Skip to main content

Full text of "The American species of Marchantia"

See other formats 


Science 

Q 

11 

.  C9 

21:3 


TRANSACTIONS  OF  THE 

NNECTICDT  ACADEMY  OF  ARTS  AND  SCIENCES 


Incorporated  a.  D.  1799 


VOLUME  21,  PAGES  201-313 


MARCH,  1917 


The  American  Species 


of 


Marchantia 


BY 


ALEXANDER  W.   EVANS 


NEW   HAVEN,   CONNECTICUT 
1917 


TRJBSACTIONS  OF  THE 

CONNECTICUT  ACADEMY  OF  ARTS  AND  SCIENCES 

Incorporated  A.  D. 1799 

VOLUME  2i,  PAGES  201-313  MARCH,  1917 


The  American  Species 

of 

Marchantia 


BY 

ALEXANDER  W.  EVANS 


NEW  HAVEN,  CONNECTICUT 

1917 


Sc.^e^o^^^-^'^'^Y 


TABLE    OF    CONTENTS 


I.     INTRODUCTION 

II.     MORPHOLOGICAL  NOTES  ON  THE  GENUS 
I.     Epidermis  and  Epidermal  Pores 
Compact  Ventral  Tissue 
Ventral  Scales 
Rhizoids 
Receptacles    . 
Sporophyte 
Cupules  . 


III.     DESCRIPTION  OF  SPECIES 

Section  I.     Astromarchantia 

1.  AI.  polymorpha  L.   , 

2.  M.  plicata  Nees  &  IMont. 

3.  M.  Berteroana  Lehm.  &  Lindenb. 

Section  II.     Chlamidium 

4.  M.  paleacea  Bertol. 

5.  M.  hreviloha  Evans 

6.  M.  domingensis  Lehm.  &  Lindenb. 

7.  M.  papillata  Raddi 

8.  M.  Bescherellei  Steph. 

9.  M.  chenopoda  L.  ... 

Doubtful  Species 


PAGE 

205 

208 
210 
217 
218 
221 
221 
227 
228 

229 

232 
232 

241 
246 

253 
253 
265 
269 
280 
287 
290 

312 


LIST    OF    ILLUSTRATIONS 


PAGE 

Fig.    I.     Marchantia  polymorpha,  appendages    of    ventral 
scales  ....... 


Fig.    2.     Marchantia  polymorpha,  anatomical  details 
Fig.    3.     Marchantia  plicata,  appendages  of  ventral  scales 
Fig.    4.     Marchantia  plicata,  anatomical  details 
Fig.    5.     Marchantia  Berteroana,    appendages    of    A-entral 

scales  and  other  anatomical  details    . 
Fig.    6.    Marchantia  paleacea,  ventral  scales 
Fig.    7.     Marchantia  paleacea,  portions  of  ventral  scales 
Fig.    8.     Marchantia  paleacea,  anatomical  details 
Fig.    9.     Marchantia  hreviloha,   plants,    natural   size,    and 

various  anatomical  details 
Fig.  10.     Marchantia  domingensis,  appendages   of   ventral 

scales  ....... 

Fig.  II.     Marchantia    domingensis,    epidermal    pores     of 

thallus         ....... 

Fig.  12.     Marchantia  domingensis,  anatomical  details 
Fig.  13.     Marchantia  papillata,  appendages  of  ventral  scales 

and  other  anatomical  details 
Fig.  14.     Marchantia  Bescherellei,  anatomical  details 
Fig.  15.     Marchantia  chenopoda,    appendages    of    ventral 

scales 

Fig.  16.     Marchantia  chenopoda,    appendages    of    ventral 

scales  ....... 

Fig.  17.    Marchantia  chenopoda,    appendages    of    ventral 

scales . 

Fig.  18.    Marchantia  chenopoda,    appendages    of    ventral 

scales  ....... 

Fig.  19.     Marchantia  chenopoda,   epidermal    structures   o 

thallus         ....... 

Fig.  20.  Marchantia  chenopoda,   anatomical  details  . 


236 
239 
243 

245 

250 
257 
259 
262 

267 

275 

277 
279 

284 
289 

294 
298 
301 
303 

305 
309 


I.     INTRODUCTIOX 

The  genus  Marchantia  is  almost  world-wide  in  its  distribution 
and  includes  some  of  the  largest  and  most  conspicuous  of  the 
Hepaticae.  According  to  the  current  rules  of  nomenclature  the 
genus  was  not  definitely  established  until  1753,  when  Linnaeus 
published  it  in  the  first  edition  of  his  Species  Plantarum,  but 
the  use  of  the  name  Marchantia  dates  from  the  year  1713. 
Linnaeus  recognized  seven  species,  only  the  first  two  of  which, 
M.  polymorpha  and  M.  chenopoda,  are  now  retained  in  the  genus. 
The  type  species,  M.  polymorpha,  he  cites  from  Europe  only, 
and  gives  Martinique  as  the  habitat  of  M.  chenopoda.  At  the 
present  time  M.  polymorpha  is  known  to  be  almost  cosmopolitan, 
while  the  range  of  M.  chenopoda,  although  apparently  restricted 
to  tropical  America,  is  likewise  very  extended. 

For  a  long  time  M.  polymorpha  was  the  only  species  recognized 
in  Europe.  In  1817,  however,  a  second  species,  M.  paleac-ea,  was 
described  by  Bertolini^  from  material  collected  in  Italy.  This 
species  had  been  distinguished  and  figured  by  Micheli^  nearly  a 
century  earlier  but  had  not  been  accepted  by  Linnaeus.  It  is 
now  known  to  have  a  wide  distribution  in  tropical  and  subtropical 
regions,  its  range  extending  far  beyond  the  confines  of  Europe. 
Other  European  species  which  have  been  proposed  from  time 
to  time,  such  as  M.  macrocephala  Corda  and  AI.  Sykorae  Corda, 
have  never  received  wide  acceptance  and  undoubtedly  represent 
mere  forms  of  M.  polymorpha. 

The  history  of  the  genus  in  America,  when  the  entire  continent 
is  considered,  is  very  much  involved.  This  is  due  partly  to  the 
full  representation  of  the  genus  and  partly  to  the  confusion 
which  has  arisen  in  the  interpretation  of  certain  species. 
Before  the  publication  of  Gottsche,  Lindenberg  and  Nees  von 
Esenbeck's  Synopsis  Hepaticarum,  in  1847,  the  following  species 
of  Marchantia  had  been  recorded  from  North  and  South  Amer- 
ica: M.  papillata  Raddi  (1823)  from  Brazil,  M.  platycnemos 
Schwaegr.  (1827)  from  Brazil,  M.  Szuart::ii  Lehm.  &  Lindenb. 
(1832)   from  Jamaica,  M.  squamosa  Raddi  (1832)   from  Brazil, 

^Opus.  Sci.  Bologna  1:242.     1817. 

^  Nova  Plant.  Gen.  2.  pi.  i,  f.  4.    Florence,  1729. 


2o6  Alexander  W.  Evans, 

M.  cartilaginea  Lehm.  &  Lindenb.  (1832)  from  St.  Vincent,  M. 
hrasiliensis  Lehm.  &  Lindenb.  (1832)  from  Brazil,  M.  Berteroana 
Lehm.  &  Lindenb.  (1834)  from  Juan  Fernandez,  M.  domingensis 
Lehm.  &  Lindenb.  (1834)  from  Santo  Domingo,  M.  tholophora 
Bisch.  (1835)  from  Mexico,  M.  inflexa  Mont.  &  Nees  (1838) 
from  Martinique,  M.  plicata  Nees  &  Mont.  (1838)  from  Bolivia, 
M.  quinqucloha  Nees  (1838)  from  the  West  Indies,^  M.  peru- 
viana (Nees  &  Mont.)  Nees  (1839,  ^s  Grimaldia  peruviana)  from 
Bolivia.  In  the  Synopsis  Hepaticarum  these  species  are  all 
recognized  with  the  exception  of  M.  Swart sii,  which  is  made  a 
synonym  of  M.  chenopoda,  and  M.  platycnemos,  which  is  made  a 
synonym  of  ikf.  papillata.  Two  other  species,  M.  pusilla  Nees  & 
Mont,  from  Chile  and  M.  lamellosa  Hampe  &  Gottsche  from 
Venezuela,  are  described  as  new;  a  third  species,  M.  linearis 
Lehm.  &  Lindenb.  (1832),  originally  described  from  Nepal,  is 
quoted  from  several  of  the  Lesser  Antilles ;  while  both  il/.  poly- 
morpha  and  M.  chenopoda  are  cited  from  numerous  American 
localities.  The  Synopsis,  therefore,  recognizes  sixteen  species 
in  all  from  North  and  South  America. 

During  the  period  from  1847  to  1899  comparatively  little  was 
added  to  our  knowledge  of  the  genus  in  America.  The  follow- 
ing species,  however,  were  described  as  new :  M.  flabellata  Hampe 
(1847)  from  Venezuela,  M.  Notarisii  Lehm.  (1857)  from  Chile, 
M.  Dillenii  Lindb.  (1883)  from  Jamaica,  M.  subandina  Spruce 
(1885)  from  Peru,  M.  Bescherellei  Steph.  (1888)  from  Brazil, 
and  M.  oregonensis  Steph.  (1891)  from  Oregon.  Two  of  the 
most  noteworthy  papers  on  Marchantia  appearing  during  this 
time  were  by  Schififner.  In  the  first  he  brought  out  the  fact 
that  M.  hrasiliensis  and  M.  cartilaginea  were  synonyms  of 
M.  chenopoda* ;  in  the  second  he  showed  that  M.  tahularis  Nees, 
a  South  African  species,  was  a  synonym  of  the  older  M.  Berte- 
roana.^ Another  reduction  to  synonymy  was  suggested  by 
Howe,®  who  showed  that  M.  oregonensis  was  based  on  very 
uncertain  characters  and  that  it  could  not  be  well  separated  from 
M.  polymorpha. 

*No  station  is  cited  for  this  species  in  the  original  publication;    the 
Synopsis,  however,  gives,   "in  India  occidentali." 
*  Nova  Acta  Acad.  Leop.-Carol.  60  :  287,  288.     1893. 
'  Oesterr.  Bot.  Zeitschr.  46 :  41-44,  100-103.     1896. 
°  Mem.  Torrey  Club  7  :  62.     .1899. 


American  Species  of  Marchantia.  207 

In  1899  Stephani"  published  his  monograph  on  Marchantia  in 
the  first  volume  of  his  Species  Hepaticarum.  He  describes 
eighteen  species  in  all  from  America,  six  of  which  are  confined 
to  North  America  and  eight  to  South  America.  Of  these 
eighteen  species  M.  Elliottii  of  Dominica  and  M.  caracensis  of 
Venezuela  and  Mexico  are  described  as  new,  while  M.  cepha- 
loscypha  Steph.  (1883),  originally  described  from  New  Zealand, 
is  quoted  from  Chile  and  Patagonia.  He  accepts  Schififner's 
reduction  of  M.  cartilaginea  to  synonymy  but  maintains  both 
M.  brasiliensis  and  M.  oregonensis  as  valid.  Under  M.  tabularis 
he  cites  AI.  Berteroana  as  a  synonym  (on  the  authority  of 
Schiffner)  but  gives  no  American  localities.  Under  M.  domin- 
gensis  he  gives  M.  inflexa  as  a  synonym  and  states  further  that 
the  American  stations  for  M.  linearis  (as  given  in  the  Synopsis) 
belong  to  M.  domingensis  instead.  He  includes  M.  Dillenii 
among  the  synonyms  of  M.  chenopoda  and  considers  that 
M.  peruviana  and  M.  Notarisii  are  very  close  to  this  species  and 
may  be  merely  forms  of  it.  Two  species  recognized  by  the 
Synopsis,  M.  quinqueloba  and  M.  piisiUa,  he  gives  up  altogether, 
because  they  were  based  on  fragmentary  specimens,  and  he  makes 
no  mention  whatever  of  M.  flahellata. 

