﻿^°1- 
  53-] 
  SIGILLARIA 
  AND 
  GLOSSOPTERIS 
  IN 
  SOUTH 
  AFRICA. 
  331 
  

  

  is 
  very 
  considerable 
  variation 
  as 
  regards 
  the 
  proximity 
  of 
  the 
  leaf-scars 
  

   and 
  the 
  size 
  of 
  the 
  leaf-cushions. 
  In 
  Weiss's 
  paper 
  on 
  this 
  subject 
  

   he 
  gives 
  a 
  figure 
  of 
  a 
  single 
  cushion 
  and 
  scar 
  of 
  a 
  form 
  spoken 
  of 
  as 
  

   Sigillaria 
  wettinensis, 
  Weiss, 
  1 
  which 
  cannot 
  be 
  separated 
  specifically 
  

   from 
  S. 
  Brardi 
  ; 
  in 
  this 
  form 
  the 
  leaf-cushion 
  is 
  well 
  marked 
  and 
  

   prominent, 
  and 
  the 
  relative 
  size 
  of 
  the 
  leaf-scar 
  and 
  cushion 
  agrees 
  

   closely 
  with 
  that 
  shown 
  in 
  fig. 
  2f. 
  If 
  this 
  specimen 
  were, 
  however, 
  

   specifically 
  identical 
  with 
  that 
  shown 
  in 
  PL 
  XXIIL, 
  we 
  should 
  expect 
  

   a 
  greater 
  equality 
  in 
  the 
  size 
  of 
  the 
  leaf-scars. 
  On 
  the 
  whole 
  I 
  

   incline 
  towards 
  the 
  view 
  that 
  this 
  impression 
  may 
  be 
  that 
  of 
  a 
  

   Sigillaria, 
  but 
  the 
  preservation 
  is 
  too 
  imperfect 
  to 
  admit 
  of 
  any 
  accu- 
  

   rate 
  determination. 
  A 
  comparison 
  of 
  PI. 
  XXII. 
  fig. 
  4 
  a, 
  PI. 
  XXIV. 
  

   fig. 
  3, 
  and 
  text-fig. 
  2/ 
  with 
  such 
  a 
  drawing 
  as 
  that 
  given 
  by 
  Potonie 
  

   of 
  Lepidodendron 
  Volhmannianum, 
  Sternb., 
  2 
  shows 
  a 
  very 
  close 
  

   agreement. 
  This 
  species, 
  however, 
  characterizes 
  a 
  much 
  lower 
  

   horizon 
  than 
  that 
  at 
  which 
  Sigillaria 
  Brardi 
  is 
  usually 
  found. 
  

  

  The 
  same 
  piece 
  of 
  rock 
  as 
  that 
  on 
  which 
  this 
  Lepidodendroid 
  im- 
  

   pression 
  occurs 
  contains 
  several 
  fragments 
  of 
  Glossopteris, 
  one 
  of 
  

   which 
  is 
  seen 
  in 
  PI. 
  XXII. 
  fig. 
  4 
  at 
  c, 
  and 
  on 
  a 
  larger 
  scale 
  in 
  

   text-fig. 
  1 
  6 
  at 
  G 
  (p. 
  324). 
  

  

  In 
  the 
  important 
  posthumous 
  monograph 
  by 
  Weiss, 
  3 
  completed 
  

   by 
  Sterzel, 
  the 
  name 
  Sigillaria 
  mutans, 
  Weiss, 
  is 
  adopted 
  as 
  the 
  

   designation 
  of 
  a 
  specific 
  type 
  which 
  includes 
  a 
  very 
  great 
  variety 
  

   of 
  forms. 
  In 
  this 
  species 
  Sigillaria 
  Brardi, 
  S. 
  wettinensis, 
  and 
  nu- 
  

   merous 
  other 
  examples 
  of 
  both 
  Leiodermarian 
  and 
  Clathrarian 
  forms 
  

   are 
  included. 
  The 
  reasons 
  advanced 
  by 
  Weiss 
  for 
  the 
  substitution 
  

   of 
  the 
  specific 
  term 
  mutans 
  for 
  such 
  well-marked 
  types 
  as 
  Brardi 
  

   and 
  others 
  are 
  hardly 
  convincing. 
  The 
  imperfect 
  specimen 
  from 
  

