated. 

Thamnodictya  Newbeuuyi,  Hall. 

Platk  l,  Figs.  1-6 ;  Plate  lii,  Fia.  1. 

1863.     Dictyophytmi Newberry i,  Hall,     Sixteenth  Ann.  Kept.  N.  Y.  State  Cab. 

Nat.  Hist.,  p.  87,  pi.  iv,  figs.  1,  2,  4,  (not  fig.  3). 
1882.     Phragiiwdictya  Netoherryi,  Hall.  Notes  on  the  Family  DictyospongidsB  ; 

Expl.  pi.  17,  figs.  10,  11. 
1884.     Thamnodictya  Newherryi,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State 

Mus.  Nat.,  Hist.,  p.  477  (partim)  jjl.  (17)  18,  figs.  10,  11. 

Sponge  of  moderate  size,  attenuate  and  subcylindrical  toward  the  base, 
rapidly  expanding  above,  forming  a  broad  vase  supported  upon  a  long  pedicel.' 
Cross-section  at  any  point  of  the  vase  circular ;  the  pedicel  however  ajjpears 
to  have  been  obscurely  prismatic. 

Heticidum,  composed  of  strong  rectangularly  intersecting  bands,  the 
primary  series  being  broad  and  forming  trapezoidal  quadrules,  averaging, 
over  the  main  portion  of  the  vase,  about  6  mm.  in  diameter  and  9  mm.  in 
height.  The  vertical  strands  diverge  very  rapidly  outward  with  the  growth 
of  the  cup,  losing  their  conspicuous  size  toward  the  aperture.  The  subordi- 
nate net-work  is  sharp  and  fine,  especially  in  the  apertural  region  where  the 
prevalence  of  the  finer  strands  obliterates  the  coarser  meshes.  The  pedicel, 
which  is  not  well  preserved  in  any  of  the  specimens  studied,  appears  to  have 
borne  coarse  vertical  ridges  toward  its  upper  part. 

Dimensions.     The  original  example  of  this  species  is  the  most  complete 

in  the  collections  examined.     Its  length  from  the  lower  end  of  the  pedicel, 

which  appears  to  be  nearly  complete,  to  the  aperture,  is  115  mm. ;  the  median 

width  of  the  pedicel,  7  mm. ;  the  diameter  of  the  aperture,  90  mm.     Portions  of 

2?^ 


102  DlCTYOSPONGIDJE, 

other  specimens  of  about  the  same  size  have  been  observed,  but  the  majority  oi 
these  are  of  small  size. 

Under  this  specific  name  we  have  included  only  the  slender  fonns  pos- 
sessing the  expanded  aperture.  In  previous  descriptions  some  much  larger 
subcylindrical  b(,)dies  have  been  regarded  as  pedicels  of  the  same  species,  but 
theu-  size  and  general  aspect,  supplemented  by  some  additional  structural 
details,  indicate  that  these  latter  forms  are  quite  distinct  from  typical  examples 

of  TlIAMNODICTYA. 

Localities.  In  the  beds  of  the  Waverly  group.  The  original  example,  that 
shown  on  Plate  1,  fig.  1,  is  from  a  shaly  limestone  filled  with  Feneatella  and 
Productus,  at  Richfield,  Ohio.  Other  specimens  are  from  sandy  shales  and 
limestones  at  Cuyahoga  Falls,  Ohio. 

Thamnobictya  Ortoiti,  sp.  nov. 

Plate  uii,  Figs.  1,  2. 

This  species  is  represented  by  an  internal  cast,  in  a  compact  ferruginous 
sandstone,  of  the  upper  or  vase-shaped  portion  of  a  Thamnodictya.  It  is  a  large, 
some^vhat  compressed  individual  preserving  a  considerable  portion  of  the 
aperture.  At  its  lo\ver  point,  which  represents  the  opening  of  the  pedicel,  the 
surface  shows  several  somewhat  unequal  clusters  of  the  longitudinal  spicules 
Avhich  have  been  changed  to  limonite.  The  impression  of  the  reticulum  is  fine 
and  rather  irregular.  There  are  no  predominant  vertical  and  horizontal 
spicular  ridges  as  in  Thaimwdictya  Newherryi,  but  the  entire  surface  is  cov- 
ered by  small  quadrules  about  2  mm.  square  and  these  are  again  divided  by 
minute  subordinate  bands.  The  coiirse  of  the  vertical  bands  is  not  radial  from 
the  base,  but  they  appear  to  have  made  a  broad  simple  curve  in  extending 
toward  the  aperture,  a  feature  which  may  be  exaggerated  by  the  compression 
of  the  specimen.  Along  the  apertural  margin  the  net- work  is  much  finer  than 
elsewhere. 

The  specimen  measures  from  base  to  aperture,  on  one  side,  80  mm.  and  on 
the  other,  97  mm..  Its  greatest  diameter  is  from  one  edge  of  the  folded 
apertural  margin  to  the  other ;  each  of  these  lateral  extremities  is  somewhat 
broken  but  the  full  measurement  was  about  150  mm.  The  lower  sides  of  the 
cup  slope  at  an  angle  of  about  60  degrees. 

Locality.  In  the  Cuyahoga  shale  of  the  Waverly  group,  Moot's  run, 
Licking  county,  Ohio.     (Named  for  Dr.  Edwakd  Okton,  of  Columbus,  Ohio.) 


Species  of  the  Wavekly  Group.  16.3 

CLEODICTYA,  Ila^ 

1884.     Ckodi^tya,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State  Muh.  Nat.  Hist 
pp.  467,  479.  ■' 

Vase-shaped  Dictyosponges,  broadly  expanded  near  the  base  into  a  sin.de 
horizontal  row  of  strong  low  nodes,  thence  gradually  contracted  and  again 
widening  to  a  broad  aperture.  Surface  without  evidence  of  prismatic  faces 
or  projecting  spicular  lamellae. 

Type,  CleodUitya  gloriom.  Hall.     Keokuk  group. 

Cleodictta  Claypolei,  sp,  nov. 

Plate  li,  Fig.  1. 

Sponge   comparatively    small,    subturbinate,    expanding   rather   rapidly 
from  the  basal  extremity  into  a  single  horizontal  row  of  nodes  which  begin  a 
short  distance  above  the  base  and  extend  through  fully  one-third  the  length 
of  the  cup.     These  nodes,  though  some  are  lost  on  one  side  of  the  sponge, 
appear  to  have  been  eight  in  number,  and  were  low  and  elongate,  their  length 
measuruig  twice  their  width;  they  are  separated  by  narrow  furrows  which  do 
not  extend  down  to  the  unswollen  surface  of  the  sponge.     Above  these  nodes 
the  surface  is  abruptly  constricted  and  thence  widens  very  gradually  upward, 
the  apertural  margin  not  being  retained.     The  proportions  of  this  species  are 
different  from  those  of  C.  ghriosa,  the  form  being  more  slender,  the  expan- 
sion more  gradual  and  the  basal  nodes  much  more  elongate. 

The  reticulum  is  very  fine  and  the  primary  divisions  somewhat  obscured. 
A  series  of  transverse  bands  3-4  mm.  apart,  crosses  the  cup,  without  vertical 
bands  of  corresponding  strength.  This  gives  to  the  reticulum  the  aspect  of 
that  in  CaUtlMspongia  RedjkUi  and  Q.  CarlU.  Minor  subdivisions  to  the  fifth 
series  are  discernible. 

Dimensions.  The  specimen  described  is  somewhat  imperfect  at  both 
extremities;  it  does  not,  however,  seem  probable  that  much  has  been  lost  from 
either.  The  entire  lengtt  of  the  portion  retained  is  102  mm.  Its  basal 
extremity  has  a  diameter  of  25  mm. ;  across  the  row  of  nodes  the  diameter  is 
70  mm. ;  just  above  these  nodes  it  is  45  mm.  and  at  the  upper  extremity, 
50  mm. 

Loaditi/.     From  the  sandstones  of  the  Waverly  group  at  Akron,  Ohio. 
(Received  from  Prof.  E.  W  Claypole.) 


1  •>4  DiCTYOSPONGIDiE. 

(?)  ECTENODICTYA,  Hall. 

1884.     EctenodicUja,  Hall  (partiin).     Thirty-fifth  Ann.  Rept.  N.  Y.  State  Mus. 
Nat.  Hist.,  p.  46  fi. 

The  name  Ectenodictta  was  introduced  for  certain  forms  of  apparently 
more  or  less  irregular  growth,  which  seemed  to  have  expanding  or  unenclosed 
fronds.  Both  in  the  upper  Devonian  and  the  lower  Carboniferous  faunas 
sponges  of  such  aspect  have  been  found,  with  a  surface  usually  free  of  nodes 
or  other  ornamental  characters.  The  absence  of  a  well  defined  form  in  these 
species  was  the  principal  reason  for  placing  them  together  under  a  single  generic 
tenn,  although  the  division  could,  by  the  very  nature  of  its  composition,  have 
little  more  than  a  temporary  value.  In  1884,  two  species,  E.  implexa  and  E. 
expansa  wei"e  described  from  the  Waverly  sandstone,  E.  Burllngtonensis  from 
the  Burlington  group  and  E.  eccentrica  from  the  Keokuk  group.  Fossils  of 
like  character  are  also  abundantly  known  in  the  Chemung  group  but  no 
names  have  been  aj^plied  to  them.  The  accession  of  material  has  shown  that 
such  Chenuing  specimens  indicate,  by  one  or  another  set  of  characters,  relations 
to  some  of  the  larger  sponges  of  the  group,  many  of  these  imperfect  fronds 
probably  i-epresenting  the  species  Prismodlctya  clwanea,  I).  Ahnondensis, 
or  some  similar  sponge  in  which  the  expanse  of  surface  is  large  and  the  pris- 
matic aspect  obscured.  Thus,  also,  with  most  of  the  described  species  of 
EoTENODicTYA ;  E.  BurlinAjtanensis  seems  a  probable  representative  of  the 
genus  Lykodictta  ;  E.  expansa  undoubtedly  represents  a  species  of  Thysano- 
DiCTYA  and  E.  eccentrica*,  as  now  known,  is  the  basal  diaphragm  of  a 
sponge  like  Pubagmodictya,  but  having  a  structure  necessitating  its 
removal  to  another  and  new  genus,  AcL(EODtcTYA.  There  remains,  then,  but 
the  type-species  of  Ectenobiotya,  E.  implexa,  whose  apparent  structure  is 
here  described, 

Ectenodictya  ntfPLEXA,  Hall. 

Pl-ATK  LIV,  FlOS.  3,  4. 

1884.     Ecteiiodictya  implexa,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State  Mus. 
Nat.  HiHt.,  p.  475,  pi.  (18)  19,  fig.  1. 

"  Frond  a  reticulate  expansion,  assuming  a  variety  of  form  from  pressure  or 
other  causes ;  the  original  form  has  been  apparently  broadly  funnel-shaped  or 
ovoid.     Base  imknown. 

*  Tbose  speolas  wer*  first  d«sorlb«d  as  Pbk  AaHODicTT&. 


Spectks  of  the  Biiulinoton  Grottp.  165 

*'  Surface  cancellate  by  strong  concentric  and  vertical  striae ;  the  inter- 
mediate spaces  finely  reticulate  by  filiform  striae,  which  cross  each  other 
rectangularly.  The  body  presents  prominences  or  protuberances,  which  are 
not  sufficiently  elevated  to  be  termed  nodes. 

"  A  large  imperfect  specimen,  which  is  distorted  by  compression,  has  a 
length  of  180  mm.  and  a  width  of  about  93  mm. 

"  The  specimens  of  this  species  are  all  more  or  less  distorted ;  they  appear 
as  broad  funnel-shaped  expansions  without  evidence  of  a  distinct  tubular  base, 
and  are  usually  fragmentary  ". 

The  foregoing  is  the  original  description  of  these  fossils,  founded  upon 
very  imperfect  material.  Upon  re-examination  of  these  specimens  and  compari- 
son with  the  more  completely  known  species  from  the  Waverly  group  of 
Pennsylvania  we  are  disposed  to  conclude  that  all  are  fragments  of  CalatJio- 
spongia  Medfieldi  or  its  close  ally,  O.  Oarlli.  It  has  already  been  observed  that 
the  cups  of  these  species  were  very  delicate  and  fragile,  easily  liable  to  dis- 
tortion and  fracture ;  and  there  are  none  of  the  specimens  upon  which  the 
description  of  Ectenodictya  implexa  was  based  which  do  not  show  the  charac- 
teristic expression  of  the  reticulum  produced  by  the  predominance  of  the  hori- 
zontal spicular  bands.  The  specimen  from  which  the  original  illustration  wae 
drawn,  is  the  apertural  portion  of  a  large  cup  somewhat  infolded  at  the  mar- 
gin. It  was  incorrectly  oriented  in  the  drawing ;  the  right  margin  represents 
the  apertural  edge. 

Localities.  The  specimens  have  been  found  in  the  Waverly  group  at 
Warren,  Pennsylvania,  in  association  with  Syringotliyris  RandalU,  and  in  the 
same  formation  at  Oil  City,  in  that  state. 

SPECIES  OF  THE  BURLINGTON  GROUP. 

LYKODICTYA,  Hall. 
(For  description  see  page  170.) 

LyEODIOTTA  (?)  BURLINOTONENSIS,  Hall  (sp.). 
Plate  liii,  Via.  6. 

1884.     Ectenodictya  BurlingUynensis,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y, 
State  Mus.  Nat.  Hist.,  p.  476. 

The  single  specimen  representing  the  only  species  of  Dictyosponge  yet 
found  in  this  formation,  consists  of  a  quite  imperfect  outer  impression  of  a  cyath- 
iform  frond,  having  a  smooth  exterior  and  showing   a  strong  development 


166  DlOTYOSPONGID^. 

of  the  vertical  spicular  Inuidles  toward  the  base  and  extending  upward  in  one 
or  two  broad  and  irregular  strands.  The  reticulum  is  fine-meshed  and  over 
most  of  the  surface  the  prevailing  quadrille  has  a  diameter  of  about  3  mm., 
being  frecpieutly  subdivided  by  subordinate  series  of  spicules.  The  general 
aspect  of  the  specimen  is  like  that  of  Lybodictya,  and  the  species  is  therefore 
provisionally  referred  to  this  genus. 

Dimensions.     Length  of  the  fragment  135  mm.,  greatest  width  130  mm. 

Locality.  In  the  yellow  sandstone  below  the  Burlington  limestone,  Bur- 
lington, Iowa. 

SPECIES  OF  THE  KEOKUK  GROUP. 

DICTYOSPONGIA,  gen.  nov. 
(For  description  see  page  72,  part  1.) 

DiCTYOSPONGIA  CYLINDRIOA,  Whitfield  (sp.). 
Plate  lv,  Fig.  3;  Plate  lxi,  Fig.  6. 

1881.     Dictyophyton  cyliiulriciim,  Whitfield.     Bull.  No.  1,  Amer.  Mus.  Nat. 

Hist.,  p.  19,  pi.  iv,  fig,  3. 
1884.     Dictyophyton  cylindrimim,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  Stat© 

Mus.  Nat.  Hist.,  p.  475. 

The  original  specimen  of  tliis  species  is  a  fragment  of  a  flattened  sub- 
cylindiical  or  elongate  obconical  tube,  exposing,  for  the  most  part,  the  inner 
portions  of  the  reticulum  but  also  showing  that  the  exterior  of  the  sponge 

was  devoid  of  ornamental  features.  The  ex- 
posed surface  exhibits  fine  smooth  horizontal 
and  vertical  spicular  rods  disposed  at  numerous 
and  someAvhat  unequal  intervals  in  small  bundles. 
The  outer  or  dei-mal  quadrules  are  formed  by 
cruciform  spicules,  and  measure  about  .5  mm.  on 
„      .«,„..     .  ^  each   side.      Microscopic   examinations    of   the 

FlODUl  33.    SplculM  ot  Dietyotpongia  oyUn.  -C 

<ir«>o.x«oo.  (J.M.C.)  reticulum  reveal  among    the  parenchymalla  or 

dermalia,  umbels  of  a  form  somewhat  different  from  those  in  Cleodictya, 
curved  diactins  and  fragments  of  minute  echinate  hexactins. 

This  species  seems  to  be  of  rare  occurrence ;  besides  the  original  speci- 
men, there  are,  in  the  material  in  hand,  only  one  or  two  small  fragments,  so  that 
the  species  is  still  quite  imperfectly  known.  The  typical  example  has  a  length 
of  60  mm.  and  a  width  of  56  mm. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 


Spkcies  of  the  Keokuk  Group.  167 

DiCTTOSPONQIA  (?)  STTLmA,  Sp.  nOV. 
Platk  lvi,  Fia.  2. 

Among  the  material  from  Crawfordsville  is  a  slender  elongate  fossil 
whose  sponge  nature  is  strongly  suggested  by  a  comparison  with  the  graceful 
Chemung  species,  Dictyospongia  lophura.  It  is  therefore  noticed  here, 
although  its  precise  relations  may  still  be  somewhat  obscure. 

The  fossil  is  narrow  and  subcylindrical,  expanding  from  the  basal  point 
to  the  full  diameter  of  the  cup  in  about  one-eighth  the  length  of  the  specimen. 
The  surface  is  smooth  and  covered  with  fine,  closely"  set  vertical  lines.  The 
evidence  of  transverse  lines  is  not  very  satisfactory;  there  is,  however,  a 
series  of  comparatively  broad  and  low  transverse  depressions  which  are 
plainly  developed  over  the  upper  part  of  the  specimen,  and  these  contract 
toward  the  upper  extremity,  producing,  with  the  vertical  lines,  the  effect  of 
an  elongate  rectangular  reticulation.  Below  this  extremity  the  transverse 
depressions  produce  a  somewhat  undulating  surface. 

The  specimen  has  a  length  of  91  mm.  and  a  diameter  of  about  6  mm.  for 
seven-eighths  of  its  length.  There  is  no  evidence  of  a  spicular  tuft  at  the 
basal  extremity. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 

DiCTTOSPONGIA    (MaSTODICTYa)    080ULATA,  Sp.   nOV, 
PuLTK  LTi,  Fig.  6. 

The  figure  cited  represents  one  aspect  of  a  sponge  of  whose  complete 
form  it  is  not  possible  to  obtain  an  accurate  conception  from  the  material  in 
hand.  If  the  peculiar  shape  presented  by  it  is  normal  or  essentially  so,  as  it 
seems  to  be,  then  this  species  represents  a  distinct  type  of  generic  structure 
which,  if  established  by  future  investigations,  may  be  known  as  Mastodiotya  ; 
but  it  is  still  possible  that  the  shape  of  the  sponge  is  more  or  less  due  to  com- 
pression or  other  casual  cause.  Hence  the  characters  of  the  fossil  are  here 
described  from  the  single  example  known,  while  its  generic  and  specific  values 
are  left  contingent  upon  the  discovery  of  additional  material. 

The  lower,  broken  extremity  of  the  specimen  has  a  width  of  20  mm.  and 
thence  it  gradually  widens  upward  into  the  bulbous  swellings.  Here  the 
diameter  of  the  cup  has  increased  to  24  mm.  The  swelling  at  the  right  con- 
tracts quite  abruptly  and  apparently  terminates  in  an  osculum  or  excurrent 
orifice  at  7  mm.  above  its  gi-eatest  diameter ;  the  portion  on  the  left  however 
is  much  more  produced,  contracts  more  gradually,  extends  to  a  distance  of 
23  mm.  above  the  swelling  and  at  the  ai^erture  has  a  diameter  of  4  mm. 


168 


DlCTTOSPONGrD.(E, 


The  general  surface  is  smooth,  gracefully  expanding  to  the  bulbous  swell- 
ings, thence  sloping  with  gentle  concavity  to  the  oscula. 

The  reticulum  is  fine-meshed.  Since  only  the  outer  portion  is  exposed, 
the  vertical  spicular  bundles  are  not  \dsible,  but  a  fragment  of  a  very  large  rod 
is  seen  in  the  accompanying  figure.     At  the  angles  of  the  prevailing  quad- 


FiaoBE  23.    Spicules  of  Uattodictya  onmlata,  Z400     (J.  U.  C.) 


rules,  which  measure  about  1.5  mm.  on  a  side,  there  are  very  strong  pentactins 
or  hexactins  with  modified  outer  arm,  and  lateral  arms  extending  for  nearly 
the  entire  length  of  the  quadrule.  An  extremital  fragment  of  one  of  these 
is  represented  in  the  above  figure  which  also  shows  a  many-toothed  umbel 
and  a  siliquiform  diactin. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 


PRISMODICTYA,  gen.  nov. 

(For  description  see  page  79,  part  1.) 

Prismodictya  polyhedra,  sp.  nov. 

Plate  lv,  Figs.  1,  2. 

Sponge  large,  subcylindrical,  obscurely  polyhedral,  slightly  expanding 
toward  the  top.  Surface  regular  and  uniformly  free  of  asperities,  nodes  or 
protuberances  except  such  as  are  produced  by  the  slight  outward  extension 
of  the  principal  spicular  bundles.  In  the  best  preserved  of  the  compressed 
specimens  each  .side  bears  five  broad  vertical  faces,  and  at  the  edges  are  traces 
of  two  more,  twelve  in  all.  These  prism-faces  make  low  angles  with  one 
another,  becoming  obsolete  at  the  even  and  regular  margin  of  the  osculum. 


Spkcies  of  the  Keokuk  Gkoup.  169 

Reticulum.  The  quadrate  meshes  are  remarkably  uniform  in  size  and 
arrangement.  Assuming  that  the  vertical  spicular  bands  of  the  first  order 
lie  along  the  angles  made  by  adjoining  prismatic  faces,  a  given  square  of  the 
first  order  measures  about  20  mm.  on  a  side,  varying  with  the  slight  curvature 
of  the  surface,  and  with  the  upward  expansion  of  the  prism-faces.  The  sub- 
division of  these  quadrules  is  carried  out  with  regularity  to  the  fifth  degree, 
and  even  in  some  of  these  pentameres  there  is  evidence  of  a  division  into  ulti- 
mate quadrules.  Some  of  the  specimens  indicate  that  the  surface,  in 
its  original  condition,  was  fenestrated  by  the  slight  projection  of  the 
principal  lamellae. 

Dimensions.  The  apertural  width  of  the  specimen  figured  is  120  mm. ; 
its  diameter  where  narrowest  is  aboiit  90  mm.  and  the  incomplete  cup  has  a 
length  of  115  mm.  which  was  probably  somewhat  more  than  one-half  its 
entire  length. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 

LEBEDICTYA,  gen.  nov. 

Large  obconical  cups,  perhaps  somewhat  unsymmetrical  by  reason  of  more 
rapid  growth  on  one  side,  probably  expanding  from  a  subacute  base.  Surface 
obscurely  prismatic  and  cancellated  by  short  projecting  spicular  lamellae. 
Aperture  crowned  by  an  erect  fringe  of  long  marginal  spicules. 

Type,  Lebedictya  crinita,  sp.  nov. 

Lebedictya  cbinita,  sp.  nov. 

Plate  lviii,  Fioa.  1,  3 ;  Plats  lls,  Fios.  1,  2 ;  Platb  lxi,  Fio.  5. 

Sponge  cyathiform,  expanding  with  apparently  slight  asymmetry  from  a 
narrow  base.  This  asymmetrical  growth  is  especially  evident  in  one  specimen 
which  not  only  shows  a  notable  difference  in  the  size  of  the  two  sides,  but 
along  the  apertural  margin,  which  is  distinctly  retained,  indicates  a  rapid 
multiplication  of  the  spicular  net- work  on  the  longer  side  of  the  cup.  The 
surface  is  rendered  obscurely  prismatic  by  the  predominance  of  certain  ver- 
tical skeletal  ridges  which,  near  the  aperture,  lie  about  25  mm.  apart.  The 
horizontal  ridges  do  not  attain  so  great  size,  but  the  prevailing  quadrules  are 
nevertheless  large,  measuring  from  12  to  15  mm.  on  a  side,  and  are  variously 
subdivided.  All  these  vertical  and  horizontal  spicular  bands  are  erect  and 
form  a  moderately  deep  surface  reticulation.  There  may  also  have  been 
minute  tufts  at  the  intersections  but  the  evidence  of  them  is  not  very  clear. 
The  fenestration  of  the  exterior  of  the   cup  is   similar  to,  but   much  less 


170 


DiOITOSPONGIDjE. 


decided  than  that  in  the  genus  CiiATHKospoNoiA,  and  upon  internal  casts  is 

much  obscured  or  even  lost. 

About  the  aperture  the  vertical  spicules  are  extended  into  a  long  and 

narrow  marginal  tuft,  apparently  a  single  row  of  coarse  and  fine  rods  {ina/r- 

ginalia),  such  as  occurs  in  the  living  species  Bathydorus  Jhnbriatus* 

In  figure  24  are 
shown  some  of  the  skele- 
tal elements  of  this 
sponge,  the  dermalia 
being  represented  by 
umbels,  micrumbels,  tri. 
pinulus  and  echinate 
pentactins,  and  the  par- 
enchymalia  by  smooth 
„.   „  .  ,     .,  ^  ^, ,     _,  ,.     ,^  siliquiform  diactins. 

FiQUBE  24.    Sploulw  Of  LebtiMya  crinita,  s400.  -1- 

The  figure  at  the  bottom  Is  a  fragment  of   one  of  the  main  rods,  partly  decorticated  j  JJllftlBTlSiOTlS.  1  llC 

among  the  others  are  umbels  and  micrumbels  in  various  attitudes,  smooth  diactins,  an 

Incomplete  tripinuius,  etc.  (J.  M.  c.)  best     prescrved     speci- 

men, which  is  incomplete  at  the  lower  end,  has  a  length  from  aperture 
downward  of  155  mm.  Its  lower  diameter  is  75  mm.;  its  apertural  diameter 
in  its  compressed  condition  150  mm.  The  marginal  fringe  in  some  places  has 
a  length  of  30  mm.  A  somewhat  larger  specimen  which  has  been  com 
pressed  almost  vertically  has  an  apertural  diameter  of  180  mm. 

Locality.  In  the  shales  of  the  Keokuk  group  at  Crawfordsville,  Indiana. 
(Collection  of  A.  S.  Tiffany.) 

LYRODICTYA,  Hall. 

Cyathiform  Dictyosponges  with  regularly  expanding,  generally  smooth 
exterior,  fine  net-work,  low,  erect  tufts  at  ^vide  intervals  and  very  broad, 
thick  vertical  bundles  of  rods  and  clemes,  with  no  horizontal  bundles  of  cor. 
responding  size. 

Type,  Lyrodictya  Romingeri,  Hall. 

Lyeodictya  Romingeri,  Hall. 

Plxtb  lvi,  Fiq.  1. 
1884.     I/yrodictya  Romingeri,  Hall.     Thirty-fifth   Ann.  Kept.   N.  Y.   State 
Mus.  Nat.  Hist.,  p.  476. 
Sponge  broadly  expanding ;  its  form  imperfectly  known.     Surface  proba- 
bly covered  originally  with  low  and  erect  lamellae ;  ridges  with  short  tufts  at 


•  See  ScHULZE,  IlexactlnelUda,  pi.  IvllI,  flg.  1. 


Species  of  tiik  Kkokuk  Gkoijp. 


171 


the  angles  of  some  of  the  quadrules ;  but  in  its  general  aspect  smooth,  without 
prism-faces,  nodes,  or  other  irregularities. 

Reticulum  characterized  by  very  broad  and  strong  vertical  bundles  or 
lateraUa.  On  the  exposed  surface  of  the  specimen  there  are  seven  of  these 
at  wide  but  unequal  intervals. 
These  bundles  reach  almost  or 
quite  to  the  lower  extremity  of 
the  specimen,  spreading  gradually 
outward.  An  interesting  struc- 
tural feature  is  the  intercalation 
of  other  bundles  in  two  of  the 
interspaces  at  a  considerable  dis- 
tance above  the  base,  and  both 
these  and  the  primary  bundles 
increase  in  width  upward.  All 
the  bundles  are  composed  of  a 
large  number  of  stout,  cylindrical, 
continuous  rods,  varying  some- 
what in  size,  and  with  them  are 
numerous  clemes  of  so  great  size 
as  to  be  visible  to  the  naked  eye. 
The  latter  appear  to  be  restricted 
to  the  middle  and  upper  extent  of 
the  bundles,  no  trace  of  them  hav- 
ing been  found  on  the  lower  part 
of  the  specimen,  and  on  some  of 
the  intercalary  bundles  they  are 
not  to  be  seen  at  all.  These 
clemes  are  similar  to  those  occur- 
ring in  Physospongia  Dawsoni 
and  P.  Colleti,  though  very  much 
larger.  The  direction  of  their 
teeth  is,  however,  directly  the 
reverse  of  that  in  all  other  known 
instances.  This  peculiar  fact  at 
first  suggested  the  possibility  of 
an  error  in  the  orientation  of  some 
of  the  specimens  but  repeated  re- 

,  FIOURX  25.    Lyrodictya  R&mingeri, 

views  of  our  observations  seem  to  Fragmmt  of  one  of  the  urge  revereedclemee,  •  dlactlne  rod,  crncl- 

, .  ,  1         1    .       i>  i  1      •  form  iplculei  of  varioiu  elzee  and  tbe  upper  lurfaee  of  a  very  aniall  umbeL 

dispel  any  doubt  ot  their  accuracy,  ^t^,  ij.m.cj 


172  DlCTYOSPONGID^ 

There  are  no  horizontal  bands  corresponding  with  these  vertical  bundles 
The  rest  of  the  skeleton  is  pretty  much  lost ;  here  and  there  a  few  interlacing 
horizontal  spicules  are  to  be  found  at  the  intersections,  but  the  broad  vertical 
interspaces  are  regularly  divided  by  impressions  of  small  quadrules  which 
measure  from  2  to  3  mm.  on  a  side,  without  further  evidence  of  coarser  mark- 
ings. At  one  side  of  the  specimen  the  edge  of  the  cup  is  exposed  and  shows 
the  fact  that  the  interstitial  vertical  and  cross  spicules  projected  for  a  short 
distance  beyond  the  surface  of  the  cup,  and  also  that  at  wide  intervals  there 
were  small  projecting  tufts  similar  to  those  in  Physospongia. 

In  the  matrix  taken  from  the  interspaces  between  the  lateralia  have  been 
found  large  echinate  hexactins  or  pentactins  with  curved  arms,  also  some 
minute  hexactins  and  micrumbels. 

Dimensions.  The  single  known  specimen  of  this  species  has  a  length  of 
98  mm.  and  a  maximum  width  of  105  mm. 

Locality.  In  the  calcareous  shales  of  the  Keokuk  group,  Crawfords- 
ville,  Indiana. 

PHRAGMODICTYA,  HaU. 

1881.  Dictyophyton,  Whitfield.   Bull.  No.  1,  Amer.  Mus.  Nat.  Hist.,  p.  18. 

1882.  Phragniodictya,  Hall.     Notes  on  the  Family  Dictyospongidse ;  Expl 

pis.  17,  19,  20. 
1884.    Phragmodictya,  Hall.    Thirty-fifth  Ann.  Kept.  N.  Y.  State  Mus.  Nat. 
Hist.,  pp.  466,  477,  478. 

Sub-cylindrical  or  slightly  expanding  cups  abruptly  contracted  at  the 
base  to  form  a  smooth,  broadly  obconical  or  nearly  transverse  plate  or 
diaphragm.  The  edge  of  this  basal  plate  bears  a  broad  peripheral  frill. 
Attachment  was  probably  effected  both  at  the  apex  of  the  diaphragm  and  by 
the  basal  frill.  Surface  covered  with  vertical  ridges  and  nodes.  Reticulum 
very  fine  and  without  tufts. 

Type,  Phragmodictya  catillij'ormis,  Whitfield  (sp.). 

This  genus  differs  from  TirrsAKODioTYA  in  the  absence  of  a  coarse  regular 
quadration  and  fenestration  of  the  surface  and  in  the  distinctly  radiate  net- 
work of  the  basal  diaphragm,  and  from  Aclceodictya  in  the  convergence  of  the 
radial  bands  of  the  diaphragm  to  a  well-defined  and  probably  tufted  apex. 


Spkciks  of  tiik  Kkdkitk  Group. 


173 


PlIRAGMODICTYA  CATILLIFOUMIS,  Whitfield  (sj).). 
Platk  lxiv,  Fios.  1-5;  Platb  lxv,  Fios.  1.  2;  Plate  lxvi,  Figs.  1-9;  Plate  lxvji,  Fios.  1-4; 

Plate  lxvui,  Fios.  1-4. 


1881. 


Bull.  No.  1,  Amer.  Mus.  Nat. 


1882. 


1884. 


Dictyophyton  catllUfm-me,  Whitfield. 
Hist.,  p.  18,  pi.  iii,  fig,  1. 

Phragmodictya  scyphus,  Hall.     Notes  on  the  Family  Dictyospongidse. 

Expl.  pi.  17,  figs.  12,  13;  pi.  19,  figs.  2,  3;  pi.  20,  figs.  1-6. 
Phragmodictya  catilUformis,  Hall.     Thirty-fifth  Ann.   Kept.  N.   Y. 

State  Mus.  Nat.  Hist.,  p.  477,  pi.  18,  figs.  12-14;  pi.  20,  figs.  2,  3; 

pi.  21,  figs.  1-6. 

Sponge  sometimes  of  large  size  but  usually  of  medium  proportions ;  subcyl- 
indrical  or  with  a  somewhat  flaring  aperture ;  base  broad,  slightly  expanded, 
its  diameter  increased  by  the  projection  of  the  periloph.  General  fonn  stout, 
the  length  being  less  than  twice  the  median  diameter.  Specimens  having  the 
aperture  clearly  retained  show  a  slight  contraction  of  the  upper  margin  but  it 
is  not  fully  determined  whether  this  is  a  normal  feature. 

The  base  is  a  very  shallow  obcone  with  a  slightly  eccentric  apex ;  its  margins 
appear  to  make  an  angle  of  about  1108  with  the  lateral  walls  of  the  cup,  and 
its  converging  surfaces  to  be  gently  concave  except  about  the  apex. 

Vertical  walls  of  the  sponge  generally  devoid  of  ornament  except  for  a 
series  of  irregularly  disposed  vertical  ridges  and  elongate  nodes.  These  lie  at 
subecpial  intervals  and  occasionally  a  single  ridge  may  be  traced  for  nearly  the 
full  length  of  the  cup.  The  nodes  seldom  occur  about  the  base,  while  they 
multiply  toward  the  top,  the  ridges  sometimes  being  resolved  int^  a  series  of 
disconnected  nodes,  but  frequently  the  ridge-like  nodes  appear  in  the  inter- 
spaces between  the  ridges,  and  without  order  or  arrangement.  At  the  edge 
of  the  basal  disc  the  ridges  are  abruptly  transsected,  making  the  margin 
nodose,  as  it  is  in  the  genus  Thysanodictya.  The  surface  of  the  periloph 
shows  a  more  or  less  regular  continuation  of  the  ridges.  The  surface  of  the 
basal  obcone  or  disc  is  wholly  smooth  and  in  this  respect  is  in  marked  distinc- 
tion to  the  generally  exposed  or  vertical  walls  of  the  cup. 