If  M.  Berteroana  is  reinstated  as  an  American  species  and 
if  M.  flahellata  is  added,  Stephani's  total  of  eighteen  species 
would  still  be  maintained,  even  if  il/.  brasiliensis  and  M.  oregon- 
ensis are  considered  synonyms.  It  will  be  seen  that  this  total  is 
scarcely  different  from  the  total  of  sixteen  species  given  in  the 
Synopsis  Hepaticarum.  The  writer  hopes  to  show,  however,  that 
these  numbers  are  much  too  high  and  that  further  reductions  to 
synonymy  are  necessary.  In  his  opinion  there  are  only  nine 
species  based  on  characters  which  seem  trustworthy,  and  it  is 
possible  that  two  of  these  will  not  be  considered  distinct  when 
they  become  more  fully  known.  There  remain  five  species  which 
are  doubtful,  either  because  the  published  descriptions  are  incom- 
plete or  because  the  original  material  is  immature  or  fragmentary. 
Two  of  these,  as  noted  above,  are  discarded  altogether  by 
Stephani,  and  it  is  probable  that  the  other  three  deserve  the 
same  fate.  The  doubtful  species,  however,  will  be  alluded  to 
briefly  at  the  close  of  the  paper. 

'Bull.  Herb.  Boissier  7:383-407,  518-533.     1899. 


II.     MORPHOLOGICAL  NOTES  ON  THE  GENUS 

No  other  liverwort  has  been  so  much  discussed  and  described 
as  Marchantia  polymorpha.  According  to  Lindbergh  it  attracted 
the  attention  of  naturahsts  at  a  very  early  date  and  was  know^n 
to  both  Aristotle  and  Theophrastus.  Within  more  recent  times 
it  has  repeatedly  been  the  subject  of  morphological  researches 
and  has  served  in  numerous  text  books  as  a  typical  representa- 
tive of  the  thallose  Hepaticae.  Over  eighty  years  ago  Mirbel- 
published  the  first  extensive  account  of  its  morphology.  He 
brought  out  the  essential  features  of  the  thallus  and  of  its  various 
tissues  and  gave  a  clear  description  of  the  receptacles  and  the 
gemmae.  Of  the  later  works  dealing  with  the  morphology  of 
the  species  those  by  Leitgeb,^  Kny,*  Ikeno,^  and  Durand''  may  be 
particularly  mentioned.  The  first  two  deal  with  the  plant  in  a 
general  way,  very  much  as  Mirbel's  memoir  did,  although  they 
include  many  original  observations.  The  last  two  are  much  more 
specialized  and  deal  with  the  cytolog}'  and  development  of  the 
reproductive  organs.  Although  M.  polymorpha  itself  has  been 
treated  so  exhaustively  the  other  species  of  the  genus  have 
been  but  little  studied  by  morphologists.  In  one  of  his  earlier 
papers  Schiffner'^  published  a  series  of  interesting  observations 
on  the  Javan  M.  geminata  R.  Bl.  &  N. ;    but  aside  from  this, 

^  Hepat.  Utveckling  15.    Helsingfors,  1877. 

^  Recherches  anatomiques  et  physiologiques  sur  le  Marchantia  poly- 
morpha. Mus.  Hist.  Nat.  Nouv.  Ann.  1:92-130.  pi.  s-7-  1832.  For  a 
reprint  of  this  paper,  with  a  few  slight  alterations,  and  a  Complement 
des  observations  sur  le  Marchantia  polymorpha,  see  Mem.  Acad.  Sci. 
13:337-436.  pi.  1-8.  1835.  For  a  translation  into  German  by  Von  Flotow, 
see  Nees  von  Esenbeck,  Naturg.  Europ.  Leberm.  4 :445-494.    Breslau,  1838. 

^Unters.  iiber  Leberm.  6:  114-123.  pi.  9.    Graz,  1881. 

*  Bau  und  Entwickelung  von  Marchantia  polymorpha  L.  Hot.  Wand- 
tafeln  364-401.  pi.  84-go.    Berlin,  1890. 

^Beitrage  zur  Kenntnis  der  pflanzlichen  Spermatogenese :  Die  Sper- 
matogenese  von  Marchantia  polymorpha.  Beih.  Bot.  Centralbl.  15 :  65-88. 
pi.  3  +  f-  I-     1903- 

"  The  development  of  the  sexual  organs  and  sporogonium  of  Marchan- 
tia polymorpha.    Bull.  Torrey  Club  35 :  321-335.  pi  21-25.     1908. 

^  Ueber  exotische  Hepaticae.  Anhang  I.  Morphologische  Bemerkungen 
iiber  Marchantia.    Nova  Acta  Acad.  Leop. -Carol.  60 :  279-284.  pi.  ig.    1893. 


American  Species  of  Marchantia.  209 

records  of  morphological  importance  are  mostly  in  the  form  of 
scattered  notes,  and  these  are  often  to  be  found  in  taxonomic 
treatises. 

In  the  present  paper  the  morphology  of  Marchantia  will  be 
treated  largely  from  the  standpoint  of  the  taxonomist.  In  other 
words  the  parts  of  the  plant  which  yield  the  most  distinct 
and  constant  specific  characters  will  be  primarily  considered. 
These  parts  include  the  epidermis  and  the  epidermal  pores,  the 
compact  ventral  tissue,  the  ventral  scales,  the  rhizoids,  the 
receptacles,  and  the  cupules.  The  photosynthetic  layer,  the  sexual 
organs,  and  the  sporophyte,  although  yielding  important  generic 
characters,  are  less  helpful  when  the  individual  species  are  con- 
sidered. For  the  sake  of  completeness,  however,  a  brief  account 
of  the  sporophyte  will  be  included. 

The  flat  thallus  of  Marchantia  is  of  the  usual  prostrate  dorsi- 
ventral  type  and  branches  repeatedly  by  forking.  It  varies  con- 
siderably in  size  and  in  thickness  in  certain  species,  so  that 
measurements  of  its  various  dimensions  have  to  be  employed  with 
caution.  At  the  same  time  some  of  the  species  are  distinctly 
larger  than  otliers.  The  growth  of  the  thallus  is  normally  unlim- 
ited until  the  sexual  branches  or  receptacles  (see  Fig.  9,  A,  B) 
are  produced.  These  represent  the  erect  prolongations  of 
prostrate  branches  and  are  limited  in  growth.  The  inflorescence 
is  dioicous  throughout  the  genus.  Vegetative  reproduction  is 
carried  on  by  means  of  discoid  gemmae,  which  may  be  formed 
on  either  male  or  female  individuals  and  which  apparently  do  not 
interfere  with  the  growth  of  the  plant. 

The  thallus  shows  clearly  the  usual  differentiation  into  an 
epidermis,  a  photosynthetic  layer  and  a  compact  ventral  tissue 
bearing  scales  and  rhizoids.  The  photosynthetic  tissue  consists 
of  a  single  layer  of  large  air-chambers  separated  from  one  another 
by  continuous  plates  of  cells.  Each  air-chamber  is  connected 
with  the  outside  by  a  single  pore  in  the  epidermal  roof.  From 
the  floor  of  the  chamber  arise  numerous  short  rows  of  green 
cells,  subspherical  in  form  and  freely  exposed  to  the  air  of  the 
chamber.  The  rows,  which  are  simple  or  branched,  are  mostly 
from  two  to  five  cells  long  and  the  uppermost  cells,  except  in 
the  vicinity  of  the  pore,  are  usually  attached  to  the  epidermis. 
The  air-chambers  vary  greatly  in  size,  not  only  in  different  species 
but  often  in  different  parts  of  an  individual  thallus. 


2IO  Alexander  W.  Evans, 

'     I.     Epidermis  and  Epidermal  Pores 

The  ordinary  epidermal  cells  are  fairly  uniform  tliroughout 
the  genus  and  it  is  doubtful  if  they  offer  any  very  trustworthy 
differential  characters.  Their  size  often  varies  markedly  on  an 
individual  thallus  and  may  be  directly  affected  by  differences  in 
external  conditions.  Although  the  cells  are  usually  colorless  or 
pale  they  sometimes  produce  chloroplasts  in  abundance.  In  the 
majority  of  cases  they  are  arranged  in  a  single  layer,  but  in 
certain  species  at  least,  such  as  M.  chenopoda  (Fig.  19,  E)  and 
M.  pal  caeca  (Fig.  8,  D),  the  epidermis  may  be  two  cells  thick  in 
parts  of  its  extent.  The  walls  may  vary  considerably  in  thick- 
ness, but  they  are  rarely  very  firm  and  are  destitute  of  distinct 
trigones. 

Cells  containing  oil-bodies,  cells  containing  slime,  and  minute 
surface  papillae  are  sometimes  found  in  the  epidermis.  The  cells 
containing  the  oil-bodies  are  usually  distinctly  smaller  than  the 
neighboring  cells  and  are  easily  distinguished  by  their  granular 
contents,  which  nearly  or  quite  fill  the  cell  cavities.  In  M.  cheno- 
poda these  cells  are  not  infrequent  and  do  not  seem  to  be 
restricted  to  any  definite  part  of  the  tliallus ;  in  M.  polymorpha 
they  occur  near  the  margin  and  seem  to  be  absent  elsewhere ; 
while  in  certain  other  species  there  are  apparently  no  cells  of 
this  character  in  the  epidermis. 

Epidermal  cells  containing  slime  are,  according  to  our  present 
knowledge,  restricted  to  M.  chenopoda.  The  slime-cells  are 
scattered  about  in  the  epidermis  and  always  occur  in  regions 
where  the  epidermis  is  two  cells  thick,  being  situated  in  the  inner 
layer  (Fig.  19,  L).  They  are  much  larger  than  the  surrounding 
epidermal  cells  and  strongly  compress  those  of  the  outer  layer. 
When  a  piece  of  the  epidermis  is  examined  from  above  the  slime 
cells  are  seen  to  be  covered  over  by  these  compressed  cells. 
Apparently  Voigt®  was  the  first  to  observe  the  slime-cells, 
although  he  failed  to  recognize  their  true  character.  The  much 
larger  sHme-canals  in  Conocephalmn  conicum  (L.)  Dumort. 
were  soon  afterwards  described  by  Goebel,''  and  Leitgeb^" 
pointed  out  that  the  slime-cells  of  M.  chenopoda  were  of  the  same 


Bot.  Zeit.  37  :  733-     1879. 
'  Arb.  Bot.  Inst.  Wiirzburg  2  :  531.     1880. 
"Unters.  iiber  Leberm.  6:16.     1881. 


American  Species  of  Marchantia.  211 

nature.  He  showed  that  they  occurred  not  only  under  the 
epidermis,  as  he  expressed  it,  but  also  in  the  compact  ventral 
tissue  and  in  the  partitions  between  the  air-chambers,  and  he 
emphasized  the  fact  that  they  were  especially  abundant  in  the 
female  receptacles.  The  distribution  of  the  slime-cells  in  Mar- 
chantia was  a  little  later  discussed  at  length  by  Prescher.^^  He 
found  no  trace  of  them  in  M.  Berteroana,  M.  papillata,  M.  emar- 
ginata  R.  Bl.  &  N.,  or  M.  linearis;  he  found  them  restricted  to 
the  compact  tissue  of  certain  definite  regions  in  M.  polymorpha 
and  M.  paleacea;  and  it  was  only  in  M.  chenopoda  (including 
M.  cartilaginea)  that  he  found  them  in  the  epidermis. 