   Yereeniging 
  bears 
  a 
  very 
  close 
  resemblance 
  to 
  S. 
  mutans 
  forma 
  

   Brardi 
  var. 
  sublwvis 
  figured 
  by 
  Weiss 
  & 
  Sterzel 
  in 
  their 
  pi. 
  xvi. 
  

   Compare 
  also 
  S. 
  mutans 
  forma 
  wettinensis 
  var. 
  convexa 
  (pi. 
  xii) 
  ; 
  

   and 
  S. 
  mutans 
  forma 
  urceolata 
  (pi. 
  xiii). 
  

  

  Coxites 
  sp. 
  (PI. 
  XXII. 
  fig. 
  2.) 
  

  

  Sternberg, 
  ' 
  Flor. 
  Vorwelt,' 
  fasc. 
  iii. 
  (1823) 
  p. 
  36 
  ; 
  Seward, 
  Cat. 
  Mesozoic 
  Plants 
  

   Brit. 
  Mus., 
  Wealden 
  Flora, 
  vol. 
  ii. 
  (1895) 
  p. 
  113. 
  

  

  This 
  specimen, 
  2 
  cm. 
  in 
  length 
  and 
  1 
  cm. 
  broad, 
  occurs 
  on 
  the 
  

   same 
  slab 
  of 
  sandstone 
  as 
  the 
  Gangamopteris-le&l 
  of 
  PI. 
  XXII. 
  fig. 
  1. 
  

   The 
  surface 
  shows 
  several 
  distinct 
  four- 
  to 
  six-sided 
  areas, 
  measuring 
  

   2 
  mm. 
  in 
  length 
  and 
  from 
  1 
  to 
  1*5 
  mm. 
  in 
  breadth, 
  which 
  appear 
  

   to 
  be 
  the 
  proximal 
  ends 
  of 
  thick 
  scales 
  which 
  expanded 
  distally, 
  as 
  

   seen 
  on 
  the 
  sloping 
  side 
  of 
  the 
  fossil 
  at 
  s. 
  The 
  proximal 
  ends 
  appear 
  

   to 
  be 
  solid 
  ; 
  the 
  fractured 
  appearance 
  of 
  a 
  few 
  of 
  the 
  areas 
  is 
  

   probably 
  accidental, 
  and 
  does 
  not 
  indicate 
  originally 
  hollow 
  scales. 
  

   Had 
  the 
  scales 
  been 
  hollow, 
  the 
  specimen 
  would 
  have 
  presented 
  a 
  

  

  1 
  Zeitschr. 
  Deutsch. 
  geol. 
  Gesellsch. 
  vol. 
  xl. 
  (1888) 
  p. 
  569. 
  (Included 
  by- 
  

   Weiss 
  in 
  a 
  later 
  work 
  as 
  a 
  ' 
  form' 
  of 
  S. 
  mutans, 
  Weiss.) 
  

  

  2 
  Abh. 
  k. 
  Preuss. 
  geol. 
  Landesanst. 
  n. 
  s. 
  pt. 
  xxi. 
  (18915) 
  p. 
  43, 
  fig. 
  43. 
  

  

  3 
  ' 
  Die 
  Sigillarien 
  der 
  Preuss. 
  Steink. 
  Koth. 
  Gebiete,' 
  pt. 
  ii., 
  Weiss 
  & 
  Sterzel, 
  

   Abb. 
  k. 
  Preuss. 
  geol. 
  Landesanst. 
  n. 
  s. 
  pt. 
  ii. 
  (1893). 
  

  

  