Mode  of  attachment.  The  convergence  of  the  vertical  spicular  bands  to 
the  apex  of  the  basal  diaphragm  and  the  usual  imperfection  of  this  apex  are 
evidences  of  the  attachment  of  the  sponge  at  this  point.  Such  attachment, 
however,  was  restricted  to  the  apex  of  this  obcone,  and  though  similar  to  that  in 
most  of  the  Dictyosponges  here  considered,  would  have  been  ineflScient  in 
holding  the  sponge  firmly  in  position.     It  is  evident  that  important  accessory 


174 


DlCTYOSPONOID^. 


FloiTRE  26.    Phraffmodictya  catitliformU. 
X  minor  mesh  bounded  by  roda  and  pentactlus,  with  curved  pentactlnf 
or  heXfttlDS  over  the  Interspace.    (J.  M.  C.) 


aid  was  contributed  by  the  spicular  bands  of  the  periloph  extending  downward 
beyond  the  periphery  of  the  disc.      This  extension  bears  in  some  degree  the 

ornamental  features  of  the 
lateral  walls  of  the  cup  and 
many  of  the  stronger  sjnculai 
bundles  of  those  Avails  were 
continued  into  it.  The  tejm- 
inal  or  lower  edge  of  the  peri- 
loph is  not  regular,  but  where 
the  ridges  come,  down  from 
above  it  is  produced  into  short 
radiciform  extensions. 

Heticuluvi.  The  net-work 
of  the  sponge  is  exceedingly 
fine,  and  the  subdivision  very 
uniform.  There  are  no  strong 
bands  of  vertical  or  horizontal 
rods,  and  it  is  often  difficult  to 
find  traces  of  them  except  near  the  base.  On  the  surface  of  the  basal  obcone, 
however,  the  vertical  or  rather  radial  bands  are  very  distinct,  while  the  hori- 
zontal or  concentric  bands  are  less  so. 

Slceleton.  The  search  for  the  spicular  elements  of  the  skeleton  has  shown 
that  the  rhabdus  of  the  gastral  and  anchoring  spicules  are  small  and  few  in 
comparison  with  the  other  species  whose   spicular  structure  is  known.     The 

fine   texture  of  the  re- 
f?  .  I  ticulum  is  due  to  quad- 

rules,  which  for  the  most 
part,  seem  to  be  outlined 
by  strong  hexactins  at 
the  angles,  their  arms 
overlapping  (fig.  26). 
These  have  long,  smooth 
and  acute  branches. 
There  also  appear  to 
have  been  pentactins  or  hexactins  of  great  size  with  echinate  arms,  as  indicated 
by  the  large  fragment  shown  in  figure  29,  but  had  they  been  abundant  we 
should  probably  know  more  of  their  form.  The  only  other  spicules  of  nota- 
ble size  are  indicated  by  the  slender  fragments  bearing  long  spinules,  shown 


Figures  27,  28.    Spicules  of  Phragmodictya  catilUformU,  x400. 
In  figure  37  are  curved  pentactins  like  those  seen  In  figure  26.    In  one  of  these  spicules 
the  arms  end  in  knobs. 

Figure  28  shows  other  modifications  of  the  parenchymal  spicules.  (J.  M.  C.) 


Speciks  of  tiik  Kkokitk  Group. 


17f 


in  the  same  figure.  These  may  be  parts  of  large  pinuli,  or  still  another  style 
of  cruciform  spicule.  Characteristic  of  this  species  are  the  curious  pentactins 
or  hexactins  with  curved  arms,  which  lie  scat- 
tered abundantly  among  the  quadrules  (see  ^ft  W  il 
figs.  26,  27).  Other  modifications  of  the  pen-  ^^'^  ^  *^ 
tactin  are  shown  in  figs.  28  and  30.  Consider-  ^^-^.^^^  ^~. 
able  diversity  of  foi-m  is  found  among  the  W^/S^  if  ^O 
diactins ;  some  are  elongate,  pod-shaped,  some 
stout  and  bean-shaped,  others  have  a  strong 
single  curvature  or  a  double  curve.  A  few 
fragments  have  been  seen  which  indicate  diac- 

+\na      ,.,;+!,       „1  i  1        •       1  ...  FioDR«S9.    FraKtiMiitt of  «elifiiat««plenlM  o( 

ims    witn    clavate    or    sphencal    extremities.  Phragmodtctvacatmiformu.^m.  (j.  m.c.) 
These  are  represented  in  figure  30.      The   only  trace  of   anchorate  spicules 
observed  is  a  minute  form  shown  in  the  same  figure. 

The  elements  of  the  skeleton,  taken  as  a  whole,  are  (piite  distinct  from 
those  of  the  other  species  here  described.  The  umbels  and  clemes,  characteriz- 
ing the  genera  Piiyso- 

SPONOIA,    ClEODICTYA, 

LYKODicTYA,etc.,  Seem 
to  be  wholly  wanting. 

Dimensions.  This 
species,  which  is  the 
most  abundant  of  all 
the  forms  occurring 
in  the  Crawfordsville 
shales,  varies  greatly 
in  size.  A  young  and 
essentially  entire  in- 
dividual which  prob- 

fjl^ly    retains  the    Dro-  figure  so.    Spteal«8of  Fhra(7mo<iicfvacaf{l/(/V>rmi«,  Z400. 

Here  are  shown  variouj  f omu  of  dlactlne  and  cruciform  Bplculeo.  and  amonir  tbem  U  a 
portions     normal      for    'man  anchorate  olavule.    (J.  M.  C.) 

full  growth,  has  a  slightly  curved  cup,  its  length  from  the  aperture  to  the 
distal  edge  of  the  periloph  measuring  60  mm.  The  width  of  its  aperture 
is  50  mm.,  its  median  diameter  is  28  mm.,  and  its  basal  width  about  25  mm., 
the  periloph  having  a  length  of  5  mm.  A  nearly  entire  specimen  of  average 
size,  slightly  incomplete  at  the  aperture,  has  a  length  of  95  mm.,  the  periloph  is 
15  mm.  in  length  where  longest,  and  the  apex  of  the  basal  obcone  is  9  mm.  below 
the  plane  of  its  base.     The  largest  cup  observed  has  an  apertural  diameter  of 


176  DiCTYOSPONGIDvE. 

230  mm.      A  basal  diaphragm,  140  mm.  in  greatest  diameter,  bears  a  frill 
45  mm.  iu  width. 

Locality.  Keokuk  group.  In  the  calcareous  shales,  and  rarely,  in  the 
overlying  sandstone,  at  Crawfordsville,  Indiana. 

Pheagmodictta  patelliformis,  Hall. 

Plate  lxv,  Fig.  3. 

1884.     Phragmodictya  patelliformis,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y. 
State  Mus.  Nat.  Hist.,  p.  478. 

The  original  specimen  of  this  species  is  a  large  oval  basal  obcone  Avith  a 
highly  eccentric  apex.  Its  major  diameter  is  132  mm.  and  the  apex  lies  32 
mm.  from  the  nearest  margin.  The  minor  diameter  through  the  aj^ex  is  100 
mm.,  through  the  center  115  mm.  The  impressions  of  the  radiating  spicular 
bundles  are  strong,  and  the  surface  of  the  plate,  which  has  been  preserved 
in  a  calcareous  nodule  with  its  contour  undisturbed,  is  gently  convex.  The 
aspect  of  this  basal  disc  is  \inlike  that  of  Phragmodictya  catilUformis,  in  its 
much  more  eccentric  apex  and  convex  surface. 

Locality.     Keokuk  group.    In  the  calcareous  shales  at  Crawfordsville, 

Indiana. 

Phragmodictya  (?)  crebristriata,  Hall. 

Plate  lxi,  Fig.  4. 

1884      Phragmoilictya  (?)  crehri'itriata,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y. 
State  Mus.  Nat.  Hist.,  pi.  21,  fig.  7. 

This  fossil  which  has  been  illustrated  in  the  place  cited  but  not  before 
described,  appears  to  be  an  impression  of  a  part  of  the  flaring  aperture  of  a 
Dictyosponge  with  an  extremely  fine  reticulum,  or  part  of  an  outer  mould  of 
a  basal  obcone  with  obscure  nodes  at  the  periphery.  The  radial  spicular 
impressions,  though  fine,  are  distinct,  while  the  reticulating  bands  are  highly 
obscure.  It  probably  represents  a  species  unlike  any  of  the  others  here 
described,  but  its  generic  characters  are  still  uncertain. 

Locality.     Keokuk  group.    In  the  calcareous  shales  at  CraAvfordsville, 

Indiana. 

(?)  Phragmodictya  lineata.  Hall. 

Plate  lxviii,  Fig.  5. 

1884.     Phi'agmodictya  lineata,  Hall.      Thirty-fifth  Ann.  Rept.   N.  Y.  State 
Mus.  Nat.  Hist.,  p.  478,  pi.  21,  fig.  8. 
This  name  was  given,  in  the  work  cited,  to  an  imperfect  specimen  of  a 
small  subcylindrical  cup,  with  moderately  coarse  reticulation  and  a  surface 


Speciks  ok  the  Keokuk  Group. 


177 


devoid  of  the  ornamental  nodes  and  ridges  occuring  in  Phragniodictya  catillir 
formis.  The  specimen  was  described  as  having  an  obli(^ue,  convex  basal  cone 
with  a  strong  eccentric  cicatrix,  but  upon  a  reexamination  this  supposed 
structure  seems  to  be  a  portion  of  the  lateral  wall  of  the  cup  which  has  been 
broken  and  somewhat  irregularly  compressed.  The  characters  of  this  speci- 
men are  not  retained  with  sufficient  clearness  to  demonstrate  its  specific  value. 

Locality.  Keokuk  group.  In  the  calcareous  shales  at  Crawfordsville, 
Indiana. 

ACLCEODICTYA,  gen.  nov. 

1882.     Phragmodictya,  Hall.     Notes  on  the  Family  Dictyospongidjp ;  EzpL 
pi.  19,  fig.  1. 

1884.     JEctenodlctya,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State  Mus.  Nat 
Hist.,  p.  476. 

Subcylindrical  sponges  with  strongly  fenestrated  exterior,  as  in  Clathbo- 
SPONGIA,  and  with  abruptly  obconical  base,  as  in  Phragmodictya.  There  is 
no  periloph  as  in  the  latter  genus  and  in  Tiiysanodictya,  and  the    vertical 


FlOHEK  SI.    Large  hexactln  or  pentactln  of  AclmOtctva  mar><puj,  i400.   (J.  M.  C.) 

spiciilar  bands  converge  irregularly  to  or  about  a  broad  apical  point.     The 
radial  bands  from  one  side  are  continuous  across  the  disc  and  are  reticulated 
by  another  set  of  radial  bands  from  the  other  two  quadrants  of  the  disc. 
Type,  Adveodictya  marsipus,  sp.  nov. 


178 


DlClYOSPONU  ID.E. 


ACLCEODICTTA   MARSIPUS,    Sp.  nOV. 
Plats  lv,  Figs.  4,  5;  Plate  lx,  ftas.  3-5 ;  Platb  lxi,  Figs.  1-3;  Plate  lztiii,  Figb.  7,  8. 

Sponge  robust,  elongate,  subcylindrical.  Basal  obcone,  when  uncom- 
pressed, expanding  at  an  angle  of  about  50  degrees;  continuing  for  not  more 
than  one-lifth  the  length  of  the  cup,  whence  the  surface,  with  a  sharp  but  not 
angular  bend,  becomes  abruptly  erect  in  its  growth.  Near  the  basal  disc  the 
cup  is  at  fii'st  broadly  constricted,  thence  upward  gradually  expands  and 
again  narrows  to  the  aperture.  Thus  the  form  of  the  sponge  is  that  of  a  broad, 
medially  inflated  tube  resting  upon  a  shallow  obcone. 

Reticulum.  The  vertical  and  horizontal  strands  foi-m  meshes  which  are 
divisible  into  various  sei'ies,  though,  with  the  usual  preservation  of  the  fossil, 

these  differences  are  obscure.  The 
primary  series  form  quad  rules  Avhich 
are  about  10  mm.  on  each  side,  and 
these  are  divided  by  subordinate 
series  in  the  usual  maimer  Over 
the  body  of  the  sponge,  both  hor- 
izontal and  vertical  spicular  bauds 
were  pi'oduced  into  erect  reticulat- 
ing lamellae  which  form  a  series  of 
fenestrated  areoles.  The  broadest 
of  these  lamellae  may  have  a  margin 
of  5  mm.  A  single  poorly  preserved 
specimen  in  which  a  portion  of  the 
pyritized  skeleton  is  I'etained  shows 
the  presence  of  very  large  anchor- 
ate  basalia  with  broadly  divergent, 

FioHRE 33.  deme, anchorate clavole >nd  stauractlns of  AcUxodictya  ,^        n     ^  i  i  i 

marsipus,  i4oo.  (J.  M.  c.)  smooth    flukes    and    convex    head. 

There  are  also  great  pentactins  with  curved  arms,  and  clemes  with  short  inter- 
vals between  the  acuminate  lateral  processes.  These  are  forms  similar  to  those 
occuri'ing  in  Piiysospongia  and  Cleodictya.  A  peculiar  pentactin  (or  tetractin) 
with  short  club-shaped  ai-ms,  a  foi-m  not  observed  in  other  species,  occurs 
here  among  the  jiareuchymalia.  A  figure  is  also  given  of  a  very  small  pent- 
actin with  long,  straight  and  simple  arms. 

Dimensions.  The  largest  and  best  preserved  example,  in  which  the 
sjjonge  is  compressed  but  essentially  entire,  has  a  length  of  1 40  mm. ;  length 


Speciks   of   thk  KKOKtrK  Guoirp.  17H 

of  bascal  obcone,  30  mm. ;  diameter  at  base  of  the  cylindrical  portion,  70  mm., 
at  the  middle,  85  mm. ;  at  the  aperture,  75  mm. 

Localities.  In  the  calcareous  shales  at  Crawfordsville  and  Indian  Creek, 
Indiana.     (Largely  from  the  collection  of  A,  S,  Tiffany.) 

ACLCEODICTTA  (?)  ECCENTKICA,  Hall  (sp.). 
Plate  uv,  Fiqs.  1,  2. 

1882.     Phragmodictya    eccentrica,   Hall.      Notes    on    the    Family    Dictyo- 

spongidse;  Expl,  pi.  19,  fig.  1. 
1884.     Ectenodictya  eccentrica,  Hall.     Thirty-fifth  Ann.  Rept.  N.  Y.  State 

Mus.  Nat.  Hist.,  p.  476,  pi.  20,  fig.  1. 

This  species  was  based  upon  certain  discoid  bodies,  marked  by  irregu- 
larly radiating  and  concentric  spicular  impressions,  which  converge  about  the 
central  area.  The  aspect  of  the  specimens  suggests  the  basal  obcone  in 
Acloeodictya  niarsipus,  but  the  reticulation  is  much  finer  than  in  that  species 
and  the  surface  toward  the  periphery  shows  undulations  with  a  tendency  to 
plication.  Two  of  the  specimens  indicate  that  a  portion  of  the  vertical  or  radiat- 
ing spicules  from  opposite  quadrants  of  the  disc  are  continuous  over  the 
apical  region,  Avhile  they  are  crossed  by  the  spicules  from  the  other  quadrants, 
the  radial  spicular  bands  thus  reticulating  with  each  other.  The  concentric 
or  horizontal  bands  also  appear  to  be  present  on  this  apical  area.  No  cicatrice 
is  observable  in  the  specimens. 

The  specimens  which  represent  this  species  difEer  considerably  in  size, 
one  having  a  semidiameter  of  50  mm.,  the  other  of  80  mm.  The  entire 
diameter  of  another  is  55  mm. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 

GRIPHODICTYA,  gen.  nov. 

Elongate  subcylindrical  sponges  with  subequally  expanded  base  and 
aperture.     No  diaphragm  or  basal  disc  present  (?). 

Surface  smooth.  Reticulum  very  fine.  Skeleton  bearing  a  great  number 
of  hexactins  with  modified  arms  (oxyhexasters)  and  umbrella-shaped  clavules 
of  various  forms. 

Type,  Griplwdictya  epiphanes,  sp.  nov. 

The  external  characters  of  this  sponge  are  not  fully  known,  but  its 
spicular  composition  is  so  totally  unlike  that  of  other  forms  that  upon  this 


180 


DlOTTOSPONGID^. 


character  both   genus   and   species  are  founded.     Sufficient   is   also  known 

of  the  shape  of  the  body  to  permit  its  recognition  from  the  accompanying 

description. 

Griphodictya  epiphanes,  sp,  nov. 

Plate  lv,  Fig.  6. 

SpoTfOE    slender,   originally    subcylindrical  or  with   the    tube    gently 

expanding  to  both  extremities,  the  incurvature  of  the  vertical  Avails  being  a 


FianBE  es.  Spicules  of  Qriphodictya  epiphanes,  x266.  1-1,  umbels  of  different  form ;  5,  ozyhezaster  with  the 
arms  divided  Into  throe  prongs  each  ;  6,  7,  8,  oxyhexasters  with  the  arms  of  the  horizontal  axes  unmodified  and  the 
others  variously  branched  ;  9,  oxyhexaster  with  four  prongs  on  two  of  the  arms  and  three  on  the  other  two-, 
10,  simple  hexact  or  pentact ;  11,  fragments  of  echlnate  spicules.   (J.  M!.  0.) 

very  broad  arc  without  interruption  throughout  its  extent.  The  base  of  the 
sponge  terminates  abruptly  in  a  broadly  undulated  margin  by  which  attach- 
ment was  evidently  effected  or  aided,  as  in  the  case  of  the  periloph  of  Phkao 


Spfcies  of  thk  Kbokitk  GrRorip. 


181 


MODicTYA.  Apertural  region  somewhat  more  expanded  than  the  base ;  margin 
of  aperture  regular  and  without  tufts.  Surface  smooth,  or  with  traces  of 
obscure,  discontinuous  vertical  ridges. 

Reticulum,  very  fine.  The  specimen  is  so  preserved  as  to  expose  for 
nearly  its  entire  length  the  inner  surface  of  the  wall,  this  being  somewhat 
abraded  about  the  base.  At  the  summit  a  portion  of  the  internal  cast  adheres, 
so  that  the  entire  thickness  of  the  skeleton  is  here  retained.  Toward  the 
lower  part  of  the  sponge  may  be  seen  remnants  of  two  small,  widely  separated 
vertical  bundles  of  stout  smooth  rods.  No  other  vertical  rods  are  apparent, 
while  the  horizontal  rods,  though  minute,  are  so  abundant  and  so  matted 
together  as  to  form  a  felt  of  spicules  without  any  division  into  bundles.  The 
prevalence  of  these  quite  obscures  all  traces  of  reticulation.  The  flesh  spicules, 
which  are  extremely  abundant  over  the  upper  part  of  the  specimen,  are  large 
oxyhexasters  with  their  rays  variously  modified;  sometimes  four  of  these 
rays  are  simple,  while  those  of  the  third  axis  are  divided  each  into  three 
prongs ;  in  other  cases  the  rays  of  a  single  axis  are  simple  and  those  of  the 
other  axes  divided  into  three  or  four  prongs.  The  prongs  are  not  always  of 
the  same  number  in  a  hexaster,  some  of  the  rays  bearing  three,  some  four, 
and  the  prongs  themselves  being  variously  subdivided  and  often  j)roducing 
very  complicated  forms.  There  is  some  variation  in  size  in  these  spicules  as 
shown  in  the  accompanying  figures.  Besides  these  hexasters,  which  seem  to 
compose  the  greater  part  of  the  spicular  mass,  there  are  occasional  fragments 
of  regular  hexactins,  and  numerous  fragments  of  hexactin  rays  showing  very 
finely  echinate  extremities.  There  are  also  two  styles  of  umbrella-shaped 
clavules,  one  having  a  sharply  tapering  head  with  broad  divisions,  eight  in 
number,  the  other  considerably  larger,  with  apparently  seven  highly  divergent, 
narrow  and  acuminate  divisions. 

The  specimen  bears  no  little  resemblance  to  a  small,  or  young  individual 
of  Phragmodictya  catilliformis,  though  showing  no  evidence  of  basal  plate  or 
frill,  and  it  was  regarded  as  that  species  until  the  examination  of  its  skeletal 
structure  demonstrated  the  impropriety  of  referring  it  either  to  this  or  any 
other  known  species  of  Dictyosponges. 

Dimensions.  Length,  54  mm. ;  basal  width  20  mm. ;  median  width 
15  mm.;  apertural  width  25  mm. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 


182  DicryospoNoiD^. 

CALATHOSPONGIA,  gen.  nov. 
(For  description  see  page  155.) 

Calathosponoia  AMPnoRiNA,  sp.  nov. 

Plate  lxviii,  Fig.  6. 

Sponge  small,  abruptly  and  broadly  expanded  at  the  aperture ;  body 
comparatively  short,  subcylindrical,  having  considerably  less  than  one-half  the 
width  of  the  aperture;  basal  portion  somewhat  expanded,  probably  terminat- 
ing abruptly.     Surface  without  ridges,  nodes  or  other  irregularities. 

Reticulum  composed  of  very  fine  spicular  bands,  producing  uniformly 
small  meshes.  On  the  internal  cast  there  are  no  sharply  defined  quadrulesbut 
there  is  a  notable  predominance  of  the  vertical  spicular  bundles  which  are 
individualized  near  the  base  but  in  ascending  they  become  broadened  and 
diffuse. 

The  single  specimen  has  its  flaring  apertural  portion  bent  over  upon  the 
body  of  the  cup,  but  the  original  form  of  the  sponge  is  very  clearly  shown. 
Piart  of  the  base  is  missing  but  the  slight  expansion  of  the  cup  in  this  region 
indicates  a  stout  and  abrupt  termination. 

Dlmeasionx.  The  diameter  of  the  aperture  in  the  specimen  described  is 
54  mm. ;  that  of  the  body  of  the  cup  at  the  base  of  the  aperture,  20  mm.,  and 
the  entire  length  of  the  specimen  70  mm. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 

CaLATIIOSPONGIA  (?)  MAGNIFICA,    Sp.    nOV. 
Plate  lti,  Fig.  6 ;  Plate  lvu.  Fig.  1. 

Sponge  of  large  size,  funnel-shaped,  probably  subcylindrical  about  the 
body,  rapidly  and  abruptly  expanding  toward  the  aperture.  Form  incom- 
pletely known.  Surface  more  or  less  obscurely  prismatic  and  without  nodes 
or  other  irregularities. 

Reticulum  divided  into  large  quadrules  by  horizontal  intersections  with 
the  obscure  prism-edges  or  spicular  ridges  corresponding  therewith.  These 
quadrules  measure  about  24  mm.  on  a  side,  over  the  body  of  the  ciip,  but 
widen  considerably  toward  the  aperture.  There  are  four  or  five  subsidiary 
series  of  meshes  in  each  quadrule  of  the  first  order,  the  reticulation  of  the 
entire  surface  being  thus  complete  and  regular.  Probably  the  principal  spicu- 
lar bands  were  more  or  less  elevated  into  erect  lamellae. 


► 


Spkciks  of  the  Keokuk  Group, 


183 


Dimensions.  We  have  represented  two  large  fragments  of  this  Hi)ecie8, 
both  showing  the  expanded  upper  portion  and  a  part  of  the  body  of  the  cup. 
The  larger  of  these  has  a  length  of  240  ram.,  a  width  below  of  140  ram.,  and  at 
the  upper  end  the  diameter,  if  entire,  would  be  about  300  mm.  The  second 
specimen  is  smaller  and  shows  a  somewhat  more  rapid  expansion  of  the  cup. 
Both  of  these  specimens  have  been  compressed.  Fragments  of  the  species  are 
not  uncommon  but  are  usually  very  imperfect. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana, 


CLEODICTYA,  Hall. 
(For  description  see  page  163.) 

Cleodictya  gloriosa.  Hall, 

Plate  lxix,  Fio.  1 ;  Plate  lxi,  Fios.  1,  3, 

1884.     Cleodictya  gloriosa,  Hall.     Thirty-fifth  Ann,  Rept.  N,  Y.  State  Mu8; 
Nat.  Hist.,  p.  479, 

Sponge  large,  urceolate  in  form;  rapidly  expanding  from  a  probably 
broad  and  flattened  base  into  the  single  basal  row  of  nodes,  where  it  attains 
its  greatest  width ;  above  this  it  is  abruptly  contracted,  thence  grailually  expands 
with  a  smooth  surface  and  graceful  outward  curvature  toward  the  aperture. 
From  the  summit  of  the  nodes  downward,  which  was  less  than  one-half  the 
length  of  the  cup,  the  surface  is  gently  convex ;  above  them  the  surface  is 
concave. 

The  nodes  are  large,  strongly  convex,  obtuse,  wider  vertically  than  hori- 
zontally and  are  directed  obliquely  upward  resting  on  the  crest  of  a  horizontal 
ridge  produced  by  the  constriction  of  the  cup.  The  number  of  these  nodes  is 
somewhat  variable  ;  the  finest  example  bears  ten  of  about  equal  size  and  shows 
an  additional  but  incipient  node  in  one  of  the  dividing  grooves.  A  second  and 
smaller  specimen  has  eleven  fully  developed  nodes  with  an  incipient  twelfth. 
These  facts  indicate  the  probability  of  increase  in  the  number  of  nodes  with 
the  growth  of  the  individual  as  well  as  their  numerical  variation  in  different 
individuals.  The  external  cast  indicates  that  the  nodes  were  not  tufted  nor 
any  other  part  of  the  surface  elevated  into  spicular  lamellae.  The  grooves 
separating  the  nodes  are  broad  and  moderately  deep,  not,  however,  interrupt- 
ing the  general  elevation  of  the  ridge  upon  which  the  latter  rest. 

Shortly  above  the  upper  base  of  the  nodes  the  surface  contracts  for  about 
one-fourth  the  lower  diameter  of  the  cup,  and  its  gradual  expansion  from  this 
point  upward  continues,  it  is  believed,  ^vithout  interruption  to  the  aperture ; 


184  DlClTOSPONGID^. 

thu8  making  the  upper  part  of  the  sponge  regularly  vase-shaped  with  a  circu 
lar  cross-section.     The  actual  extent  of  this  vase  and  the  precise  form  of  the 
aperture  remain  unknown. 

The  reticiduin  is  fine-meshed  throughout  and  in  no  place  is  there  evidence 
of  conspicuous  reticulating  bands,  the  aspect  of  the  net-work  being  very 
similar  to  that  in  species  of  Calathospongia  (e.  g.  O.  RedfieUi,  C.  Carlli), 
The  larger  quadrules  are  minutely  subdivided  and  vary  in  size  and  form  as 
the  curvatures  of  the  surface  vary.  The  fossil  is  in  a  sandstone  and  the 
spicules  are  not  preserved. 

This  magnificent  sponge  attained  commanding  dimensions.  The  best  of 
the  specimens,  which  has  lost  something  from  each  extremity,  has  a  height  of 
190  mm.  It  has  been  somewhat  compressed  laterally  and  its  greatest  width 
at  the  lower  end  is  145  mm. ;  its  width  across  the  basal  nodes  is  215  mm.  At 
its  narrowest  diameter,  above  the  nodes,  it  measures  120  mm.,  and  at  the  top 
•about  140  mm.  A  smaller  specimen  representing  the  basal  portion  only, 
measures  115  mm.  across  the  row  of  nodes. 

Localitij.  From  a  sandstone  of  the  age  of  the  Keokuk  group,  overlying 
the  calcareous  shales  at  Crawfordsville,  Indiana.     (Collection  of  E.  B.  PIall.) 

Cleodictya  Moiiri,  Hall. 

Plate  ixx,  Fig.  3. 

1884.     Cleodictya  f  Mohri,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State  Mus. 
Nat.  Hist.,  p.  479. 

Sponge  elongate,  swollen  below,  broadly  constricted  and  gradually 
expanding  above.     The  base  is  broad  and  the  expansion  thence  to  the  hori. 

zontal  row  of   nodes  is  convex  and 
^^    <j»*   j/  1%^  rapid.     This  expanded   portion   of 


I 


the  sponge  is  relatively  much  longer 
than  in  other  species  of  the  genus, 
the  nodes  themselves  being  very 
obscure,  elongate  and  but  slightly 
elevated.  Over  one  portion  of  the 
elevation  on  which  they  lie,  where 
FiauBB»4.  st,imu»ot  Cleodictya  xohri.  ^\^q    reticiilum   Is    fuUy  preservcd, 

I,  Group  of  umbels,  x60;  2,  compound  pfnulus,  X200;  S,   hexsetln  ,  . 

turrouudedbyumbelsandfragmsnUof  rh«bds,  i80.  (J.  M.  C.)  they    are     SCarCCly     disCCmible.        Ou 

the  exposed  half  of  the  cup  there  are  evidences  of  four  nodes  with  a  possible 
fifth.     Above  the  nodiferous  expansion,  the  contraction  of  the  cup  is  gradual 


Species  of  the  Keokuk  Group, 


185 


for  one-half   the  remaining  distance,  thence   upward    the   expsAision  begins 
again  and  is  continued  to  the  aperture. 

The  reticulum  is  smooth  and  in  its  general  aspect  like  that  of  C.  gloriosa 
and  O.  QLaypolei.     Over   the  upper  portion  of   the  cup  there  ia  a  series  of 


Figure  SS.    Splculea  of  Cleodietya  Stohri,  x400.  1,  larg*  b*zact  ;  2,  extremity  of  rhcbd  or  hezact ;  S,  echlnsU 
bexact ;  i,  extremity  of  Urge  echlnate  hezact;  5,  umbel ;  (,  mlcrumbel.    (J.  M.  C.) 

horizontal  bands  of  spicules  which  are  more  conspicuous  than  the  rest  and  are 
not   crossed  by  vertical   bands  of    corresponding   size  at   regular  intervals. 


FiouRE  s;.    Spicules  of  CUocUetya  XohH,  z400.    1, 1,  hezaeU  ;  i,  •mall  cleme  ;  4,  rhabd  ihowlns  concentric  layerl ;  B,  «,  umbel 
and  mlcrumbel.    (J.  U.  C.) 

Above  the  basal  expansion  the  vertical  bundles  of  spicular  rods  or  rhabds 
lose  their  close  definition  and  are  spread  over  the  quadrules  somewhat  to  the 


186 


DlCTYOSPONGID^. 


obscuration  of  the  latter.  There  is  ranch  variation  in  the  size  of  these  rods, 
large  and  small  ones  being  bundled  together.  In  figure  37  is  shown  the 
tenninal  portion  of  one  such  rod.  A  close  examination  seems  to  establish 
the  fact  that  the  majority  of  these  lateralia  terminate  before  reaching  the 
basal  expansion  and  those  which  continue  to  the  base  of  the  cup  are  mainly 
the  rods  of  larsrest  diameter.  The  horizontal  bundles  are  of  similar  size, 
but  individual  rods  are  frequently  as  large  as  any  occuriing  in  the  A'ertical 


Figure  37.  Spicules  of  Cleodictya  Mohri,  T400,  1 ,  Terminal  portion  of  vertical  rod  ;  8,  umbel ;  8,  the  same  from 
beneath  ;  4,  mlcrumbel ;  5,  C,  the  two  sides  of  an  entire  trlplnulus  ;  7,  9,  10,  11,  13,  rarloufl  fragments  of  plnulus 
forms  ;  8,  Incomplete  amphiaster  ;  12,  echinate  hexact ,  H,  15,  tpieuUn  innominatcE  ;  16,  anchorate  clavule  , 
17,  rhabd  showing  concentric  layers.    (J.  M.  C.) 

series.  Tlie  ultimate  quadrules,  which  measure  about  .5  mm.,  have  at  each 
angle  regular  hexactins  whose  arms  traverse  more  than  one-half  the  length  of 
the  contiguous  squares,  the  extremital  portions  of  adjacent  spicules  overlap, 
ping.  The  outer  vertical  arm  is  sometimes  modified  into  a  blunt  node.  Such 
hexactins  are  shown  in  figures  34,  35  and  36.  There  are  also  other  regular 
hexactins  of  large  and  small  size  with  echinate  arms,  as  shown  by  the  frag- 


Specos  of  the  Keokuk  Grolp. 


187 


ment3  represented  in  figure  35.  The  smaller  of  these  probably  belong  to  the 
parenchymalia  hut  the  larger  has  not  been  located. 

There  is  a  very  considerable  variety  in  the  forms  which  may  be  referred 
to  the  parenchymalia  and  dermalia.  Perhaps  the  most  striking  of  these  are  the 
large  and  small  umbels  (umbels  and  micnimbels ;  figs.  34,  35,  36  and  37).  These 
are  like  the  dermal  clavules  which  have  been  figured  by  F,  E.  Sciiulze  in  the 
recent  species  Farrea  occa*  except  that  here  the  shafts  are  always  short  and 
smooth,  gently  swollen  beneath  the  umbels,  and  the  teeth  of  the  umbels 
smooth.  They  evidently  have  nothing  to  do  with  the  anchoring  spicules  or 
basalia,  and  the  same  is  probably  true  of  the  anchor-shaped  fragment 
shown  in  fig.  37  (16)  which  was  taken  from  the  upper  part  of  the  cup. 
To  the  dermalia  may  also  be  referred  the  peculiar  compound  tri-pinulus, 
both  sides  of  the  complete  form  of  which  are  shown  in  figure  37  (5,  6). 
The  morphology  of  this  spicule  is  not  readily  apprehended.  There  are  also 
other  peculiar  pinulus  forms  as  shown  in  figures  34  (2)  and  37  (7,  9,  10,  11, 
13),  which  are  as  yet  incompletely  known.  In  figure  37  (8)  is  an  imper- 
fect amphiaster  belonging  to  the  parenchymalia. 

Dimensions.  The  original  specimen  measures  130  mm.  in  height, 
70  mm.  in  diameter  at  the  base,  100  mm.  across  the  nodose  expansion,  65  mm. 
where  narrowest,  and  about  75  mm.  at  the  upper  end  which  is  imperfect. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 

PHYSOSPONGIA,  Hall. 
1881.    UphankBma,  Whitfield.    American  Journal  of  Science,  vol.  xxii,  p.  132. 
1881.    Uphankenia,  Dawson.     American  Journal  of  Science,  vol.  xxii,  p.  132. 

1881.  UpTiantcmia,  Whitfield.     Bull.  No.  1,  Amer.  Mus.  Nat.  Hist.,  p.  15. 

1882.  Fhyso.'ipongia,  Hall.      Notes  on  the  Family  Dictyospongidae ;    Expl. 

pi.  19. 
1884.  Physospongia,  Hall.  Thirty-fifth  Ann.  Kept.  N.  Y.  State  Mus.  Nat 
Hist.,  pp.  467,  479-481. 
The  fossils  of  this  genus  have  the  form  of  a  diminutive  basket  with  broad 
strands  and  coarse  meshes.  The  upright  body  has  a  very  gradual  expansion 
from  the  base  upward  and  its  surface  was  probably  regular,  as  far  as  appears 
from  the  usual  mode  of  preservation,  though  it  is  possible  that,  when  com- 
pressed, the  strong  vertical  strands  of  the  spicules  may  have  given  it  a  some- 
what prismatic  form.     Nonnally  the  surface  bears  two  series  of  these  vertical 

•  Haxactlnellldae,  pi  Uxl,  flc>.  S,  9,  ». 