Surface  papillae  have  been  figured  very  accurately  by  Kny 
in  the  case  of  M.  polymorpha.  They  are  minute  appendages 
of  the  epidermis,  which  are  cut  off  by  walls  and  rounded  or 
bluntly  pointed  at  their  free  ends  (Fig.  2,  J,  L,  O,  P).  Sometimes 
a  papilla  is  situated  over  a  single  cell  and  sometimes  over  the 
partition  between  two  cells,  showing  in  the  latter  case  that  an 
epidermal  cell  had  divided  after  the  papilla  had  been  formed. 
Papillae  of  this  type  seem  to  be  rare  on  vegetative  branches  and 
confined  to  certain  species.  So  far  they  have  been  reported  m 
two  East  Indian  species,  M.  emarginata  and  M.  Treubli  Schirfn., 
but  they  seem  to  be  absent  from  all  the  American  species  except 
M.  polymorpha.  In  this  last  species,  as  shown  by  Schiffner,^* 
the  median  portion  of  the  thallus  is  always  free  from  papillae, 
while  the  marginal  regions  sometimes  show  them  clearly.  The 
distribution  is  very  different,  however,  in  M.  Trenhii,  where  the 
papillae  are  most  abundant  in  the  median  portion  and  gradually 
decrease  toward  the  margins.  Whether  papillae  of  this  character 
form  a  constant  feature  of  any  of  the  species  where  they  have 
been  found  is  perhaps  doubtful.  In  one  specimen  of  M.  emar- 
ginata, for  example,  in  the  writer's  collection  (Schiffner,  Iter 
Indicum  57),  the  plants  seem  to  have  developed  no  papillae,  and 
they  are  frequently  absent  from  the  vegetative  branches  in  M. 
polymorpha.  When  they  occur  on  receptacles  or  cupules,  as  in 
this  same  species,  they  seem  to  be  more  constant. 

"  Die  Schleimorgane  der  Marchantieen.     Sitzungsb.  Kais.  Acad.  Wissen. 
Wien,  Math.-naturw.  CI.  86' :  132-158.  pi.  i,  2.     1882. 
"^Bot.  Wandtafeln  pi.  84,  f.  2,  3.     1890. 
"  See  Schififner,  FI.  de  Buitenzorg  4:  32,  35.     I.eiden,  1900. 
"  Lotos  49  :  93.    1901. 


2  12  Alexander  IV.  Evans, 

The  complex  epidermal  pores  of  Marchantia  are  of  much 
interest.  They  are  of  the  dolioforrh  or  barrel-shaped  type,  that 
is,  the  opening  of  the  pore  is  surrounded  by  two  series  of  cells 
arranged  in  concentric  rows,  one  series  projecting  more  or  less 
above  the  surface  of  the  thallus,  the  other  projecting  into  an 
air-chamber.  Although  pores  of  this  type  are  found  on  the 
sexual  branches  of  most  of  the  Marchantiaceae,  the  only  genera 
where  they  occur  on  the  vegetative  branches  are  Marchantia, 
Preissia,  and  Bucegia.  Even  in  Marchantia,  as  shown  by  Kamer- 
ling,^^  immature  shoots  sometimes  produce  pores  of  the  simple 
type  found  in  most  of  the  other  members  of  the  group. 

The  first  attempt  to  utilize  the  structural  features  of  the  pores 
for  taxonomic  purposes  seems  to  have  been  made  by  Voigt.^®  He 
studied  eight  species  of  the  genus,  and  showed  that  the  number 
of  pores  in  a  given  area,  the  number  of  rows  of  cells  surround- 
ing a  pore,  and  the  number  of  cells  in  a  row  were  fairly  constant 
for  each  species.  Stephani  also  has  drawn  specific  characters 
from  the  pores,  but  certain  of  his  distinctions,  as  will  be 
shown  below,  are  subject  to  variation  and  must  be  used  with 
caution. 

In  the  case  of  M.  polymorpha  the  pores  have  been  repeatedly 
figured,  although  the  published  illustrations  are  not  all  of  the 
same  degree  of  excellence.  Among  recent  figures  those  by  Voigt, 
Kny,  and  Miiller^'  bring  out  most  of  the  essential  points.  Accord- 
ing to  Voigt,  whose  account  of  the  pores  is  unusually  full,  the 
opening  is  surrounded  by  five  circular  rows  of  cells,  three  belong- 
ing to  the  upper  and  two  to  the  lower  series,  but  both  Kny  and 
Miiller  state  that  the  upper  series  is  normally  composed  of  only 
two  rows  making  four  rows  in  all,  a  statement  which  agrees 
with  the  writer's  observations  (see  Fig.  2,  A,  B).  Under  some 
conditions  the  number  of  rows  may  be  reduced  to  three  or  even 
to  two.  In  the  upper  series  each  row  is  composed  (in  most 
cases  at  least)  of  four  cells  (Fig.  2,  A,  B),  and  immediately 
surrounding  the  pore  a  circular  membranous  ridge  is  present, 
probabl}^  representing,  as  in  the  simple  pores  of  Targionia,^^  a 
collapsed  series  of  cells.     This  ridge  is  shown  by  Voigt  (/.  i), 

"Flora  84  (Erganzungsb.)  :    57-     1897. 

''  Bot.  Zeit.  37  :  74i-     i879- 

"  Rabenhorst's  Kryptogamen-Flora  6:/.  187.    Leipzig,  1907. 

"  See  Deutsch,  Bot.  Gaz.  53  :  494.  /.  p.    1912. 


American  Species  of  Marchantia.  213 

but  is  not  brought  out  in  the  figures  of  Kny  and  Miiller.  It  is 
sometimes  very  narrow  and  absolutely  colorless  and  can  then 
be  demonstrated  only  with  difficulty.  In  the  lower  series  each 
row  is  likewise  composed  in  most  cases  of  four  cells,  those 
bounding  tlie  inner  opening  being  distinctly  dii^'erentiated.  Their 
usual  appearance  is  clearly  shown  by  Kny  {pi.  84,  f.  2),  each 
cell  being  in  the  form  of  a  narrow,  curved,  four-sided  figure 
with  a  rounded  median  projection  extending  toward  the  center 
of  the  pore.  All  the  cell-walls  immediately  bounding  the  pore 
are  shown  covered  over  with  a  granular  deposit  of  some  resinous 
substance,  which  hinders  or  prevents  the  entrance  of  water 
through  the  pore.  Kny  comments  on  the  fact  that  the  pores  vary 
greatly  in  size  and  that  the  projections  from  the  cells  bounding 
the  inner  opening  sometimes  meet.  In  his  opinion  these  projec- 
tions probably  make  still  more  difficult  the  entrance  of  water 
through  the  pore.  This  view  is  upheld  by  Ruge,"  who  finds  the 
pores  almost  completely  closed  by  the  projections  in  a  submerged 
form  of  M.  polymorpha.  In  Fig.  2,  D-I,  some  of  the  variations 
shown  by  the  cells  bounding  the  inner  opening  are  brought  out. 
In  Fig.  2,  E,  the  projections  are  only  slightly  developed,  although 
the  upper  cell  on  tlie  left  approaches  the  condition  portrayed 
by  Kny;  in  Fig.  2,  D,  F,  I,  the  projections  are  well  developed 
but  not  sharply  defined  from  the  rest  of  tlie  cell ;  in  Fig.  2,  G,  H, 
the  projections  are  both  well  developed  and  sharply  defined. 
These  last  figures,  drawn  from  a  plant  growing  in  a  very  wet 
locality,  support  the  statements  of  Ruge  and  agree  with  the  fig- 
ures published  by  Miiller.  The  cells  drawn,  however,  seem  to  be 
nearly  or  quite  destitute  of  the  resinous  deposit  so  conspicuously 
shown  in  the  remaining  figures  and  in  Miiller's  figures  also. 

Although  the  inner  openings  of  the  pores  in  M.  polymorpha  are 
subject  to  so  much  variation,  Stephani  insists  that  important 
specific  characters  in  the  genus  Marchantia  are  yielded  by  the 
inner  openings.  He  recognizes  four  types-''  and  states  that  they 
are  not  connected  by  transitional  conditions.  In  the  first  type  the 
four  cells  bounding  the  opening  are  narrow  and  not  materially 
changed  in  shape  by  increased  turgidity,  the  opening  itself  exhib- 
iting a  quadrate  form.     In  the  second  type  the  four  bounding 


"  Flora  77  :  294.  /.  11.     1893. 

■"  Bull.  Herb.  Boissier  7  :  385.  /.  a-d. 


214 


Alexander  W.  Evans, 


cells  bulge  into  the  opening  in  the  form  of  rounded  projections, 
the  opening  itself  showing  an  outline  with  four  strongly  concave 
sides  and  four  sharp  angles;  by  increasing  the  turgidity  this 
opening  can  be  almost  completely  closed.  In  the  third  type 
(which  is  essentially  the  same  as  the  pores  of  Preissia)  the  four 
cells  likewise  bulge  into  the  opening  but  the  bulging  portions  are 
more  sharply  defined  and  the  opening  appears  in  the  form  of 
a  four-sided  figure  with  very  concave  sides  but  with  rounded 
dilations  at  the  angles;  this  opening,  which  Stephani  describes 
as  cruciate,  can  be  completely  closed  by  an  increase  of  turgidity. 
In  the  fourth  type  the  opening  is  very  large  and  bounded  by 
many  cells  (fifteen  in  Stephani's  figure),  each  cell  bulging  into 
the  opening  in  the  form  of  a  longer  or  shorter  cylindrical  pro- 
jection, the  opening  itself  thus  acquiring  a  very  irregular  outline. 
To  the  first  type  Stephani  assigns  (among  others)  M.  poly- 
morpha,  M.  plicata  and  M.  domingensis;  to  the  second  type, 
M.  disjuncta;  and  to  the  third  type,  M.  cephaloscypha  and 
M.  paleacea.  The  only  representative  of  the  fourth  type  is 
M.  macropora  Mitt,  of  New  Zealand. 

Schiffner,-^  however,  had  already  called  attention  to  the  danger 
of  placing  too  much  confidence  in  the  peculiarities  of  the  cells 
bounding  the  inner  openings.  According  to  his  account  these 
cells  in  most  species  of  Marchantia  bulge  more  or  less  into  the 
opening,  the  form  of  which  may  vary  accordingly,  and  his  state- 
ments would  support  the  view  that  there  was  no  sharp  distinction 
between  the  first  and  second  types  of  Stephani.  Even  in 
M.  Berteroana,  which  Stephani  would  assign  to  his  third  type, 
Schiffner  finds  only  an  insignificant  modification  of  the  usual 
condition.  He  adds  that  the  number  of  bounding  cells  in  this 
species,  although  usually  four,  may  vary  from  three  to  six  on  an 
individual  thallus,  and  that  the  walls  of  the  cells  commonly  lack 
the  resinous  deposit  found  in  M.  polymorpha.  GoebeP-  is  like- 
wise inclined  to  recognize  a  single  type  of  pore  in  Marchantia 
with  respect  to  the  inner  opening,  and  he  sees  no  essential  dif- 
ference between  Stephani's  fourth  type  and  the  others.  He 
expresses  no  positive  opinion  on  this  last  point,  however,  because 
he  had  no  material  of  M.  macropora  at  his  disposal.  He  con- 
siders that  the  pores  are  plastic  structures,  subject  to  modifica- 

"'  Nova  Acta  Acad.  Leop. -Carol.  60 :  286.  pi.  ig,  f.  8.  20.     1893. 
"  Flora  96 :  193.     1906. 


American  Species  of  Marchantia.  215 

tion  through  external  conditions,  and  he  emphasizes  the  fact 
that  the  pores  of  xerophilous  forms  can  often  be  more  or  less 
completely  closed  by  an  increased  turgidity  of  the  bounding  cells. 

It  is  clear  from  the  observations  of  Schiffner  and  Goebel  that 
the  pores  in  Marchantia  (excepting  perhaps  in  M.  macropora) 
conform  to  one  general  t}^pe  and  that  the  distinctions  relied  upon 
by  Stephani  are  less  constant  than  he  supposed.  This  is  espe- 
cially well  seen  in  M.  polymorpha,  where  the  inner  opening  shows 
all  gradations  from  a  quadrate  to  a  cruciate  form  and  thus 
exemplifies  all  three  of  the  conditions  upon  which  the  first  three 
of  Stephani's  types  were  based.  M.  polymorpha,  however,  is. an 
exceedingly  plastic  species  and  it  is  doubtful  if  any  of  the  other 
members  of  the  genus  exhibit  the  same  wide  range  of  variation 
in  the  inner  opening.  Schift'ner's  figures  of  M.  geminata,  for 
example,  although  illustrating  conditions  connecting  the  first 
and  second  of  Stephani's  types,  show  no  approach  to  the  third; 
while  in  M.  paleacea,  according  to  the  information  at  hand,  the 
inner  opening  is  always  cruciate  and  thus  does  not  deviate  from 
the  third  type.  For  purposes  of  taxonomy,  therefore,  the  writer 
would  still  consider  it  expedient  to  recognize  two  types  of  pore 
among  the  American  species,  the  distinctions  between  the  types 
breaking  down  in  the  case  of  M.  polymorpha.  In  the  first  type 
(which  includes  Stephani's  first  and  second  types)  the  inner  open- 
ing is  bounded  by  three  to  six  cells,  the  usual  number  being  four, 
and  shows  all  gradations  between  a  polygon,  commonly  four- 
sided,  with  slightly  convex  sides  and  one  with  strongly  concave 
sides  and  sharp  angles.  In  the  second  type  (which  is  the  same 
as  Stephani's  third  type)  the  inner  opening  is  distinctly  cruciate 
with  four  rays  dilated  at  the  apex  or,  in  the  rare  cases  where 
the  number  of  bounding  cells  is  less  or  greater  than  four,  with 
fewer  or  more  rays.  Stephani's  fourth  type,  which  does  not  occur 
in  America  so  far  as  known,  need  not  be  further  considered. 

As  an  example  of  the  first  type  of  pore  M.  chenopoda  may  be 
selected.  In  this  tlie  opening,  as  pointed  out  by  Voigt,  is  sur- 
rounded by  about  seven  rows  of  cells,  shown  clearly  in  cross- 
section  (Fig.  19,  C-G),  four  of  the  rows  usually  belonging  to  the 
upper  and  three  to  the  lower  series.  The  walls  bounding  the 
opening  are  either  smooth  or  with  a  resinous  deposit.  In  the 
upper  series  (Fig.  19,  A,  B)  the  innermost  row  is  usually  com- 
posed of  four  narrow  cells  and  the  second  row  of  the  same  num- 


2i6  Alexander  W.  Evans, 

ber,  but  the  third  row  commonly  shows  twice  as  many  and  the 
fourth  row  a  much  larger  number.  The  ridge  immediately  around 
the  opening  is  clearly  marked.  In  the  lower  series  (Fig.  19,  H-K) 
the  innermost  row  lies  almost  'directly  beneath  the  second  row, 
so  that  only  the  first  and  third  rows  show  clearly  from  below. 
The  first  and  second  rows  are  usually  composed  of  four  cells 
each,  the  walls  bounding  the  pore  being  more  or  less  strongly  con- 
vex. The  third  row"  usually  contains  more  cells  than  the  first  and 
sometimes  twice  as  many,  but  it  rarely  contains  as  many  as  the 
fourth  row  of  the  upper  series,  where  the  cells  are  essentially 
like  the  ordinary  epidermal  cells.  Of  course  the  numbers  just 
given  are  subject  to  variation,  the  number  of  cells  bounding  the 
outer  and  imier  openings  being  often  more  than  four. 

As  an  example  of  the  second  type  of  pore  M.  paleacea  may  be 
selected,  and  the  illustrations  given  in  the  present  paper  (Fig. 
8,  A-H)  may  be  compared  with  the  one  published  by  Miiller.-^ 
The  descriptions  given  by  Voigt  may  likewise  be  consulted.  The , 
cells  bounding  the  pore  are  usually  in  six  rows,  three  belonging  to 
each  series,  and  the  rows  are  commonly  composed  of  four  cells 
apiece.  The  cell-walls  bounding  the  pore  are  smooth  through- 
out. In  the  upper  series  the  ridge  around  the  opening  is  distinct 
and  the  cells  are  very  narrow,  standing  in  sharp  contrast  to  the 
neighboring  epidermal  cells.  In  the  lower  series  the  cells  bound- 
ing the  inner  opening  are  much  broader  than  the  others  and  project 
so  strongly  that  they  often  touch  in  the  center  and  almost  occlude 
the  cruciate  opening.  Sometimes  one  or  more  cells  of  the  second 
row  project  also  (Fig.  8,  D),  but  the  cells  of  both  the  second 
and  third  rows  are  usually  narrow,  resembling  in  this  respect  the 
cells  in  the  upper  series. 

According  to  the  account  given  by  Kamerling,-*  the  size  of  the 
inner  opening  in  a  pore  of  the  first  type  is  not  decreased  to  any 
great  extent  by  an  increase  of  the  turgidity  of  the  surrounding 
cells,  while  in  a  pore  of  the  second  type  the  decrease  is  very 
marked.  In  his  opinion  the  cells  surrounding  the  inner  opening 
act  independently  of  the  cells  in  the  other  rings.  As  a  definite 
example  of  a  species  with  pores  that  can  be  closed  he  quotes  M. 
nitida  Lehm.  &  Lindenb.,  a  species  which  is  to  be  regarded  as  a 
synonym  of  M.  paleacea. 

^  Rabenhorst's  Kryptogamen-Flora  6 :  /.  188.    1907. 
°*  Flora  84  (Erganzangsb.)  :  46.     1897. 


American  Species  of  Marchantia.  217 

2.     Compact  Ventral  Tissue 

The  ventral  tissue  in  Marchantia  gradually  thins  out  from  the 
thickened  median  portion  until  it  is  frequently  only  two  or  three 
cells  thick  along  the  margins  of  the  thallus.  It  consists  primarily 
or  even  wholly  of  parenchyma  and  its  chief  function  apparently  is 
to  act  as  a  storage-tissue  for  water  and  organic  food.  In  some 
parts  of  the  thallus  it  is  usually  possible  to  demonstrate  the 
presence  of  elongated  pits  in  the  cell-walls,  and  a  purplish  pig- 
mentation of  the  walls  is  often  apparent.  Cells  containing  oil- 
bodies  are  usually  conspicuous  among  the  other  parenchyma  cells 
(Fig.  20,  A)  and  seem  to  be  present  in  all  the  species.  In 
herbarium  material,  however,  it  is  not  always  easy  to  demonstrate 
them.  Cells  containing  mycorrhiza  are  likewise  very  frequent. 
In  M.  chenopoda,  as  noted  by  Leitgeb  and  Prescher,  the  ventral 
tissue  and  the  partition  walls  between  the  air-chambers  contain 
scattered  slime-cells  similar  to  those  found  in  the  epidermis. 
Slime-cells  of  tliis  character  occur  also  in  tlie  compact  tissue  of 
M.  paleacea,  M.  breviloba  sp.  nov.  and  the  East  Indian  M.  emar- 
ginata,  but  have  not  yet  been  detected  in  other  species  except  in 
connection  with  the  reproductive  organs.  According  to  Cavers^^ 
the  slime-cells  and  slime-canals  of  Conocephalum  conicum  fail  to 
develop  when  the  plants  are  cultivated  under  water,  and  it  is 
therefore  possible  that  slime-cells  may  not  always  be  present  in 
the  species  of  Marchantia  just  listed.  In  fact  Prescher  reported 
that  they  were  absent  from  M.  cmarginata,  and  specimens  of  M. 
paleacea  and  of  M.  chenopoda  might  be  cited  where  they  are 
very  infrequent  or  perhaps  not  present  at  all. 

The  only  cells  found  in  the  ventral  layer  which  are  not 
parenchymatous  in  their  nature  are  the  more  or  less  elongated 
sclerotic  cells  with  yellow  or  brown  walls,  which  occur  in  certain 
species.  Cells  of  this  character  were  first  demonstrated  by 
GoebeP^  in  the  case  of  Preissia  quadrata  (Scop.)  Nees,  and  the 
same  author  has  called  attention  to  their  occurrence  in  the  New 
Zealand  M.  foliacea  Mitt."  In  this  species,  according  to  his 
account,  the  sclerotic  cells  are  variable  in  length  and  are  usually 
scattered  singly  among  the  parenchyma  cells.     Occasionally  two 


"  Ann.  Bot.  18 :  93.     1904. 

'"  Arb.  Bot.  Inst.  Wiirzburg  2  :  533.     1882. 

='  Flora  96 :  194.  /.  I42-     1906. 

Trans.  Conn.  Acad.,  Vol.  XXI  15  1917 


2i8  Alexander  W.  Evans, 

cells  will  occur  end  to  end  and  sometimes  even  longer  groups 
or  strands  are  formed,  perhaps  corresponding  with  Stephani's 
"strands  of  sclerenchyma."  The  walls  of  the  cells  are  thick 
and  pigmented,  showing  that  their  fimctions  are  primarily 
mechanical,  but  Goebel  finds  that  the  cavities  sometimes  contain 
starch-grains.  Sclerotic  cells  occur  in  several  American  species, 
such  as  M.  paleacea,  M.  chenopoda  (Fig.  20,  A,  B)  and  M.  domin- 
gensis,  and  agree  closely  with  Goebel's  description.  Whether 
they  are  always  produced  by  the  species  where  they  have  been 
detected  is  perhaps  a  question.  Cavers-^  calls  attention  to  the  fact 
that  Preissia  quadrata,  when  grown  indoors  in  a  moist  atmos- 
phere, fails  to  develop  thick-walled  cells,  and  in  all  probability 
the  formation  of  the  similar  cells  in  Marchantia  is  influenced  by 
environmental  conditions.  In  any  case,  however,  the  presence 
of  sclerotic  cells  is  associated  with  certain  definite  species. 

3.     Ventral  Scales 

The  ventral  scales  in  Marchantia  exhibit  considerable  diversity, 
not  only  when  different  species  are  compared  but  also  when  an 
individual  species  is  considered.  This  is  due  to  the  fact  that 
each  species  produces  at  least  two  distinct  kinds  of  scales,  only 
one  of  which  bears  appendages.  In  the  other  genera  of  the 
Marchantiaceae  the  scales  with  appendages  are  tlie  only  kind 
produced. 

Taylor-''  was  apparently  the  first  to  observe  that  the  scales  in 
M.  polymorpha  were  not  all  alike.  He  distinguished  three  dif- 
ferent kinds,  and  these  are  described  at  length  by  Leitgeb,^"  who 
designates  them  as  median,  laminar  and  marginal  scales,  respec- 
tively. The  median  scales  are  attached  by  a  long  line,  which 
begins  near  the  axis  of  the  thallus,  then  extends  almost  longi- 
tudinally and  finally  curves  gently  outward,  reaching  perhaps 
half  way  to  the  margin.  The  scales  are  at  first  very  narrow 
but  become  abruptly  dilated  in  the  outer  part ;  here  on  each  scale 
the  characteristic  appendage  is  attached,  strongly  contracted  at 
its  junction  with  the  scale  and  then  abruptly  dilated  into  an 
orbicular  expansion,  rounded  to  apiculate  at  the  apex  (Fig.  i). 

^^  Contrib.  to  the  Biol,  of  the  Hepat,  28.    Leeds  and  London,  1904. 
^*  Trans.  Linn.  See.  17:377.     1835. 
'"Unters.  iiber  Leberm.  6:  114.     1881. 


American  Species  of  Marchantia.  219 

The  laminar  scales  are  more  numerous  than  the  median  scales 
and  form  a  series  about  midway  between  the  median  scales  and 
the  margin.  They  are  attached  by  a  much  shorter  line  and 
broaden  out  at  once  into  lunulate  or  ovate  structures,  rounded 
at  the  apex  and  destitute  of  appendages.  The  marginal  scales 
are  still  more  numerous  and  attached  by  even  shorter  lines,  but 
they  resemble  the  laminar  scales  in  lacking  appendages  and  in 
most  other  respects.  They  are  situated  near  the  margin  and 
some  of  them  extend  beyond.  The  arrangement  of  the  scales  is 
shown  clearly  in  a  figure  by  Goebel/^  who  brings  out  the  fact 
that  the  laminar  scales  are  arranged  in  an  irregular  row  and  that 
the  marginal  scales  are  still  more  irregular  in  their  arrangement, 
although  a  linear  series  is  approximated. 