188  DiCTYOSPONGIDiE. 

spicular  bands,  one  broad  and  conspicuous,  its  members  alternating  with  a 
series  of  narrow  bands.  In  the  type-species,  P.  Dawsoni,  where  all  generic 
features  are  most  clearly  defined,  these  bands  are  generally  placed  at  ecjual 
intervals,  but  there  may  be  in  other  cases,  and  even  in  this  species,  consider- 
able variation  in  this  respect.  By  the  crossing  of  the  concentric  or  horizontal 
spicular  bands,  Avhich  are  narrow,  of  equal  breadth  and  nearly  equidistant, 
the  area  enclosed  by  any  two  of  the  broader  vertical  bands  is  divided  into 
subequilateral  quadrules,  and  each  of  these  (piadrules  is  again  divided  into 
four  (piadrules  by  the  intersection  of  a  vertical  band  of  the  second  order  Avith  a 
horizontal  band.  Of  the  foiir  quadrules  thus  formed  two  are  elevated  into 
bullate  expansions  or  pouches,  the  other  two  are  depressed  into  regularly  con- 
cave basins;  nodes  and  depressions  alternating  in  position  so  that  normally 
every  node  is  surrounded  by  four  depressions  and  every  depression  by  four 
nodes,  each  at  the  same  time  being  separated  from  every  other  by  spicular 
bands.  Considerable  irregularity  in  the  disposition  of  these  elevated  and 
depressed  areas  is  sometimes  shown  by  P.  Colletti,  while  the  little  known  P. 
mi(ltib)trsarki  in  this  structure  quite  fails  to  conform  with  our  conception 
of  the  genus. 

As  usually  preserved  the  tops  of  the  surface  nodes  are  broken  off,  or  by 
adherence  to  the  matrix  give  the  impression  of  a  vertical  series  of  areoles,  but 
normally  the  surface  of  these  nodes  and  depressions  is  continuous  and  finely 
reticulated  by  regularly  disposed  interlacing  pentacts.  All  specimens 
observed  have  been  more  or  less  subjected  to  compression  and  it  is  possible  to 
ascertain  the  full  elevation  of  the  nodes  only  along  the  edges  of  the  cup.  In 
such  marginal  sections  we  also  find  evidence  of  extensions  from  the  horizontal 
spicular  bands  into  slender  erect  tufts.  These  always  occur  at  the  intersec- 
tion of  the  horizontal  and  vertical  bands,  and  it  is  probable  that  they  Avere 
developed  at  every  such  intersection.  The  spicules  of  the  vertical  bands  do 
not  appear  to  have  been  complicated  with  those  of  the  horizontal  bands  in  the 
formation  of  these  tufts.* 

There  is  certainly  a  striking  similarity  in  external  structure  between 
Physospongha.  and  the  great  explanate  or  saucer-shaped  bodies  already, 
described  as  HYPHAm'^NiA,  and  which  occur  in  the  form  of  impressions  in  the 
sandstones  of  the  Chemung  group.  Our  knowledge  of  HypHANTiEisriA  is  still 
incomplete  but  we  know  that  the  intervals  between  the  vertical  and  hori- 
zontal or  concentric  spicular  bands  were  partially  if  not  wholly  covered  by 
spongin. 

*  la  the  brief  ori^loftl  description  of  tbU  genua  these  tufts  were  regarded  as  originating  from  the  summits  of  the  nodes. 


RpKciKfl  OK  TiiE  Keokuk  (tkoup.  189 

Similar  skeletal  structure  is  also  to  be  found  in  the  genus  Koemeri- 
BPONGiA  of  the  middle  Devonian  of  Germany  {R.  Gerohtelnemiii,  F.  Roemer ; 
see  part  1,  page  67). 

In  Ameiican  faunas  species  of  Piiysospongia  are  known  only  from  the 
Keokuk  group,  all  specimens  having  been  derived  from  the  calcareous  shales 
in  the  vicinity  of  Crawfordsville,  Indiana, 

Physospongia  Dawsoni,  Whitfield  (sp.). 

PtATE  Lxii,  Figs.  1-10. 

1881.      Uphantcenia  Daiosoni,   Whitfield.       American   Journal   of  Science, 
vol.  xxii,  3d  ser.,  p.  132. 

1881.  TTphantoBnia  Dawsoni,  Whitfield.     Bull.  No.  1,  Amer.  Mus.  Nat.  Hist., 

p.  15,  pi.  iv,  figs.  1,  2. 

1882.  Physospongla  Dawsoni,  Hall.     Notes  on  the  Family  Dictyospongidae  ; 

Expl.  pi.  19,  figs.  4-6,  8  (not  fig.  7). 
1884.     Physospongla  Dawsoni,  Hall.     Thirty-fifth  Ann.  Kept.  N.  Y.  State 
Mus.  Nat.  Hist.,  p.  479,  pi.  19  (20),  figs.  4-6,  8. 

Sponge  subcylindrical  over  the  body  and  lower  portion,  expanding  some- 
what more  rapidly  toward  the  aperture.  In  the  majority  of  specimens  the 
cup  seems  to  taper  quite  rapidly  toward  the  base,  but  the  base  itself  was 
evidently  broadly  obtuse,  and  in  some  instances  slightly  expanded.  This  por- 
tion of  the  cup  is  seldom  retained.  Size  comparatively  small,  rarely  exceeding 
a  length  of  75  ram.  with  a  nearly  equal  width  at  the  aperture  when  under 
compression.  As  usually  preserved  the  width  of  the  cup  at  the  upper  end  is 
about  one  third  greater  than  at  the  lower  end ;  occasionally  the  former  is 
twice  that  of  the  latter. 

Metioulum.  The  primary  reticulation  of  the  surface  is  very  coarse  and  ia 
produced  by  a  double  series  of  vertical  spicular  bundles  intersecting  a  single 
series  of  horizontal  bands.  Of  the  vertical  bundles  the  principal  series  ia 
broad  and  flat,  its  width  being  from  three  to  five  times  that  of  the  secondary 
series.  There  is  a  notable  increase  in  width  from  below  upward  in  the  prin. 
cipal  bundles,  as  occurs  in  P.  Colletti.  The  principal  and  secondary  vertical 
bundles  alternate  in  position  and  are  equidistant,  the  latter  equally  dividing 
the  area  set  off  by  the  former.  The  horizontal  bands  are  narrow,  caiTying 
about  the  same  number  of  spicular  rods  aa  the  secondary  vertical  bundles. 
The  intersection  of  these  with  the  principal  vertical  bundles  divides  the  sur- 
face into  equal  quadrules,  each  of  these  being  sub-divided  into  four  equal 


190 


DlCTYOSPONGID^. 


FiatTRE  3S.    Physospongta  Dawaoni,    A  group  of  an- 
Chora  and  a  cleme.    x60.    (J.  U.  C.) 


])arts  by  the  intersection  of  the  horizontal  with  the  secondary  vertical  l)ands. 
Two  of  the  quadrilles  thus  formed  are  elevated  into  regular  convex  pouches  or 
nodes,  the  other  two  are  concave,  and  it  seems  probable  that  the  curvature  of 

the  concavity  was  originally  equal  to  that 
of  the  node.  These  elevations  and  depres- 
sions alternate  in  position  and  thus  each 
node  is  surrounded  by  four  depressions 
and  each  depression  by  four  nodes.  The 
general  surface  effect  produced  by  this 
arrangement  is  that  of  a  series  of  double 
vertical  rows  of  nodes  separated  by  the 
broad  vertical  spicular  bands.  The  num- 
ber of  the  primary  vertical  bands  is  usually 
twelve,  though  some  specimens  seem  to 
have  had  not  more  than  ten. 

Growth.  The  younger  parts  of  the 
specimens  usually  show  a  diminution  in 
the  size  and  in  the  development  of  the  surface  nodes  and  depressions,  and 
when  the  actual  basal  parts  of  the  cup  become  known  they  probably  will  be 
found  comparatively  free  of  the  usual  surface  conformation.  Near  the  aper- 
tural  margin  also  the  nodes  become  less  distinct,  that  is,  less  elevated,  though 
often  of  greater  area.  The  duplication  of  the  nodes  and  depressions  as  an 
accompaniment  of  growth  in  the  expansion  of  the  cup,  such  as  occurs  fre- 
quently in  P.  CoUetti,  is  very  rarely  observed  in  this  species. 

Sheleton.  In  the  remarks  already  given  upon  the  structure  of  the  skele- 
ton of  the  thin-fleshed  Dictyosponges,  it  has  been  observed  that  the  tuft  of 
basal  spicules  or  basalia  is  but  a  continuation  and  union  of  the  long  vertical 
spicular  bundles  which,  with  the  horizontal  bands,  produce  the  characteristic 
reticulation  of  the  surface.  Further  evidence  from  analogy  with  living 
hexactinellids,  as  well  as  from  the  facts  themselves,  is  that  these  lateralia  or 
upward  extensions  of  the  basalia  lay  withm  the  sponge  and  probably  bounded 
the  surface  of  the  great  gastral  cavity.  The  flesh- wall  of  the  sponge,  however, 
was  so  exceedingly  tenuous  that  in  the  best  preserved  of  the  Dictyospoxgid^ 
it  is  usually  extremely  difficult  to  distinguish  the  gastral  from  the  dermal 
surface  as  far  as  shown  by  differences  of  structure. 

In  Physospangia  Dawsoni  the  innermost  or  true  gastral  surface  of  the 
principal  lateralia  bears  only  long  cylindrical  rods,  varying  notably  in  size. 
These  may  frequently  be  followed  for  the  length  of  two  or  three  quadrules 


SPF/IIKS    OF    THK    KkOKITK    GrOITP. 


191 


without  interruption,  and  probably  were  originally  continuous  to  the  base  of 
the  sponge.  The  upper  or  outer  portions  of  these  spicular  bundles  contain 
numerous  small  cylindrical  rods  which  terminate  in  a  two-pronged  anchor, 
each  barb  or  prong  tapeiing 
backward  toward  the  aperture 
into  a  slender  rod  ending  in  a 
point.  The  head  of  the  anchor 
is  considerably  inflated  on  the 
smooth  surfaces  and  its  apex  or 
point  is  blunt.  The  size  of  the 
anchors  varies  somewhat,  as 
shown  in  the  accompanying  fig- 
ure, both  enlargements  to  the 
same  degree.  These  anchorate 
spicxdes  are  scattered,  often 
abundantly,  throughout  the  en- 
tire length  of  the  lateralia  and 
could  have  had  nothing  to  do 
with  the  fixation  or  anchoring 
of  the  sponge. 

Together  with  the  anchorate  sj^icules  or  just  above  them  and  on  the 
outermost  layer  of  the  bundles,  is  a  series  of  parallel  twigs  or  clemes  which 
have  been  described  as  somewhat  flattened  rods  expanded  alternately  fii-st  on 
one  lateral  margin  and  then  on  the  other,  into 
elongate  triangular  surfaces  whose  outer  or  mar- 
ginal angle  is  acute  and  continued  a  short  dis- 
tance backward  into  a  very  slender  rod-like  exten- 
sion. Between  this  and  the  edge  of  the  spicules 
the  margin  of  the  triangle  is  gently  incurved. 
The  intervals  between  the  triangular  expansions 
vary  somewhat,  and  the  spicules  themselves  are 
generally  more  slender  than  the  rods  of  the  gas- 
tral  surface,  though  they  are  seen  to  be  of  con- 
siderable length.  One  specimen,  which  happens 
to  be  the  original  of  Uphantcenia  Dawsoni, 
Whitfield,  is  so  broken  that  a  portion  of  these  vertical  bundles  remains 
on  the  interior  or  gastral  cast  and  the  rest  or  outer  portion  on  the 
enclosing  rock.     The  former  of   these  fragments   is  figured    by  Whitfield 


FlauRB  89.    Phyaotpongia  Dawaoni.     Tb6  head!  of    two  anchorate  splo- 
ulM.     X400      (J.  M.  C.) 


FlQURB  40.  Phyaospongia  Dawaoni,  A 
group  of  tpicules.  The  stauractln  In  th«  circia 
l8  enlarged  30  diameter* ;  the  otheri  3S0  di- 
ameters,   (J,  U.  C.) 


192 


DlGTYOSPONGID.fi. 


FiGUnE  41.  Spicules  of  Physospongia  Dawsoni,  X400. 
Fragments  of  eohinate  hexacts,  a  smooth-rayed  peutact, 
a  diact  and  part  of  a  c'leme.    (J.  M.  C.) 


(loc.  cit.  fig.  1).  The  separation  is  of  such  a  nature  that  only  the  cylindrical 
rods  adhere  to  the  gastral  cast  while  the  anchorate  spicules  and  denies 
are    beautifully  displayed  in  their    original  position    upon  the  enveloping 

matrix.  The  secondary  vertical  lateralia 
and  the  horizontal  reticulating  bands  are 
composed  of  comparatively  few  cylindrical 
spicules,  some  of  which  attain  a  greater 
size  than  any  observed  in  the  jjrincipal 
lateralia,  but  neither  of  these  series  has 
shown  any  trace  either  of  the  anchorate  or 
of  the  clemate  spicules. 

Mention  has  been  made  of  the  radiate 
tufts  produced  by  the  extension  of  the 
horizontal  spicular  rods  at  their  intersec- 
tion with  the  vertical  latei-alia  of  both 
series.  The  spicules  of  the  horizontal 
lateralia  ap[)ear  to  lie  nearer  the  gastral 
surface  than  do  the  principal  and  secondary  vertical  bundles.  At  the  intersec- 
tion of  the  horizontal  and  vertical  bundles  are  occasionally  seen  large  pentacts, 
sending  a  ray  along  eaeh  bundle  departing  from  that  point,  the  fifth  ray 
passing  inward.     These  pentacts  are  the  heaviest  parts  of  the  skeleton,  and 

as  they  lie  abt)ve  or  outside  of 
the  lateralia  they  probably  belong 
to  the  dermal  surface. 

The  dennal  surface  of  the 
quadrilles  formed  by  the  intei-sec- 
tion  of  the  lateralia  is  regularly  and 
very  finely  reticulated  by  smooth- 
rayed  pentacts  lying  in  apposi- 
tion.    These  vary  in  size  but  are 

Figure  42.   Physospongia  Dawstmi.    A  somewhat  diagrammatic  fig-  .|  i_        lI  ^ 

ure  of  one  of  thcspicuiar  tufts,  xio.  (J.  M.  c.)  ncver  iiiorc  than  ouc- tenth  as  large 

as  the  smooth  pentacts  at  the  intersection  of  the  lateralia.  Among  the  spicules 
\vhich  evidently  belong  to  the  parenchyma  of  the  sponge  are  numerous  frag- 
ments of  echinate  branches,  some  of  which  undoubtedly  are  parts  of  echinate 
hexacts,  as  that  represented  in  figure  41.  The  same  figure  shows  a  fragment 
of  a  large  spicule  in  which  the  surface  sj)inules  are  more  produced  than  in  any 
of  the  echinate  spicules  of  other  species.  This  form  of  spicule  seems  to  be  of 
rare  occun-ence  in  the  species,  and  from  analogy  with  jP,  Colletti,  Cleodlctya 


Spkcijb  ok  the  Kkokuk  Group.  198 

ghrioHd,  and  other  species,  we  have  to  conclude  that  it  belongs  to  a  large 
hexact  or  pentact  whose  precise  position  in  the  skelet<)n  still  remains  in  douht 

Smooth,  elongate  8ili<piate  diacts  (figure  -41)  of  small  size  are  occasionally 
f(nmd  among  the  parenchymalia. 

After  careful  search  tliis  species  has  afforded  no  evidence  of  tirahels  such 
as  occur  in  J*.  Colletti  and  Clewllctija  (jloriuna.  Nevertheless  it  seems  prol>- 
able  that  they  exist,  though  they  may  be  of  great  raiity. 

Dimensions.  The  size  of  this  species,  in  ctunparison  with  its  closest  ally, 
P.  Colletti,  is  always  small.  An  individual  of  rather  large  dimensions  has  a 
length  of  70  mm.,  an  apertural  diameter  of  80  mm.,  and  a  basal  diameter  (»f 
30  ram.  Fragments  of  somewhat  larger  examples  have  been  observed.  An 
average  specimen  which  seems  to  be  complete,  is  45  mm.  in  length,  has  a  bjisal 
width  of  50  mm.,  contracting  above  this  to  a  width  of  44  ram.  and  expanding 
again  to  an  apertural  diameter  of  00  ram. 

Localitij.  Physosjnnigia  Dawsoni  is  known  only  from  the  calcareous  shales 
of  the  Keokuk  group  at  Crawfordsville  and  Indian  Creek,  Indiana. 

PiiYsospowGiA  Colletti,  Hall. 

Plate  i-xiii,  Flos.  1-7. 

1884.     Physosponijia    Colletti,  Hall.     Thirty-fifth  Ann.    Rept.   N.    Y.    State 
Mus.  Nat.  Hist,  p.  480,  pi.  xx,  fig.  7. 

Sponge  large,  rapidly,  and  sometimes  unsymmetrically  e.vpanding  from  a 
broad  base  to  a  wide  aperture. 

Surface  highly  nodose,  similar  in  structure  and  aspect  to  that  of  P. 
Dawsoni,  but  with  nodes  of  much  greater  size,  and  with  more  frequent  irreg- 
ularities in  their  ari\angement. 

Peticuliim.  The  2)riraary  and  secondary  vertical  bundles  are  of  very 
unequal  size.  The  former  are  broad  and  ct)mpact  over  the  lower  portion  of 
the  sponge  but  become  diffuse  above,  spreading  into  a  fan-like  brush  near  the 
aperture  and  obscuring  thereby  some  of  the  main  quadrules.  Over  the  median 
and  lower  portion  of  the  cup  the  quadrules  are  nearly  square  but  toward  the 
aperture  both  nodes  and  concavities  become  transverse,  the  horizontal  diame- 
ter increasing  and  the  vertical  diameter  lessening,  until  each  division  becomes 
very  narrow.  Thereupon  ensues  a  multiplication  in  the  number  of  vertical 
rows  of  nodes  and  depressions,  the  two  rows  of  any  one  of  the  main  vertical 
divisions  of  the  surface  increasing  to  four,  and  the  area  occupied  by  each  large 
node  or  depression  over  the  body  of  the  cup  bears  two  small  nodes  and  cor- 
responding depressions.     This  duplication  of  the  vertical  rows  of  quadrules  is 


194 


D 


ICTYOSPONfUD/E. 


Figure  43.  Spicules  of  Physospongia  CoUetti ;  frag- 
ments iif  cli'iiies,  anchorate  clavulos  and  echinate  hex- 
attins,  etc.    XG5.    (J.  M.  C.) 


shown  in  figs.  5  and  6''of  the  phite  cited.  The  apertnre  of  the  cup  appears 
to  be  smooth  and  its  margin  regular  and  even,  without  processes  or  spicular 
projections. 

Sheleton.     As  in  Phijmspon<jki  Dawsoni,  the    vertical  bundles  in    this 
species  are  composed  of  a  great  number  of  smooth  rods,  there  being  associated 

with  them  anchors  and  clemes.  In  the 
material  imder  examination  the  spicular 
structure  is  not  very  satisfactorily  retained, 
but  it  Avould  seem  as  if  these  anchorate 
and  clemate  rods  were  compai-atively  less 
abundant  than  in  P.  Dawsoni,  and  they 
prove  to  be  of  somewhat  smaller  size.  The 
predominant  pentactins  are  of  large  size  and 
strongly  echinate.  It  is  presumed  that 
these  lay  at  the  angles  of  the  quadrules  and  their  abundance  is  in  contrast  to 

the  comparatively  rai'e 
occurrence  and  small 
size  of  similar  bodies 
in  P.  Dawsoni.  The 
surface  of  the  nodes 
and  depressions  is 
very  minutely  and 
retangulai'ly  reticu- 
lated but  so  little  of 
the  spicular  fi-ame- 
work  over  those  areas 
has  been  preserved 
that  only  the  large 
imibel  (fig.  44)  can 
be  referred  thereto. 
It  has  been  observed 
that  although  the 
specimens  of  P.  Daw- 
soni have  afforded 
much  moi'e  satisfac- 
tory means  of  study- 

FIODEK  44.    Spicules  of  P;ii/soiil)on{/ia  CoHc»(.     x400.     (.f.  M.  C.)  ing      thc     SplCUlar    clC- 

ments  of  the  skeleton,  no  umbels  have  been  found  in  that  species.  Presum- 
ably in  P.  Colletti   spicular  tufts   were   present  at  the  intersection  of   the 


Species  ok  tuk  Kkokitk  Group.  195 

horizontal    ami    vertical   spicules,    as    in    1\   Dawmmi,    but  no  satisfactory 
evidence  of  them  lias  been  observed. 

DinienHions.  A  specimen  which  2)reserve8  the  up[)er  j)ortion  of  the  cup 
for  a  length  of  82  mm.  has  an  apertural  width  of  150  ram.  and  a  diameter  at 
the  lower  end  of  80  ram.  Another  example  which  represents  the  basal  and 
median  parts  of  the  spcmge,  is  90  mm.  long,  105  mm.  wide  at  the  top,  and 
55  mm.  wide  at  the  base.  Both  of  these  sjKicimens  have  Ijeeu  laterally 
compressed. 

Locality.     In  the  calcareous  shales  at  Crawfordsville,  Indiana. 

PlIYSOSPONOIA    ALTERNATA,    Hall. 
Plate  i.xii,  Kio.  II. 

1882.     Pliysospongia  alternata,  Hall.      Notes  on  the  Family  DictyospongidaB. 
Expl.  pi.  19.  fig.  9. 

1884.     Physospoii'jki  alternata,   Hall.     Thirty-fifth  Ann.   Rept.  N.  Y.  State 
Mus.  Nat.  Hist.,  }).  481,  pi.  xx,  fig.  9. 

The  few  specimens  of  this  species  which  have  been  observed  show  a 
narrow  subcylindrical  tube  with  a  length  of  about  40  mm.  and  a  width  of 
25  mm.  That  these  undoubtedly  represent  a  specific  form  distinct  from  the 
others  here  described  is  shown,  first,  in  the  subequal  width  of  the  vertical 
lateralia,  and  second,  in  the  comparatively  slight  development  of  the  alternat- 
ing elevated  and  depressed  quadrules. 

The  vertical  and  horizontal  spicular  bands  are  narrow,  subequal  in  width, 
about  equally  distant  from  one  another  and  hence  the  quadrules  formed  by 
their  intei'sectiim  are  nearly  square.  The  number  of  double  I'ows  of  squares  is 
but  six  at  the  lo^\  er  part  of  the  specimen,  but  towards  its  upper  part  there  is 
a  duplication  of  one  of  the  rows  by  the  development  of  a  new  vertical  spicular 
bundle,  such  as  frequently  appear  in  P.  Colletti.  It  has  not  been  possible  to 
fully  investigate  the  spicular  skeleton  of  this  sponge.  Some  of  the  vertical 
lateralia  bear  clemes  similar  to  those  of  the  other  species  of  Physospongia,  but 
no  anchorate  rods  have  been  observed.  Large  pentacts  lie  at  the  angles  of 
the  meshes  ."ind  small  pentacts  and  smooth-rayed  hexacts  are  found  over  the 
quadrules. 

Locality.     Keokuk  group,  Ci-awfoi-dsville,  Indiana. 


11(6 


DlCTYOSPONGlD^. 


PlIYSOSPONGIA  MULTIBUR8ARIA,  Sp.  nOV. 
PiATK  ua.  Fig.  7. 
This  is  an  aberrant  form,  referred  to  the  genus  Piiysospongia,  pending  a 
more  complete  knowledge  of  its  structure.  The  single  specimen  represents  a 
portion  of  one  side  of  what  must  have  been  a  very  large  sponge  of  whose 
original  proportions  it  is  not  possible  to  form  an  accurate  conception.  The 
surface  is  flat  and  measures  175  ram.  in  length  and  110  mm.  in  its  greatest 
diameter.  This  expansion  is  covered  with  a  great  number  of  small,  rounded 
elevations  having  tlie  form  of  drooping  pouches,  the  smaller  of  them  resemb- 
ling the  nodes  of  Physospongi-a  Dawsoni,  but  the  longer  appear  to  have  been 
of  the  penduh)us  nature  of  the  lobes  in  Botryodictya  though  of  much  smaller 
size.  These  nodes  ai-e  arranged  in  vertical  rows  although  there  are  portions 
of  the  surface  where  the  order  is  somewhat  obscured  by  the  overlapping  of 


FiQURE  45.    Spicules  of  Physospongia  muUihuraarla,  z400.     Showing  a  largo  Irregular  pentact,  fragments  of  ectiiuate 
hezacU,  smooth  diact  and  a  minute  cleme.    (J.  M.  C.) 

the  longer  nodes.  At  the  bottom  of  the  specimen,  thirteen  of  these  rows  may 
be  counted,  while  toward  the  top,  where  the  surface  is  broadest,  there  are  as 
many  as  twenty.     No  arrangement  into  transverse  rows  can  be  distinguished. 

A  few  of  these  processes  show  traces  of  reticulation  about  their  basal 
portions  or  over  their  summits,  but  where  the  interspaces  are  uncovered  or 
the  nodes  have  been  removed,  the  impression  of  a  fine  spicular  net- work  is 
everywhere  seen.  The  unbroken  marginal  portion  at  the  right  of  the  specimen 
is  without  nodes,  and  here  the  reticulum  is  partially  preserved  in  pyrite. 

The  meshes  and  spicular  bands  appear  to  be  in  tv.'o  not  very  clearly 
marked  seines,  the  principal  bands^  so  far  as  can  be  seen,  being  from  3  to  5  mm, 


Spkctes  ok  TiiK  Keokuk  Ououp.  197 

apart  which  is  the  width  of  the  nodes.  The  bands  themselves  are  extremely 
narrow.  There  is  no  such  prominent  development  of  bundles  of  lateralia  as 
in  the  typical  species  of  Piiysosponoia,  a  fact  which,  of  itself,  is  suggestive  of 
probable  different  generic  relations  in  this  species.  Some  small  portions 
of  the  skeleton  which  have  been  submitted  to  microscopic  examination  show 
very  large  pentactins  with  blunt  extremities,  stich  as  are  shown  in  figure 
45.  These  were  probably  at  the  junction  of  the  larger  transverse  and  hori- 
zontal })ands.  There  are  also  small,  echinate  pentactins  or  tetractins,  some 
larger  fragments  indicating  similar  spicules  of  more  conspicuous  size,  and 
small  sausage-shaped  diactins.  The  little  cleme  represented  in  the  figure 
shows  the  extreme  minuteness  of  certain  elements  of  the  skeleton.  Throtigh- 
out  the  mass  examined  are  clusters  of  minute  spheres  of  pyrite,  which  at 
first  impression  would  seem  of  concretionary  origin ;  it  may,  however,  be  pos- 
sible that  these  are  modified  spicules  (diactins  ?).  No  evidence  has  been  found 
of  the  umbels  (clavules),  anchors  and  plumes  of  Physospongia  Dawsoni. 

LocaUtij.     In    the    soft   shales    of  the    Keokuk  group,  Crawfordsville, 
Indiana.    - 


EXPLANATIONS  OF  PLATES. 


200  DiCTVOSPONGID/E. 

PLATE  I. 

CYATHODICTYA,  nora.  propos. 

Page  24. 

Cyathodiotya  reticulata,  Walcott  (sp.). 

Page  21. 

Figures  1-6.  A  series  of  young  specimens,  all  of  natural  size,  showing  grada- 
tion in  form,  and  most  of  them  retaining  the  reticulum  more  or  less 
completely. 

Figure  7.  An  unusually  large  example,  its  spicular  structure  somewhat  con- 
ventionally reconstructed  in  places,  showing  the  arrangement  of  the 
spicular  elements  and,  beneath  the  reticulum,  the  irregularly  dis- 
tributed areoles  outlined  in  an  unresolvable  mass  of  pyrite. 

Figure  8.  The  apertural  portion  of  an  average  cup,  torn  at  one  side ;  enlarged 
to  show  the  finer  spicular  net- work  about  the  margin,     x  3. 

Figure  9.     An  individual  of  somewhat  less  than  average  size. 

Figures  10,11.  The  apertural  extremities  of  two  individuals.  The  apparent 
thickness  of  the  spicular  wall  is  due  to  slight  incurvature  about  the 
apei'ture.     x  3. 

Figure  1 2.  The  basal  extremity  of  an  average  specimen  ;  showing  the  exten- 
sion of  the  reticulum  into  a  short  basal  tuft,     x  3. 

Figure  13.  An  enlargement  of  a  portion  of  the  reticulum,  showing  the  differ- 
ent series  of  simple  cruciform  spicules,  and  the  underlying  layer  with 
its  areoles.  This  inner  layer  is  unresolvable,  its  substance  as 
preserved  being  granular  pyrite.  It  will  be  observed  that  the 
areoles  are  irregularly  disposed  and  are  covered  by  spicules  of  the 
outer  frame- work,  x  5. 
Utica  slate.    Holland  Patent,  New  York. 

SPH^RODICTYA,  gen.  nov. 

Page  26. 

Sph^eodiotya  suBSPHiERicA,  Walcott  (sp.). 

Page  26. 

Figure  14.  A  cluster  of  young  individuals  of  various  sizes,  showing 
peripheral  spicules  on  some  of  the  specimens. 


i>icTvosi»o>rcTii>is. 


Memoirs  Geological  Survey  of  New  York. 


Plate  I. 


G.B  Simpson  del 


Phalip  Ast  hth 


Explanations  of  Platks.  201 

Figure  15.  One  of  the  two  largest  of  the  specimens  in  this  gioup,  enlarged; 
showing  the  ])eripheral  spicules,  and  the  cruciform  spicules  over  a 
part  of  the  wall.  The  surface  is  a  continuous  pyrite  film  with  an 
in-egular  bi-eak  near  the  centei",  due  to  the  adhesion  of  a  part  of 
the  film  to  the  opposite  side  of  the  matrix.  Tlie  specimen  from 
which  these  two  drawings  have  been  mad(^  was  loaned  for  the 
purpose  by  the  U.  S.  National  Museum  and  is  one  of  the  speci- 
mens upon  which  Rauff  based  his  conception  of  the  genus 
Teganium.     X  3. 

Figure  10.     Another  group  of  young  examples. 

Figure  17.  One  of  these  enlarged,  showing  long  peripheral  spicules  and  a 
continuous  surface  film,     x  3. 

Figure  18.  A  normal  mature  specimen,  natural  size.  This  specimen  is  in 
the  condition  of  ]  )reservation  usual  for  this  species,  the  sponge 
having  separated  in  such  a  manner  as  to  expose  the  inner  poriferous 
wall  of  the  reticulum,  and  the  outer  wall  in  peripheral  section. 
The  rest  of  the  latter  adheres  to  the  separated  portion  of  the 
matrix. 

Figure  19.  An  enlargement  of  a  portion  of  the  inner  wall,  showing  the 
areoles  enclosed  by  a  net-work  of  irregularly  arranged  cruciform 
spicules.     X  3. 

Figure  20.  A  portion  of  the  outer  wall  of  the  sponge,  showing  a  regular 
reticulation,  the  matrix  beneath  bearing  obscure  traces  of  the  areoles 
of  the  inner  layer,     x  3. 

Figure  21.  An  enlargement  of  a  portion  of  the  surface  similar  to  that 
shown  in  figure  19.     x  3. 

Figure  22.     Part  of  a  segment  of  one  of  these  spheres,  showing  a  portion  of 
the  inner  wall,  and  the  section  of  the  outer  wall,     x  3. 
All    the   specimens   here    figured    are  from    the    Utica     slate    at 
Holland  Patent,  New  York. 


202  DictyospongidjE. 

PLATE  II. 

HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plate.  Ill,  IV,  V.  VI,  VH.  VIII,  IX.  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII, XXXIX.  XLVI ) 

Hydnoceras  Avoca,  sp.  nov. 

Page  103. 
(See  Plates  IX,  X,  XI.) 

Figure  1.     Tlie  upper  portion  of  a  large  individual,  an  internal  cast  with  veiy 
sharply  defined  reticulation  ;  showing  the  broad,  subcircular  outline 

of  the  nodes. 

Chemung  group.     Avoca,  Nero  Yorh. 
Figure  3.     An  incomplete  and  slender  specimen  with  four  horizontal  rows  of 
nodes,  three  of  these  being  greatly  extended. 
Chemung  group.     Jmks  q^carry,  Bath  JVew  Yorh. 

Hydnoceras  kodostjm.  Hall  (sp.). 

Page  109. 

Figure  2.     A  view  of   a  gutta-percha  squeeze  of  the  original  impression; 
showing  the  small  size  of  the  species  and  its  prominent  nodes  in  at 

least  five  rows. 

Chemung  group.     Cattaraugus  county,  New  York 

Hydnoceras  botrcedema,  sp.  nov. 

Page  102. 

Figure  4.  A  specimen  of  average  size,  showing  the  extended,  flattened  and 
slightly  drooping  nodes,  completely  circular  body  and  fine  reticula- 
tion. This  species  is  almost  invariably  found  in  a  highly  imperfect 
condition,  and  consequently,  in  this  drawing  the  tips  of  the  nodes 
have    been    some\vhat    restored    from   data   furnished   by    several 

specimens. 

Chemung  group.     Irish  Ml,  near  Bath,  New  Yorh. 

Hydnoceras  tuberosum,  Conrad. 

Piifte  97. 
(See  Plates  III,  IV,  V,  VI,  VII,  VIII,  XXXVIII.) 

Figure  5.     View  from  the  upper  end  of  an  average  internal  cast. 
Figure  6.     An  enlargement  of  the  surface  of  the  internal  cast^ 

Chemung  group.     Broken  hill,  near  Coliocton,  New  Yorh. 


nicTvo  spoxrwii>A\ 


Memoirs  Geological  Survey  of  New  York. 


Platen 


G  D  Simpson  del 


CFaiisel  iith. 


Explanations  of  Pf-atks.  203 

Figure  7.  A  somewhat  restored  aiul  conventional  figure  of  this  species, 
reproduced  from  the  Thirty-fifth  Annual  lleport  of  the  N.  Y.  State 
Museum,  pi.  17  (18),  fig.  7. 

Hydnockras  ANTiiRACia,  sp.  nov. 

Page  109. 
(See  Plates  VI  and  XXXIII.) 

Figure  8.     A  fragment  showing  the  upper  portion  of  a  small  individual  with 
fine  reticulation,  broad  apertural  surface  and  four  low  subcircular 
nodes, 
Chemung  group.      Wellsville,  New  York. 


204  DicryospoNGiD^ 

PLATE  HI. 
HYDNOCERAS,  Courad. 

Page  %. 
fSc.PUteill,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XLVI.) 

Hydnocebas  tuberosum,  Conrad. 

Pnee  97. 
(See  Plates  II,  IV,  V,  VI,  VII,  VIII,  XXXVIII.) 

Figure  1.  The  lower  portion  of  a  rather  large  individual  in  which  the  first 
horizontal  row  of  nodes  has  developed  with  regulai-ity,  but  the 
second  and  third  rows  have  been  disordered  by  the  appearance  of 
three  nodes  out  of  their  normal  position.  It  wil]  be  observed  that 
owing  to  this  malformation  and  a  defect  in  the  net-work  near  the 
center  of  the  specimen,  the  vertical  strands  at  the  left  of  the  drawing 
become  continuous  with  and  form  the  horizontal  strands  on  the  more 
remote  node,  lying  next  to  the  uppermost  one  on  that  side.  This 
peculiar  structure  is  very  clearly  shown  on  the  specimen. 

Figure  2.  An  incomplete  specimen  which  has  been  somewhat  compressed 
vertically  and  the  depth  of  the  horizontal  constrictions  thereby 
increased.  The  nodes  have  here,  also,  been  somewhat  irregular  in 
their  appearance,  one  in  the  third  row  occurring  out  of  its  vertical 
position,  and  its  representative  in  the  fourth  row  being  omitted. 