The  account  of  the  scales  just  given  is  somewhat  at  variance 
with  the  description  and  figures  of  Muller.^-  According  to  this 
author  the  innermost  scales  are  long  and  very  narrow;  they  are 
attached  almost  longitudinally  throughout  their  entire  length  and 
are  destitute  of  appendages.  These  are  said  to  be  borne  instead  on 
the  scales  of  the  next  outer  row,  which  agree  in  all  respects  with 
the  median  scales  as  described  by  Leitgeb.  The  third  type  of  scale 
recognized  by  Miiller  includes  both  the  laminar  and  marginal 
scales  of  Leitgeb.  An  interpretation  of  the  scales,  closely  agreeing 
with  Mijller's,  has  recently  been  published  by  Massalongo.^^  The 
present  writer,  however,  has  been  unable  to  demonstrate  the 
narrow  innermost  scales  without  appendages.  According  to  his 
observations  the  statements  of  Leitgeb  are  essentially  correct. 

The  scales  in  AI.  polymorpha,  as  well  as  in  the  other  species, 
are  delicate  in  texture  and  are  sometimes  more  or  less  pigmented ; 
in  most  cases,  however,  the  pigmentation  is  of  short  duration  and 
the  scales  become  bleached  and  transparent.  The  cells  tend  to 
be  wavy  and  irregular,  especially  toward  the  margin  (Figs.  7,  A; 
20,  C) .  The  cell-walls  are  thin,  although  trigones  may  sometimes 
be  demonstrated  on  the  marginal  scales.  Scattered  about  among 
the  other  cells  are  cells  containing  oil-bodies  (Fig.  20,  D)  and 
rhizoid  initials,  the  latter  giving  rise  to  tuberculate  rhizoids  (Fig. 
20,  E).  In  the  appendages  rhizoid  initials  are  absent,  but  cells 
containing  oil-bodies  can  often  be  distinguished  (Figs,  i,  3,  etc.). 


*^  Organographie  der  Pflanzen  /.  138.    Jena,  1898. 

^^  Rabenhorst's  Kryptogamen-Flora  6:17.  /.  12.     1905. 

^^Atti  R.  1st.  Veneto  75:696.  pi.  4,  f.  12-15.    1916. 


2  20  Alexander  W.  Evans, 

According  to  Leitgeb  the  scales  in  certain  species,  such  as 
M.  domingensis  and  M.  nitida,  are  all  of  the  median  type  while 
in  M.  chenopoda  the  laminar  scales  are  less  numerous  than  the 
median  scales  and  the  marginal  scales  are  absent  altogether.  In 
Goebel's  figure  of  M.  chenopoda^^  a  single  laminar  scale  is  shown 
among  sixteen  median  scales  and  the  implication  is  made  that  the 
number  of  laminar  scales  is  very  small.  The  writer  has  exam- 
ined numerous  specimens  oi  M.  pale  ace  a  (which  includes  M. 
nitida),  of  M.  chenopoda  and  of  other  species  and  finds  laminar 
scales  always  present  (see  Fig.  6,  H-O).  They  differ  from  the 
laminar  scales  in  M.  polymorpha,  however,  in  being  situated 
much  closer  to  the  median  scales.  On  account  of  their  shorter 
lines  of  attachment  they  extend  only  a  small  part  of  the 
distance  toward  the  median  line.  The  laminar  scales  usually 
alternate  with  the  median  scales  but  occasionally  there  may 
be  two  laminar  scales  between  two  successive  median  scales. 
Under  these  circumstances  one  of  the  laminar  scales  is  often 
reduced  in  size. 

The  species  just  noted  will  give  some  idea  of  the  differences  in 
arrangement  which  the  ventral  scales  may  show.  These  dif- 
ferences can  often  be  utilized  in  separating  species,  but  the  best 
differential  characters  yielded  by  the  scales  are  those  drawn  from 
the  appendages.  These  are,  with  very  rare  exceptions,  borne 
singly  and,  as  has  been  shown,  are  confined  to  the  median  scales. 
Although  the  appendages  vary,  within  wide  limits  in  certain 
species,  they  nevertheless  present  striking  and  distinctive  features. 
In  comparing  them  the  form,  the  character  of  the  margin  and 
apex,  the  size  of  the  cells,  and  the  presence  or  absence  of  cells 
containing  oil-bodies  should  be  taken  into  consideration.  With 
respect  to  size  the  cells  may  be  approximately  the  same  through- 
out the  entire  extent  of  an  appendage  (see  Fig.  i6,  F,  G).  It 
is  much  more  usual,  however,  for  the  median  cells  to  be  much 
larger  than  the  marginal  cells  and  the  gradation  from  one 
to  the  other  may  be  either  gradual  (see  Fig.  7)  or  very  abrupt 
(see  Fig.  5,  A-D).  The  texture  of  the  scales,  aside  from 
the  appendages,  is  much  the  same  throughout  the  genus.  The 
scales  and  their  appendages  will  be  again  considered  in  connection 
with  the  various  species  discussed  below. 

'*L.c.f.i57. 


American  Species  of  Marchantia.  221 

4.     Rhizoids 

The  rhizoids  in  the  genus  Marchantia,  as  in  practically  all  of 
the  Marchantiales,  are  of  two  types,  the  smooth  and  the  tuber- 
culate.  In  the  smooth  type  the  walls  are  thin  or  uniformly 
thickened ;  in  the  ttiberculate  type  numerous  local  thickenings 
of  the  wall  extend  into  the  lumen  of  tlie  rhizoid  in  the  form  of 
cylindrical  or  bluntly  conical  projections.  In  some  of  the  tuber- 
culate  rhizoids  the  projections  are  discrete  and  irregular  in  their 
distribution ;  in  others  they  are  more  or  less  coalescent  and  show 
a  spiral  arrangement.  Kamerling^^  has  shown  that  these  spiral 
tuberculate  rhizoids  are  abundant  in  M.  polymorpha  and  Schiff- 
ner^^  has  examined  this  and  other  species  of  the  genus  with 
reference  to  these  peculiar  structures.  He  confirms  Kamerling's 
statements  about  their  occurrence  in  M.  polymorpha  and  finds, 
so  far  as  American  species  are  concerned,  that  they  are  equally 
abundant  and  typical  in  M.  chenopoda;  that  they  still  occur, 
although  in  less  typical  form,  in  M.  domingensis  and  its  allies; 
and  they  are  wanting  altogether  in  M.  paleacea. 

Most  of  the  rhizoids  in  Marchantia  run  in  parallel  bundles 
under  the  scales  and  converge  to  form  a  single  large  median 
bundle.  There  are,  however,  numerous  rhizoids  in  the  thickened 
median  portion  which  spread  at  right  angles  to  the  surface,  and 
Schiffner  has  made  a  number  of  interesting  observations  on  these. 
In  forms  of  M.  polymorpha  where  a  definite  dorsal  band  lacks 
air-chambers,  the  rhizoids  in  question  are  smooth;  in  forms 
where  the  air-chambers  extend  across  the  median  region,  the 
rhizoids  are  tuberculate.  In  M.  plicata  the  spreading  rhizoids 
are  smooth;  in  M.  chenopoda,  smooth;  in  M.  Berteroana, 
tuberculate ;  in  M.  paleacea,  smooth  or  with  scattered  tubercles. 
These  differences  may  sometimes  be  of  help  in  distinguishing 
species. 

5.     Receptacles 

The  receptacles  in  Marchantia  are  strikingly  different  from  the 
vegetative  branches  and  attain  a  higher  degree  of  complexity  than 
in  any  other  genus  of  the  group.  Two  distinct  portions  may  be 
distinguished,  the  erect  stalk  and  the  horizontal  disc  (or  recep- 

^' Flora  84  (Erganzungsb.)  :  31.  pi.  i,  2,  f.  7.    1897. 

''Ann.  Jard.  Bot.  Buitenzorg  2  (Suppl.  3)  1489,  490.     1909. 


222  Alexander  W.  Evans, 

tacle  proper),  which  bears  the  sexual  organs.  In  some  cases  the 
disc  shows  clearly  that  it  has  but  one  plane  of  symmetry.  In 
other  cases  it  presents  the  appearance  of  being  radial ;  but  even 
here,  as  recently  emphasized  by  Goebel,^^  there  is  actually  but  a 
single  plane  of  symmetry,  a  fact  made  clearly  evident  when  the 
structure  and  development  of  the  receptacle  are  considered.  The 
stalk,  likewise,  looks  superficially  as  if  it  were  radial,  but  here 
again  a  single  plane  of  symmetry  is  present,  and  the  stalk  main- 
tains its  dorsiventrality  (or  zygomorphy)  in  spite  of  its  erect 
position. 

It  has  already  been  noted  that  the  receptacles  represent  pro- 
longations of  prostrate  branches.  These  branches  may  be  more 
or  less  elongated,  but  they  are  often  very  short,  a  receptacle 
being  developed  almost  immediately  after  a  dichotomy  has  taken 
place.  A  receptacle,  as  shown  so  clearly  by  Leitgeb,^®  is  a  branch- 
system,  the  growing  point  of  the  original  prostrate  branch  under- 
going one  or  more  divisions.  A  study  of  the  stalk  shows  that  the 
first  division  usually  takes  place  very  early  in  the  development 
of  the  receptacle.  If  a  cross-section  is  examined  (Figs.  5,  K;  8, 
5;  etc.)  the  dorsiventrality  of  the  stalk  becomes  at  once 
apparent,  and  the  side  which  represents  the  ventral  por- 
tion usually  shows  two  deep  longitudinal  furrows,  enclosed 
by  scales  and  containing  tuberculate  rhizoids,  the  dorsal  side 
being  destitute  of  such  furrows.  In  very  rare  cases  a  single 
furrow  is  present  near  the  base  of  the  stalk  (Fig.  20,  I). 
The  presence  of  two  furrows  is  evidence  that  the  grow- 
ing point  has  already  divided  once,  even  if  the  stalk  itself 
remains  undivided.  Usually  no  further  divisions  take  place  until 
the  disc  begins  to  develop,  but  in  some  cases  the  stalk  shows  three 
or  four  rhizoid  furrows,  indicating  that  one  or  two  secondary  divi- 
sions have  occurred.  This  is  seen  clearly  in  M.  hreviloha  and  M. 
domingensis  (Figs.  9,  I-K;  12,  A,  D).  In  the  first  the  stalks 
of  both  male  and  female  receptacles  show  four  furrows  apiece; 
in  the  second  the  stalk  of  the  male  receptacle  which  is  figured 
shows  three  furrows,  the  stalk  of  the  female  receptacle  show- 
ing four.  The  occurrence  of  more  than  two  furrows  has 
apparently    been    rarely    observed    in    Marchantia    and    allied 

"  Organographie  der  Pflanzen,  2d  ed.  686.     1915. 
^  See  Unters.  iiber  Leberm.  6  :  20-37.     1881. 


American  Species  of  Marchaniia.  223 

genera.  Spruce''^  mentions  the  occasional  presence  of  three 
furrows  in  the  stalk  of  the  female  receptacle  in  Marchantia,  with- 
out citing  definite  species ;  Leitgeb,**'  in  a  single  instance,  found 
four  furrows  in  the  stalk  of  the  female  receptacle  in  Preissia 
quadrata;  Stephani"  states  that  the  stalk  of  the  male  receptacle 
in  the  African  M.  Wilmsii  Steph.  has  four  furrows  but  doubts 
the  constancy  of  this  condition;  and  Schiffner*^  notes  that  the 
stalk  of  the  female  receptacle  in  Bucegia  romanica  Radian  some- 
times shows  four  furrows.  These  seem  to  be  the  only  references 
to  more  than  two  furrows  in  the  literature,  but  in  all  probability 
a  higher  number  than  two  would  occasionally  be  found  in  most 
species  of  Marchantia  if  enough  stalks  were  examined.  In  M. 
hreviloha  four  furrows  seem  to  be  the  rule  in  the  female  recep- 
tacle, although  it  would  hardly  be  safe  to  state  that  four  were 
always  present. 

In  the  case  of  M.  polymorpha  it  was  noted  long  ago  by  Mirbel 
that  the  dorsal  side  of  the  stalk  of  the  female  receptacle  showed 
a  distinct  strip  of  photosynthetic  tissue  with  air-chambers,  epi- 
dermal pores  and  short  green  filaments.  This  strip  seems  to  be 
of  constant  occurrence  throughout  the  genus.  It  commences 
close  to  the  base  of  the  stalk  and  extends  nearly  to  the  disc.  In 
most  cases  the  strip  is  continuous  (Figs.  5,  K;  8,  J ;  9,  K; 
etc.)  but  sometimes,  as  in  M.  chcnopoda,  it  may  be  separated  into 
two  strips  by  a  median  groove  (Fig.  20,  G-I).  In  the  stalk  of 
the  male  receptacle  photosynthetic  tissue  is  usually  absent,  the 
dorsal  portion  being  composed  of  compact  parenchyma.  In  cer- 
tain species,  however,  such  as  M.  domingensis,  the  photosynthetic 
tissue  is  about  as  well  developed  in  the  male  (Fig.  12,  A)  as  in 
the  female  receptacle.  In  M.  hreviloha  the  lower  part  (Fig. 
9,  I)  of  the  stalk  develops  photosynthetic  tissue  while  the  upper 
part  (Fig.  9,  J)  lacks  it  completely. 

In  the  disc  of  the  male  receptacle  the  dichotomous  branching 
usually  continues  and  a  distinct  division  into  rays  becomes 
apparent.  Although  the  number  of  rays  is  subject  to  variation, 
certain  numbers  seem  to  be  normal  or  typical  for  certain  species. 
In  M.  polymorpha,  for  example,  there  are  usually  eight  rays 

=*  Trans.  Bot.  Soc.  Edinburgh  15 :  558.     1885. 

^'Unters.  iiber  Leberm.  6:  31.     1881. 

*^  Hedwigia  31 :  196.     1892. 

*"  Beih.  Bot.  Centralbl.  23= :  282.  /.  16.     1908. 


224  Alexander  IV.  Evans, 

present  and  in  M.  chenopoda,  four,  although  deviations  from 
these  numbers  are  of  frequent  occurrence.  Goebel*^  considers 
that  the  number  of  rays  developed  is  dependent  on  nutritive  con- 
ditions. In  an  unnamed  species  from  the  Fiji  Islands,  related 
to  M.  geminata,  he  notes  a  reduction  in  the  number  of  rays  to 
two,  showing  that  only  one  dichotomy  has  taken  place,  and  he 
compares  this  extreme  condition  with  the  two-rayed  female 
receptacles  found  in  Exormotheca  and  Aitchisoniella. 

The  rays  are  in  one  plane  and  vary  greatly  in  length.  They 
are  sometimes  much  shorter  than  the  undivided  portion  of  the 
receptacle,  appearing  in  the  form  of  rounded  marginal  scallops 
separated  by  shallow  but  acute  sinuses.  This  condition  is  seen 
clearly  in  M.  polymorpha  and  its  allies.  It  is  much  more  usual, 
however,  for  the  rays  to  be  longer  than  the  undivided  portion,  the 
whole  receptacle  thereby  acquiring  a  palmate  appearance.  This 
type  of  receptacle  is  found  in  such  species  as  M.  chenopoda  and 
M.  domingensis  and  is  commonly  associated  with  a  smaller  number 
of  rays  than  the  first  type.  In  some  cases  at  the  tip  of  a  ray  a 
slight  depression  marking  the  position  of  a  growing  point  can 
be  discerned,  even  in  an  old  receptacle,  but  often  all  traces  of  the 
growing  points  disappear.  The  stalk  is  not  attached  to  the  disc 
marginally  but  peltately,  although  often  excentrically.  The 
peltate  attachment  is  due  to  intercalary  growth  taking  place  in 
the  region  where  the  dorsal  surface  of  the  stalk  and  the  dorsal 
surface  of  the  disc  would  naturally  be  continuous.  In  this  way 
a  thin  plate  of  tissue  is  formed  between  the  two  external  rays 
of  the  disc,  which  would  theoretically  be  distinct  to  their  junction 
with  the  stalk.  The  presence  of  this  plate,  similar  in  all  essen- 
tial respects  to  the  tissue  forming  the  sinuses,  intensifies  the  radial 
appearance  which  the  receptacles  of  certain  species  show. 

In  its  structure  the  disc  shows  many  of  the  features  which  are 
found  in  tlie  vegetative  thallus.  It  is  distinctly  dorsiventral  and 
the  differentiation  into  epidermis,  photosynthetic  tissue  and  com- 
pact ventral  tissue  is  clearly  marked.  On  the  ventral  surface 
of  the  rays  scales  with  appendages  and  scales  without  appendages 
can  be  distinguished  in  two  or  more  series,  and  the  appen- 
dages are  much  like  those  of  the  ordinary  scales  except  that  they 
are  smaller  and  sometimes  less  constricted  at  the  base.     Rhizoid 

"  Organographie  der  Pflanzen,  2d  ed.  699.  /.  669  II.     1915- 


American  Species  of  Marchantia.  225 

initials  are  present  among  the  cells  of  the  scales,  the  appendages 
alone  being  free  from  them. 

The  antheridia  arise  in  acropetal  succession,  the  oldest  being 
formed  near  the  center  of  the  disc.  In  many  species  each  ray 
develops  two  distinct  rows  of  antheridia,  but  in  certain  species, 
such  as  M.  polymorpha,  the  antheridia  are  more  irregular  in  their 
arrangement  and  each  ray  shows  more  than  two  indistinct  rows. 
The  antheridia  are  borne  singly  in  deep  depressions  with  small 
circular  openings.  The  depressions  extend  down  into  the  com- 
pact ventral  tissue,  and  are  surrounded  by  the  characteristic 
air-spaces  with  their  branched  rows  of  photosynthetic  cells  and 
dolioform  epidermal  pores. 

The  stalk  of  the  female  receptacle  develops  more  slowly  than 
that  of  the  male  receptacle  and  persists  in  an  active  condi- 
tion until  the  sporophytes  are  mature.  In  the  disc  the  division 
into  rays  takes  place  just  as  in  the  male  receptacle  and  the  num- 
ber of  rays  present  is  subject  to  similar  variations.  The  arche- 
gonia  form  groups  and  arise  in  acropetal  succession,  beginning 
when  the  disc  is  very  young;  but,  on  account  of  the  strong  inter- 
calary growth  in  the  median  region  of  the  dorsal  portion,  the 
archegonia  are  arched  over  and  displaced  until  they  seem  to  be 
situated  on  the  ventral  surface  of  the  disc.  In  this  way  the  oldest 
archegonia  come  to  lie  nearest  the  periphery  of  the  disc  and  the 
youngest  nearest  the  stalk.  Each  group  of  archegonia  contains 
a  variable  number,  arranged  in  two  or  three  more  or  less  definite 
radial  rows,  and  is  derived  from  one  of  the  growing  regions  of 
the  disc. 

In  the  East  Indian  M.  geminata  and  its  allies  the  groups  of 
archegonia  are  clearly  situated  underneath  the  rays  of  the 
receptacle.  These  rays,  therefore,  are  obviously  homologous 
with  the  rays  of  the  male  receptacles  throughout  the  genus. 
This  condition,  however,  is  very  exceptional.  In  most  species 
of  the  genus,  including  all  the  American  representatives,  the 
groups  of  archegonia  alternate  with  the  rays  of  the  recep- 
tacle and  are  situated  in  the  sinuses  between  them.  This 
is  caused  by  the  rapid  intercalary  growth  of  the  regions 
between  the  growing  points,  the  so-called  "middle  lobes" ;  the 
rays,  accordingly,  are  formed  by  the  middle  lobes  and  are  not 
homologous  with  the  rays  of  the  male  receptacle  but  rather  with 
the  sinuses.     These  relationships  are  discussed  at  length  by  Leit- 


2  26  Alexander  W.  Evans, 

geb,**  who  notes  also  the  fact  that  the  two  external  rays  are  to  be 
compared  with  the  "side  lobes"  in  an  ordinary  dichotomy.  In  M. 
polymorpha,  where  nine  rays  are  commonly  present,  seven  would 
represent  middle  lobes  and  tw^o,  side  lobes.  Between  the  two  side 
lobes  there  is  of  course  no  group  of  archegonia,  so  that  there  are 
eight  groups  for  the  entire  receptacle,  showing  that  three 
dichotomies  have  taken  place. 

The  rays  of  the  female  receptacles  vary  in  length,  very  much 
as  in  the  case  of  the  male  receptacles.  There  is,  however,  no 
correspondence  between  the  two.  In  M.  polymorpha,  for  exam- 
ple, the  rays  of  the  female  receptacle  are  long,  and  those  of  the 
male  receptacle  are  short,  while  in  M.  chenopoda  the  conditions 
are  reversed.  The  rays  of  the  female  receptacle  differ  also  in 
form,  being  flat  in  some  species  and  cylindrical  in  others.  When 
they  are  flat  they  are  often  retuse  or  shortly  bilobed  at  the  apex, 
and  Goebel  points  out  that  a  deepening  of  the  apical  sinuses 
would  lead  to  the  condition  found  in  M.  geminata,  where  sinuses 
instead  of  rays  are  present  between  the  groups  of  archegonia. 
In  young  receptacles  the  rays  are  strongly  curved  downward, 
but  they  gradually  straighten  out  if  fertilization  has  taken  place 
and  assume  a  horizontal  position. 

On  account  of  the  strong  intercalary  growth  which  displaces 
the  archegonia  to  the  lower  surface  of  the  disc,  the  portion  of 
the  receptacle  which  is  morphologically  ventral  is  less  extensive 
than  at  first  appears.  The  lower  surface  between  the  groups  of 
archegonia  is  ventral  in  character  and  the  same  thing  is  of  course 
true  of  the  lower  surface  of  the  rays,  especially  when  these  repre- 
sent the  middle  lobes  of  the  branch-system.  Even  here,  how- 
ever, when  the  rays  become  cylindrical  through  intercalary  dorsal 
growth,  the  ventral  surface  is  much  less  extensive  than  the 
dorsal.  In  M.  geminata  the  ventral  surface  of  the  rays  is  sit- 
uated on  both  sides  of  the  groups  of  archegonia.  The  ventral 
surface  is  characterized  by  the  presence  of  tuberculate  rhizoids 
and  slender  scales,  the  latter  being  sometimes  branched  and 
strikingly  different  from  the  ventral  scales  of  the  vegetative 
thallus.  The  dorsal  portion  (except  where  the  archegonia  are 
situated)  develops  a  complex  system  of  air-chambers  of  the  usual 
type. 

**  Unters.  iiber  Leberm.  6 :  34.     1881. 


American  Species  of  Marchantia. 


227 


Each  group  of  archegonia  is  enclosed  by  an  involucre,  which 
consists  of  a  pair  of  membranous  structures  often  toothed  or 
laciniated  on  the  margin  (Figs.  2,  M ;  4,  F,  G;  8,  K;  etc.)- 
Each  archegonium  is  further  protected  by  a  campanulate  pseudo- 
perianth  contracted  at  the  mouth  to  a  small  opening.  It  begins 
its  development  soon  after  the  archegonium  is  formed  but  does 
not  reach  full  maturity  unless  fertilization  has  taken  place.  The 
pseudoperianth  is  very  delicate  and  becomes  irregularly  torn 
when  the  stalk  of  the  sporophyte  elongates. 