Figure  4.     End  view  of  a  large  specimen  having  but  seven  nodes. 
Chemung  group.     Brown  hill,  near  Cohocton,  New  Yorh. 

Hydnoceras  bathense,  sp.  nov. 

Page  101. 
(See  Plate*  VIII  and  IX. ) 

Figure  3.     An  incomplete  individual  of  average  size,  showing  the  prominent 
nodes. 
Chemung  group.     Jenhs  quarry,  Bath^  New  York. 


DK/rVOSPONOID.V. 


Memoirs  Geological  Survey  of  New "iijrk. 


Litem. 


G.  B.Sirnpsunrlcl 


C  Fdusel  liih 


206  DlClTOSPONGIDA 

PLATE  IV. 
HYDRIODICTYA,  gen.  nov. 

Page  77. 
(SeePlatesV,  XVI,  XXI.) 

Hydbiodictya  patula,  Hall  (sp.). 

Page  78. 

Figure  1.     A  view  of  the  original  specimen ;  showing  the  smooth,  expanding 
surface,  regular  aperture  and  sharply  defined  reticulation. 
Chemung  group.     Brown  hill,  near  ColiocUni,  N.  Y. 

HYDNOCERAS,  Conrad. 

Tape  05. 
(Seo  Plates  II,  III,  V,  VI,  VII,  VIII,  IX,  X,  XI.  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XLVI  ) 

Hydnoceras  tuberosum,  Conrad. 

Piigo  97. 
(See  Plates  11,  III,  V,  VI,  VII,  VIII,  XXXVIII.) 

Figure  2.     A  plaster-cast  of  the  original  specimen  of  the  sj^ecies. 

Chemung  group.     Said  to  have  come  from  near  AMlson,  Steuben 
county,  New  Yovh. 

Figure  3.     A  very  hroad  and  somewhat  compressed  example  with  the  reticu- 
lation and  subprismatic  form  finely  preserved. 

Figure  4.  A  small  specimen  with  the  normal  number  of  four  horizontal  rows 
of  low  nodes,  the  summit  of  each  of  which  is  incomplete,  indicating 
the  presence  of  spicular  tufts  at  these  points.  The  specimen  lies  in  its 
matrix,  which  retains  the  impressions  of  the  pointed  tufts  on  the 
lateral  nodes  and  the  vertical,  erect  expansions  connecting  them. 
These  were  probably  also  present  in  varying  degrees  of  development 
on  all  of  the  principal  vertical  and  horizontal  spicular  bands.  An 
interesting  abnormality  is  shown  on  the  lo\ver  right  margin  of  the 
specimen  Avhere  a  very  strong  spicular  tuft  exists  between  two  nodes 
in  their  normal  j^osition,  and  there  is  only  very  obscure  evidence  of 
an  intercalary  node  at  this  point.  The  tuft  near  the  base  of  the 
sponge  represents  another  very  faintly  developed  node. 
Chemung  group.     Brown  hill,  near  Cohocton,  New  Yorh. 


nU  TVOSPOXOID.V. 


Memoirs  Geological  Survey  of  New  York. 


Plate  IV. 


ii.  Cm  If.,  lull  riel 


(.;|.jU.-;el.[Miv 


208  DlCTYOSPONGin.E. 


PLATE  V. 
IIYDRIODICTYA,  gen.  nov. 

Page  77. 
(See  PUtee  IV,  XVI,  XXI.) 

Hydriodictya  cylix,  sp.  nov. 

Page  78. 
(See  Plate  XVI.) 

Figures  1,  2.     Opposite  sides  of  a  small,  essentially  entire  specimen,  showing 
the  character  of   the  reticulation.      The  surface  of   the  sponge  is 
smooth  and  the  apparent  swellings  and  depressions  wholly  due  to 
compression  in  fossilization. 
Chemung  group.     Deyo  basin,  Naples,  New  Yorh. 

IIYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II,  III,  IV,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII.XXXIX,  XLVI.) 

Hydnocekas  tuberosum,  C'onrad. 

Page  97. 
(See  Plates  II,  III,  IV,  VI,  VII,  VIII,  XXXVIII.) 

Figui'e  8.  A  long,  slender  and  nearly  entire  individual,  bearing  five  horizontal 
rows  of  nodes  with  an  additional  one  near  the  base  in  an  incipient 
condition ;  showing  the  smooth  surface  in  the  apertural  region,  the 
regular  aperture  and  the  tufts  at  each  node,  which  have  been  drawn 
in  from  impressions  left  upon  the  matrix. 

Figure  4.     A  stouter,  more  sharply  prismatic  specimen  with  more  extended 
tufts,  retaining  also   the  vertical   lamellae  connecting  them.      The 
cavities  left  by  the  bases  of  these  tufts  are  shown  at  the  summits  of 
some  of  the  nodes. 
Chemung  group.     Brawn  hill,  near  CoJiocton,  New  Yorh. 


1)TC  TYOSPOXOIl>A\ 


Memoirs  Geological  Survey  of  New  York . 


PlateV 


E  Errimons  del 


Philip  Asi  hth 


210  UlC'TYoSl'UiNOlD.E. 


PLATE  VI. 
HYDNOCERAS,  Conrad. 

Pagfi  95. 
(See Plate.  II,  UI,  IV,  V,  VII,  VIII,  IX,  X,XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XLVI.) 

Hydnoceeas  antiiracis,  sp.  nov. 

Page  109. 
(See  Plates  II,  XXXIII.) 

Figure  1.  A  view  of  the  upper  portion  of  an  individual,  showing  the  small 
size  of  the  species,  its  prismatic  form,  broad  and  smooth  apertural 
surface  and  subcircular  nodes.  On  the  right  of  the  specimen  the 
extension  of  these  nodes  and  their  tufts  into  the  matrix  is  shown. 

Figure  2.     The  opposite  side  of  the  same  specimen. 
Chemung  group.     Scio,  Neio  York. 

Hydnoceras  tuberosum,  Conrad. 

Page  97. 
(See  Plates  II,  in,  IV,  V,  VU  VIII,  XXXVIII.) 

Figure  3.     Two   small  individuals   in  a   block  of   sandstone  which  retains 
impressions  of  the  spicular  tufts  and  lamellae. 
Chemung  group.     Brown  hill,  'near  Cohocton,  New  Yorh. 

Hydnoceras  phymatodes,  sp.  nov. 

Page  104. 
(See  Piute  XXXVII.) 

Figures  4,  5.  Opposite  sides  of  a  typical  specimen  having  seven  horizontal 
rows  of  low  nodes ;  in  the  second  and  third  rows  there  being  eight 
nodes,  in  the  fourth,  nine,  in  the  fifth  and  sixth,  eleven.  In  the  fifth 
horizontal  row  in  figure  5,  an  intercalary  node  begins  a  new  vertical 
row. 
Chemung  group.     Bath,  New  Yorh. 


l>TCTV(>SlM>NCiIl>A- 


Memoirs  Geological  Survey  of  New  York 


Plate  VI 


(j.B  bimpson  del 


C  Fausei  Inh. 


212  DicrYosi'uNuiu>E. 


PLATE  VII. 

HYDNOCERAS,  Conrad. 

Pago  95. 
(S«o  Plates  II,  III,  rV,  V,  VI,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XL, VI.) 

Hydnoceeas  hypastrum,  sp.  nov. 

Page  105. 

Figure  1.  A  transverse  section  viewed  from  beneath ;  showing  the  circular 
body  of  the  species,  and  its  strong,  elongate  pendant  nodes  arranged 
in  pairs  elevated  on  stout  bases. 

Chemung  group.      Middle  or    southern  part  of   Steuben   county, 
New  Yorh. 

Hydnoceras  tuberosum,  Conrad. 

Patfe  97. 
(See  Plates  II,  III,  IV,  V,  VI,  VIII,  XXXVIII.) 

Figure  2.  The  basal  portion  of  a  small,  nearly  entire  individual  with  Ioav 
nodes  and  of  irregular  growth  on  one  side ;  showing  the  abrupt 
appearance  of  an  intercalary  node  near  the  top. 

Figure  3.  An  entire  specimen  in  which  the  growth  has  been  interrupted, 
evidently  by  some  accident,  and  the  reticulum  repaired,  not  how- 
ever, without  the  disturbance  and  obliteration  of  most  of  the  nodes. 
There  is  no  actual  break  or  tear  in  the  reticulum  as  might  be  inferred 
from  the  appearance  of  the  drawing  near  the  top. 
Chemung  group.     Brourn  hill,  near  Cohocton,  New  YorJc. 

Hydnoceras  tuberosum,  Conrad,  var.  glossema,  var.  nov. 

Page  101. 
(See  Plate  IX.) 

Figures  4,  5.      Opposite  sides   of    a   nearly  entire  example,  having   a   very 
slender  form  and  six  horizontal  rows  of  sharp  nodes  which  have 
been  made  to  appear  i-ather  too  broad  at  the  lateral  mai'gins. 
Chemung  group.    Southwestern  New  Yorh  {exact  locality  uncertaiti). 


1>  IC  T  VO  S  1»  OX  O I  l>iS. 


Memoirs  Geological  Survey  of  New  York. 


PlateVll. 


G.B-  Simpson  del 


Philip  Ast  'lith 


214  DiCTYOSPONOID.E. 


PLATE  VIII. 
HYDNOCERAS,  Conrad. 

Page  95. 
(Sm  PUtee  II,  III,  IV, V,  VI,  VII,  IX,  X.XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII  XXXIX,  XLVI.) 

Hydnoceras  tuberosum,  Conrad. 

Page  97. 
(Bee  Plates  II,  III,  IV,  V,  VI,  VII,  XXXVCII.) 

Figure  1.     The  basal  extremity  of  a  large  individual,  showing  the  very  obscure 
nodes  and  sharply  defined  prism-faces. 

Figure  2.  A  nearly  entire  individual  with  four  well  defined  horizontal  rows 
of  low  and  sharp  nodes,  and  broad,  gently  and  regularly  expanded 
apertural  region  bearing  only  two  small  nodes  belonging  to  the 
median  vertical  series.  Near  the  apex  of  the  specimen  the  prismatic 
faces  are  more  obscure  than  elsewhere. 
Chemung  group.     Brown  MU,  near  Coliocton,  New  Yorh. 

Hydnoceras  sp. 

Page  111. 

Figure  3.     An  imperfect  and  irregular  fragment  representing  a  moderately 
large  undescribed  species  of  this  genus,  from  an  liorizon  where  such 
forms  are  of  very  rare  occurrence. 
Chemung  group.      Wellsville,  New  Yorh. 

Hydnoceras  Bathense,  sp.  nov. 

Page  101. 
(See  Plates  III  and  IX.) 

Figure  4.     A  nearly  entire  example,  with  four  horizontal  rows  of   strong 
nodes ;  showing  the  fine  reticulation  of  the  surface  and  the  obsolescence 
of  the  prismatic  faces. 
Chemung  group.     Jenhs  qtiarry,  Bath,  Ncio  York. 

Hydnoceras  eutiieles,  sp.  nov. 

Page  105. 
(See  Plate  XXXIII.) 

Figures  5,  6.     Two  fragments  representing  like  parts  of  two  individuals  and 
showing  the  broad  form  of  the  sponge  and  tlie  narrow,  extended 
and  drooping  nodes. 
Chemung  group.     Cotton  hill,  near  Avoca,  Nexv  YorTc. 


r>TCTYOSl»OXOII>A- 


Memoirs  Geological  Survey  of  New  York . 


PlateVm. 


G.B  Simpson  del 


Plulip  Ast  hth 


Explanations  of  Plates. 


215 


Hydnocekas  LuTiiEiii,  sp.  nov. 

Page  106. 

Figure  7.     A  young  and  slender  specimen  with  low  nodes. 

Figure  8.     An  entire  individual  with  three  well  developed  rows  of  nodes  and 

entire  aperture. 
Figure  9.     A  mature  individual  with  four  horizontal  rows  of  nodes. 
Figure  10.     A  full-grown  exainple  with  strong  nodes  in  three  rows  and  an 

accessory  node  in  the  apertural  region. 

Chemung  group.     8egur''s  gully,  Italy,  Yates  county,  New  YorJc. 


21G  DlCTYOSPONGID^E. 

PLATE  IX. 

HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II,  III,  IV,  V,  VI,  VII,  VIII,  X,  XI,Xir,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XLVI.) 

Hydnocebas  Batiiense,  sp.  nov. 

Page  101. 
(See  Plates  III,  VIII.) 

Figure  1.  A  large,  somewhat  incomplete  individual  with  six  rows  of  strong 
rounded  nodes,  showing  some  interesting  peculiarities  of  develop- 
ment. In  the  fourth  row  one  node,  instead  of  appearing  at  its  proper 
place,  stands  at  the  middle  of  the  prism-face,  and  in  the  fifth  row 
the  node  of  a  different  vertical  series  is  wanting,  while  in  the  sixth 
row,  so  far  as  it  is  preserved,  there  is  but  a  single  well  defined  node. 
Chemung  group.     Jenks  quarry^  Bath,  New  Yoi'h. 

Hydnocebas  tubeeosum,  Conrad,  var.  glossema,  var.  nov. 

Page  101. 
(See  Plate  VII.) 

Figure  2.  In  this  example  there  is  a  considerable  area  at  the  base,  where  an 
incipient  row  of  nodes  is  represented  only  by  a  series  of  broad  swell- 
ings at  the  prism-angles.  Three  horizontal  rows  are  well  developed, 
and  in  the  apertural  region  is  evidence  of  another  row  of  smaller 
nodes.  The  two  lower  rows  (not  including  the  incipient  row  near 
the  base)  have  seven  nodes,  while  that  immediately  above  has  eight, 
the  eighth  vertical  row  appealing  abiniptly,  as  shown  in  the  figure. 
The  reticulation  is  sharply  defined,  but  not  finely  divided.  The  sur- 
face shows  impressions  of  worm-like  markings  which,  as  elsewhere 
suggested,  may  be  the  tube  of  some  commensal  annelid. 
Chemung  group.     Southwestern  New  York  (exact  locality  unlcncnun). 

Hydnocebas  Avoca,  sp.  nov. 

Page  103. 
(See  Plates  II,  X,  XI.) 

Figure  3.  An  imperfect  example  with  three  fully  develojied  horizontal  rows 
of  large  and  short  nodes  with  subsemicircular  marginal  outline,  and 
an  inceptive  basal  row.  The  prismatic  faces  are  obscured,  except  over 
the  basal  region,  and  the  horizontal  constrictions  are  deep  and  naiTow. 
Chemung  group.     Near  Avoca,  New  Yo)-k. 


218  DlCTlOSPONULDJE. 


PLATE  X. 

HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II.  Ill,  IV,  V,  VI,  VII.VIII,  IX,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XLVI.) 

Hydnoceras  Avoca,  sp.  nov. 

Page  103. 
(See  Plates  II,  IX,  XI.) 

Figure  1.  A  large,  somewhat  incomplete  example  with  five  horizontal  rows 
of  fully  developed  nodes  and  an  incipient  row  near  the  base.  This 
specimen  shows  the  great  size  of  the  nodes,  their  semielliptical  or 
semicircular  outline  and  the  deep  and  narrow  horizontal  constrictions 
by  which  they  are  separated. 
Chemung  group.     Near  Avoca,  New  York. 


i)TCTVOsi»o>roir)i?. 


Memoirs  Geological  Survey  of  New  York. 


Plate  X 


G.B  Simpson  del 


Philip  Ast  hth 


290  DlCTYOSPONGID^ 


PLATE  XI. 
HYDNOCERAS,  Conrad. 

Page  95. 
(Se«  Plates  H,  HI,  IV,  V,  VI,  VII,  VHI,  IX,  X,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIH,  XXXIX  XLVl.) 

Hydnoceras  Avoca,  sp.  nov. 

Page  103. 
(See  Plates  II,  IX  and  X.) 

Figure  1.  This  specimen  represents  the  apertural  and  upper  portion  of  a 
gigantic  individual  of  this  species,  whose  probable  original  propor- 
tions are  partially  restored  in  outline.  The  sponge  is  somewhat 
irregularly  compressed,  and  the  nodes  upon  its  surface  have  been  flat- 
tened without  destroying  their  characteristic  expression  m  this 
species.  The  specimen  was  taken  from  the  foundation  wall  of  an  old 
mill  at  Avoca,  and  its  exact  locality  is  not  known,  though  doubtless 
from  the  beds  in  that  vicinity  which  have  produced  most  of  the 
known  examples  of  this  species. 
Chemung  group.     Avoca,  New  Yorh. 


< 


222  •  DlCTYOSPONOID^ 


PLATE  XII. 
HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plate*  11,  HI,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XXXIII,  XXXVI,  XXXVU,  XXXVIII,  XXXIX,  XLVI.) 

Hydn^oceras  eumeces,  sp.  nov. 

Pago  no. 

Figures  1,  2,  3.  Three  fragments  of  one  specimen  which  consists  of  a  large 
cup  greatly  compressed  along  the  vertical  axis.  Figure  1  shows  a 
portion  of  the  internal  cast,  figures  2  and  3  impressions  of  the 
exterior,  indicating  the  depth  of  the  horizontal  constrictions,  and 
showing  the  sharp,  tufted  nodes  and  the  character  of  the  reticulation. 
Chemung  group.     Alfred,  New  Yorh. 


OICTVOSPOXOIUA- 


Memoirs  Geological  Survey  of  New  York 


Plate  Xri. 


G.B  Simpson  del 


Philip  Ast  lith 


224  DiCTYOSPONGIDiE. 


PLATE  XIII. 

BOTRYODICTYA,  gen.  nov. 

Page  111. 

BoTBTODicTYA  RAMOSA,  Lesquereux  (sp.). 

Page  111. 
Figure  1.     A  compressed  and  imperfect  specimen  showing  the  three  rows  of 
compound  nodes  and  the  single  row  of  simple  nodes  in  the  apertural 
region.     These  nodes  have  all  been  much  flattened  and  considerably 
distorted.     The  pedicel  is  partially  retained. 

Figure  2  A  transverse  section  of  a  specimen  showing  a  portion  of  the 
apertural  nodes  in  their  normal  extent  and  form. 

Figure  3.     The  other  side  of  the  specimen  represented  in  figure  1. 

Figure  4.  A  laterally  compressed  specimen  viewed  from  the  base  of  the 
cup,  showing  the  outward  extent  of  the  nodes. 

Figure  5.  An  apical  view  of  a  guttapercha  cast  taken  from  the  interior  of 
the  base  of  the  cup;  showing  that  the  vertical  bands  of  the  cup 
converge  to  an  apex  and  form  a  distinct  internal  reticular  wall  or 
diaphragm  at  this  plane. 

Figure  6.  A  specimen  having  an  imusually  large  vase  and  slender  pedicel. 
This  example  shows  the  obsolescence  of  the  prismatic  faces,  the 
erect,  bilobed  or  simple  subapertural  nodes  and  the  pendulous 
compound  nodes  of  the  lower  rows. 

Figure  7.  The  opposite  side  of  the  same  individual,  only  a  portion  of  the 
surface  and  lobes  being  exposed,  more  below  than  above.  The  erect 
node  at  the  summit  is  apparently  a  division  of  one  of  the  apertural 
nodes. 

Figure  8.  The  most  complete  individual  obtained ;  showing  the  form  of  the 
cup  and  its  nodes,  and  a  portion  of  the  long,  smooth  reticulate  pedicel. 
Chemung  group.     Laivrenceville,  Tioga  county,  Pennsylvania. 


nTCTvosi»oxcrinA\ 


Memoirs  Geological  Survey  of  New  York 


Plate  XIII 


G.B  Simpson  del 


Plulip  Ast  h:h 


22G  DlOTYOSPONGIDyE. 

PLATE  XIV. 
CLATHROSPONGIA,  Hall. 

Piige  121. 
(See  Plates  XV,  XVni,  XXI,  XXV,  XLIX,  L.) 

Clathrospongia  (?)  Hamiltonensis,  Hall  (sp.). 

Pago  66. 

Figure  1.  The  original  specimen,  wliich  is  a  calcareous  replacement  of  the 
skeleton;  showing  the  strong  primary  bands  and  the  imperfectly 
retained  lesser  bands ;  also  exhibiting  the  minute  reticulation  over 
small  patches  near  the  middle  of  the  surface. 

In  the  calcareous  shales  of  the  Hamilton  group.     Canandaigua  lake, 
I^ew  York. 

Clathrospongia  fekestkata,  Hall  (sp.). 

Pago  122. 
(See  Plate  XXV.) 

Figure  2.     A  view  of  the  original  specimen,  from  the  drawing  given  in  the 
Sixteenth  Annual  Report  of  the  N.  Y.  State  Cabinet  of   Natural 
History,  1863,  pi.  3,  fig.  4. 
Chemung  group.      Chemung  narrows^  New  York. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plate*  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI,  I.XI.) 

Dictyospongia  sceptrum,  Hall  (sp.). 

Pago  TJ. 
(See  Plates  XV,  XVI,  XXVJI,  XXXVI.) 

Figure  3.  A  slender,  well  preserved  and  nearly  entire  individual,  showing 
the  prevailing  form  of  the  species  and  the  elongate  rectangular  shape 
of  the  primary  meshes  of  the  net- work. 

Figure  4.     A  specimen  of  similar  character,  but  accidentally  curved. 
Chemung  group.     Friendship,  New  York. 

Figure  5.  A  somewhat  smaller  example  preserving  the  finer  divisions  of  the 
reticulum. 

Figure  6.     Two  very  small  examples  regarded  as  the  young  of  this  species. 
Chemung  group.      Wellsville,  New  Yoi-k, 


1>IC   TVOSPONCwIUA- 


Memoirs  Geological  Survey  of  New  York. 


Plate  XIV. 


G.B  Simpson  del 


Phihp  Ast  hih 


228  DlCTYOSPONGID^^ 

PLATE  XV. 
DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(Sm  Plates  XIV,  XVI,  XXVn,  XXXIII,  XXXVI,  XXXVIII,  XU,  XLII.  XLIV,  XLVl,  LV,  LVI,  LXI.) 

DiCTYOSPONGIA    EUMORPIIA,    Sp.    nOV. 
Page  75. 

Figure  1.     A  small,  entire  individual,  showing  the  fine,  square  meshes  of  the 

net- work. 
Figure  2.     A  larger,  more  elongate   and    slender  example. 
Figure  3.     A  broader  cup  with  the  fine  reticulation  retained  over  but  a  part 

of  the  surface. 

Chemung  group.      Wellsville,  New  YorJe. 

DicnrospoNGiA  charita,  sp.  nov. 

Page  74. 

Figure  4.     An  entire  sponge  with  the  basal  tuft  retained. 
Chemung  group.      Wellsville,  New  Yorh. 

DiCTYOSPONGIA    LOPHUEA,  Sp.  nOV. 
Page  74. 

Figure  5.     The  original  specimen,  an  entire  cup  showing  the  elongate  rec- 
tangular quadrules  and  the.  extended  basal  tuft. 
Chemung  group.      Wellsville,  New  Yorh 

CLATHROSPONGIA,  Hall. 

Page  121. 
(See  Plates  XIV,  XVIII,  XXI,  XXV,  XLIX,  L.) 

ClATIIROSPONGIA   (?)  IRREGULARIS,  Hall  (sp.). 
Pago  124. 

Figures  6,  7.     Opposite  sides  of  the  original  specimen ;  showing  the  form  of 
the  sponge  and  the  character  of  its  reticulation. 
Chemung  (Portage  ?)  group.     ItJiaca,  New  Yorh. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plate*  XIV,  XVI,  XXVII,  XXXIIt,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI,  LXI.) 

DiCTYOSPONGIA   SCEPTRUM,  Hall  (sp.). 

Page  73. 

(See  Plates  XIV,  XVI,  XXVII,  XXXVI.) 

Figures  8,  9.     Opposite  sides  of  a  large,  somewhat  flattened  individual,  show- 
ing the  characteristic  surface  reticulation. 
Chemung  group.     Friendsliijj,  New  York. 


1>ICTVC>  Sl»<>XO  1 1>A« 


Memoirs  Geological  Survey  of  New  York. 


PlaleXV 


te^ 


1 


tsa 


G.B  Simpson  del 


Philip  Ast  liih 


230  DlCTYOSPONGID^ 

PLATE  XVI. 
RHABDOSISPONGIA,  gen.  nov. 

Page  116. 
(See  Plate  XLVII.) 

Rhabdosispongia  Amalthea,  Hall  (sp.). 

Page  116. 

Figure  1.     A   nearly  entire  individual   with   broad  annulations  and   sharp 
vertical  ridges.     Drawn  from  a  plaster  cast  of  the  original  specimen 
which  was  from  the 
Chemung  group.     Great  Bend,  Susquelmnna  county,  Pennsylvania. 

HYDRIODICTYA,  gen.  nov. 

Page  77. 
(See  Plates  IV,  V,  XXI.) 

Htdriodictta  cylix,  sp.  nov. 

Page  78. 
(See  Plate  V.) 

Figure  2.     The  external  impression  of  a  large  individual  somewhat  crumpled 
by  compression ;  showing  the  prevailing  coarse  and  square  meshes 
of  the  reticulum.     Near  the  middle  of  the  sponge  a  break  in  the 
net-work  has  been  repaired  by  highly  irregular  spicular  bands. 
Chemung  group.     Deyo  hasin,  Naples,  New  York. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI,  LXI.) 

DiCTYOSPONGIA    SCEPTUUM,  Hall  (sp.). 
Pago  73. 
(See  Plates  XIV,   XV,  XXVII,  XXXVI.) 

Figures  3,  4.  Opposite  sides  of  a  slender  but  imperfect  Sf)eciraen  ;  showing 
a  difference  in  the  preservation  of  the  reticulum.  In  figure  4,  the 
exposed  side  is  a  shai-p  internal  cast  and  clearly  shows  the  elongate 
rectangles  produced  by  the  intersection  of  the  primary  lamellae, 
while  the  surface  represented  in  figure  3  is  involved  in  a  muddy  or 
shaly  sediment  which  has  retained  the  finer  net-work  of  the  outer 
wall. 
Chemung  group.     Friendsliijy,  New  Yorh. 


nic  TV(>si»oxoii>i¥.. 


Meinoirs  Geological  Survey  of  New  York 


PlateXVI 


G.B  Simpson  del 


Philip  Ast  liih 


232  DlCTYOSPONGIDJS. 

PLATE  XVII. 
PRISMODICTYA,  gen.  nov. 

Page  79. 
(S««  Plates  XVUI,  XIX,  XX,  XXI,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XLI,  XLII,  LV.) 

Prismodictya  pkismatica,  Hall  (sp.). 

Page  83. 
(See  Plates  XXI  and  XLII.) 

Figure  1.     An  incomplete  specimen  showing  a  tendency  to  spiral  growth. 
Chemung  group.      Wellsville,  New  Yo?'k. 

Prismodictya  banano,  sp.  nov. 

Page  87. 

Figure  2.     The  type-specimen,  which   is  an  essentially  entire  individual  oi 
very  slender  and  gracefully  tapering  form. 
Chemung  group.     Scio,  Allegany  county,  Kew  YorTc. 

Pbismodictya  telum,  Hall  (sp.). 

Page  8C. 
(See  Plates  XXXV  and  XLII.) 

Figure  3.  Two  individuals,  one  of  them  entire  and  with  the  normal  propor- 
tions of  the  species,  the  other  somewhat  imperfect  and  having  a 
curvature  which  appears  to  be  the  result  of  natural  growth.  Both 
bear  casts  of  spiral  vermiform  tubes  or  furrows. 

Figures  4,  5.  Opposite  sides  of  an  entire  individual.  The  surface  represented 
in  figure  4,  is  somewhat  worn  but  shows  distant  obscure  horizontal 
grooves ;  that  shown  in  figure  5,  is  slightly  broken  at  the  aperture 
and  has  the  horizontal  grooves  represented  by  sharper  reticular 
furrows. 
Chemung  group.      Wellsville,  New    Yorlc. 

Figure  6.  An  entire  individual  of  small  size  and  undisturbed  proportions, 
showing  the  character  of  the  reticulation  and  the  form  and  size  of 
the  aperture.  Here  as  in  figures  5  and  7,  the  prism-faces  are  made 
to  appear  somewhat  too  concave. 

Figure  8.     Basal  view  of  the  same  specimen. 

Chemung  group.     Cattaraugus  county,  New  York. 

Figure  7.     A  small  entire  example. 


I 


1)  I C  T  YO  S  I*  ai^  0 1 1  >  A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  XVII 


G,B  Simpson  de 


PhihpAst  Jiih 


Explanations  of  Plat>».  288 

Figures  9,  10.  Opposite  sides  of  an  entire  specimen  showing  the  marks  of 
encircling  or  spiral  tubes,  probably  made  by  a  commensal  annelid. 
On  one  side  of  the  specimen  (fig.  9)  these  are  preserved  as  ridge-like 
casts  and  on  the  other  (fig.  10)  as  furrows. 

The  specimens  represented  in  figures  3,  4,  5,  7,  9  and  10  are  from 
a  single  slab  crowded  with  individuals  of  this  species. 
Chemung  group.      Wellsville,  New  Yorh. 

Pbismodictta  cobynia,  sp.  nov. 

Page  8B. 

Figure  11.     The  upper  portion  of  a  very  large  sponge  with  broadly  expanded 
pnsm-faces  and  a  coarse  square  reticulation. 
Chemung  group.      Wellsville,  New  YorK 


234  DiCTYOSPONGIDJE, 


PLATE  XVIII. 
PRISMODICTYA,  geu.  nov. 

Page  79. 
(See  Plates  XVII.  XIX,  XX,  XXI,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XLI,  XLII,  LV.) 

Pbismodictya  baculum,  Hall  (sp.). 

Page  81. 

Figures  1,  2.     Opposite  sides  of  an  incomplete  sponge  which  is  referred  to 
this  species. 
Chemung  group.     Alfred,  New  York. 

Figure  3.     The  type-specimen,  an  incomplete  cup,  slender  near  the  base  and 
expanded  toward  the  aperture.     The  specimen  is  a  sharp  internal 
cast,  showing  the  regularity  of  the  primary  reticulation. 
Chemung  group.      Wellsville,  New  York. 

Prismodictya  Coneadi,  Hall  (sp.). 

Page  91. 
(See  Plate  XIX.) 

Figure  4.     Lateral  view  of  the  original   specimen;   showing  the  regularly 
expanding   form,  the   character  of    the  reticulation  and,  over   the 
upper  part  of  the  sponge,  the  somewhat  irregular  tendency  to  the 
formation  of  nodes  upon  the  prism-faces. 
Chemung  group.     Handolph,  Cattaraugus  county.  New  Yorh. 

CLATHROSPONGIA,  Hall. 

Page  121. 
(See  Plates  XIV,  XV,  XXI,  XXV,  XLIX,  L.) 

Clatiikospongia  (?)  TOMACULUM,  Hall  (sp.). 

Page  123. 
(See  Plate  XXI.) 

Figures  5,  6.  Opposite  sides  of  the  type-specimen,  which  is  a  portion  of  a 
large,  very  gradually  expanding  sponge  with  low  and  nan-ow  prism- 
faces.  The  reticulation  is  coarse  and  over  most  of  the  exposed 
surfaces  shows  a  prevalence  of  square  meshes.  The  curvature  of 
the  specimen  is  probably  accidental. 
Chemung  group.     Alfred,  New  Yorh. 


J3ICTVOSl»OXOIl>iV^. 


Memoirs  Geological  Survey  of  New  York. 


PlateX\1II. 


G.B  Simpson  del 


Philip  Ast  hth 


236  DiCTTOSPONGIDiE. 

PLATE  XIX. 
PRISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVII,  XVIII,  XX,  XXI,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XLI,  XLII,  LV.) 

Prismodictya  choanea,  sp.  nov. 

Page  85. 
(See  Plates  XXVII  and  XXXV.) 

Figure  1.  The  basal  portion  of  a  specimen  of  medium  size,  showing  a  fine 
reticulum  and  faint  traces  of  prism-faces. 

Figure  2.     A  similar  portion  of  a  larger  specimen. 

Chemung  group.     Iscliua,  Cattaraugus  county,  New  Yorh. 

Figure  3.  The  upper  portion  of  a  large  cup,  believed  to  represent  this 
species;  having  the  prism-faces  obsolescent  and  an  extremely  fine 
reticulum.  This  specimen  is  accompanied  by  fragments  of  others 
of  the  same  species  and  impressions  of  the  brachiopod,  Spirifer 
dhjunctiis. 
Chemung  group.     Almond,  Allegany  county,  New  York. 

PlilSMODICTYA    PARALLELA,  Hall  (sp.). 

Page  82. 

(See  Plates  XX,  XXI,  XXXV.) 

Figure  4.  The  exterior  of  a  portion  of  a  cup,  showing  its  gradual  expansion, 
the  somewhat  obscure  and  rounded  prism-faces  and  the  sharply 
elevated  primary  reticular  bands.  This  is  drawn  from  a  plaster- 
cast  of  the  original  specimen,  which  is  an  external  impression  in 
sandstone. 

Chemung    group.       Southwestern     New      Yorh     (^frecise     locality 
unhnoi07i). 

Pkismodictya  Conradi,  Hall  (sp.). 

Page  91. 
(See  Plate  XVIII.) 

Figure  5.     An  apical  view  of  the  basal  extremity  of  the  specimen  shown  on 

plate  18,  figure  4. 
Figure  6.     Side  view  of  the  same  specimen  presenting  its  narrower  diameter 

and  showing  the  irregular  node  at  the  top. 

Chemung  group.     Randolph,  Cattaraugus  county,  New  Yorh. 


DICTVOSlM>NCiIl>A' 


Memoirs  Geological  Survey  of  New  York. 


Plate  XIX 


G.D  53 1 1 apsoix  de 


PlulipAst  hlh 


Explanations  ok  Plates.  237 

TIIYSANODICTYA,  gen.  nov. 

Ta-e  125. 
(See  Plates  XXIII,  XXIV,  XXV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII,  LIU.) 

TlIYSANODICTYA    KUDIS,  ILlU  (sp.). 

Page  128. 

(See  Plate  XXIII  ) 

Figui-e  7.     A  fragment  of  a  cup  showing  the  coarse,  square  primary  spicular 
bands  and  the  very  fine  subordinate  reticulation. 
Chemung  group.      Wellsville,  New  York. 


2',\S  DlCrYOSPONGID^ 


PLATE    XX. 

PRISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVir,  XVIII,  XIX,  XXI,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XW,  XLII,  LV.) 

Peismodictta  cercidea,  sp.  nov. 

Page  89. 
(See  Plate  XLII.) 

Figure  1.  An  entire  flattened,  though  symmetrical  specimen,  with  subequal 
extremities.  It  is  possible  that  the  spot  at  the  lower  end  may 
represent  the  aperture  and  the  specimen  have  thus  been  drawn  in  an 
inverted  position,  but  this  is  uncertain  as  the  margins  of  this  spot 
are  not  so  clearly  defined  as  here  represented  and  the  other  extremity 
is  not  fully  uncovered. 

Figures  2,  3.     Opposite  sides  of  another  example  which  has  been  obliquely 
compressed. 
Chemung  group.      Wellsville,  New   York. 

Pkismodictya  Allegania,  sp.  nov. 

Page  84. 

Figure  4.     A   fragment   of    a   very  large  individual   showing   the   upward 
expansion  of  the  sponge,  the  median  division  of  the  prism-faces  and 
the  fine  reticulation. 
Chemung  group.     Scio,  New  YorTc. 

Pkismodictya  speotabilis,  sp.  nov. 

Page  82. 

Figure  .5.  A  nearly  entire  specimen  representing  a  fine  sponge,  which  attains 
its  greatest  expansion  shortly  above  the  base  and  is  regularly  pris- 
matic for  the  rest  of  its  extent.  The  prism-faces  are  not  developed 
near  the  base  of  the  cup,  although  thus  represented  in  the  lithograph, 
but  the  surface  about  the  apex  is  smooth  and  round  for  a  consider- 
able distance. 
Chemung  group.      Wellsville,  New  Yoi^h. 

Pkismodictya  pakallela,  Hall  (sp.). 

Page  82. 

Figure  6.     Transverse  section  of  an  uncompressed  specimen. 


nicTvo  si><>xcii  i>iV. 


Meinoirs  Geological  Survey  of  New  York. 


Plate  XX 


G.B  Simpson  del 


Philip  Ast  lith 


Explanations  ok  Platks. 


239 


PkiSiMODictya  aulopiiia,  sp.  nov. 

Page  88. 
(See  Plates  XXXIV  and  XLI.) 

Figure  7.     The  basal  portion  of  a  weathered  specimen ;  showing  the  concave 

prism-faces. 
Figure  8.     A  fragment  of  the  median  portion  of  a  specimen ;  showing  the 

concave  prism-faces  and  the  extremely  fine  surface  reticulation. 

Chemung  group.      Wellsville,  New  York. 


240  DKriYOSlH)NGID/E. 


PLATE  XXI. 

PKISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVII,  XVIII,  XIX,  XX,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVl,  XLI,  XLII,  LV.) 

Prismodictya  prismatica,  Hall  (sp.). 

Page  83. 
(See  Plates  XVII  and  LXII.) 

Figure  1.  The  apertural  end  of  an  average  specimen;  showing  tlie  slight 
contraction  toward  the  margin. 

Figure  2.     A  portion  of  a  small  and  very  slender  example. 

Figure  3.     A  portion  of  an  individual  having  a  tendency  to  spiral  gro^vth. 

Figure  4.     A  nearly  entire  individual  of  average  size. 

Figure  5.  An  enlargement  of  the  surface  of  an  internal  cast ;  showing  the 
character  of  the  reticulation  and  the  rows  of  depressions  on  the 
prism-angles  at  their  intersection  with  the  principal  horizontal 
spicular  bands.  Such  pits  probably  represent  the  bases  of  spicular 
tufts. 

Figure  6.     Another  internal  cast  also  showing  vertical  rows  of  pits. 

Chemung  group.     Concord  station,  Erie  county,  Pennsylvania. 

Pkismodictya  FiLiTExnus,  Hall  (sp.). 

Page  86. 
(See  Plate  XXVII.) 

Figure  7.  A  gutta-percha  squeeze  of  the  original  specimen  which  is  an 
external  cast  showing  a  fine  reticulum  and  very  obscure  prism-faceSo 
Chemung  group      Steuben  county,  New  Yorh 

Pkismodictya  pakallela,  Hall  (sp.). 

Page  82. 
(See  Plates  XIX  ami  XXXV.) 

Figure  8.     A  fragment  of  a  rather  large,  uncompressed  specimen. 
Figure  9.     Transverse  section  of  the  same  example. 
Chemung  group.      Wellsville,  New  Yorh 


ni(  TVOSl»OXOIl>A- 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXI. 


b;  Em  1  nous  de 


P)iilip  Ast  lull 


Explanations  ok  Plat*s.  241 

IIYDRIODICTYA,  gen.  nov. 

Page  77. 
(See  Plates  IV,  V,  XVI,  LV.) 

IIyduiodictya  nki'iielia,  8|).  nov. 

Page  79. 

Figure  10.     An    incomplete   specimen   shomng   what   appears    to    be    the 
expanded  apertural  part   of   a  vase-shaped  cup  with  subprismatic 
faces. 
Cemung  group,     Scio,  New  YorJc. 

CLATimOSPONGIA,  gen.  nov. 

Page  121. 
(See  Plates  XIV,  XV,  XVIII,  XXI,  XXV,  XLIX,  L.) 

ClATHBOSPOKGIA  (?)  TOMACULUM,  Hall  (sp.). 

Page  123. 

(See  Plate  XVIII.) 

Figure  11.     A  sharply  defined  cast  of  a  portion  of   the  exterior  showing  the 
character  of  the  quadrulation. 
Chemung  group.     Alfred,  N  Y. 

Clathbospongia  vascellum,  Hall  (sp.). 

Page  123. 
(See  Plate  XXV.) 

Figure  12.     The  opposite  side  of  the  specimen  shown  on  the  plate  cited. 
This  surface  exhibits  traces  of  very  coarse  primary  vertical  spicular 
ridges  and  in  places  shows  some  of  the  finer  reticulation. 
Chemung  group.     Alfred,  N.  Y, 


242  Dktyosi'ongid/E. 


PLATE  XXII. 
HELICODICTYA,  gen.  nov. 

Page  114. 
(See  Plate  XXVII.) 

Heucodictya  trypania,  sp.  nov. 

Page  114. 

Figure  1 .     Lateral  view  of   a  compressed  specimen,  showing  the  spiral  ridges 

parallel  to  one  of  the  rectangularly  reticulating  spicular  series. 
Figure  2.     The  same  specimen  with  the  lower  portion  removed  to  show  the 

continuity  of  the  spiral  ridges  on  the  impression  of  the  opposite 

side. 

Chemung  group.      IVellsvllk,  New  Yoi'Jc. 

CERATODICTYA,  gen.  nov. 

Page  117. 
(See  Plates  XLI,  XLII,  XLHI.) 

Ceratodictya  axnulata,  Hall  (sp.). 

Page  120. 

Figures  3,  4.     Opposite    sides   of   a    large   specimen   with   broad   and   low 

annulations. 

Chemung  group.     Deyo  basin,  Naples,  New  Yorh. 
Figure  5.     The  original  figure  of  the  species,  showing  Hve  short  and  subequal 

aimulations. 

Figure  6.     Another  view  of  the  same  specimen. 

Chemung  group.      Western  Neio  Yorlc  (^precise  locality  lost'). 

Ceratodictya  cincta,  Hall  (sp.). 

Page  118. 

Figure  7.     One  side  of  the  original  specimen,  showing  below  the  disordered 

and  interrupted  growth  of  the  annulations. 
Figure  8.     The    opposite   side   of    the  same   fragment,    the   growth    of   the 

annulations  here  being  regular  and  undisturbed. 
Figure  9.     An  end  view  of  the  specimen,  showing  the  compression  to  which 

it  has  been  subjected. 

Chemung  group.     From  a  boulder  at  Erie,  Pennsylvania, 


1>  I C^  T  VO  S  1»  C>  X  OH)  A\ 


Memoirs  Geological  Survey  of  New  York 


Plate  XXn 


G.B  Simpson  del 


Philip  Ast  lith 


244  Dld'YOSPONGID^. 

PLATE  XXIIL 
TIIYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIV,  XXV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII,  LIII.) 

TiiysANODicrrA  rudis,  Hall  (sp.). 

Page  128. 
(See  Plate  XIX.) 

Figure  1.     An  essentially  entire  individual,  showing  the  form  of  the  aperture, 

the  cylindncal  body  and  coarse  and  line  reticulation. 

Chemung  group.     Isehua,  Cattaraugus  county,  New  York. 
Figure  2.     A  basal  disk,  probably  of  this  species,  showing  the  excentric  apex, 

the  concentric  and  apparently  i-adial  striae. 

Chemung  group.     Alma,  Allegany  coimty,  New  Yorlc. 

Figure  3.  A  portion  of  a  specimen  believed  to  be  of  this  species ;  showing 
differences  in  the  preservation  of  the  reticulum  on  various  parts 
of  the  surface. 

Figure  4.  The  opposite  side  of  the  same  specimen  lying  in  its  matrix  and 
showing  the  projection  of  both  vertical  and  horizontal  erect  lamellae. 

Figure  5.     Basal  view  of  this  specimen. 

Chemung  group.      Wdlsvtlle,  New  York 

Figure  6.     The  original  specimen  of  this  species,  Avhich  is  a  nearly  entire 
individual  showing  the  form  and  size  of  the  primary  quadrules  and 
a  part  of  the  finer  reticulation. 
Chemung  group.     Zittle  Genesee,  Allegany  county,  New   York. 

Thysanodictya  Johnstoni,  sp.  nov. 

Page  130. 

Figure  7.  An  entire  specimen,  bent  upward  at  the  base.  The  body  of  the 
cup  is  covered  with  low  nodes  of  shaly  matter  Avhich  appears  to  have 
formed  in  a  concretionary  manner  about  irregularities  or  projections 
upon  the  surface  of  the  reticulum.  The  character  of  the  basal  disk, 
its  eccentric  apex  and  nodose  j)eriphery  are  well  known. 

Figure  8.     A   basal  disk   with    a   highly  irregular   reticulum   and    without 
defined  apex.      This  appears  to   be   a  normal   and    not  accidental 
condition. 
Chemung  group.      Wellsville,  New  York, 


I>ICTYOSPONGIJ>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXm. 


G.B  Simpson  del 


Plulip  Ast  lith 


I 


246  DicrYospoNGii)^. 

PLATE  XXIV. 
THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIII,  XXV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII,  LIII.) 

TiiYSANODicTTA  Edwin-Halli,  iiom.  nov. 

Page  126. 

Figures  1,  2.  Opposite  sides  of  a  nearly  entire  cap  of  average  size;  showing 
the  broad,  oblique  aperture,  the  basal  disk  and  the  coarse  primary 
reticulation. 

Figure  3.  A  much  more  slender  specimen,  showing  both  the  primary  and 
subordinate  reticulation. 

Figure  4.  A  portion  of  the  surface  of  another  specimen,  showing  primary 
and  secondary  reticulation. 

Figure  5.     A  small  specimen  exposing  a  part  of  the  basal  disk. 

Figure  6.     The  lower  portion  of  a  specimen,  showing  tlie  basal  disk. 

Figure  7.  A  view  of  the  same  specimen  lying  in  its  matrix  and  showing  the 
broad  frill  at  the  periphery  of  the  basal  disk. 

Figure  8.  A  flattened  but  essentially  entire  sponge  in  which  only  the  strong 
primary  quadrules  are  shown. 

Figure  9.  An  incomplete  specimen  showing  the  finer  reticulation.  When  pre- 
served in  this  condition  these  specimens  are  hardly  to  be  distinguished 
from  internal  casts  of  Olathmspotigia  abacus  (see  Plate  49,  figure  8). 
Chemung  group.      Wellsville,  New    Yorh. 

Thysanodiotya  pcecilus,  sp.  nov. 

Page  127. 

Figure  10.  A  very  elongate  individual,  having  the  prevailing  form  of  the 
primary  (j^uadi-ules  nearly  scpiare;  showing,  also,  some  of  the  tufted 
intersections  of  the  erect  surface  lamellae. 

Figure  11.  A  broader  individual,  showing  the  complete  outline  of  the  body 
of  the  sponge,  its  aperture  and  base  and  the  strong  erect  vertical 
lamellae  and  their  intersections  with  similar  horizontal  lamellae.  At 
the  base,  on  one  side,  a  portion  of  the  basal  frill  is  retained.  Both 
this  and  the  preceding  specimen  are  preserved  on  the  same  slab  in 
association  with  other  forms  less  complete  but  showing  less  extreme 
outlines. 
Chemung  group.     Scio,  Allegany  county,  New  Yorh. 

TiiYSANODicTYA  Randalli,  Hall  (sp.). 

Page  129. 

Figure  12.     A  view  of   the  original  specimen  which  is  a  fragment  of  the 
upper  portion  of  a  cup  showing  the  character  of  the  reticulation. 
Chemung  group.      Warren,  Pennsylvania. 


niCTYOSPO]VOII>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXIV. 


G.B  Simpson  del 


Plulip  Ast  lith 


I 


248  DlCI'YOSPONGID^. 


PLATE  XXV. 

THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIII,  XXIV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII,  LIII.) 

Thysanodictya  saccus,  sp.  nov. 

Pago  130. 

Figure  1.     Lateral  view  of  an  entire  individual  shomng  its  rapidly  tapering 

form  and  contracted  aperture. 
Figure  2.     Basal  view  of  the  same  specimen  which  has  been  compressed  and 

shows  an  irregular  reticulum  with  a  somewhat  nodose  margin. 

Chemung  group.      Whitesville,  Allegany  county,  ]^ew  Yoi'k. 

CLATIIROSPONGIA,  Hall. 

Page  121. 
(See  Plates  XIV,  XV,  XVIII,  XXI,  XLIX,  L.) 

ClATHROSPONGIA  (J)    DESMIA,  Sp.  UOV. 
Page  125. 

Figure  3.     A  portion  of  the  external  surface  of  a  species  referred  with  doubt 
to  this  genus ;  showing  the  character  of  the  primary  quadrules. 
Chemung  group.     Alfred,  New  Yorh. 

Clatiiuospongia  vascellum,  Hall  (sp.). 

Page  123. 
(See  Plate  XXI.) 

Figure  4.     One  side  of  a  large,  somewhat  waterworn  fragment,  with  very 
large  obscure  primary  quadrules  and  a  distinct  development  of  the 
subordinate  spicular  bands. 
Chemung  group.     Alfred,  Neto  YorK 

Clathrospongia  fenestrata,  Hall  (sp.). 

Page  122. 
(See  Plato  XIV.) 

Figure  5.     An  internal  cast  of  a  nearly  entire  cup. 

Chemung  group.     Chemung  Narrwm,  New  YorTc. 


1>IC  TVOSl>OXOIl>A- 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXV 


O  B  .Simpson  del 


PbiUp  Ar.t  lull 


Explanations  of  Platks. 


249 


AGLITIIODICTYA,  gen.  nov. 

Page  145. 

Aglithodictya  ncmulina,  sp.  nov. 

Page  145. 

Figures  6-8.  Lateral,  basal  and  apertural  views  of  the  same  specimen ; 
showing  the  size  and  peculiar  form  of  the  species,  the  character  of 
the  aperture  with  the  surface  canals  on  the  oral  slope,  and  the  very 
fine  reticulum. 

Figure  9.     An  enlargement  of  the  surface  to  show  more  distinctly  the  regular 
reticulations  of  the  skeleton,     x  4. 
Chemung  group.    Near  ColwcUm,  Steuben  county.  New  York. 


250  DlClYOSPONGID.*. 


PLATE  XXVI. 

THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIII,  XXIV,  XXV,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII,  LUX.) 

TlIYSANODICTYA    APLETA,    Sp.    IIOV. 

Page  135. 

(See  Plate  XXVII.) 

Figure  1.     Portion  of  a  very  large  specimen ;  drawn  from  a  plastei-  cast  of  a 
natural  mould. 
Chemung  group.     Alfred,  New  Yorh. 


Q 

H 

0 


H 

r  ^ 

W 

H 


01 

< 

I 


CO 

d 
to 


0) 

•a 
g 

CO 

t 

C3 


252  DlCTYOSPONGID^. 

PLATE  XXVII. 
HELICODICTYA,  gen.  nov. 

Page  114. 
(See  Plate  XXII.) 

Helicodictya  (?)  Scio,  sp.  nov. 

Page  116. 

Figure  1.     A  fragment  of  a  frond  which  appears  to  have  had  a  distinctly- 
spiral  mode  of  growth,  with  strong  constrictions  or  impressions  of 
ridges  running  parallel  to  one  of  the  reticulating  series. 
Chemung  group.     Scio,  Allegany  county,  New  Yorh. 

Helicodictya  (?)  Concokdia,  sp.  nov. 

Page  115. 

Figure  2.     A  spii'ally  growing  sponge  found  associated  with  Prisinodictya 
priHmatica  in  the 
Chemung  group.     Qmcoi'd  statwii,  Erie  county,  Pennsylvania. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI,  LXl.) 

DlCTYOSPONGIA  SCEPTRUM,  Hall  (sp.). 
Page  73. 
(See  Plates  XIV,  XV,  XVI,  XXXVI.) 

Figure  ?t.     A  small  incomplete,  somewhat  compi-essed  example  with  very  fine 
reticulation.' 
Chemung  group.      Viellmille,  Neui  Yorh. 

PRISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVII,  XVIII,  XIX,  XX,  XXI,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XLI,  XLII,  LV  ) 

Peismodictya  filitextilis,  Hall  (sp.). 

Page  86. 
(See  Plate  XXI  ) 

Figure  4.     A  fragment  of  an  uncompressed  individual  showing  the  obscure 
prism-faces  and  fine  reticulation, 
Chemung  group.      Wellsville,  New   YorTc. 


DICTYOSl»ON^Omiy. 


Memoirs  Geological  Survey  of  New  York 


Plate  XXVn. 


G.B  Simpson  del. 


Philip  Ast  lith 


Explanations  ok  Platkh.  253 

PBISilODICTYA   CHOANEA,    Sp.  nOV. 

Page  85. 

(See  Plates  XIX  anil  XXXV.) 

Figure  5.     The  lower  portion  of  a  large  cup  showing  obscurely  prismatic 
faces. 
Chemung  group.     Imhaa,  Cattaraugus  county,  New  York. 

DiCTYOSPONGIA  ?    (EcTENODICTTA  ?). 

Figure  6.  A  flat  and  expanded  portion  of  a  frond,  whose  outline  may 
represent  the  original  form  of  the  cup,  although  the  edges  appear  to 
be  somewhat  ragged.  The  reticulation  is  very  fine  and  the  surface 
without  evidence  of  prism-faces.  The  specific  and  generic  relations 
of  the' specimen  are  uncertain. 
Chemung  group.      Wellsvllle,  New  YorJc. 

THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIII,  XXIV,  XXV.  XXVI,  XXXVIII,  XXXIX,  XL,  XLII,  LIU.) 

Thtsanodictya  iiermenia,  sp.  nov, 

Pago  133. 
(See  Plates  XXXIX  and  XL.) 

Figure  7.  The  surface  of  an  incomplete  individual ;  showing  the  very  strong 
primary  bands  and  large  quadrilles.  No  evidence  is  retained  of 
subordinate  reticulation  but  the  surface  of  the  specimen  has  been 
involved  in  shale  and  the  structure  obscured.  The  drawing  is  made 
from  a  plaster-cast  of  a  natural  mould  of  the  exterior. 
Chemung  group.     Alfred,  New  Yorlc. 

Thtsanodictya  apleta,  sp.  nov. 

Page  135. 
(See  Plate  XXVI.) 

Figure  8.     A   fragment    of   the   surface,   sho^ving   the   coarse    primary    and 
secondaiy  quadrules. 
Chemung  group.     Alfred,  New  York 


254  DiCTYOSPONGIDiK 


PLATE  XXVIII. 
HYPHANT^NIA,  Vanuxem. 

Page  137. 
(See  Plates  XXIX  and  XLV.) 

HypHANT^NiA  CuEMUNGENSis,  Vanuxem. 

Page  139. 
(Sea  Plates  XXIX  and  XLV.) 

Figure  1.     An  internal  cast  of  the  central  or  basal  portion  of  a  frond,  showing 

the  concentric  and  radiating  ligulate  bands  and  the  open  quadrate 

interspaces. 
Figure  2.     A    view   of   the  original    specimen    described    and   figured   by 

Vanuxem.     This  shows  the  strongly  elevated  margins  of  the  radial 

bands  which  are  maintained  at  the  intersections  of  these  with  the 

concentric  bands. 

Chemung  group.      In  tlie  vicinity  of   Owego,   Tioga  county,  Ntw 

York 


X 
X 


< 

s 

a. 


t 


o 

u 

0 
CO 

13 
u 

■So 

o 

I — I 
o 
cu 
O 


%^f' 


c 


250  I>lCl'Y<>SI><)N(JII)iK, 


PLATE  XXIX. 

HYPHANT^NIA,  Vanuxem. 

Pago  137. 
(See  Plates  XXVIII  »nd  XLV.) 

IIypiiant^nia  Ciiemungensis,  Vanuxem. 

Page  139. 
(See  Plates  XXVIH  and  XLV.) 

Figure  1.  A  fragment  of  a  specimen  whose  entire  diameter  is  probably 
represented.  This  appears  to  be  a  mould  from  an  impression  of 
the  exterior,  shows  the  expanding  saucer-shaped  form  of  the  sponge, 
the  raj)id  increase  in  the  width  of  the  radial  bands,  also  their  elevated 
margins  and  the  open  quadrangular  interspaces  made  by  the  inter- 
section of  the  two  series  of  bands.  The  center  of  the  specimen 
as  Avell  as  that  represented  in  figure  1  on  the  preceding  plate  lacks 
definition  and  lends  probability  to  the  inference  that  there  was  a 
tuft  of  basal  spicules  at  this  point. 
Chemung  group.    Near  Owego,  Tioga  county,  New  York. 


X 

X 
X 

r— 4 


\r 


■\ 


V 


-/ 


r':x 


y- 

\^'" .  -*, 

"  \    " 

X 

u 

^ 

^ 

(U 

iz; 

U-H 

o 

\ , 

§r 

\'  ■■ 

^ 

3 

CO 

V- 

r— • 

u 

•1-1 

\. 

■on 

o 

--,, 

o 

0) 

-^  A 

C> 

" 

tn 

^1 

\ 

•S 

o 

(J) 

s 

■< 

r\ 


i 


258  DlClYOSPONGlD^K. 


PLATE  XXX. 
ACTINODICTYA,  Hall. 

Page  142. 
(See  Plato  XXXI.) 

ACTINODICTYA    PLACENTA,  Hall. 

Page  143. 

(See  Plate  XXXI.) 

Figure  1.  An  irregiilai'ly  oval  frond  showing  great  differences  in  the  size  of 
the  spicular  bands.  In  this  specimen  these  bands  show  an  unusually 
regular  reticulation  although  the  various  series  may  intersect  at 
different  angles.  The  drawing  has  Ijeen  made  with  accuracy  but 
the  apparent  abundance  of  cruciform  spicules  is  wholly  due  to 
elevation  of  reticulating  spicular  bands  at  and  near  their  intersec- 
tions ;  as  these  bands  gradually  die  out,  the  more  strongly  projecting 
portions  produce  the  effect  observed. 

Figure  2.  Two  essentially  entire  fronds,  one  j)artially  overlapping  the  other 
and  somewhat  broken  near  one  extremity.  This  specimen  Avith  that 
represented  in  figure  1  and  that  in  figure  1  on  Plate  xxxi  show  the 
form  of  this  fossil  at  the  locality  where  it  has  been  found  in  the 
greatest  perfection  and  without  evidence  of  much  distortion.  Such 
specimens  indicate  that  the  bodies  were  broad,  convex  expansions, 
evidently  not  of  highly  irregular  growth  but  with  a  definite  marginal 
outline  and     uboval  form. 

Chemung  group.     Orie-lialf  mile  from  Brown  hill,  near  Coliocton, 
Meio  York. 

Figure  3.     A  fragment  of  a  specimen  the  surface  of  which  bears  irregularly 
scattered  spots  representing  subcircular  perforations  of  the  spicular 
net- work.     These  may  be  to  some  extent  due  to  accident. 
Chemung  group.     Near  Howard,  New  Yorh. 


l>ICTVOSPOXOII>A\ 


Memoirs  Geological  Survey  of  New  York 


Plate  XXX. 


v/ 


/  \ 


■y 


G.B  Simpson  del 


Philip  Ast  lith 


m 


200  DuTVOSI'OiNCilD.K. 


PLATE  XXXI. 

ACTINODICTYA,  Hall. 

Page  142. 
(See  Ph>te  XXX  ) 

ACTINODICTYA    PLACENTA,    Ilall. 

Page  143. 

(See  Plate  XXX  ) 

Figure  ] .     A  very  nearly  complete  froud  showing  the  regularity  of  much  of 

the  finer  reticulation. 

Chemung   group.       One-half  mile  from,    Broton  hill   school  house 

near  Cohocton,  New    Yorh. 
Figure  2.     A   small  frond,  with  well  defined  margins  indicating  completed 

peripheral  growth. 

Chemung  group.     Brown  hill  colony,  near  Cohocton,  New    Yorh. 

CRYPTODICTYA,  Hall. 

Pago  144. 
(See  Pl.ite  XXXII.) 

Cryptodictya  Alleni,  Hall. 

Page  144. 
(See  Plate  XXX 11  ) 

Figure  3.  A  portion  of  a  nodose  margin  of  a  specimen  which  shows  that 
these  nodes  bore  tufts  of  spicules.  Such  tufts  are  seen  only  when 
the  nodes  lie  at  the  mai-gins  of  the  surface  expansion  and  the  tufts 
extend  along  the  plane  of  lamination  of  the  matrix. 

Figure  4.  An  irregular  frond  showing  on  one  side  the  abrupt  depression  of 
the  nodose  surface  to  the  broad,  anodose  marginal  expansion. 

Figure  5.  A  large  specimen  in  \vliich  the  nodes  are  arranged  in  groups. 
This  frond  has  been  folded  along  the  i-ight  margin,  the  opposite 
side  being  of  the  same  extent  as  that  here  represented,  and  having 
the  nodes  similarly  arranged  in  clusters.  This  folding  of  the  frond 
is,  however,  probably  Avholly  due  to  compression. 
Chemung  group.     Lyndon,   Cattaraugus  county,  New    Yorh. 


l>ICTVOSPOXOII>;^. 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXI. 


G.B  Simpson  del 


Plulip  Ast  Inh 


262  DlCl'YOSPONGIDJE. 


PLATE  XXXII. 
CRYPTODICTYA,  Hall. 

Page  144. 
(See  Plate  XXXI.) 

Cryptodictya  Alleni,  HalL 

Page  144. 
(See  Plate  XXXI.) 

Figure  1.  The  nodose  portion  of  a  large  specimen.  This  appears  to  have 
been  a  lateral  or  extremital  portion  of  a  frond  where  the  surface  is 
marked  by  obliijue  angular  curved  lidges  teiTuinatiug  in  swoUen 
nodose  elevations.  Several  of  these  ridges  show  coarse  lines  or 
grooves  radiating  from  the  summit,  and  over  the  broad  general  slope 
of  the  surface  as  well  as  between  the  ritlges  are  seen  traces  of 
spicular  lines. 

Figure  2.  A  small  fragment  representing  a  partially  enfolded,  narrow  portion 
of  a  frond,  bearing  a  divided  node  on  the  surface. 

Figure  3.  A  specimen  of  subclavate  outline,  exhibiting  the  flattened  nodose 
surface,  the  sides  falling  away  vertically  in  all  directions,  except 
below.  These  specimens  all  indicate  that  the  original  form  of  the 
frond  was  a  suboval  expansion  having  a  slight  elevation,  a  nodose 
upper  or  expanded  surface  and  probably  was  attached  by  the  free 
margins  or  by  the  narrowed  extremity. 

Figure  4.     A  somewhat  similar  portion  of  a  frond  showing  the  irregularly 
nodose  surface.     The  margins  of  the  expansion  are  bent  vertically 
downward  on  all  sides  except  the  postenor  and  the  specimen  appears 
to  be  the  elevated  and  flattened  surface  of  the  frond. 
Chemung  group.     Neai-  Lyndon,  Cattaraugus  county,  New  YorJc. 


13IC  T  YO  S  P  OX  0 1 1>  A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXII. 


G.B  Simpson  del 


Plulip  Ast  hth 


2(54  DiCTYOSl'ONGIA. 

PLATE  XXXIII. 
DICTYSPONGIA,  gen.  nov. 

Page  72. 
(Seo  Plates  XIV,  XV,  XVI,  XXVII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  EV,  LVI,  LXI.) 

DiCTYOSPONGIA   IIAPLEA,  Sp  IIOV. 
Page  68. 

Figure  1.     The  larger  and  more  complete  of  the  specimens  observed  ;  showing 
the  smooth  non-prismatic  surface  and  a  contraction  of  diameter  near 
the  aperture. 
Portage  gi-oup.     JVajyles,  New  YovTc. 

PRISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVn,  XVIII,  XIX,  XX,  XXI.  XXVII,  XXXIV,  XXXV,  XXXVI,  XLI,  XLII,  LV.) 

Pkismodictya  pal^a,  sp.  nov. 

Pagp  69. 

Figure  2.     The  original  specimen ;  a  nearly  entire  individual  showing  faint 
prism-faces  and  indications  of  low  horizontal  depressions. 
Portage  group.     Naples,  Neto  Yorh. 

HYDNOCEEAS,  Conrad. 

Page  95. 
(See  Plates    II,  UI,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXVI,  XXXVII,  XXXVIII,  XXXIX,  XLVL) 

Hydnoceuas  gracile,  sp.  nov. 

Page  70. 

Figures  3,  4.     Two  specimens,  entire  in  the  extent  of  the  cup  from  base  to 
apertui'e,  but   foreshortened  by  vertical  compression.     They  show 
three  horizontal  rows  of  nodes  with  traces  of  incipient  nodes  near 
the  aperture. 
Portage  group.      Varyshurg,  New  Yorh 

Hydnoceras  legatum,  sp.  nov. 

Page  69. 

Figure  5.     A  specimen  of  medium  size  ^vith  apex  preserved  and  bearing  three 

horizontal  rows  of  nodes. 

Portage  group.     Naples,  New  York 
Figure  6.     Fragment  of  a  \'ery  large  example  with  broad  nodes,  similar  to 

those  in  II.  Avoca. 

Portage  gi-oup.      Varyshurg,  New  Yo7'h. 


Die  Tiro  Sl>OXOIX>A\ . 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXIII. 


Philip  Ast  lith 


Explanations  ok  Plates. 


265 


Hydnockbas  eutiieles,  sp.  IIOV.  (?) 

Page  70. 
(See  Plate  Vin.) 

Figure  7.     A  fragment  with  elongated,  subpendiilous  nodes,  similar  to  those 
snown  in  the  figures  of  this  species  on  Plate  viii  (figs  5  and  6). 
Portage  group.      Varysbury,  Neiv    Yorh 

Hydnocebas  antubacis,  sp.   nov. 

Page  109. 
(See  Plates  II  and  VI.) 

Figures  8,  9.     Opposite  sides  of  the  most  complete  example  observed ;  show- 
ing the  sharply  prismatic  anodate  lower  half  of  the  sponge  and  the 
three  horizontal  rows  of  low  nodes  of  the  upper  half  lying  on,  and 
ridged  by  the  pnsm-angles. 
Chemung  group.     Scio,  New   Yoi'h. 


20G  DlCTYOSPONOID^. 


FLITE  XXXIV. 
CLEPSYDROSPONGIA,  gen.  nor. 

Page  71. 

Cl^PSTDROSPONGIA   MATUITNA,    Sp.  nOV. 
Pago  71. 

Figures  1,  2.  Opposite  sides  of  the  original  specimen  ;  showing  the  subequal 
expansion  of  the  extremities  and  the  contracted  median  nodose  area. 
Figure  1  is  drawn  from  the  specimen  as  it  lies  in  its  matrix  and 
shows  the  node-tufts  even  at  spots  where  no  nodes  are  clearly 
developed  on  the  surface. 
Portage  group.     Naples,  New  Yo^'K 

PRISMODICTYA,  gen.  nov. 

Pago  79. 
(See  Plates  XVII,  XVIII,  XIX,  XX,  XXI,  XXVH,  XXXIII,  XXXV,  XXXVI,  XLI,  XLII,  LV.) 

Pkismodictya  ptionia,  sp.  nov. 

Page  88. 

Figure  8.     The  original  specimen  showing  its  great  expansion  near  the  middle 
and  its  concave  prism-faces. 
Chemung  group.     Scio,  Neio  Yorh. 

Pkismodictya  aulopiiia,  sp.  nov. 

Page  88. 
(See  Plates  XX  and  XLI  ) 

Figures  4,  5.     Two  specimens ;  showing  the  elongate  form  and  concave  faces. 
These  specimens  are  from  a  fragment  of  rock  which  contains  traces 
of  niimerous  other  individuals  of  this  species. 
Chemung  group.     Scio,  New  Yofk. 


r)ICTY0SP01»J^0II>iE. 


Memoirs  Geological  Survey  of  NewYork. 


Plate  XXXIV: 


G.B.  Simpson  del. 


Philip  Asthth. 


268  DicrYORPONoinvE. 

PLATE  XXXV. 

PRISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plate.  XVII,  XTIII,  XIX,  XX,  XXI.  XXVII,  XXXIII,  XXXIV,  XXXVI,  XLI,  XLII,  LV.) 

Prisjiodictya  citiiaba,  sp.  nov. 

Page  90. 
(See  Plate  XXXVI.) 

Figures  1,  2.  Opposite  sides  of  the  same  specimen,  complete  at  the  aperture 
but  imperfect  toward  the  base ;  showing  the  abrupt  expansion  of  the 
cup  in  the  upper  part  of  the  sponge,  with  low  nodes  at  the  elevated 
portions  of  the  prism-angles. 

Figure  3.     Another  specimen  showing  similar  characters. 
Chemung  group.     Scio,  Nev)  Yofh. 

Pkismodictya  nartiiecia,  sjj.  nov. 

Page  90. 

Figures  4,  5.     Two  views  of  the  typical  specimen,  an  elongate  and  slender 
sponge  \vith  but  slight  expansion  above  the  middle. 
Chemung  group.     Scio,  New  YorK 

Pkismodictya  choanea,  sp.  nov. 

Page  85. 
(See  Plates  XIX  and  XXVII.) 

Figure  6.     Fragment  of  a  large  individual  with  lo\v  prism-faces  (too  strongly 
rendered  in  the  drawing). 
Chemung  group.     Olean,  JSfeio  YorJe. 

Pkismodictya  telum.  Hall  (sp.). 

Page  80. 
(See  Plates  XVII  and  XLII.) 

Figures  7,  8.     Two   nearly   entire   specimens    bearing   low   prism-faces    and 
showing  the  characteristic  expansion  of  the  frond  near  the  base. 
Chemung  group.     Friendship,  New  Yorlc. 

Pkisjiodictya  paballela,  Ilall  (sp.). 

Page  82. 
(See  Plates  XIX  and  XXI.) 

Figure  9.     A  part  of  a  very  long  specimen  showing  little  increase  in  diameter 
throughout  its  length. 
Chemung  group.     Hinsdale,  New  Yor7.\ 


DICTYOSl>OXOII>i!e. 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXV. 


G.B  Simpson  del 


Philip  Ast  hth 


270  DltriYOSPONGID^. 


PLATE  XXXVI, 
HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II,  III,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVII,  XXXVIII,  XXXIX,  XLVI.) 

Hydnocekas  vakiabile,  sp.  nov. 

Page  108. 
(See  Plate  XXXVII.) 

Figure  1.     The  apical  portion  of  a  specimen  with  coarse,  strong,  subequal 
meshes.     There  is  no  indication  of  nodes  upon  the  surface  of  the 
cast,  except  near  the  broken  end,  but  the  matrix  shows  the  pres- 
ence of  nodal  tufts  in  this  region. 
Lower  Chemung.     Deyo  basin,  Naples,  New   York. 

PEISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVII,  XVIIT,  XIX,  XX,  XXI,  XXVII,  XXXHI,  XXXIV,  XXXV,  XIJ,  XLII,  LV.) 

PlUSMODICTYA  AMICITI^,  Sp.  IIOV. 
Page  87. 

Figures  2,  8.  Two  specimens  which  show  the  elongate  and  slender  form  of 
the  s])ecies,  with  the  abrupt  and  gradually  declining  expansion  of 
the  sponge  above  the  base. 