6.     Sporophyte 

The  sporophyte,  as  in  all  the  Marchantiaceae,  shows  the  usual 
differentiation  into  foot,  stalk  and  capsule.  The  foot  is  flat- 
tened and  forms  a  low  ridge  enclosing  the  base  of  the  stalk.  The 
latter  is  at  first  very  short,  but  it  elongates  sufficiently  at  maturity 
to  push  the  capsule  through  the  calyptra  and  beyond  the  mouth 
of  the  pseudoperianth.  The  capsule  constitutes  the  principal 
part  of  the  sporophyte.  It  is  nearly  spherical  in  form  and  is 
boimded  on  the  outside  by  a  wall  composed  of  a  single  layer 
of  cells.  These  cells  throughout  the  genus  have  brownish  ring- 
like thickenings  in  their  walls,  although  the  rings  are  often 
incomplete.  The  entire  cavity  of  the  capsule  is  filled  with  spores 
and  elaters.  The  spores  are  much  smaller  than  in  most  genera 
of  the  Marchantiaceae,  especially  in  J\I.  polymorpha  and  its  allies. 
In  some  cases  a  distinct  border  is  present  where  the  spherical 
face  meets  the  three  plane  faces,  and  under  these  circumstances 
low  and  irregular  surface  lamellae  are  usually  developed.  In 
other  cases  the  spores  are  destitute  of  distinct  markings  and 
become  completely  rounded  off  after  the  tetrads  break  up.  The 
elaters  are  long  and  slender  and  of  the  usual  type,  showing  two 
distinct  spiral  bands.  At  maturity  the  wall  of  the  capsule  splits 
from  tlie  apex  to  about  the  middle  into  an  indefinite  number  of 
lobes,  some  of  which  may  become  further  subdivided.  There  are 
apparently  no  very  definite  lines  of  dehiscence,  the  edges  of  the 
splits  being  irregular  and  jagged  from  projecting  cells  which 
formerly  interlocked.  Except  for  the  spores,  which  differ  in  size 
and  in  the  peculiarities  of  their  walls,  the  sporophyte  yields  very 
few  differential  characters. 


2  28  Alexander  W.  Evans, 

y.     CupuLES 

The  characteristic  gemmae  of  Marchantia  have  been  repeatedly 
described.  They  consist  of  flat  discoid  structures,  each  bearing 
two  opposite  marginal  growing  points  in  shallow  indentations. 
They  are  attached  to  the  thallus  by  a  short  stalk,  which  joins  the 
margin  of  the  gemma  midway  between  the  growing  points,  the 
gemma  in  consequence  being  vertical  in  position.  The  gemmae 
occur  in  clusters  on  the  upper  surface  of  the  thallus  and  are 
surrounded  by  a  circular  membranous  outgrowth,  forming  a  cup 
or  cupule.  Although  the  gemmae  are  very  uniform  throughout 
the  genus,  the  cupules  yield  a  few  differences  which  sometimes 
assist  in  the  determination  of  species.  Two  principal  types  occur : 
in  the  one,  the  margin  of  the  cupule  is  simply  dentate  to  ciliate, 
the  teeth  being  sometimes  scattered  and  sometimes  close  together ; 
in  the  other  type  the  margin  bears  a  series  of  triangular  pointed 
lobes,  the  edges  of  which  are  dentate  to  ciliate.  As  an  example 
of  the  first  type  M.  domingensis  (Fig.  12,  K)  may  be  cited,  while 
M.  polymorpha  (Fig.  2,  N)  shows  the  second  type  clearly. 


III.     DESCRIPTION    OF    SPECIES 

Nees  von  Esenbeck^  divided  the  genus  Marchantia  into  the  tsvo 
sections  Astromarchantia  and  Chlamidium.  The  first  included 
species  in  which  the  female  peduncle  was  "central" ;  the  second, 
which  was  first  proposed  by  Corda  as  a  genus,  included  species 
in  which  the  female  receptacle  was  "excentric."  In  the  first  sec- 
tion he  placed  M.  polymorpha,  in  the  second  M.  paleacea.  These 
two  sections  are  retained  in  the  Synopsis  Hepaticarum,  except 
that  the  first  is  renamed  Stellatae;  they  are  likewise  retained  by 
Dumortier,^  who  coined  the  name  Marchantiotypus  for  the  first 
section,  Schiffner^  follows  the  example  of  the  Synopsis,  empha- 
sizing the  radial  symmetry  of  the  female  receptacle  in  the  Stella- 
tae; while  Stephani  bases  his  two  groups,  "a"  and  "b,"  which 
he  does  not  designate  by  formal  names,  upon  differences  in  the 
symmetry  of  the  female  receptacle,  the  first  group  including  spe- 
cies with  "symmetrical"  receptacles  and  the  second,  species  with 
"unsymmetrical"  receptacles.  It  is  interesting  to  note  that  he 
includes  M.  paleacea  in  his  first  group,  although  his  predecessors 
had  placed  it  definitely  in  the  section  Chlamidium. 

The  fact  has  already  been  brought  out  that  the  female  recep- 
tacles throughout  the  genus  are  always  symmetrical  with  respect  to 
one  plane  of  symmetry  but  never  with  respect  to  more  than  one. 
Stephani's  distinction,  therefore,  falls  to  the  ground,  and  the  dis- 
tinction in  the  position  of  tlie  stalk,  emphasized  by  Nees  von 
Esenbeck,  is  not  much  more  trustworthy.  In  certain  species, 
where  the  two  basal  rays  are  sometimes  distinctly  shorter  than 
the  others  and  sometimes  about  as  long,  it  breaks  down  alto- 
gether; in  the  first  case  the  stalk  would  be  "excentric,"  in  the 
second  "central."  At  the  same  time  the  sections  Astromarchantia 
and  Chlamidium  represent  natural  groups  of  species  and  can  still 
be  maintained  if  different  characters  are  used  to  distinguish 
them.  In  Astromarchantia,  for  example,  there  are  no  sclerotic 
cells  in  the  thallus,  and  the  rays  of  the  female  receptacle  are 
terete,  at  least  in  the  outer  part;   in  Chlamidium,  sclerotic  cells 

*  Naturg.  Europ.  Leberm.  4:60.     1838. 

'  Bull.  Soc.  Bot.  Belgique  13  :  150.     1874. 

'  Engler  &  Prantl,  Nat.  Pflanzenfam  i^ :  37-     1893. 


23°  Alexander  W.  Evans, 

are  normally  present  in  the  thallus,  and  the  rays  of  the  female 
receptacle  are  flat  or  convex.  In  both  these  sections  the  invo- 
lucres (and  clusters  of  archegonia)  alternate  with  the  rays. 
A  third  group,  typified  by  M.  geminata,  in  which  the  involucres 
are  situated  beneath  the  rays,  also  seems  worthy  of  sectional  rank, 
but  since  this  group  is  not  represented  in  America  (at  any  rate 
according  to  our  present  knowledge),  it  need  not  be  further 
considered  here. 

In  the  preparation  of  this  paper  the  writer  has  had  the  privi- 
lege of  examining  the  large  collection  of  Marchantiae  in  the 
herbarium  of  the  New  York  Botanical  Garden  (N.  Y.),*  which 
includes  the  Mitten  and  Underwood  herbaria.  This  has  been 
supplemented  by  the  specimens  in  the  Cryptogamic  Herbarium 
of  Harvard  University  (H.),  which  includes  the  Taylor  and 
Sullivant  herbaria,  and  by  the  material  in  the  United  States 
National  Herbarium  (U.  S.),  the  private  herbarium  of  Miss 
C  C.  Haynes  (C.  C.  H.),  and  the  herbaria  at  Yale  Univer- 
sity (Y.),  the  last  including  the  Eaton  herbarium  and  the  writer's 
private  herbarium.  Several  specimens  from  the  Montagne  (M.) 
and  Boissier  (B.)  herbaria,  including  a  number  of  types,  have 
hkewise  been  available  for  study,  through  the  courtesy  of  MM. 
Paul  Hariot  and  G.  Beauverd,  respectively.  The  writer  would 
extend  his  sincere  thanks  to  all  who  have  aided  him  in  his  work. 

Key  to  the  species 

Thallus  destitute  of  sclerotic  cells :  stalk  of  male  receptacle  destitute  of 
air-chambers,  with  two  rhizoid-furrows ;  rays  short  and  broad:  stalk 
of  female  receptacle  with  a  single  band  of  air-chambers  and  two 
rhizoid-furrows;  rays  mostly  nine  or  more,  terete,  at  least  in  outer 
part ;  involucre  with  dentate  or  ciliate  lobes :  cupules  with  dentate 
lobes,  bearing  papillae  on  outside. 

Section  I.    Astromarchantia, 
Epidermal  pores  usually  surrounded  by  four  rows  of  cells,  never  dis- 
tinctly cruciate:    marginal   scales   present;    appendages   of   median 
scales  irregularly  crenulate  or  denticulate. 
Rays  of  female  receptacle  bearing  papillae;    basal  sinus  scarcely  or 
not  at  all  wider  than  the  others. 

1.  M.  polymorpha. 
Rays  of  female  receptacle  destitute  of  papillae;    basal  sinus  usually 

distinctly  wider  than  the  others. 

2.  M.  plicata. 

*  The  letters  in  parentheses  are  abbreviations  used  below  in  the  citation 
of  specimens. 


American  Species  of  Marchantia.  231 

Epidermal  pores  usually  surrounded  by  six  rows  of  cells,  distinctly 
cruciate;  marginal  scales  not  present;  appendages  of  median  scales 
minutely  and  regularly  crenulate  or  denticulate;  rays  of  female 
receptacle  destitute  of  papillae;  basal  sinus  scarcely  or  not  at  all 
wider  than  the  others. 

3.  M.  Berteroana. 
Thallus  with  sclerotic  cells :    epidermal  pores  usually  surrounded  by  five 

to  seven  rows  of  cells :  marginal  scales  not  present :  rays  of  female 
receptacle  mostly  five  to  nine,  rarely  more,  flat  to  convex  on  upper 
surface,  never  terete,  destitute  of  papillae;  basal  sinus  usually  dis- 
tinctly wider  than  the  others:    cupules  destitute  of  papillae. 

Section  II.  Chlamidium. 
Stalk  of  female  receptacle  with  a  single  band  of  air-chambers. 
Epidermal  pores  cruciate:  appendages  of  scales  entire  or  slightly 
toothed:  stalk  of  male  receptacle  destitute  of  air-chambers,  with 
two  rhizoid-furrows ;  rays  short  and  broad :  stalk  of  female 
receptacle  with  two  rhizoid-furrows;  rays  long  and  narrow; 
involucre  with  ciliate  lobes :    cupules  with  dentate  lobes. 

4.  M.  paleacea. 
Epidermal  pores  not  cruciate:    stalk  of  male  receptacle  vith  a  single 

band   of   air-chambers    and   two   to    four   rhizoid-furrows ;     rays 
long  and  narrow   (at  maturity)  :    stalk  of  female  receptacle  with 
two   to    four   rhizoid-furrows ;    involucre  vaguely   or   not  at  all 
lobed,  entire  to  ciliate :    cupules  not  lobed,  dentate  to  ciliate. 
Appendages   of   ventral   scales   sparingly   crenulate   or   denticulate: 

rays    of    female    receptacle    short    and    broad,    mostly    eleven ; 

involucre  ciliate. 

5.  M.  breviloba. 
Appendages  of  ventral  scales  usually  closely  denticulate  or  cilio- 

late:   rays  of  female  receptacle  long  and  usually  narrow. 
Epidermal  pores  mostly  90-130  x  70-80,61:    rays  of  female  recep- 
tacle slightly  or  not  at  all  dilated  at  the  apex,  rarely  emar- 
ginate;    involucre  crenulate  to  ciliate. 

6.  M.  doniingensis. 
Epidermal  pores  mostly  50-70  x  40-45/x :  rays  of  female  receptacle 

distinctly  dilated  at  the  apex,  usually  emarginate;    involucre 
entire  to  crenulate. 

7.  M.  papillata. 
Stalk  of    female  receptacle  with  two  bands  of  air-chambers  and  two 

rhizoid-furrows;  rays  normally  five,  short  and  rounded,  not  dilated; 
involucre  dentate  to  ciliate  or  laciniate:  stalk  of  male  receptacle 
with  two  rhizoid-furrows :  epidermal  pores  not  cruciate :  append- 
ages of  ventral  scales  entire  to  sparingly  dentate :  cupules  not  lobed, 
dentate  to  ciliate. 
Thallus  thin  and  very  delicate. 