Figure  4.     This  specimen,  found  in  association  with  numerous  individuals  of 
this  species,  shows  a  spiral  twisting  of  the  frond,  resulting  from 
abnormal  growth. 
Chemung  group.     Frietidship,  New  York 

Pkismodictya  cithaea,  sp.  nov. 

Page  90. 
(See  Plate  XXXV.) 

Figure  5.     A  large  and  somewhat  distorted  specimen  of  this  species  in  which 
the  nodes  up{m  the  expanded  portion  are  rendered  conspicuous  by 
compression  of  the  prism-faces. 
Chemung  group.     Scio,  New  YorTc. 


DTCTVOSl>OXOII>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXVI. 


G.B  Simpson  del. 


Phihp  Ast  hth 


Exi'LANATIONS   OK    PlATKS.  271 

Pkismodicita  sp.  ? 
Figure  6.     A   compressed,    expanded   and    somewhat   twisted    frond    whose 
specific  relations  have  not  been  determined.     It  bears  some  relation 
to  P.'ptionia  but  differs  therefrom  in  its  less  abrupt  expansion  and 
broader,  shorter  basal  region. 
Chemung  group.     Scio,  New  Yoi'k. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVIII,  XIJ,  XLII,  XLIV,  XLVI,  LV,  LVl,  LXI.) 

DlCTYOSPONGIA   SCEPTBUM,  Hall  (sp.). 
Page  73. 
(Seo  Plates  XIV,  XV,  XVI,  XXVII.) 

Figure  7.     The   basal  end    of   a  specimen,  showing   a   portion   of  the  long 
anchoring  tuft. 
Chemung  group.     Ischua,  New  Ya)'k. 

IIALLODICTYA,  gen.  nov. 

Page  140. 
(See  Plate  XLI.) 

Hallodictya  Sciknsis,  sp.  nov. 

Page  140. 

Figure  8.  Tlie  original  specimen,  showing  the  expanded,  irregular  and  nodose 
character  of  the  sponge.  The  reticulation  is,  in  itself,  regularly 
constructed  but  conforms  to  the  highly  uneven  surface.  This 
specimen  is  a  mould  of  the  exterior  surface  and  the  depressions 
represented  on  the  drawing  indicate  nodes  of  tlie  outer  surface. 
Chemung  group.     Scio,  New  Yorh 


272  ■  I)urrYOspoN(iii>.i-:. 


PLATE  XXXVII. 
IIYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II,  III,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVIII,  XXXIX.  XLVI.) 

Hydnoceras  bhopalum,  sp.  nov. 

Page  107. 

Figures  1,  2.  Opposite  sides  of  a  very  slender,  somewhat  malformed  sj^ecimen 
with  faint  indications  of  nodes  over  the  lower  half  of  the  surface. 
The  mode  of  growth  here  exhibited  and  the  general  aspect  of  the 
sponge  is  (piite  distinct  from  that  of  11.  tuherosum,  in  association 
with  which  this  sjiecimen  ^vas  found. 
Chemung  group.     Brown,  Jiill  colony,  near  Cohodon,  New  Yoi'h. 

Hydnocekas  phymatodes,  sp.  nov. 

Page  104. 
(See  Plate  VI.) 

Figure  3.     A  small  nearly  entire  specimen  with  five  well  defined  horizontal 
rows  of  rather  low  nodes  and  two  faintly  developed  additional  rows, 
one  at  the  base  and  one  at  the  aperture. 
Lower  Chemung.     Italy,  Yates  county,  New  Yoi'h. 

Hydnoceuas  variabile,  sp.  nov. 

Page  108. 
(See  Plate  XXXVI  ) 

Figures  4,  6.     Opposite  sides  of  a  specimen  preserving  the  apertural  portion 

of  the  sponge  and  showing  the  indistinct  nodes  and  irregular  form 

which  characterize  the  species. 
Figure    5.     Another   specimen    showing   similar   characters   and    the    coarse 

reticulation  of  the  cast. 

Lower  Chemung.      Deyo  basin,  Naples,  New  Yorh. 


»tctyosi>o:noij>a\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXVII, 


G.B  Simpson  del 


PMip  Ast  lith. 


( 


274  DlCTYOSPONOID.«. 


PLATE   XXXVIII. 
HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II,  III,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXIX,  XLVI.) 

Hydnoceras  tuberosum,  Conrad. 

Page  »7. 
(See  Plates  II,  III,  IV,  V,  VI,  VII,  VIII. ) 

Figures  1,  2.     Opposite  sides  of  a  short  compressed  specimen  of  somewhat 
irregular  growth  and  unusual  in  the  number  of  nodes  (eleven)  in  the 
second  and  third  horizontal  rows. 
Lower  Chemung.     Brown  liill  colony,  near  Cohoaton,  New  Yorh. 

THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIII,  XXIV,  XXV,  XXVI,  XXVII,  XXXIX,  XL,  XLIt,  LIII.) 

TlIYSANODICTYA    QUASILLUM,    Sp.  UOV. 
Page  131. 

Figures  3,  4.  Opposite  sides  of  a  specimen  figured  to  illustrate  the  extreme  dif- 
ference in  retention  of  the  reticulum  possible  in  the  same  specimen. 
In  figure  3  the  finer  net- work  is  wholly  obliterated  and  only  the  coarse 
primary  meshes  have  left  their  impression  on  the  sandstone  cast. 
On  the  other  side  (figure  4)  the  primary  bands  are  but  obscurely 
visible  while  the  minute  ultimate  meshes  of  the  reticulum  have  left 
their  impression  on  the  thin  shaly  coating  of  the  sandstone  cast. 
Chemung  group.      Wenton  Mills,  Cattaraugus  county,  New  Yorh. 

TlIYSANODICTYA    TURRICULA,  Sp.  nOV. 
Page  135. 

Figure  5.     A  coarsely  reticulate  specimen  lying  in  its  matrix  and  showing  the 

basal  diaphragm  wth  some  of  the  long  and  slender  lateral  tufts. 

The  species  is  somewhat  variable  in  shape  and,  in  this  instance,  is 

swollen  about  the  aperture. 
Figure  6.     A  specimen  which  is  essentially  entire  and  gently  tapers  from  base 

to  aperture.     It  retains  the  basal  disc   and  portions  of  the  lateral 

tufts. 

Chemung  group.     Isckna,  New  Yorh 


DICTYO  si*o:xoii>i¥. 


Memoirs  Geological  Survey  of  New  York. 


Plate  XXXVm. 


G.B  Simpson  del. 


PlubpAstlith 


Exi'I.ANATIONH    OK    PlATI<«. 


275 


DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI,  LXI.) 

DlCTYOSPONOIA  (?)  UACTEKIA,  Hp.  IIOV. 
Page  77. 

Figures  7,  8.     Two  views  of  a  supposed  sponge  with  smooth  surface,  Ix^aring 
very  faint  horizontal  lines  which  are  exaggerated  in  tlie  drawings  in 
order   to   make  theni  visible.     The  fossil  does  not  clearly  show  a 
structure  which  conclusively  demonstrates  it  spongeous  nature. 
Chemung  group.     Steuben  county  (?'),  New  York. 


276 


DlCTYOSPONGn>.iE. 


PLATE  XXXIX. 

HYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  H,  III,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXIII,  XXXVI,  XXXVII,  XXXVIII,  XLVL, 

Hydnocekas  multinodosum,  sp.  nov. 

Page  106. 

Fi<'-ure  1.     Fragment  of  a  large  individual  bearing  three  horizontal  rows  of 
nodes  and  showing  the  unusual  number  of  nodes  in  each  row. 
Chemung  group.     Cotton  hill,  near.Avoca,  New  Yorh. 

TIIYSANODICTYA,  gen.  nov. 

Page  135. 
(See  Plates  XIX,  XXIII,  XXIV,  XXV,  XXVI.  XXVII,  XXXVIII,  XL,  XLII,  LIII.) 

Thysanodictya  iiermenia,  sp.  nov. 

Page  133. 
(See  Plates  XXVII  and  XL.) 

Figure  2.  A  large  basal  disc  with  its  undulated  periloph  or  peripheral 
expansion.  The  reticular  net- work  is  sharply  defined  and  it  is  seen 
that  the  two  sets  of  reticular  bands  here  intersecting  are  those  which, 
passing  upward  over  the  vertical  surface  of  the  frond,  produce  both 
horizontal  and  vertical  bands. 
Chemung  group.     Hiiisdale,  Cattaraugus  county,  New  Yorh. 


l>ICTYOSl*ONOII>iy. 


Memoirs  Geological  Survey  of  New  York 


Plate  XXXIX. 


G.B  Simpson  del 


',  Fausei.  litK 


278  DlClTfOSPONCilD^. 


PLATE  XL, 

THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Plates  XIX,  XXIII,  XXIV,  XXV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XLII,  LIII.) 

Thysanodictya  heemenia,  sp.  nov. 

Page  133. 
(See  Plates  XXVII  and  XXXIX.) 

Figures  1,  2.     Basal  discs  without  the  periloph  but  preserving  more  or  less 

completely  the  coarser  and  finer  reticulation. 
Figure  3.     Side  view  of  a  rather  small  specimen  with   the  basal  disc    poorly 

retained.     The  sponge  has  been  crushed  toward  the  upper  end  and 

a  part  of  its  diameter  with,  doubtless,  a  considerable  j)ortion  of  the 

length  has  been  lost. 

Chemung  group.     Hinsdale,  Cattarauyus  county,  New  Yorh. 


I>ICTY0S1>0>"0IJDA\ 


Memoirs ,  Geological  Survey  of  New  York. 


Plate  XL. 


G.B  Simpson  del 


-C  Fausel.  lit,- 


280  DlCl'YOSPONGIDiE. 

PLATE  XLI. 
CERATODICTYA,  gen.  nov. 

Page  117. 
(See  Plates  XXII,  XLII,  XLIII.) 

Ceratodictya  centeta,  sp.  nov. 

Page  120. 

Figure  1.     A  very  young  individual. 

Figure  2.     A  fragment  of  a  large  example. 

Figures  3,  4,  5.     Specimens  of  medium  size,  showing  some  irregularity  in  the 

size  of,  and  intervals  between  the  annulations. 

Chemung  group.     Ischtm,  New  York. 

Ceratodictya  zonata,  sp.  nov. 

Page  119. 

Fio-ure  G.     The  type-specimen,  showing  equidistant  and  broad  annulations. 
Chemung  group.     Found  loose  at  Erie,  Pennsylvania. 

PRISMODICTYA,  gen.  nov. . 

Page  79. 
(See  Plates  XVII,  XVIIF,  XIX,  XX,  XX[,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XLII,  LV.) 

PuiSMODICTYA  AULOPIIIA,  Sp.  UOV. 

Page  88. 

(See  Plates  XX  and  XXXIV.) 

Figure  7.     Fragment  of  a  large  specimen ;  showing  the  concave  prism-faces 
and  finely  preserved  net-work. 
Chemung  group.      Wellsville,  New  York. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI,  LXI.) 

DiCTYOspoNGiA  (?)  Marcellia,  Clarke  (sp.). 

Page  65. 

Figure  8.     A  portion  of  the  suface  of  a  piece  of  bituminous  shale  covered  Avith 
pyritized  spicules  of  this  species. 

Figure  9.     An  enlargement  from  the  same,  showing  the  cruciform  spicules  and 
the  broad  ligulate  spicules  whose  position  in  the  skeleton  is  not  yet 
understood. 
Marcellus  shale.     In  the  Livonia  salt  sliaft,  Livonia,  New  York. 


l>ICTYOSl»OI^OIl>A\. 


Memoirs  Geological  Survey  of  NewYork. 


'  Plate  XLI. 


G.B  Simpson  del. 


Pkihp  Ast  lith 


Explanations  of  Plates.  281 

ARYSTIDICTYA,  gen.  nov. 

Page  i:»6. 
(See  Plate  XLII.) 

ARYSTIDfCTYA    KLEOANS,  Sp.  nOV. 

Page  136. 

(See  Plate  XLII.) 

Figure  10.  One  side  of  the  typical  specimen  which  is  preserved  from  the 
base  or  basal  disc  to  the  aperture.  The  drawing  has  intensified  the 
wrinkles  of  the  reticulum  resulting  from  a  slight  foreshortening, 
and  fails  to  give  with  proper  force  the  regularly  reticulating  bands 
which  are  clearly  apparent  especially  over  the  lower  part  of  the 
sponge.  The  marginal  radiating  channels  about  the  edge  of  the  vase 
are  well  shown. 
Lower  Chemung  group.     Cotton  hill,  iiea/r  Avoca,  New  Yai'Jc. 

Aeystidictya  nodifeba,  sp.  nov. 

Page  137. 

Figure  11.     The  specimen  described;  a  portion  of  a  vase-shaped  sponge  like 
A.  elegans,  having  the  sui-face  covered  with  elongate,  irregularly 
disposed  nodes. 
Lower  Chemung  group.     Cotton  hill,  near  Avoca,  New  Yoi'h. 

HALLODICTYA,  gen.  nov. 

Page  140. 
(See  Plate  XXXVI.) 

Hallodictya  Cottoniana,  sp.  nov. 

Page  142. 

Figure  12.     A  fragment  of  a  frond  having  a  sharply,  somewhat  irregularly 
reticulate  surface,  free  of  nodes. 
Lower  Chemung  group.     Cotton  hill,  tiear  Avoca,  New  Yorh. 


282  DlCTYOSPONOID^E. 

PLATE  XLII. 
CERATODICTYA,  gen.  nov. 

Page  117. 
(8«e  Plates  XXII,  XLI,  XLni.) 

OeE.\I0DICTTA    CABPENTEBIA^fA,  Sp.  nOV. 

Page  117. 
(See  Plate  XLIII.) 

Figure  1.     A  finely  preserved  sjjecimen  of  a  part  of  the  sponge,  having  an 
apparently  normal   curvature,  and   showing   the   character  of   the 
reticulum  and  the  duplicate  annulations. 
Chemung  group.     Near  Ischua,  New  York. 

PRISMODICTYA,  gen.  nov. 

Page  79. 
(See  Plates  XVH,  XVlil,  XIX,  XX,  XXI,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVT,  XLI,  LV.) 

Prismodictya  telttm,  Hall  (sp.). 

Page  80. 
(See  Plates  XVII  and  XXXV.) 

Figure  2.     A  small  cigar-shaped  individual  shoAving  a  twisted  or  spiral  mode 
of  growth. 
Chemung  group.      Welhville,  New  YorJc. 

Pkismodictya  cercidea,  sp.  nov. 

Page  !W. 
(See  Plat«  XX.) 

Figure  3.     A  flattened  but  entire    example,  sho^^■ing  the  expansion  of    the 
sponge  above  the  middle  and  the  low  nodes  on  the  prism-angles. 
Chemung  group.      WellwilU,  Neio  York. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(Se«  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXAa,  XXXVIII,  XLI,  XLIV,  XL\1,  LV,  LVI,  LXI.) 

Dictyospongia  sir^a,  sp.  nov. 

Page  75. 

Figures  4,  5.     Two  individuals  showing  the  size  and  general  form  of  the 
species,   one  of  them  (figure  4)  retaining  the   long   tuft  of  basal 
spicules. 
Lower  Chemung  group.     Cotton  hiU,  near  Avoca,  Neio  York. 

Prismodictya  cf.  prismatica.  Hall. 
Figure  6.     Part  of  a  cup  which  may  prove  to  belong  to  this  species. 
Lower  Chemung  group.     Italy,  Yates  county,  Nero  York. 


13Xctyosi*o:n^oij>a\ 


Memoirs  Geologicail  Siirvey  of  New  York 


Plate  XLII, 


G-B  Simpson  del 


Philip  Ast  lith 


Explanations  of  Plaii<x  283 

ARYSTIDICTYA,  gen.  nov. 

Page  136. 
(See  Plate  XLI.) 

Ajiystidictya  eleoans,  sji.  nov. 

Page  136. 
(See  Plato  XLI.) 

Figure  7.     The  other  side  of  the  specimen  figured  on  Plate  xli,  showing  the 
regular  reticulation  over  the  lower  part  of  the  cast  and  the  radiating 
peripheral  channels. 
Lower  Chemung  group.     Cotton  Jiill,  near  Avoca,  New  York. 

THYSANODICTYA,  gen.  nov. 

Page  125. 
(See  Platig  XIX,  XXni,  XXIV,  XXV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII,  LIII.) 

Thysanodictya  scypuina,  sp.  nov. 

Page  132. 

Figure  8.     The  typical  specimen,   showing  the  undulated  basal  periloph  or 
fringe  of  the  diaphragm. 
Chemuug  group.     Olean,  New  York. 


284  ■  DiCTVOSPONOIDiE. 


PLATE  XLIII. 
CEKATODICTYA,  gen.  nov. 

Page  117. 
(See  Plates  XXU,  XLI,  XLII.) 

Cebatodictya  Carpenteeiana,  sp.  nov. 

Page  117. 
(See  Plate  XLII.) 

Figure  1.  A  large  fragment  of  this  species,  complete  at  neither  extremity. 
This  drawing  has  been  made  from  several  pieces  of  the  same  individ- 
ual and  as  -these  did  not  perfectly  fit,  the  figure  does  not  represent 
the  entire  length  of  the  fragment.  The  specimen  shows  the  very 
gradual  expansion  of  the  sj)onge,  indicating  that  it  must  have 
attained  great  length  at  full  growth. 

Figure  2.     Another  fragment  preserving  the  apical  portion  entire  and  showing 
the  gradual  change  with  growth  in  the  aspect  of  the  annulations. 
Chemung  group.     Near  Ischua,  JVew  York. 

Figure  8.  A  very  imperfect  fragment  from  near  the  aperture  of  a  full  groAvn 
specimen.  This  specimen  indicates  the  great  length  which  the 
species  must  have  attained  and  shows  the  ultimate  modification  in 
the  form  of  the  annulations  which  are  here  brought  so  close  together 
that  their  duplicate  character  is  obscured. 
Chemung  group.     Randolph,  New  Yorh. 


IH 

H 

H 


286 


DlcrYOSI'ii.NfilM.K. 


• 


PLATE  XLIV. 
DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(Seo  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLVI,  LV,  LVI  LXI.) 

DicTYospoNGiA  Almondensis,  sp.  nov. 

Page  76. 

Figure  1.  A  large  fragment  representing  the  upper  part  of  a  frond  of  great 
size.  Fragments  of  this  character  abound  at  the  locality  from  which 
this  was  obtained,  but  so  large  and  fragile  was  the  sponge  that  no 
entire  individual  has  been  obtained  and  but  very  few  specimens 
which  aiford  a  clew  to  the  original  form  of  the  species.  Elsewhere 
is  given  a  drawing  of  the  apical  portion  of  one  of  these  sponges 
^vhich  shows  that  it  was  doubtless  a  Dictyospongia. 
Chemung  group.     In  a  bed  of  grey  mndsbone,  at  Almond,  New  York. 


288  DicTYOsPONGiD.*:. 


PLATE  XLV. 

HYPHANT^NIA,  Vanuxem  (emend.). 

Page  137. 
(See  Plates  XXVIII  and  XXIX.) 

HYFiTATfi^NiA  CirEMTjNGENsis,  Vanuxem. 

Page  139. 
(See  Plates  XXVIII  and  XXIX.) 

Figure  1.  Fragment  of  the  largest  iutlividual  observed,  slightly  fractured 
near  the  middle.  This  specimen  differs  from  those  elsewhere  figured 
in  having  the  apertures  left  by  the  intersection  of  the  ligulate  bands 
circumscribed  by  an  encroachment  of  the  spicular  tissue.  The 
diameter  of  this  specimen  when  entire  was  not  less  than  two  feet. 
Chemung  group.      Union  township,  Broame  county,  New  Yorh. 


\ 


^ 


X^^' 


i.. 


"^. 


I 


I 


290  1  )i(rrY()SPON«iD^ 

PLATE  XLVI. 

Species  from  the  Psammit^  du  Cmidroz,  mhmitted  ly  Professor  Ciiaeles 
Bakrois,  of  Lille,  Frarwe. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  LV,  LVI,  LXI.) 

DicTTOspoNGiA  MoKiNi,  Barrois  (sp.). 

Page  147. 

Figui'es  1,  2.     Views  of  the  original  specimen  showing  the  smooth,  regularly 
expanding  cup. 
Psammites  du  Condroz.     Jeurnont,  France. 

IIYDNOCERAS,  Conrad. 

Page  95. 
(See  Plates  II,  III,  IV,  V,  VI,  VII,  VIII,  IX,  X,  XI,  XII,  XXXHI,  XXXVI,  XxXVII,  XXXVIII,  XXXIX.) 

Hydnocebas  Babboisi,  nom.  nov. 

Page  147. 

Figures  ?>,  4.  The  original  specimen  of  IHctyophyton  tiiherosiim,  Barrois. 
This  is  a  curved,  strongly  constricted  cup,  with  moderately  strong 
nodes  upon  the  sharply  developed  prism-angles. 

Psammites  du  Condroz.     Jeiimont,  France. 

Hydnocekas  Jeumontense,  sp.  nov. 

Page  148. 

Figure  5.     A  specimen  with  three  horizontal  rows  of  nodes  and  showing  very 

gradual  expansion  of  the  cup. 
Figure  6.     Another  fragment  of  the  same  species  showing  some  variation  in 

the  number  of  nodes  in  each  horizontal  row.     An  additional  piece  of 

this  specimen  increases  its  length  by  about  60  mm.  and  has  traces  of 

two  rows  of  nodes. 

Psammites  du  Condroz.     Jeiimont,  France. 


i>ictyospo:jvoii>jv.. 


Memoirs  Geological  Survey  of  New  York. 


Plate  XLVI 


292  DictyosponoidjE. 


PLATE  XLVII. 

Species  from  tlie  Psammites  dii  Oondroz,  submitted  hij  Professor  Charles 
Barkois,  of  Lille,  Prance. 

EHABDOSISPONGIA,  gen.  nov. 

Page  116. 
(See  Plate  XVI.) 

RlIABDOSISPONGIA    CoNDROZIANA,  Sp.  nOV. 
Pago  149. 

Figures  1,  2.     Opposite  sides  of  the  original  specimen  showing  its  strong, 
unequal  annulations,  producing  a  highly  irregular  form,  and  also  the 
sharply  developed  prism-faces. 
Psammites  du  Condroz.     Jeamoiit,  Prance. 


13ICTYOSPO]VGin?E. 


Memoirs  Geological  Survey  of  New  York. 


Plate  XLVII 


C  Fausel,  iith 


294  DlCTY08HONOID-<E. 


PLATE  XLVIII. 
CALATHOSPONGIA,  gen.  no^^ 

Page  155. 
(See  Plates  XLIX,  L,  LI,  LIT,  LVI,  LVII,  LX,  LXVIII.) 

Calatiiospongia  Redfieldi,  Hall  (sp.). 

Page  lo5. 
(See  Plate  XLIX.) 

Figures  1,  2.  Opposite  sides  of  a  fine,  large,  nearly  entire  sponge,  with  the 
aperture  complete  on  one  side.  Tlie  median  swelling  of  this 
cup  seems  to  be  due  to  distortion  from  vertical  compression,  and 
the  base,  though  not  actually  retained  was  probably  not  greatly 
different  from  the  present  condition  of  the  lower  extremity.  The 
specimen  shows  the  characteristic  surface  features  with  great  clear- 
ness, and  over  the  apertural  region  as  well  as  upon  the  swollen  por- 
tion of  the  cup  in  fig.  2  may  be  seen  obscure  traces  of  the  prismatic 
faces. 
Waverly  group.     Akron,  Ohio, 


Q 

H 

0 

H 

H 


296 


DlClVOSl'ONOID^. 


PLATE  XLIX. 


CALATHOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  L,  LI,  LII,  LVI,  LVII,  LX,  LXVIII.) 

Calatiiospongia  Redfieldi,  Hall  (sp.). 

Page  155. 
(See  Plate  XLVHI.) 

Figure  1.     A    fragment  of  a   small  specimen 
Waverly  group,     Jiichjield,  Ohio. 

Figures  2,  3.  The  opposite  sides  of  a  specimen  which  is  ^^rtually  entire,  the 
apertural  portion  having  been  distorted  by  the  vertical  compression 
and  rupture  of  the  reticulum.  The  surface  of  this  specimen  has  been 
somewhat  abraded  on  the  side  shown  in  fig.  2,  and  the  primary 
transverse  bands  of  the  reticulum  are  not  as  clearly  displayed  as  in 
the  example  rejiresented  on  the  preceding  plate.  In  fig.  3,  the  finer 
supei-ficial  net-work  is  predominant. 

Figure  4.     An  enlargment  of  the  surface  from  the  same  specimen,  taken  from 
near  the  top  of  the  side  shown  in  fig.  3.     These  figures  are  from  the 
type-specimen  of  the  species. 
Waverly  group.     Harrisville,  Ohio. 

CLATIIROSPONGIA,  Hall. 

Page  121. 
(See  Plates  XIV,  XV,  XVIII,  XXI,  XXV,  XLIX,  L.) 

Clatheospongia  abacus.  Hall. 

Page  153. 

Figure  5.  A  nearly  complete  individual  which  shows  the  obconical  form  of 
the  cup  and  the  great  development  of  the  erect  vertical  and  horizon- 
tal spicular  lamellae. 

F'igure  6.  A  restoration  of  a  segment  of  the  surface  showing  the  relative 
development  of  the  successive  series  of  reticulating  lamellae,  Avliich 
produce  a  deep  fenestration  of  the  outer  spouge-wall. 

Figure  7.  A  restoration  showing  only  the  primary  vertical  and  horizontal 
bands  with  the  marks  of  subordinate  lamellae  upon  their  surface. 

Figure  8.     An  internal  cast  showing  the  aspect  of  the  fossil  ^vheu  divested  of 
its  exterior  lanlelhie. 
Waverl}'  group.      Warren, ^Pennsylvania. 


I 


Die  TYO  SI*0:N^CTir>iSu 


Memoirs  Geological  Survey  of  New  York. 


Plate  XLDC. 


R  P  V/  del 


Philip  Ast  lith. 


298  DicriYospoNGiD.*. 

PLATE  L. 
TIIAMNODICTYA,  Hall. 

Page  160. 
(See  Plates  LII  ami  LIII.) 

TlIAMNODICTYA    NeWBEKKYI,    Hall. 

Page  161. 
(See  Plate  LII.) 

Figure  1.  An  entire  frond  showing  its  broad  vase-shaped  expansion  and 
coarsely  reticulate  surface.  This  specimen  is  largely  an  internal  cast 
in  soft  shale  but  retains  the  features  of  the  exterior. 

Figures  2,  3.  Opposite  sides  of  an  internal  cast  of  a  small  frond  which  has 
been  ruptured  along  one  side,  btit  preserves  a  portion  of  the 
apertural  mai'gin  and  shows  the  disappearance  of  the  coarse  spicular 
bands,  and  tlie  predominance  of  the  finer  net-work  in  this  region. 

Figure  4.     An  enlargement  of  a  portion  of  the  surface  of  an  internal  cast. 

Figure  5.     An  internal  cast  of  the  expanded  portion  of  a  specimen. 

Figure  6.     An  enlargement  of  a  portion  of  the  surface  shoAvn  in  fig.  1. 
Waverly  group.     Cuyahoga  Falls,  Oliio. 

CALATIIOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  XLIX,  LI,  LII,  LVI,  LVII,  LX,  LXVIII.) 

Calatiiospongia  (?)  SACCULUS,  Hall  (sp.). 

Page  160. 

Figure  7.     A  view  of  the  specimen,  showing  its  shoi't  subcylindrical,  iinex- 
panded  form  and  the  character  of  its  reticidation. 
Waverly  group.     Jiiclifiekl,  Ohio. 

CLATHROSPONGIA,  Hall. 

Page  121. 
(See  Plates  XIV,  XV,  XVIII,  XXI,  XXV,  XLIX.) 

Clathrospongia  caprodonta,  sp.  nov. 

Page  154. 

Figures  8,  9.     Opposite  sides  of  an  essentially  complete  internal  cast,  which 
represents  a  long  slender  cup  with  a  subprismatic  surface  and  traces 
of  l(jw  nodes  at  the  angles  of  large  quadrules..    Only  the  coai'ser 
reticulation  of  the  skeleton  is  appai-ent. 
Waverly  group.     Portsmoufh,  Ohio. 


DTc  TYO  sr»  o:n^gii>  A. . 


Memoirs  Geological  Survey  of  NewYork. 


Plate  L 


GB  Simpson  del. 


Plulip  Ast  hth. 


300  DicTYOSpoNorn^. 

PLATE  LI. 
CLEODICTYA,   HaU. 

Page  163. 
(See  Plates  LXIX  and  LXX.) 

Cleodictya  Clatpolei,  sp.  nov. 

Page  163. 

Figure  1.     An  incomplete  example,  showing  the  small  size  of  the  sponge,  the 
basal   row  of  elongate  nodes  and  the   gradual,  vase-like    expansion 
above. 
Waverly  group.     Ah'on,   Ohio. 

CALATHOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  XLIX,  L,  LII,  LVI,  LVII,  LX,  LXVIII.) 

Calatiiospongia  caroeralis,  sp.  nov. 

Page  157. 
(See  Plates  LII  and  LX.) 

Figures  2,  3.  Opposite  sides  of  a  specimen,  evidently  entire,  but  somewhat 
irregular  at  the  aperture,  and  nearly  complete  at  the  base.  The 
specimen  shows  the  strong  reticulation  and  the  prominence  of  the 
primary  bands,  in  which  resjiect  it  is  unlike  other  species  referred  to 
the  genus,  though  agreeing  ^\  ith  them  in  its  form  and  broad  base. 
The  apertural  margin,  as  shown  in  fig.  1,  is  entire  and  regular  but 
the  growth  of  the  sponge  seems  to  have  been  abruptly  stopped  on 
this  side,  as,  upon  the  opposite  surface,  it  is  continued  for  a  consider- 
able distance  further  upward.  This  is  evidently  not  due  to  a  distor- 
tion of  the  skeleton  for  the  horizontal  bands  may  be  traced 
continuously  about  the  surface  without  deviation  from  their  plane. 

Figure  4.  A  larger,  nearly  complete  specimen.  This  is  a  figure  which  was 
used  in  the  original  illustration  of  Dictyophyton  Nemherryi  (Six- 
teenth Annual  Report  of  the  N.  Y.  State  Cabinet,  pi.  iv,  fig.  3),  but 
the  surface  represented  shows  much  less  distinctly  than  the  opposite 
side  of  the  specimen  the  coarse  primary  quadrules  and  spicular  bands. 
Waverly  group.     Michfield,  Ohio. 


13TCTY0S1^0X0IJ>i¥. 


Meinoirs  Geological  Survey  of  New  York. 


Plate  LI. 


G.B  Simpson  del 


Philip  Ast  lith 


Exi'I.ANATIONH    OK    PlATES  301 

Calatuosi'ongia  TiFFANYi,  sp.  nov. 

Page  168. 

Figures  5,  6.     Views  of  a  somewhat  compressed  specimen  showing  the  rate 
of  expansion  of  the  sponge  and  its  obscurely  prismatic  surface. 
Waverly  group.     Ohio  {precise  locality  not  hnowri). 


.JQ2  DicTYospoNoinie. 


PLATE  LII. 
THAMNODICTYA,  HalL 

Page  160. 
(See  PUtei Land  LIU.) 

Thamnodictya  Newbebryj,  Hall. 

Page  161. 
(See  Plats  L.) 

Figure  1.     The  upper  vase-shaped  part  of  a  large  iudividual,  the  apertural 
portion  being  bent  down  and  flattened. 
Waverly  group.     JRiehfield,  Ohio. 

CALATHOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  XLIX,  L,  LI,  LVI,  LVII,  LX,  LXVIII.) 

Calathospongia  carcekalis,  sp.  nov. 

Page  157. 
(See  Plates  LI  and  LX.) 

Figure  2.     An    essentially    entire  example  with  well  developed  prism-faces. 
Figure  3.     A  small  individual  with  obscure  prismatic  division. 
Waverly  group.     Richfield,  Ohio. 

Calathospongia  Caklli,  sp.  nov. 

Page  158. 

Figures  4,  6.     Two  views  of  a  nearly  entire  but  somewhat  distorted  specimen, 
showing  the  narrow  basal  part  and  wide  apertural  expansion. 

Figures  6,  7.     Opposite  sides  of  a  small  example,  essentially  entire  at  both 
extremities. 
Waverly  group.    Nea/r  Pleasantville,  Venango  county,  Pennsylvania. 


DTCTYOSl^OXOIDA^ 


Memoirs  Geological  Survey  of  New  York 


Plate  LII. 


G.B  Simpson  del. 


Philip  Ast  lith 


304  DurryosPONoii)^. 

PLATE  LIII. 
THAMNODICTYA,  Hall. 

Page  160. 
(See  Plates  L  and  LII.) 

TlIAMNODICTYA    OrTONI,    Sp.   nOV. 
Page  162. 

Figures  1,  2.  Opposite  sides  of  a  specimen  whicli  is  an  internal  cast  of  the 
expanded  apertural  portion  of  the  sponge,  the  smooth  lower  point 
representing  the  place  of  attachment  of  the  pedicel.  The  species  is 
characterized  by  the  fineness  of  the  reticulation,  the  peculiar  curva- 
ture of  the  vertical  bands  and  its  large  size. 
Waverly  group.     Moo^s  Ran,  Licking  county,  Ohio. 

TIIYSANODICTYA,  gen.   nov. 

«  Page  126. 

(Sea  Plates  XIX,  XXIII,  XXIV,  XXV,  XXVI,  XXVII,  XXXVIII,  XXXIX,  XL,  XLII.) 

TlIYSANODICTYA    EXPANSA,    Hall  (sp.). 
Page  154. 

Figure  3.     A  large  basal  disc  or  diaphragm  characterized  by  its  irregularly 
radiating  spicular  bands  and  nodose  periphery.     This  represents  an 
individual  of  great  size. 
Waverly  group.      Warren,  Pennsylvania. 

TYLODICTYA,   gen.  nov. 

Page  151. 

Tylodictya   ( ? )  TENUIS,  Hall  (sp.). 

Page  152. 

Figure  4.     A  small  portion  of  a  frond  showing  two  double  nodes,  one  more 
distinctly  divided.     Drawn  from  a  cast  of  the  original  impression. 

Figure  5.     The  original  illustration  of  this  species,  showing  the  impression  of 
one  of  the  double  nodes  and  the  extremely  fine  reticulation  of  the 
surface,  but  with  the  specimen  oriented  differently  than  in  fig.  4. 
Waverly  group.      Warren,  Pennsylvania. 


Explanations  ok  Vlatks.  305 

LYRODICTYA,   Hall. 

Page  170. 
(See  Plate  LVI.) 

LyRODICTYA    (?)    BURLINGTONENSIS,    Hall    (sp.). 
Page  165. 

Figure  6.     A  view  of  the  original  specimen,  which  is  a  fragment  of  a  broadly 
expanded  frond  with  strong  bundles  of  vertical  spicules,  compacted 
at  the  base  but  diffused  toward  the  apertural  portion.     The  reticula- 
tion is  uniformly  fine. 
Burlington  group  (Yellow  sandstones).     Burlinijton,  Iowa. 


306  DlCTYOSPONOID^.. 

PLATE  LIV. 

ACLCEODICTYA,  gen.  nov. 

Page  177. 
(See  Plates  LV,  LX,  LXI,  LXVIII.) 

AcLfEODICTYA  ECCKIfTRICA,  Hall  (sp.). 
Page  179. 

Figure  1.     A  basal  disc,  somewhat  compressed  on  one  side  but  shoAving  the 

continuity  of  the  radial  striae  across  the  apical  region,  in  two  sets 

reticulating  with  each  other. 
Figure  2.     A  portion  of  a  larger  disc  showing  a  tendency  to  the  formation  of 

elongate  nodes  or  ridges  near  the  periphery. 

Keokuk  group.     Crawfordnville,  Indiana. 

(?)  ECTENODICTYA,  Hall. 

Page  164. 

(?)  ECTENODICTYA  IMPLEXA,  Hall. 
Page  164. 

Figure  3.  A  large  irregularly  enfolded  frond  which  has  been  much  weathered. 
The  net-work  is  very  fine  but  is  ci'ossed  by  stronger  horizontal  bands 
like  those  in  fig.  4,  though  these  are  much  obscured  in  the  specimen 
and  are  not  represented  in  the  drawing.  Enclosed  in  this  specimen 
is  the  original  example  of  Clathrospongia  abacus,  figured  upon 
Plate  xlix. 
Waverly  group.      Warren,  Pennsylvania. 

Figure  4.     Another  fragmentary  sponge,  showing  the  horizontal  bands  of  the 
reticulum. 

Waverly  group.     Oil  City,  Pennsylvania. 

Both  of  these  specimens  have  a  reticulum  very  similar  to  that 
characterizing  CalatJiospongia  Redfieldi  and  C.  GarUi,  and  it  is 
probable  that  such  incomplete  examples  are  to  be  referred  to  one  or 
the  other  of  these  species. 


13IC  T  vo  s  i»  o:Nr  oi  i>  a\ 


Memoirs  Geological  Survey  of  New  York . 


Plate  LIV. 


■.'iiH^'^'?^^'^"'' 


/ 


E  Enriinons  del 


Philip  Ast  lith 


3Qg  Du  TYOSPONC.ID.E. 

PLATE  LV. 
PRISMODICTYA,  gen.  nov. 

rage  79. 
(See  Plates  XVII,  XVIII,  XIX,  XX,  XXI,  XXVII,  XXXIII,  XXXIV,  XXXV,  XXXVI,  XLI,  XLH.) 

Pkismodictya  polyiiedra,  sp.  nov. 

Page  168. 

Figures  1,  2.     Opposite  sides  of  the  original  specimen. 

Keokuk  group.     In  the  shales  at  Crawfordsville,  Indiana. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVI[t,  XLI,  XLII,  XMV,  XLVI,  LVI,  LXI  ) 

DlCTYOSPONGIA   CYLINDRICA,   Whitfield  (sp.). 
Page  166. 

Figure  3.     View  of  the  original  specimen  retaining  the  pyritized  skeleton. 
Keokuk  group.     Crawfordsville,  Indiana. 

ACLCEODICTYA,  gen.  nov. 

Pago  177. 
(See  Plates  LIV,  LX,  LXI,  LXVIII.) 

ACL(E0DICTYA   MAESIPUS,    Sp.    UOV. 

Page  178. 

(See  Plates  LX,  LXI,  LXVIII.) 

Figure  4.     Portion  of  the  specimen   with  pyritized   skeleton ;   showing  the 
extent  of   one  of  the  vertical  lamellae  of  the  outer  ^vall.  of   the 
sponge. 
Keokuk  group.     Crawfordsville,  Indiana. 

(?)  Lyrodictya,  Hall. 
Figure  5.     Enlargement  of  a  part  of  the  surface  of  a  young  sponge  ^vhich 
probably  belongs  to  a  species  of  this  genus ;  showing  the  pyritized 
spicular  rods  of  two  of  the  vertical  bundles.     The  quadrnles  bear 
faint  traces  of  a  fine  reticulation.     x3. 
Keokuk  gi-ouji.     Crawfordsville,  Indiana. 


DTCTirOSPOI«^GIJ>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  LV. 


■ 

1 

R      ''^M 

y 

1 

|,i' 

'm 

1 

Y                 ■' 

'1 
'"4 

* 

W;               '\ 

r; 

% 


U*-:- 


,'f/ 


G.B  Simpson  del. 


Pkilip  Ast  hth 


Explanations  ok  Platks,  309 

GRIPHODICTYA,  gen.  nov. 

Page  179. 

Gripiiodictya  epipiianes,  sp.  nov. 

Page  180. 

Figure  6.     The  specimen  from  which  the  spicules  represented  on  page  180 
were   taken.     This   sponge   has    to    some    degree    the    aspect    of 
Phragmodictya  catilliformis,  but  its  characteristic  spicules  have  not 
been  found  in  any  other  species. 
Keokuk  group.     OrawfordsvilU,  Indiana. 

HYDRIODICTYA,  gen.  nov. 

Page  77. 
(See  Plates  IV,  V,  XVI,  XXI.) 

Hydriodictya  ctlix,  sp.  nov. 

Page  78. 
(See  Plates  V  and  XVI.) 

Figure  7.     An  enlargement  of  a  portion  of  the  specimen  given  on  Plate  xvi, 
fig.  2,  showing  a  part  of  the  repaired  net-work  of  the  skeleton. 
Chemung  group.     Deyo  hasin,  Naples,  N.  Y. 


310  DicrYospoNoiD^ 

PLATE  LVI. 
LYRODICTYA,  Hall. 

Page  170. 
(See  Plafeg  LIII  and  LV.) 

Lyrodictya  Rominoeri,  Hall. 

Page  170. 

Figure  1.     A  view  of  the  original  specimen;  showing  the  strong  bundles  of 
rod-like,  vertical  spicules  which  are  increased  in  number  upward,  by 
intercalation,  and  also  the  finer  reticulation  over  the  broad   inter- 
spaces, in  which  the  horizontal  impressions  predominate. 
Keokuk  group.     Cratvfordsville,  Imliana. 

DICTYOSPONGIA,  gen.  nov. 

P»ge  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LXI.) 

DiCTYOSPONGIA    (?)    STYLINA    Sp.  UOV. 
Page  167. 

Figure  2.  A  long,  slender,  sponge-like  body  bearing  horizontal  and  vertical 
striae  near  the  top.  The  true  character  of  this  fossil  is  not  fully 
determined  but  it  is  illustrated  here  on  account  of  its  resemblance  to 

forms  of  DiCTYOSPONGIA. 

Keokuk  group.     Crmiifordmille,  Indiana. 
PHYSOSPONGIA,  Hall. 

Page  187. 
(See  Plates  LXI,  LXII,  LXIII.) 

Physospongia,  sp.  ? 

Figure  3.  A  portion  of  the  upper  part  of  a  small  example  in  Avhich  the 
primary  reticulatiug  bundles  of  spicules  are  retained  in  a  pyritized 
condition,  while  the  outer  portions  of  the  skeleton  are  lost. 

Figure  4.  An  enlargement  of  the  same  specimen  to  three  diameters,  showing 
the  square  and  regular  quadrules  made  by  the  intersecting  l)undles. 
The  specimen  is  interesting  in  showing  that  in  youth  the  surface  of 
the  sponge  is  without  nodes  or  depressions,  and  it  may  be  compared 
with  the  similar  condition  exhibited  by  the  senile  stage  shown  in 
figure  4,  Plate  Ixiii. 
Keokuk  group.     Crawfordsville,  Indiana. 


Die  TYO  Sl^  o:xGrii>is. 


Memoirs  Geological  Survey  of  New  York 


Plate  LVI. 


-^ 

>•• 

3 

!^ 

V        1 

■4^.  -^""^ 

G.B  Simpson  del 


Pliilip  Ast  hth 


Explanations  of  Platks. 


311 


CALATTIOSPONGIA,  gen.  nov. 

Page  155, 
(See  Plates  XLVIII,  XLIX,  L,  LI,  LII,  LVII,  LX,  LXVIIL) 

CaLATIIOSPONGIA  (?)  MAGNIFICA,  Sp.  nOV. 

Page  182. 

(See  Plate  LVII.) 

Figure  .5.     The  expanded  apertural  portion  of  a  large  sponge  showing  sub- 
prismatic  faces. 
Keokuk  group.     Ormjjfordsville,  Iruliana. 


MASTODICTYA,  gen  nov. 

Pngo  167. 

DicTYOspoNGiA  (Mastodictya)  osculata,  sp.  nov. 

Page  167. 

Figure  6.     A  view  of   the   specimen   described ;    showing  the   conical    pro- 
longation  at   the   left,  terminating   in   an    osculum,   the   imperfect 
portion  at  the  right  probably  representing  a  similar  process.     The 
reticulum  is  very  fine,  its  meshes  being  of  uniform  size. 
Keokuk  group.     Crawfordsville,  Indiana, 


;112  J3lCTV(».SP0NGIDA 


PLATE    LVII. 

CALATIIOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  XLIX,  L,  LI,  LII,  LVI,  LX,  LXVIII.) 

CaLATIIOSPONGIA  (?)  MAGNIFICA,    Sp.    nOV. 
Page  182. 
(See  Plate  LVI.) 

Figure  1 .     The  upper  portion  of  a  very  large  example,  retaining  a  part  of  the 
body  and  enough  of  the  apertural  e.xpansion  to  show  its  extent. 
Keokuk  group.     Grawfordsville,  Indiana. 


+-■ 

^^^^^^^^^^^^^^Ef    Y- 

- '«'(.' '— '^'^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^fe^^^^^^^^ 

< 

p. 

1 

■ 

f-l 
;^ 
en 

13 
o 

•ob 
o 

o 


o 

i 


O 

to 

CO 

o 


gj^  DlCTYOSPONGID^E. 


PLATE  LVIII. 
LEBEDICTYA,   gen.  nov. 

Page  169. 
(Se«  Plates  LIX  and  LXI.) 

Lebedictya  crinita,  sp.  nov. 

Page  169. 
(See  Plates  LIX  .-ind  LXI.) 

Figures  1,  2.  Opposite  sides  of  a  specimen,  probably  representing  most  of 
tlie  sponge,  retaining  the  aperture  complete  and  shoAving  its  mar- 
ginal fringe  of  long  spicules.  The  cup  has  an  essentially  smooth, 
gently  expanding,  slightly  prismatic  surface ;  the  face  exposed  in  fig.  2 
showing  the  finer  net-work  and  the  uneven  growth  of  the  reticulum 
at  the  margin,  while  the  surface  represented  in  fig.  1  shows  traces  of 
the  erect  outer  spicular  lamellae  which  form  coarse  quadrules,  and  at 
the  right,  where  exfoliated,  a  large  internal  bundle  of  vertical 
spicules. 
Keokuk  group.     Crawfordsville,  Indiana. 


M 

I 

h 

H 


f 


.•{IG  DlClYOSPONOIDiE. 


PLATE  LIX. 


LEBEDICTYA,  gen.  iiov. 

Page  169. 
(See  Plates  LVIII  aud  LXI.) 


LeBEDICTYA  CRINITA,  sp.  11 OV. 

Page  169. 

(See  Plates  LVIII  and  LXI.) 

Figure  1.     The  apertiiral  portion  of  a  cuji  showing  the  marginal  fringe. 

Figure  2.  A  large,  obliquely  compressed  cup  exhibiting  about  one-half  of  the 
apertural  margin  and  its  spicular  fringe.  In  both  of  these  specimens 
it  is  seen  that  there  is  a  difference  in  the  strength  of  the  spicules 
composing  the  fringe,  the  coarser  seeming  to  lie  in  the  line  of  the 
coarser  vertical  bundles  of  the  reticulum, 

Keoliuk  group,     OrawfordsvUle,  Indiami. 


I 


DTCTYOSl>OXOmi¥. 


Memoirs  Geological  Survey  of  New  York . 


Plate  LIX. 


<       /,- 


G  B  Simpson  del 


Plulip  Ast  htl-i 


a  18  DlOTYOSPONGID^. 

PLATE  LX. 

CALATIIOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  XLIX,  L,  LI,  LII,  LVI,  LVII,  LXVIII.) 

Calathospongia  carcebalis,  sp.  nov. 

Page  167. 
(See  Plates  LI  and  LII.) 

Figures  1,  2.  Opposite  sides  of  a  complete  specimen,  the  side  represented  in 
fig.  1  having  been  somewhat  abraded.  This  specimen  shows  the 
form  of  the  entire  sponge  from  aperture  to  base,  the  abrupt  trunca- 
tion of  the  latter,  without  evidence  of  basal  disc,  the  stout  body  and 
gradual  expansion  t()\vard  the  summit.  The  major  quadrules  of 
the  reticulum  are  somewhat  less  conspicuous  than  those  on  the  speci- 
mens from  the  Waverly  sandstone  represented  on  Plates  li  and  lii, 
but  this  difference  is  largely  due  to  the  mode  of  preservation. 
Keokuk  group.     Ci'awfordsville,  Indiana. 

ACLCEODICTYA,  gen.  nov. 

Page  177. 
(See  Plates  LIV,  LV,  LXI,  LXVIII.) 

ACLCEODICTYA   MARSIPUS,  Sp.  JIOV. 

Page  178. 

(See  Plates  LV,  LXI,  LXVIII.) 

Figure  3.  Latei-al  aspect  of  an  incomplete  specimen  showing  a  coarsely 
reticulated  surface  and  the  deep  basal  obcone  with  its  somewhat 
irregular  reticulation. 

Figure  4.  Side  view  of  a  small  and  somewhat  distorted  specimen  with  a 
finer  reticulation. 

Figure  5.  An  entire  individual,  showing  a  very  stout  siibcylindrical  cup  with 
a  coarse,  square  reticulation.  The  basal  obcone  is  relatively  deep, 
forming  a  very  obtuse  angle  with  the  walls  of  the  sponge,  and  shows 
the  absence  of  any  peripheral  fringe  and  also  the  irregular  converg- 
ence and  intersection  of  the  radial  bands. 
Keokuk  group.    Indian  Greek,  Indiana. 


DICTYOSl>OXOIl>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  LX. 


k 


G.B  Simpson  del. 


Philip  Ast  lith 


320  DiCTYOSPONGIDiE. 

PLATE  LXI. 
ACLCEODICTYA,  gen.  nov. 

Page  177. 
(See  Plates  LIV,  LV,  LX,  LXVIH.) 

AcLfEouicrYA  5IARSIPUS,  sp.  nov. 

Page  178. 
(See  Plates  LV,  LX,  LXVIII.) 

Figures  1,  2.     Opposite  sides  of  the  lower  portion  of  a  cup  which  is  referred 

Avith  doubt  to  this  species. 
Figure  3.     A  portion  of  the  body  of  a  cup  which  has  retained  impressions  of 

tlie  erect  surface  lamellae. 

Keokuk  group.     Crmvfordsville,  Indlaim. 

PIIRAGMODICTYA,  Hall. 

P.-ige  172. 
(See  PUatesLXIV,  LXV,  LXVI,  LXVII,  LXVIII  ) 

PlIltAGJlODICTYA    (?)    (JltEBUISTUIATA,    Hall. 
Page  176. 

Figure  4.     A  view  of  the  original  specimen  ^\•llich  appears  to  be  a  fragment 
of  the  apertural  portion  of  a  very  finely  reticulated  sponge. 
Keokuk  group.     Crawfordsville,  Indiana. 

LEBEDICTYA,  gen.  nov. 

Page  169. 
(See  Plates  LVIII  and  LIX.) 

Lebedictya   ckinita,  sp.  nov. 

Page  169. 
(See  Plates  LVIII  and  LIX. ) 

Figure  5.     A  portion  of  a  small  specimen  showing  the  aj)ertural  fringe. 
Keokuk  group.     Crawfordsville^  Indiana. 

DICTYOSPONGIA,  gen.  nov. 

Page  72. 
(See  Plates  XIV,  XV,  XVI,  XXVII,  XXXIII,  XXXVI,  XXXVIII,  XLI,  XLII,  XLIV,  XLVI,  LV,  LVI.) 

DiOTYOSPONGIA    CYLINDKICA,  Whitfield  (sp.). 

Page  166. 

(See  Plate  LV.) 

Figure  G.     A  fragment  of  a  small  example  ^vith  pyritized  spicides. 
Keokuk  group.     Crawfordsville,  Indiana. 


\\ 


13ICTYOSl»OKOIl>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  LXI. 


G.B  Simpson  del 


Philip  Ast  lith. 


Explanations  of  Plateh.  Ji^l 

PIIYSOSPONGIA,  IlalL 

Page  187. 
(See  Plates  LVI,  LXIl  an<l  LXIII.) 

PlIYSOSPONGIA  MULTIBURSARIA,  Sp.  DOV. 
Page  196. 

Figure  7.  A  portion  of  a  frond  which  is  a  flat,  undefined  expansion  showing 
a  finely  reticulate  surface.  The  vertical  areas  lying  between  the 
principal  spicular  bands  are  more  or  less  completely  covered  by 
pendulous,  overlapping  nodes  in  the  form  of  internal  casts  without 
much  evidence  of  reticulation.  No  other  example  of  this  species 
has  been  observed. 
Keokuk  group.     Crawfordsville,  Indiana. 


go2  DlOTYOSPONGIDiB. 

PLATE  LXII.     . 
PIIYSOSPONGIA,  Hall. 

Pago  187. 
(See  Plates  LVI,  LXI  nud  LXIII.) 

PiiYSospoNGiA  Dawsoni,  Whitfield  (sp). 

Page  189. 

Figure  1.  A  iM)i"tion  of  a  slender  cup  showing  the  difference  in  size  of  the 
primary  and  secondary  vertical  bundles,  the  alternation  of  the  nodes 
and  concavities  and  the  projection  of  a  few  spicular  tufts  at  the 
lateral  margins. 

Figure  2.  A  nearly  entire  specimen  of  average  size,  abnormal  in  having  one 
of  the  primary  vertical  strands  of  no  larger  size  than  those  of  the 
secondary  series.  The  I'apid  multiplication  of  the  qiiadrules  is 
observable  near  the  summit  of  the  specimen.  Some  of  the  nodes 
are  presented  in  profile  at  the  sides  of  the  specimen. 

Figure  8.  The  lateral  portit)n  of  an  individual  which  shows  in  the  matrix  the 
long  spicular  tufts  extending  from  the  interaections  of  the  vertical 
and  horizontal  bands. 

Figui-e  4.  The  median  portion  of  a  somewhat  worn  specimen  sho^ving  the 
alternating  vertical  strands,  nodes  and  concavities. 

Figure  5.  A  very  regularly  developed,  nearly  entire  cup  Avhich  I'etains  the 
fine  spicular  markings  of  the  quadiiiles,  shows  the  form  of  some  of 
the  nodes  in  profile  and  two  projecting  spicular  tufts. 

Figure  6.  A  specimen  which  has  been  somewhat  obliquely  compressed 
shows  the  broad  vertical  primaiy  spicular  bands  and  the  profile  of 
the  nodes  and  spicular  tufts. 

Figure  7.  A  larger  example,  showing  an  unusual  development  of  the  nodes, 
and  the  rapid  increase  of  nodes  and  concavities  toward  the  aperture, 
by  horizontal  division. 

Figure  8.  An  enlargement  of  the  lateral  portion  of  a  frontl  showing  the 
spicular  tufts  arising  at  the  intersection  of  the  vertical  and  horizon- 
tal bundles.  The  specimen  also  shows,  as  a  discoloration  of  the  matrix, 
an  erect  lamellar  film  connecting  the  spicular  tufts.  This  has  not 
been  observed  on  other  examples.     x3. 


dtctyospo:ngij>a\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  LXn. 


E  Emmons  del 


Philip  Asthth 


Exi'I.ANATIONS    OK    P[,ATKH.  323 

Figure  9      An  inctnnplete  exainple  of  medium  nize  in  which  the  priniaiy  ver 
tical  bands  are  broad  and  the  secondary  bands  exceedingly  narrow. 

Figui-e  10.     A   specimen    of  small  size;  showing  the  regularly    alternating 
ai-i-angement  of  the  nodes. 
Keokuk  group.     CrawfordnviUe,  Indiana. 

pHYSOSPONdfA    ALTKUNATA,    Hall. 
Page  195. 

Figure  11.     A  view  of  the  original  specimen;  showing  the  narrow,  subcylin- 
drical  form  with  large  quadrules  and  subequal  vertical  strands.     In 
the  latter  there  is  some  difference  in  size  although  not  clearly  shown 
on  the  somewhat  abraded  sui-face  here  represented. 
Keokuk  group.     Oraufordsville,  Indiana. 


324  DicTYospoNoiDJi:. 

PLATE  LXIII. 
niYSOSPONGIA,  Hall. 

Pago  187. 
(See  Plates  LVI,  LXI  and  LXH.) 

PlIYSOSPONGIA    COLLEITI,  Hall. 
Page  193. 

Fi<'ure  1.  A  nearly  entire  specimen  of  rather  small  size  showing  the  iisual 
aspect  of  the  species.  The  nodes  have  undoubtedly  been  somewhat 
flattened,  and  the  secondary  vertical  spicular  bundles  are  represented 
as  rather  too  broad. 

Figure  2.  An  enlargement  of  a  portion  of  the  surface  in  the  same  specimen, 
showing  two  nodes  and  their  corresponding  concavities,  each  pair 
being  separateil  by  a  major  vertical  spicular  l>aiid.  The  surface  of 
the  quadrules  shows  indications  of  a  finer  reticulation. 

Figure  3.  A  fragment  of  a  large,  somewhat  irregular  example,  showing  the 
obsolescence  of  the  nodes  near  the  aperture. 

Figure  4.  An  enlargement  of  the  surface  in  the  smooth  apertural  region  of 
the  foregoing  specimen,  showing  the  diminution  in  the  size  of  the 
sjiicular  bundles.     x2. 

Figure  5.  The  apertural  major  portion  of  a  sponge ;  showing  the  dift'usicju  of 
the  vertical  spicular  bauds  of  the  first  order,  and  the  multiplication 
of  the  (piadrules  by  increase  in  the  number  of  l)oth  horizontal  and 
vertical  bundles.  At  the  aperture  all  nodes  and  depressions  have 
become  obsolete. 

Figure  0.  A  portion  of  the  same  specimen  redrawn  to  show  more  distinctly 
the  multiplication  of  the  nodes  by  horizontal  and  vertical  division. 
Natural  size. 

Figure  7.     A  portion  of  a  very  large  specimen,  natural  size,  in  which  the  fine 
reticulation  of  the  quadrules  is  distinctly  retained. 
Keokuk  group.     Crawfo)-dsville,  Indiana. 


DICTYOS1»ON^OI1>A\. 


Memoirs  Geological  Survey  of  New  York 


Plate  LXra. 


G.B  Simpson  del 


Pluhp  Ast  lith 


:v2() 


DiCTYOSPONOIDJB. 


PLATE  LXIV. 

PIIllAGMODICTYA,  Hall. 

Page  172. 
(See  Plates  LXV,  LXVI,  LXVII,  LXVIH.) 

Phragmodictya  catilliformis,  Whitfield  (sp.). 

Page  173. 
(See  Plates  LXV,  LXVI,  LXVII,  LXVIII ) 

Fi<i'iire  1.  A  .small  specimen  which  lias  been  compressed  vertically,  disturb- 
ing the  form  of  the  cup  but  showing  the  transverse  basal  plate. 

Figure  2.     A  view  of  the  basal  surface  of  the  same  specimen ;  showing  the 
radiating  spicular  bauds. 

Figure  3.  The  basal  extremity  of  a  larger  example  to  >vhicli  the  peripheral 
frill  or  periloph  is  attached.  This  specimen  sliows  the  ridged  and 
creuulated  surface  of  the  frill,  the  eccentric  apex  of  the  base,  and 
retains  a  part  of  the  vertical  portion  of  the  sponge  though  in  a 
compressed  condition. 

Figure  4.  An  incomplete  specimen,  showing  the  approximate  outline  of  the 
entire  sponge.  This  example  has  the  surface  ridges  unusually  strong 
and  regular  and  some  of  them  are  evidently  continued  on  to  the 
periloph.     The  edge  of  the  basal  plate  is  very  sharply  defined.    ■ 

Figure  5.     An  imperfect  but  very  large  example,  exposing  the  basal  area, 
a  portion  of  the  broad  periloph  and  the  vertical  cup  in  a  compressed 
condition.     The  convergence  of  the  radial  spicular  striae  of  the  base 
is  very  clearly  defined.     This  is  the  original  specimen  of  the  species. ' 
Keokuk  group.     Crmvfurdsville,  Indiana. 


DICTYOSl»0:NOir>iS. 


Memoirs  Geological  Survey  of  NewYork. 


Plate  LXIV. 


E. Emmons  del 


Plulip  Ast  hth. 


328  DlCTYOSPON<JID.«. 


PLATE  LXV. 
PIIRAGMODICTYA,  Hall. 

Page  172. 
(See  Plates  LXIV,  LXVI,  LXVII,  LXVIII.) 

Phbaomodictya  catillifoemis,  Whitfield  (sp.). 

Pago  173. 
(See  Plates  LXIV,  LXVI,  LXVII,  LXVIII.) 

Figure  1.  The  lower  side  of  a  large  basal  disc  with  the  entire  periloph 
retained ;  showing  the  fine  radial  spicular  striations  of  the  former 
and  the  ridged  surface  and  somewhat  irregular  margin  of  the  latter 

Figure  2.     A  natural  half  section  of  a  basal  disc  with  a  highly  elliptical  form, 
undoubtedly  due  to  lateral  compression. 
Keokuk  group.     Crawfordsville,  Indiana. 

PlIUAGMODICTYA    PATKLLIFORMIS,  Hall. 
Page  176. 

Figure  3.  The  lower  surface  of  a  large  basal  disc  with  a  highly  eccentric 
apex. 

Keokuk  group.      From  the  sandy  calcareous  layei^s  overlyitig  tlie 
calcareous  shales,  at  GraiofordsviUs,  Indiana. 


M 

0 


>< 


U 


t 

I 


380  DlCTYOSPONOID^. 

PLATE  LXVI. 
PHRAGMODICTYA,  Hall. 

Page  172. 
(See  Plates  LXIV,  LXV,  LXVII,  LXVIII.) 

PlIUAGMODICTyA    CATILLIFOBMIS,  Whitfield  (sp.). 

Page  173. 

(See  Plates  LXIV,  LXV,  LXVII,  LXVIII.) 

Figure  1.     A  small  l);isal  disc  with  sharp  and  fine  radial  striae. 

Figure  2.  A  portion  of  a  large  specimen ,  show^ing  the  extremely  fine 
reticulum,  the  discontinuous  vertical  ridges  and  the  convexity  of  the 
basal  disc. 

Pigure  3.  A  portion  of  a  very  broad  periloph,  showing  the  fine  net-work,  the 
ridged  surface  and  the  irregular  free  margin. 

Figure  4.  A  part  of  the  reticulum  about  the  apertural  margin ;  showing  scat- 
tered nodes  and  nodiform  ridges,  and  a  smooth  surface  at  the  margin 
itself. 

Figure  .5.     A  small,  nearly  entire  individual,  with  broadly  expanded  aperture. 

This  specimen  retains  the  probable  outline  of  the  species,  except  for 

the  obli(piity  of  the  lower  pai't  of  the  cup. 
Figure  fi.     The  lower  jiortion  of  an  internal  cast,  showing  the  convexity  of 

the  basal  plate  and  the  regular  convergence  of  the  vertical  striae. 
Figure  7.     A  small  specimen,  entire  from  the  basal  disc  to   the  aperture; 

showing  the  expanding  form  of  the  cup. 
Figure  8.     A    specimen  of  moderately  large    size,   entire  except   about  the 

aperture,  the  parts  being  detached  at  the  basal  disc,  showing  the 

form  and  depth  of  the  latter  and  the  length  and  character  of  the 

j)erilopli. 

Figure  9.     The  same  specimen  with  the    parts  brought  together  into  their 
normal  condition. 
Keokuk  group.     CfmofordsvilU,  Indmna. 


13ictvosi»o:ngij>a\. 


Memoirs  Geological  Survey  of  New  York. 


Plate  LX  VI. 


G.B  Simpson  del 


Philip  Ast  hth. 


■ 


332  DlCTYOSl*<>NGIDJi. 


PLATE  LXVII. 
PHRAGMODICTYA,  Hall. 

Page  172. 
(See  Plates  LXIV,  LXV,  LXVI,  LXVIII.) 

Phragmodictya  catilliformis,  Whitfield  (sp.). 

Page  173. 
(See  Plates  LXIV,  LXV,  LXVI,  LXVIII.) 

Figures  1,  2.     Opposite  sides  of  a  slender  specimen,  whose  basal  disc  has  an 

elongate  apical  scar  of  attachment. 
Figure  3.      A    characteristic   example   of   average   dimensions   showing  the 

aperture  and  basal  disc. 
Figure  4.     The  apertural   portion   of  a  very  large  individual,  showing  the 

expanded  cup,  with  the  vertical  ridges  and  scattered  nodes  of  the 

surface. 

Keokuk  group.     CrmofordsvilU,  Indiana. 


nicTvo  si»o:Nroir>is. 


Memoirs  Geological  Survey  of  New  York. 


Plate  LXVII. 


E.Emmons  del 


Plulip  Ast  hth 


334  DiCI  YOSPONOin.E. 


PLATE   LXVIII. 
PIIRAGMODICTYA,   Hall. 

Page  172. 
(See  Plates  LXIV,  LXV,  LXVI,  LXVII.) 

PlIBAGMODICTYA  CATILLIFOBMIS,  Whitfield  (sp.). 

Page  173. 

(See  Plates  LXIV,  LXV,  LXVI,  LXVII.) 

Figure  1.  An  average  individual  retaining  the  skeleton  from  the  basal  disc 
n]n\'ard  and  preserving  the  usual  configuration  of  the  surface. 

Figure  2.     A  small  specimen,  exposing  the  basal  disc  and  part  of  the  periloph. 

Figure  3.  A  portion  of  a  moderately  large  example  showing  the  disc  and  the 
strong  vertical  surface  ridges. 

Figure  4.     The  concave  or  lower  surface  of  the  basal  disc  with   the  broad 
periloph  attached. 
Keokuk  group.     Crawfordsville,  Indiana. 

(?)  Phkagmodictya  lineata,  Hall. 

Page  176. 

Figure  5.     The  original  specimen,  which  is  a  fragment  of  a  smooth,  slender 
subcylindrical  cuj),  somewhat  expanded  about  the  lower  end.     It  is 
probably  not  a  species  of  Phragmodictya. 
Keokuk  group.     Grmjofordsville,  Indiana. 

CALATIIOSPONGIA,  gen.  nov. 

Page  155. 
(See  Plates  XLVIII,  XLIX,  L,  LI,  LII,  LVI,  LVII,  LX.) 

Calathosponqia  amphorina,  sp.  nov. 

Page  182. 

Figure  6.  A  specimen  with  the  flaring  upper  portion  somewhat  deflected  but 
showing  the  general  aspect  of  the  sponge.  The  surface  is  smooth 
and  the  reticulum  fine  -with  prominent  vertical  spicular  bundles. 
Keokuk  group.     Craicfui-dsvlUe,  Indiana. 


DICTYOSI>OXGIJ>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  LXVm. 


G.B  Simpson  del. 


Philip  Ast  hth 


EXIM-ANATI(»NK    OK    PlA'II-K.  335 

AC;L(E()DICTYA,  gen.  nov. 

Page  177. 
(See  Plates  LIV,  LV,  LX,  LXI.) 

AcLOiODICTYA    MARSIPITS,  Sp.  noV. 

PjiKO  178. 
(Seu  Plates  LV,  LX,  LXL) 

Figures  7,  8.     The  })asal   parts  of  two  specimens,  showing  the  character  of 
the  reticulum  and  the  broad  cui'vature  at  the  biusal  edge. 
Keokuk  group.     ludiait  Creek,  India/rM. 


;{.-{»;  DirryospONoiiht;. 


PLATE  LXIX 
CLEODICTYA,  HaD. 

Page  163. 
(See  Plates  LI  and  LXX.) 

Cleodictya  gloriosa.  Hall. 

Page  183. 
(See  Plate  I.XX.) 

Figure  1.  Lateral  view  of  the  original  specimen  which  shows  the  lower 
expansion  with  its  broad  obtuse  nodes,  and  retains  a  part  of  the 
upper  vase-shaped  extension. 

Keokuk  group.     From  ilie  sandstone  overlying  tlve  calcaremis  shales 
at  Crawfordsville,  Indiana. 


DICTYOS1*OXOI1>A\ 


Memoirs  Geological  Survey  of  New  York. 


Plate  LXIX, 


E  Emmons  del 


Philip  Ast  hth 


338  DurrY<)SH)N(iii)yK. 


PLATE  LXX. 

CLEODICTYA,  Hall. 

Page  163. 
(See  Plates  LI  aud  LXIX. ) 

Cleodictya  gloriosa,  Hall. 

Page  183. 

(See  Plate  LXIX.) 

Figure  1.     The  specimen  figured  on  the  preceding  plate,  viewed  from  above; 

showing  the  slight  lateral  compression.     The  expanded  portion  bears 

ten  well  developed  nodes  and  an  eleventh  in  an  incipient  state. 
Figure  2.     Basal  view  of  the  inferior  nodose  expansion  of  a  smaller  specimen, 

bearing    twelve    nodes,   some   of    them  obscurely    or    incipiently 

developed. 

Keokuk  group.     Crmvfordsville,  Indiana. 

Cleodictya  Moiiki,  Hall. 

Page  184. 

Figure  8.     The  original  specimen,  showing  the  low  basal  nodiferous  expansion, 
the  contraction  above  and  gradual  expansion  to  the  aperture.     The 
sj)icular  skeleton  in  this  specimen  is  largely  pyritized. 
Keokuk  group.     Oratvfordsville,  Indiana. 


JD IC  T  YO  S  1>  OX  Oil  >  A\ . 


Meinoirs  Geological  Survey  of  New  York 


Plate  LXX. 


G.B  Simpson  del 


PlulipAsthih 


I 


INDEXES 


To    Personal    Names,    Localities,  Families,  Genera 

AND  Species. 


INDEX 

To   Personal   Names. 


Allen,  J.,  r.,  105,  145. 

Ast,  Ph.,  0. 

Barrois,  C,  5,  45,  72,  95,  07,  100,  101,  147,  148, 
290,  202. 

Beecher,  C.  E.,  0,  48,  Oi5,  152. 

Carll,  J.  F.,  5.  158. 

Carpenter,  F.  H.,  117. 

Clarke,  J.  M.,  4,  0,  7,  51,  05,  100,  108,  170,  171, 
174,  175,  177,  178,  180,  184,  185,  18G,  190,  191, 
102,  104,  100,  280. 

Claypole,  E.  W.,  0,  103. 

Cole,  F.,  118. 

Conrad,  T.  A.,  3,  13,  41,  42,  95,  00,  97,  101,  ICl, 
202,  204,  206,  208,  210,  212,  214,  210,  218,  220, 
222,  204,  270,  272,  274,  270,  200. 

Cotton,  Thomas,  100,  103,  104,  IOC. 

Dawson,  J.  W.,  5,  23,  24,  25,  42,  44,  47,  48,  187. 

Doederlein,  22. 

Eichwald,  72. 

Emmons,  E.,  G. 

Freeh,  F.,  51. 

Girty,  G.  H.,  51,  04. 

Goode,  G.  B.,  0. 

Gray,  20,  21. 

Haeckel,  E.,  15. 

Hall,  E.  B.,  4,  74,  75,  77,  79,  82,  83,  85,  86,  87, 
88,  89,  00,  92,  108,  110,  114,  115,  110,  117,  118. 
127,  128,  130,  131,  132,  134,  136,  137,  140,  141, 
184. 

Hall,  James,  7,  42,  43,  44,  45,  40,  47,  40,  60, 
72,  73,  77,  78,  79,  80,  81,  82,  83,  86,  91,  95,  06. 
97,  109,  112,  110,  118,  110,  120,  121.  122,  123, 
124,  125,  120,  128,  129,  137,  1,30,  142,  143,  144, 
152,  1.53,  154,  155,  1.57,  160,  161,  163,  164,  165, 
160,  170,  172,  173,  170.  177,  179,  183,  184,  187, 
189,  193,  105,  202,  206,  226,  228,  230,  232,  234, 
230,  237,  238,  240,  241,  242,  244,  240,  248,  252, 
258,  260,  202,  208,  271,  282,  204,  200,  208,  300, 
302,  304,  305,  300,  308,  310,  320,  321,  322,  323, 
324,  326,  328,  330,  332,  334,  336,  338. 

Hinde,  G.  J.,  22,  23,  45,  47,  48,  03,  64,  72,  95,  07. 

Hyatt,  A.,  26. 

Johnston,  J.  S.,  131. 

Kemp,  J.  F.,  6. 


Larkin,  E.  P.,  42,  95. 

Lesley,  J.  P.,  83.  Ill,  139.  155. 

Lesquereux,  L.,  4(i,  111,  224. 

Luther,  D.  D.,  5,  OS,  6S),  70,  71,  72,  100,  105,  107. 

109,  143. 
Marsh,  O.  C,  02. 
McCoy,  F.,  41,  4.5,  63,  (M,  72. 
McKee,  J.  H.,  6. 
Morris,  41. 
Murchison,  B.  I.,  41. 
Neumayr,  M.,  48,  90. 
Newberry,  J.  S.,  3,  0. 
Orton,  E.,  162. 
Owen,  R.,  14,  28. 
BaufC,  H.,  15,  20,  21,  22.  23,  24,  25,  26,  27,  33. 

34,  49,  .50,  126,  127,  201. 
Bedfield,  W.  C,  156. 
Riley,  A.  H.,  117 

Roemer,  P.,  43.  45,  51,  03,  C7,  72,  95,  97,  189. 
Roth,  J.,  14.3. 
Russell,  I.  C,  (!. 
Salter,  J.  W.,  41. 
Schimper,  W.,  4.3,  95. 
Schlueter,  C,  49,  68. 
Schmidt,  O.,  21. 
Schuchert,  C,  6,  20,  49. 
Schulze,  F.  E.,  15,  18,  21,  28,  33,  40,  47,  142,  170. 

187. 
Sherwood,  A.  S.,  112,  143, 
Simonds,  F.  W.,  125. 
Simpson,  G.  B.,  0. 
Thomson,  W.,  14.  21,  22. 
Tiffany,  A.  S.,  5.  157,  160,  170,  179. 
TJlrich,  E.  O.,  48. 
Van  Deloo,  J.,  6.  142.  14.3. 
Vanuxem,  L.,  3,  13,  41,  42,  137,  139,  254,  250, 

288. 
Walcott,  C.  D.,  24.  25.  26.  200. 
Ward,  H.,  44. 
Whitfield,  R.  P.,  ,3,  6,  43,  44,  46,  9.5.  160,  172, 

173,  187,  189,  191,  308,  320,  32C,  328,  330,  332, 

334, 
Winchell,  N.  H.,  49. 
Wright,  A.  A.,  6. 
Zittel,  K.  von,  36,  4,5,  51,  95,  90,  146. 


»il 


INDEX 

To   Localities. 


Addison,  N    Y.,  41,  206. 

Akron,  O.,  15G,  Ift"?,  294,  300. 

Alfred,  N.  Y.,  5,  42,  73.  Ill,  12.3,  124, 

145,  222,  234,  241,  248,  2!)0,  2.53. 
Alma,  N.  Y  ,  80,  129,  244. 
Almond,  N.  Y.,  70,  77,  85,  230,  280. 
Andover,  N.  Y.,  73. 
Avoca,  N.  Y.,  38,  70,  70,  100,  101,  103, 
137.  142,  145,  202,  214,  210,  218,  220, 
282,  283. 
Bath,  N.  Y.,  38,  40,  101,  102,  103,  104, 

210,  2i4,  210. 
Benson,  England,  05. 
Erigsteer,  England,  64. 
Brittany,  France,  147,  148,  149,  150. 
Brown  Hill,  N.  Y.,  38,  45,  78,  99,  101, 
202,  204,  200,  208,  210,  212,  214,  258, 
274. 
Burlington,  Iowa,  166,  305. 
Canandalgua  lake,  N.  Y.,  07,  226. 
Cassadaga  croek,  N.  Y.,  127. 
Charleston,  Pa.,  112. 
Chemung  narrows,  N.  Y.,  123,  120,  220, 
Cohocton,  N.  Y.,  38,  40,  44,  45,  78,  202, 

214,  249,  258,  200,  272,  274. 
Concord  station.  Pa.,  84,  110,  240,  252. 
Condroz,   France,   147,  148,  149,  150. 
Cotton  hill,  N.  Y.,  38,  70,  70,  106,  137, 

270,  281,  282,  283. 
Crawfordsville,  Ind.,  3.  5,  13,  29,  44, 
108,  100,  170.  172,  175,  170,  177,  179, 
183,  184,  187,  180,  193,  195,  197,  .306. 
310,  311,  312,  314,  316,  318,  320,  321, 
326,  328,  330,  332,  334,  330,  338. 
Cuba,  N.  Y.,  38,  39,  73. 
Cuyahoga  Falls,  O.,  162,  298. 
Doyo  basin,  N.  Y.,  40,  70,  109,  120,  208, 

270,  272,  309. 
Eifel,  Germany,  67. 


125,  la-j, 


104,  106, 
276,  281, 

202,  204, 


108,  143, 
260,  272, 


248. 
208,  210, 


142,  214, 

166,  167, 
181,  182, 
308.  .300, 
323,  324, 


230,  242 


Erie,  Pa.,  119.  242,  280. 

Eriendslilp,  N.  Y.,  73,  81,  88.  226,  228,  230,  208, 

270. 
GemUnden,  Germany.  49. 
Gerolsteln,  Germany,  43,  45,  68. 
Gotland,  72. 

Great  Bend,  Pa.,  117,  2.30. 
Hamlin's  Farm,  Naples,  N.  Y.,  38,  40. 
Ilarrlsvllle,  Ohio,  150,  296. 
Hill  creek.  Pa.,  112. 

Hinsdale.  N.  Y.,  83,  129.  134,  135,  208,  276,  27a 
Holland  Patent,  N.  Y.,  24,  50,  200,  201. 
Howard,  X.  Y.,  101.  2.56. 
Indian  Civek,   Indiana,  29,   157,   179,   193,  318. 

335. 
Irish  hill,  N.  Y.,  38.  103,  202. 
Ischua,  N.  Y.,  a5.  118,   121,  127.  129,  135,  23«5, 

244,  253,  271,  274.  280,  282,  284. 
Italy,  N.  Y.,  101,  107,  215,  272,  282. 
Italy  hollow.  N.  Y.,  105. 
Ifliaca,  N.  Y.,  125,  228. 
Jamestown,  N.  T.,  73. 
Jenks  quarry,  N.  Y.,  101,  102.  103,  104.  202,  204. 

214,  216. 
Jeumont,  France,  45,  147,  148,  149,  150,  290,  292. 
Kendal,  England,  64. 
Kenwood,  N.  Y..  46. 
LawrenceviUe,  Pa.,  114.  143.  146,  224. 
Little  Genesee.  N.  Y.,  129,  2}4. 
Little  Sletis,  Canada,  23,  24.  25,  47,  48. 
Livonia,  N.  Y.,  51.  65,  00,  280. 
Lyndon,  N.  Y..  145,  260,  202. 
ilansfield.  Pa.,  40. 
Montrose.  Pa.,  117. 
Moots  run.  Ohio,  162,  304. 
Naples,  N.  Y.,  38.  40.  68.  69.  70,  72,  79,  100,  101, 

109,  120,  208,  204,  200,  309. 


StS 


344 


DlCnTOSPONOrDiE. 


XIIp.  X.  T.,  73,  \Xi. 

on  city.  Pn.,  la'i,  3oa 

on  cm'k.  Ph.,  158. 

Oloan,  N.  Y.,  73.  83,  84,  85,  127,  129,  132,  135, 

2C8,  283. 
Oweito,  N.  y..  140,  2.'>4,  250. 
Pirn-  Crook.  Pa..  l.W. 
PU-asantville,  Pa.,  15C,  158,  .302. 
Portsmouth,  Ohio,  154,  298. 
Potter  hill.  N.  Y..  101. 
RaiKlolph,  N.  Y.,  92,  118,  234,  230,  284. 
Klohfleld,  Ohio,  15C,  157.  160,  162,  296,  298,  300, 

.302. 
Solo,  N.  Y..  T.3.  8.5,  87,  88,  89,  90,  110,  114,  llC 

12.1.  127.  128.  1.35,  141,  210.  232,  238,  240,  24G 

252.  265,  266.  268.  270,  271. 


Spgur's  gnlly,  N.  T.,  215. 

Underbarrow,  England,  04. 

Union,  N.  Y.,  288. 

Vnrysburg,  N.  Y.,  70,  71,  106,  264,  265. 

Wallace,  N.  Y.,  99,  108. 

Warren,  Pa.,  84,  130,  151,  152,  153,  155,  158,  159, 

165,  240,  296,  304,  306. 
Wellsville.  N.  Y.,  4,  .38.  73,  74,  7.5,  79,  81,  82,  S3. 

84,  87,  88,  89,  92,  110,  111,  11.5,  117,  127,  129, 

1.30,  131,  132,  135,  136,  140,  203,  214,  226.  228. 

232.  233,  234,  237,  238,  239,  240,  242,  244,  246, 

252,  2.53,  280,  282. 
Westmoreland,  England,  41,  64,  72. 
Weston  Mills,  N.  Y.,  132,  274. 
Whitesville,  N.  Y.,  130,  248. 


INDEX 

To   Families,  Genera,   Etc. 


Numbers  in  black  type  indicate  place  of  description. 


Acloeodictya,  77.  and  C,  164,  172,  177,  .SOfi   308, 

318,  320,  33.5. 
Actinodictya,   Hall,    17,  49,   141,   142,   143,   144, 

258,  2C0. 
Aglithodictya,  77.  and  C,  145,  146,  249. 
Aglithodictyinse,  H.  and  C,  53,  50. 
Arystidictya,  77.  and  C,  136,  281,  283. 
AsconematidEO,  50. 
Astylospongla,  Roemer,  15. 
Botrj-odictya,  B.  and  C,  33,  95,  111,  114,  148, 

151,  190,  224. 
Brachiospongidoe,  48,  50. 
Calathospongia,  H.  and  C,  35,  30,  151,  152,  155, 

160,  182,  184,  294,  290,  300,  302,  311,  312,  318, 

334. 
Calathospongiinje,  77.  and  C,  53,  56. 
Ceratodlctya,  H.  and  C,  17,  35,  40,   117,  242, 

280,  282,  284. 
Clatlirospongia,  77n»,   19,  31,  66,  98,   121,  125, 

153,  170,  177,  220,  228,  234,  241,  248,  29C,  298. 
Cleodictya^  Hall,  32.  35,  30,  40,   163,   106,   175. 

178,  183,  300,  336,  338. 
Clepsydrospongia,  H.  and  C,  71,  266. 
Corynella,  45. 

Cryptodictya,  Hall,  17.  49,  141,  144,  200,  262. 
CyatUodlctya,  H.  and  C,  24,  35,  200. 
Cvathophycus,  Walcott,  23,  24,  47,  48,  50. 
DiagoiioUa,  Dawson,  23. 
Dlctyonina,  21,  22. 
Dicti/ophyton,  Hall,  13,24,  42,  43,  45,  46,  47,  48, 

72,  77,  79,  95,  90,  155,  1(!0,  101. 
Dlctyophytra,  Ratiff,  50. 
Dictyospongia,  B.  and  C,  17,  35,  30,  63,  65,  08. 

72,  70,  77,   147,   166,  226,  228,  230,  252,  253. 

264,  271,  275,  280,  282,  280,  290,  308,  310,  320. 
Dictyospongldio,  77((;/,  3,  4,  5,  7,  14.  18.  19,  21, 

22,  23,  28,  29,  32,  33,  45,  40,  47,  48,  49,  50,  54, 

147,  190. 


Dlctyospongllntc,  B.  and  C,  52,  54. 

(?)  Ettonodlclya,  Ball.  4«,  164,  177. 

Ectenodictya,  Hall,  177. 

Euplectella,  Owen,  17,  20,  21,  22,  24,  25.  27,  28, 

29,  40,  43,  45,  134. 
EuplectelUdae,  19,  25,  40,  47,  50. 
Oongylospongia,  77.  and  C,  92,  95. 
Grlphodlctya,  B.  and  C,  32,  179,  .309. 
Hallodlctya,  77.  and  C,  17,   140,  141,  144.  271, 

281. 
Hallodlctylnce,  B.  and  C,  53,  56. 
HoUcodlctya.  B.  and  C,  114,  242,  252. 
HexactlnelUda,  21,  54,  170,  187. 
Hlndia,  Duncan,  15. 
Holascua,  Schulze,  17. 
Hyalonema,  Orai/.  17,  21,  22. 
Hyalonematlda;,  50. 
Hydnoccras.  Conrad,  3.  19.  27.  32,  40,  41.  47.  48. 

63,  69,  71,  96,  96,  97,  108,  110,  111,  112.  113. 

141,  151,  154,  161,  202.  204,  206,  208,  210.  212. 

214,  216,  218,  220,  222,  264,  270.  272.  274.  270, 

290. 
Hydrlodlctya.  77.  and  C,  40,  77,  200,  208,  230, 

241,  309. 
Hyphantwnia,    T'onujrm    {emend.},   19,   20,    137, 

139,  188,  254,  256,  288. 
IlyphanUcnlina?,  77.  and  C,  .5.3,  .50. 
Ix?bodictya,  77.  and  V.,  32,  30,  99.  136,  169,  314, 

316,  320. 
Lyrodictya,  Ball,  31,  36,  46,  164,  165,  166,  170, 

175,  305,  308,  310. 
Lysactlnella,  Oirty.  51,  64. 
Lyssaclna,  21,  22.  54. 
Mastodlctya,  77.  and  C,  167,  310. 
.MonactiiM'lllda.  21. 
Oncosella.  Raun.  126. 
I'aphytrlchisnia.  146. 
Pheronema,  Leidy,  17. 


IMS 


346 


DlCrV  OSPONGIB^. 


Phorinowlla.  ff/iirfr.  23,  47,  50. 
rhruKnuMlUtja.  Hall.  17,  28,  32,  35,  40,  71,  125, 
130,  KM.  172,  177.  180,  320,  320,  328,  330,  332, 

334. 
PhmRnKMllcfya  ?,  155. 
Phragmodiclua.  160,  177,  181. 
Phrat/tmxIiHi/um.  Ilall.  4.'>. 
Physospongia.  nail.  19,  25,  27,  30,  31,  32,  35,  40, 

68,  139,  172,  175.  178,  187,  188,  189,  195,  196, 

107,  310.  .321.  .322.  .324. 
Pliysospongllnw,  H.  and  C,  53,  56. 
Pleotoderma,  Hlnde,  23.  47.  50. 
PlocttHlorniatldip,  worn,  propon.,  30. 
Plettosi)ongldH>,  Hinde.  23,  25,  36,  50.  51. 
Polylophus,  8chiil:e.  17. 
Prlsniodlctya,  H.  and  €..  19,  35,  36,   63,  69,  79, 

92,  110,  114,  108,  2.32.  2.34,  236,  238,  240,  252, 

253,  264,  266,  268,  270,  280,  282,  308. 
Protosjiongia,  SSaltrr.  13,  23,  45,  47,  49. 
Proto8i)ongida>,  22,  23,  30,  47,  50,  51. 
Ithatxlosispongia.  H.  and  C,  35.  116,  149.  230, 

292. 


Rhombodiotyon,  Whitfield,  46. 

nopmerlspongia.  fl.  and  C,  19  46,   07,  C8,  189. 

Bossellldse,  50. 

Slllcea.  20,  54,  144. 

Sphaerodlctya,  H.  and  C,  26,  200. 

StaurodermidCR.  22,  45. 

Stelligpongia,  45. 

Tpganium,  Rauff,  24.  50.  201. 

Tetractlnelllda,  21. 

Tetragonin,  MoCoy,  43,  45,  48,  72. 

Thamnodietya,  Hall,  35.  46,  160,  161,  162,  298, 
302,  304. 

Thysanodietya,  H.  and  C,  17,  19,  28,  36,  63,  71, 
121,  125,  126,  136,  154,  157,  164,  172,  173,  177, 
237,  244,  246,  248,  250,  253,  274,  276,  278,  283. 

Thysanodictyinee,  H.  and  C,  52,  56. 

Tylodictya,  H.  and  C,  95,  114,  151,  153,  304. 

Vphanto'nia,  Vanuxem,  3,  26.  43,  137.  187. 

Ventrlculitidse,  146. 

Verticillites,  45. 

Walteria,  Schulze,  143. 


INDEX 

To  Species. 


Numbers  In  black  type  indicate  place  of  deacrtpUon. 


Aoloeodictya  (?)  eccentrica.  Hall  (sp.),  179,  300. 
A.  marsipus.  H.  and  C,  31,  177,  178,  179,  308, 

318,  320,  335. 
Actinodictya   placenta,  Hall.  49,  100,  143,  258, 

260. 
Aglithodlctya  numulina,  H.  and  C,  145,  249. 
Arystidk-tya  elegans,  H.  and  C,  70,  100,   136, 

137,  142,  281,  283. 
A   nodifera,  H.  and.  C,  76,  106,  137,  142,  281. 
Bathydorus  flmbriatus,  Schulze,  170. 
Botryortk tya  ramosa,  Lisqiiereux  (sp.),  27,  40,  4G, 

111,  224. 
B   (or  Tylodietya)  sp.,  114. 
CalathosponRia  amphorina,  H.  and  C,  182,  334. 
C  carcei-alis,  H.  and  C,  157,  100,  300,  302,  318. 
C.  Carlli,  H.  and  C,  5,  158,  159,  163,  165,  184, 

302,  306. 
C.  (?)  magnifica,  H.  and  C,  182,  311,  312. 
C.  Redfleldi,  Hall  (sp.),  155,  158,  160,  163,  1&5, 

184,  294,  296,  306. 
C.  ?  sacculus.  Hall  (sp.),  160,  298. 
C.  Tiffany!,  H.  and  C.  158. 
Ceratodictya  annulata.  Hall  (sp.),  79,  109,  120, 

242. 
C.  Carpenterlana,  H.  and  C.  117,  119,  282,  284. 
C.  centeta,  H.  and  C,  120,  280. 
C.  cincta.  Hall  (sp.),  118,  242. 
C.  zonata,  H.  and  C,  119,  121,  280. 
Clathrospongia  abacus,  Hall,  44,  153,  296,  306. 
C.  caprodonta,  H.  and  C,  123,  154,  298. 
C.  (?)  desmla,  H.  and  C,  125,  248. 
C.  fenestrata,  Hall  (sp.),  122,  124,  22(i,  248. 
C.  (?)  Hamiltonensis,  Hall  (sp.),  36,  66,  226. 
C.  (?)  irregularis.  Hall  (sp.),  124,  228. 
C.  (?)  toraaeulum.  Hall  (sp.),  123,  124,  234,  241. 


C.  vascellum.  Hall  (sp.),  123,  241,  248. 
Clewlktya  Claypolel.  //.  and  C.  0.  163,  185. 
C.  glorlosa.  Hall,  46,  163,  183,  185,  192,  330,  .3.38 
C.  Mohri,  Hall,  30,  32,  4f!,  184,  18.->.  180,  .^3S. 
ClepsydrospoDgia  matutiua.  H.  and  ('..  71,  2»k». 
Cryptodlctya  AUenl,  HaU.  49,  105.  144,  260.  262. 
Cyathodictya  reticulata,    Walcott    (sp.),  25.  26, 

200. 
Cyathophycus  reticulata,  Walcott,  23,  24,  50. 
C.  subuphwricus,  Walcott,  24,  25,  26. 

C.  (Cvathodlctya)  retlculatus,  Walcott  (sp.),  24. 
Cyatliospongia  Qnelx'crnsin,  Dawson,  24. 
Diagonella  coronata,  Dawson,  23. 

D.  cyathiformis,  Datcmn,  23. 
Dietyophyton  abaciin.  Hall,  121,  122. 
D.  Amalthea,  Hall,  49,  116,  149. 

D.  angulatum,  Larkln,  42. 

D.  annulatum.  Hall,  42,  116,  120. 

D.  bacitlum.  Hall.  81. 

D.  catilliforme,  Whitfield,  173. 

D.  cinclnm,  Hall.  118,  119. 

D.  Conradi.  Hall.  42,  91. 

D.  cylindriciim.  Hall.  166. 

D.  cylindricum,  Whitfield,  166. 

D.  cylindricum,  I>arkin.  42. 

D.  Danbyi.  .McCoy  (sp.),  45.  4S.  63. 

D.  fenestratum.  Hall,  42,  122. 

D.  fllitrxtilr.  Hall,  42,  83.  80. 

D.  Ocrolstiinensix,  F.  Roemer,  45,  67. 

D.  Ham,  Hall.  125. 

D.  Hamiltonense.  Hall,  66. 

D.  hastatum,  Larkin,  42. 

n.  irregulare.  Hail,  124. 

D.  lobatum,  Larkin,  42. 

/).  ilorini,  Barrois,  147. 


M8 


DlClTf  OSPONO ID^. 


D.  Xrieherrvl.  Hall.  42,  157.  ICl,  300. 

D.  mxliMium.  Hall,  42.  109. 

U.  parallrlum.  Hall.  82. 

D.  piilulum.  I  lull.  77. 

D.  prismattcum.  Hall,  83. 

I),  ramoium,  Losqueroux.  46,  111. 

I).  Ramliilli,  Hall,  49.  129. 

D.  Redfieldi,  Hall,  42,  155,  160. 

/).  rude.  Hall.  42,  12S. 

I),  tacculum.  Hall.  160. 

D.  Knptnim.  Hall,  49. 

D.  scitiim.  Hall,  49,  128. 

D.  terratiim,  Larkin,  42. 

D.  tetum.  Hall,  79. 

D.  tenue.  Hall,  l.'>2. 

D.  tomacitliim.  Hall,  49,  123. 

D.  tuberosum,  Conrad  (sp.),  42,  43,  45,  97,  100, 
112,  147. 

D.  rnxcfUum,  Hall,  123. 

D.  [Calathospongia?]  sacculus.  Ball  (sp.),  46. 

D.  [CeratcMllctya]  clnotum,  Ball  (sp.),  46. 

D.  (Clathronpoiiyia)  abacus.  Hall,  1,')3. 

D.  (C]   Hauiiltonense,  Ball  (sp.),  46. 

D.  [C.J  Irregulare,  Ball  (sp.),  46. 

D.  [DIctyospongIa]  cyllndrlcum,  Whitfield  (sp.), 

44. 
D.  [D.]  Danbyl,  JtfeCoj/  (sp.),  47. 
D.  [D.]  Morlnl,  Barrois  (sp.),  45. 
D.  [Ectenwlictya]  impU'xum,  Boll  (sp.),  44. 
I>.  [Hydriodlctya]  patuluni.  Ball  (sp.),  46. 
D.  [Phragmodlctya]  catilllforme,  IF/ii/^eM  (sp.), 

44. 
D.  ?  IV.]  Halll,  Ball,  49,  126. 
D.  IPrlsuiodictya]  baculum.  Ball  (sp,),  46. 
D.  (P.J  parallelum,  Ball  (sp.),  4C. 
n.  (P. I  prlsiiiatleuni,  Ball  (sp.),  46, 
I>.  I  P.]  ti'luni,  Ball  (sp,),  46. 
D.  [Tyloilletya]  tenue.  Ball  (sp.),  44. 
I)l<fy«>pliytra  (?)  WalcottI,  Rauff,  50. 
Oirliiophyliim   liedflrldi,  Lesley,   155. 
DIctyoHpoiigla  Aliuoudensis,  B.  and  t'.,  76,  85, 

104,  286. 
D.  (?)  l>a<torin,  //.  and  C,  77,  275. 
D.  ebarita,  B.  and  ('.,  74,  228. 
D.  cylliidrlca,  Whitfield  (sp.),  166,  308,  320. 
D.  Danbyl,  McCoy  (sp.),  35.  63. 


D.  euniorpha,  B.  and  C,  75,  228. 

D.  baplea,  B.  and  C,  68,  204. 

D.  lophura.  B.  and  C,  74,  167,  228. 

D.  (?)  Marcellia,  Clarke  (sp.),  .SO,  31,  65,  139,  280. 

D.  Morlui,  Barrois  (sp.),  147,  290. 

D.  sceptnim.  Ball  (sp.),  38,  39,  72,  73,  125,  220, 

228,  230,  252,  271. 
D.  slrrea.  B.  and  €.,  75,  282. 
D.  (?)  stylina,  H.  and  C,  167,  310. 

D.  ?  (Ectenodictya?)  253. 

Dlctj'ospongia   (Mastodictya)    osculata,    B.   and 

C,  167,  311. 
Ectmodlctya  f  Burlingtonensis,  Hall,  164,  1C5. 

E.  eccentrica.  Hall,  164,  179. 
E.  expansa.  Hall,  154,  164. 

E.  implexa.  Ball,  153,  158,  164,  165,  306. 

E.   [Lyrodictya]  BurlinKtonensis.  Bull  (sp.),  4t>. 

Euplectella  a.spergillum,  Owen,  14,  40. 

E.  erassi.stellata,  lichiilze,  40. 

Euryp'.egraa  aurlcularis,  ^chiilzr,   142. 

Farrea  occa,  Bowerbank,  187. 

Gongylospongia  Marshi,  //.  and  C,  92,  93,  94. 

95. 
Grlphodlctya  epiphanes,  B.  and  ('.,  33,  179,  180, 

309. 
Hallodictya  Cottonlan.i,  B.  and  C.  70,  10(5,  137, 

142,  281. 
H.    Sciensls,    B.    and   C,    140,    141,    271. 
Helicodict.va  (?)  Concordia,  B.  and  ('.,  115,  2.")2. 
H,  (?)  Scio,  B.  and  C,  116,  2.">2. 
H.  trypania,  B.  and  C,  114,  115,  242. 
Byalostella  f  Marcellia,  Clarke  (sp.),  51,  05, 
Ilydnoceras  autliracis,  //.  and  ('.,  109,  203,  210, 

205. 
H.  Avoca,  //.  and  C,  09,  100,  102,  103,  202,  2H>, 

218,  220, 
H.  Barroisi,  B.  and  C,  45,  100,  147,  149,  290. 
n.  Bathense,  //.  and  C,  101,  102,  103,  204,  211, 

210. 
H.  botroedema,  B.  and  C,  .38,  102,  105,  109,  202. 
II,  eumeces,  B.  and  C,  110,  222. 
H.  eutheles,  B.  and  C,  70,  70,  105,  137,  214,  205. 
H.  gracile,  B.  and  C,  70,  264. 
H.  hypastrum,  //.  and  ('.,  105,  212. 
II.  Jeuinontense,  B.  and  C,  148,  290. 
H,  legatum,  B.  and  C,  69,  264. 


Indkx  to  Species. 


349 


n.  Lutlicil,  //.  and  C,  106,  215. 

H.  inultinoclosiun,  //.  and  ('.,  76,  106,  137,  142, 

276. 
H.  noflosuni.  Hull  (sj).),  109,  i;i7,  202. 
11.  pli.viiiMtoilcs,  //.  ai:d  ('..  104,  148,  210,  272. 
II.  rhopaliini,  //.  and  C,  100.  107,  272. 
H.  sp..  Ill,  214. 
H.  tuberosum,  Conrad,  13,  16,  38,  41,  42,  00,  76. 

96,  97,  !)8,  i)9,   102,    104,   106,   107,    108,    109, 

14:5,   202.   204,  206.   208,  210,   212,  214,  274. 
II.  tul)('ri)siim.  Coiiriul, 

var.  Klossoma,  //.  and  C,  22,  81,  101,  212,  216. 
H.  varlabile,  H.  and  C,  79,  108,  120,  270,  272. 
Hytlriodietya  cylix,  H.  and  C,  78,  120,  208,  230, 

309. 
H.  nephelia,  H.  and  C,  79,  241. 
H.  patula.  Hall  (sp.),  78,  100,  206. 
Hyphautenia   Chemungeusis,    //.   and   C,   139, 

254,  256,  288. 
Lebedictya  criuita,  ff.  and  C,  169,  170,  314,  316, 

320. 
liyrodictya  (?)  Burllngtonensls.  Hall  (sp.),  165, 

305. 
L.  Uoiningeri,  Hall,  6,  40,  170,  171,  310. 
Lysactinella  Gebhardl.  Girt//,  64. 
L.  pei'i'legans,  Girty,  64. 
Mastodictya  oseulatii.  H.  and  C,  16,  30,  167,  168, 

311. 
Oncosella  catinum,  Rauff,  126. 
I'alythoa  fatua,  21,  22. 
Phniginodiotya  catilliformis,  Whitfield  (sp.),  30, 

32,  134,  172,  173,  174,  175,  176,  177,  181,  300, 

326,  328,  330,  332,  334. 
P.  (?)  crebristriata.  Hall,  44,  176,  320. 
P.  eccentrica,  Hall,  179. 
P.  ej-pansa,  Hall,  154. 
(?)  r.  lincata.  Hall.  44,  176,  334. 
/'.  \iirb<Tri/i,  Hall,  1(!1. 
P.  patelllforuiis.  Hall,  44,  176,  328. 
P.  f  Rcdfteldi,  Hall,  155. 
/'.  seyphus.  Hall.  44.  173. 
P.  [Acla-odktya]  eccentrica.  Hall  (sp.),  44. 
P.  [Thysanodictya]  e.\p:insa.  Ha  I  (sp.>,  44. 
Physospougia  alternata,  //«.7,  4;,  195,  323. 
P.  Collettl,  Hall,  32,  46,  171,  188,  189,  190,  192, 

193,  194,  IOC,  324. 


P.   UawHonl,   Whltfiel<t  (8|»>.  ^L   ITl.   18S,   189, 

190,  101,  102.  193,  194.  106.  1»«.  107.  322. 
P.  inulllbumirlH.  //.  and  C.  188.  196,  321. 
Prlsniodlctya  AlliKailla,  //.  and  C.  84,  238. 
P.  amIcltlH!,  H.  and  C,  87,  270. 
P.  aulophla,  H.  and  V.,  88,  2:«».  2m.  280. 
P.  baculuiii.  Hall  (sp.).  81,  85,  234. 
P.  bauaiio,  //.  and  C,  87,  2.T2. 
P.  cercldea,  //.  and  ('..  H8.  89,  !H),  238.  282. 
P.  chonnea,  //.  and  v.,  85,  86.  \M.  236.  2.-..'$,  208. 
P.  cltbara,  //.  and  ('.,  88,  89,  90,  2<W,  270. 
P.  Conrndl,  Hall  (sp.).  81,  91,  05.  234.  236. 
P.  corynla.  B.  and  C.  85.  86,  233. 
P.  filltextllls.  Hall  (sp),  86,  240,  252. 
P.  nartheela,  //.  and  C.  90,  2(W. 
P.  paltea,  B.  and  C,  69,  2<U. 
P.  parallela.  Ball  (sp.),  81,  82,  84,  236.  238.  240, 

268. 
P.  iwlyhedra,  H.  and  C,  168,  3()tj. 
P.  prlsmatlca.  Ball  (sp.).  83,  84.  110.  232,  240. 
P.  cf.  piisiuatlca,  282. 
P.  ptiouia,  H.  and  C.  88,  80,  00,  266. 
P.  sijectabllls,  B.  and  C,  82,  238. 
P.  sp.?.  271. 

P.  teluui.  Ball  (sp.),  22,  80,  87,  88,  232,  208,  282. 
ProtosiHjngla  corouata,  Datcson,  23. 
P.  delleatula,  Datcson,  25. 
P.  mouonema,  Daicnon,  23. 
P.  rhenana.  Hchliitei;  49,  08. 
P.  tetranema,  Dawson,  23. 
Rnuffrlla  fllona,  Ulrlch.  48 

Rhabdosispoiigia  Araalthea,  Hall  (sp.),  116,  2.30. 
R.  Condrozlaun,  B.  and  C.  149,  202. 
Roenierlspongla  Gerolstelneusis.  F.  Roemer  (sp.), 

35,  67,  189. 
Spliiprodictya  subsphaerica,  Walrott  (sp.),  200. 
Teganium  subtphwricum,  Walcott  (sp.).  24.  25.  2«5. 
Tetragonis  ttanhyi,  McCoy,  41,  43,  45,  03,  64,  72. 
T.  Kifflcnuiif,  K.  Rot'iuer,  43. 
r.  Murchlsonl,  Eichwald,  43,  45,  72. 
T.    [Prlsmodlctya]    Elfelensls,   F.   Roemer  (sp.). 

45,  07. 
Thamnodlctya  Newberryl,  Ball,  161,  102,  298, 

302. 
T.  Ortoni,  B.  and  C,  162,  304. 
Thysanodictya  apleta,  U.  and  C,  135,  250,  25T 


;J50 


DlCTYOSPONGID^ 


T.  Mwln-llalll.  //.  iitid  C,  38,  126,  12.S,  mi.  IST., 

T.  oximiisa.  Halt,  (sp.),  154,  304. 

T.  lii-riiH-nla.  //.  and  C  133,  134.  253,  270.  278. 

T.  Johiistoiil.  //.  and  C,  130,  244. 

T.  iHi-cllus,  a.  and  C,  127,  157,  246. 

T.  (imisUliim.  /7.  mtd  C.  131,  l.^^.  274. 

T.  ItniKl.illl.  Hall  (sp.).  12.-.,  129,  1.32,  24C. 

T.  ni.Ils,  Hall  (sp.),  125,  127  128,  129,  130,  131, 

132,   155,  237,  244. 
T.  sact-us,  a.  and  C,  130,  248. 
T.  sc.rphina,  n.  and  C,  132,  283. 


T.  turriculn.  n.  and  C.  135,  274. 

TylotUctja   Warrfiionsis,   //.  and  V.,    151,   152, 

153. 
T.  (?)  tenuis,  Uall  (sp.),   152,  304. 
Vphantwnia  Chemunyriisin,  \'aiuixcm,  13,  42,  137, 

139. 
U.  Dawsoni,  Wliltfleld.  44,   189,   191. 
U.  [Ilypliantipnia]      (llieinungeusis,       VaHuxcm 

(sp.),  41. 
U.   [Physosi)ongla]  Dawsoni.  Whitfield  (sp.),  44, 

48. 
I  Walteria  Flominingl,  Schul::c,  142. 


^iT: 


-,.'i>,-;''T. 


':'^. 


•>'  /■■ 


y^Lfi-ii^i 


'^■ 


'■'r-'l^'i^UU'^v'i'';,'^^ 


■Ia.'J  !,'  ,>/' 


'■/>:'■:  ■■).y.->:. 


uy'-