8.  M.  Bescherellei. 
Thallus  usually  thick  and  firm. 

9.  M.  chenopoda. 


232  Alexander  IV.  Evans, 

Section  I.     Astromarchantia 

I.     March  ANTiA  polymorph  a  L. 
Marchantia  polymorpha  L.  Sp.  Plant.    1603.    1753. 
Marchantia  stellata  Scop.  Fl.  Carn.  24  ed.  353,     1772. 
Marchantia  umbellata  Scop.  /.  c.    354.     1772. 
Marchantia  coarctata   Corda;    Opiz,   Beitr.   zur   Naturg.     647. 

1828  (nomen  nudum). 
Marchantia  elliptica   Corda,   /.   c.  6^y.     1828    {nomen   nudum). 
Marchantia  Kablichiana  Corda,  /.  c.  647.     1828  (nomen  nudum). 
Marchantia    macrocephala    Corda,    /.    c.    64"/.     1828.     (nomen 

nudum)  ;   Sturm,  Deutschl.  Flora  2  :  63.    pi.  //.     1832, 
Marchantia  vittata  Raddi,  Mem.  Soc.  Ital.  Modena  20:  45.    1829. 
Marchantia  Syckorae    Corda;     Nees    von    Esenbeck,    Naturg. 

Europ.  Leberm.  4:97.     1838, 
Marchantia  oregonensis  Steph. ;    Roll,  Bot.   Centralbl.  43 :  203. 

1891. 

Thallus  pale  to  dark  green,  not  glaucous,  sometimes  with  a 
brownish  or  purplish  median  band  on  the  upper  surface,  often 
more  or  less  pigmented  with  purple  on  the  lower  surface,  usually 
0.75-1.25  cm.  wide  and  4-6  cm.  long,  repeatedly  dichotomous, 
the  successive  forks  usually  2  cm.  or  less  apart;  texture  some- 
times delicate,  sometimes  firm,  but  never  leathery,  margin  entire 
or  minutely  denticulate ;  epidermis  composed  of  cells  with 
thin  or  slightly  thickened  walls,  mostly  20-6o/a  long  (averag- 
ing about  29/a)  and  i2-20yu.  wide  (averaging  about  16/*), 
papillae  present  near  the  margin  or  absent  altogether ;  pores 
(with  their  surrounding  cells)  mostly  6o-75;u,  long  and  40-60/* 
wide,  sometimes  measuring  as  much  as  90  x  6^11,  surrounded 
usually  by  four  rows  of  cells  (two  in  each  series),  each  row  being 
usually  composed  of  four  cells,  inner  opening  usually  four-sided, 
the  sides  rarely  concave  throughout,  each  bounding  cell  usually 
projecting  inward  in  the  form  of  a  rounded  papilla  with  sub- 
parallel  or  converging  sides,  mostly  with  a  resinous  deposit ;  air- 
chambers  low,  more  or  less  elongated,  their  boundaries  indistinct 
when  viewed  through  the  epidermis,  usually  present  everywhere 
(except  close  to  the  margin)  but  sometimes  absent  from  the 
median  region,  rows  of  photosynthetic  cells  sometimes  three  cells 
long  but  often  shorter ;  compact  ventral  tissue  mostly  twelve  to 
twenty  cells  thick  in  the  median  portion,  destitute  of  slime  cells 
and  sclerotic  cells,  the  cell  walls  slightly  thickened  and  showing 
distinct  pits ;  ventral  scales  in  three  rows  on  each  side  of  the 
thallus,  median  and  marginal  scales  in  distinct  rows,  laminar 
scales  in  a  more  indefinite  row,  scales  often  more  or  less  pigmented 


American  Species  of  Marchantia.  233 

with  purple,  the  marginal  scales  close  together  though  scarcely 
imbricated,  usually  projecting  beyond  the  margin;  appendages 
of  median  scales  broadly  orbicular,  mostly  0.5-0.75  mm.  long  and 
0.6-0.8  mm.  wide,  rounded  to  very  bluntly  pointed,  sometimes 
apiculate,  margin  usually  minutely  and  irregularly  denticulate, 
sometimes  (in  hygrophilous  forms)  tending  to  be  crenulate,  cells 
showing  a  gradual  decrease  in  size  toward  the  margin,  median 
cells  subisodiametric,  mostly  35-40^  in  diameter,  marginal  cells 
mostly  20-25/A  long  and  14-16/x  wide,  sometimes  smaller  (io-i6/iX 
lOfi),  cells  containing  oil-bodies  about  20/a  in  diameter,  usually 
from  five  to  ten  on  each  appendage,  restricted  to  submarginal 
(and,  rarely,  marginal)  portions:  male  receptacle  borne  on  a 
stalk  1-3  cm,  long  with  two  rhizoid-furrows,  destitute  of  dorsal 
air-chambers,  the  disc  mostly  0.7-1  cm.  broad,  shortly  lobed  or 
merely  crenate,  the  lobes  or  rays  mostly  eight  (rarely  nine  or  ten), 
2  mm.  long  or  less,  rounded  at  the  apex  with  thin  wavy  mar- 
gins, covered  ventrally  with  densely  imbricated  scales  in  several 
rows :  female  receptacle  borne  on  a  stalk  2-7  cm.  long,  with  two 
rhizoid-furrows  and  a  single  broad  dorsal  band  of  air-chambers, 
the  disc  mostly  0.8-1.3  cm.  broad,  deeply  lobed,  the  lobes  or  rays 
spreading  at  maturity,  mostly  nine  (sometimes  ten  or  eleven), 
3-5  mm.  long,  separated  by  subequal  sinuses,  terete,  covered  over 
with  epidermal  papillae;  involucre  deeply  and  irregularly  lobed, 
the  lobes  long-acUminate  and  ciliate  on  the  margins:  spores 
yellow,  I2-I5ju,  in  diameter,  nearly  smooth;  elaters  3-5/^  wide, 
bispiral :  cupules  deeply  lobed,  the  lobes  acute  to  acuminate, 
usually  dentate  to  short-spinose  on  tlie  sides,  outer  surface  with 
epidermal  papillae.     (Figs,  i,  2.) 

Throughout  the  greater  part  of  Europe  and  in  the  northern 
parts  of  Asia  and  North  America  M.  polymorpha  is  the  only 
representative  of  the  genus  and  is  exceedingly  abundant.  It 
grows  in  swamps  and  bogs,  on  rocks  and  walls  near  the  ground, 
on  banks  and  the  sides  of  ditches,  in  gardens  and  greenhouses, 
and  on  the  earth  in  fields  and  woods.  It  is  perhaps  most  luxuriant 
in  bogs  and  on  steep  rocky  hillsides  where  a  liberal  supply  of 
water  is  available.  In  the  woods  it  is  especially  likely  to  occur 
where  a  fire  has  left  a  supply  of  charcoal  behind.  Toward  the 
south  other  species  of  Marchantia  make  their  appearance  and 
M.  polymorpha  becomes  less  abundant.  In  many  places  it  pre- 
sents the  appearance  of  being  an  introduced  plant. 

A  search  through  the  literature  sho'yys  that  the  occurrence  of  M. 
polymorpha  south  of  the  equator  has  been  doubted  or  denied 
by  certain  authors.    Hooker,^  for  example,  about  fifty  years  ago, 

'  See  Handb.  New  Zealand  Fl.  545.     1867. 

Tkans.  Conn.  Acad.,  Vol.  XXI  16  1917 


2  34  Alexander  W.  Evans, 

stated  that  M.  tahiilaris  (i.  e.,  M.  Berteroana)  was  the  southern 
representative  of  the  northern  M.  polymorpha,  thus  implying  that 
the  latter  species  was  absent  from  antarctic  regions.  Stephani® 
is  even  more  definite  when  he  describes  the  habitat  of  the  species 
as  "Europa,  Asia  et  America  septentrionalis."  Other  authors, 
however,  cite  definite  stations  for  M.  polymorpha  from  the 
Southern  Hemisphere.  It  will  be  sufficient  to  mention  in  this 
connection  the  recent  record  by  Schiffner^  for  Kerguelen  Island, 
that  by  Kaalaas^  for  the  Crozet  Islands,  and  that  by  Howe® 
for  South  Georgia.  In  the  opinion  of  the  present  writer  the 
occurrence  of  the  species  in  South  America  has  been  clearly 
established.  Specimens  from  Ecuador,  Bolivia  and  Patagonia 
have  been  carefully  studied  and  have  been  found  to  agree  in  all 
essential  respects  with  European  and  North  American  material. 
Specimens  from  Kerguelen  Island,  collected  by  the  Challenger 
Expedition,  and  the  specimens  from  South  Georgia  cited  above 
have  likewise  been  examined,  and  have  been  found  equally  con- 
vincing, but  no  further  statements  can  be  made  from  personal 
knowledge  regarding  the  distribution  of  the  species  in  other  parts 
of  the  Southern  Hemisphere. 

On  account  of  the  abundance  of  M.  polymorpha  in  the  United 
States  and  northward  it  seems  inadvisable  to  give  a  full  list  of 
the  North  American  specimens  which  have  been  examined.  It  is 
enough  to  state  that  the  species  has  been  collected  in  Greenland 
and  other  parts  of  arctic  America,  in  Alaska,  in  nearly  every 
Canadian  province  and  territory  and  in  nearly  every  state  of  the 
Union.  The  specimens  cited  below  are  from  tropical  North 
America  and  from  South  America. 

Federal  District  of  Mexico  :  Canada  San  Magdalena,  Con- 
treras,  October,  1908,  Barnes  &  Land  455,  438  (Y.). 

Oaxaca:  near  Miahuatlan,  1895,  E.  IV.  Nelson  23JO  (U.  S.). 

Puebla:  banks  along  Avenida  Hidalgo  and  path  to  barranca, 
Tezuitlan,  October,  1908,  Barnes  &  Land  541,  542  (Y.)  ;  Santa 
Barbara,  near  Puebla,  November,  1909,  Frere  Nicolas  20  (Y.). 

*Bull.  Herb.  Boissier  7:393-     1899. 

'  Deutsch.  Siidpolar-Exped.  8 :  64.     1906. 

^Njt  Mag.  Naturv.  49:86.     1911. 

'See  Taylor,  N.  Sci.  Bull.  Mus.  Brooklyn  2:62.     1914. 


American  Species  of  Marchantia.  235 

Vera  Cruz:  Orizaba,  1855,  F.  Miiller  224^  (N.  Y.,  listed  by 
Gottsche  in  Mex.  Leverm.  268.  1863)  ;  Mirador,  April,  1857, 
C.  Mohr  (N.  Y.)  ;  Orizaba,  January,  1892,  /.  G.  Smith  (N.  Y.). 

Jamaica:  near  Hardware  Gap,  July,  1903,  A.  IV.  Evans 
176  (Y.). 

Ecuador:  near  Bancs,  R.  Spruce  (distributed  in  Hepaticae 
Spruceanae). 

Bolivia:  Sorata,  February,  1886,  H.  H.  Rusby  300 j  in  part 
(N.  Y.,  listed  by  Spruce  as  "AI.  nova  species?"  in  Mem.  Torrey 
Club  1 : 140.     1890). 

Chile:  Renca,  near  Santiago,  February,  1901,  G.  T.  Hastings 
J18  (U.  S.)  ;  Straits  of  Magellan,  1866-67,  -"^-  Cunningham  lOj, 
IS5  (N.  Y.)  ;  Punta  Arenas,  November,  1895,  P.  Dusen  (N.  Y., 
first  determined  by  Stephani  as  M.  Berteroana;  afterwards 
listed  as  ''.1/.  tahularis"  in  Bihang  t.  K.  Sven.  Vet.-Akad.  Handl. 
26^,  Xo.  17:8.  1901)  ;  Cordilleras  of  Patagonia,  February, 
1897,  /.  B.  Hatcher  (Y.,  listed  by  the  writer  in  Bull.  Torrey 
Club  25  :  424.  1898)  ;  Punta  Arenas,  February,  1906,  R.  Thaxter 
64,65  (H.,  Y.). 

The  following  additional  stations,  recorded  in  the  literature, 
are  likewise  of  interest: 

Costa  Rica:  Volcano  of  Poas.  H.  Pittier  6021  (listed  by 
Stephani  in  Bull.  Soc.  Bot.  Belgiq.ue  31 :  180.     1892). 

Colombia:  Bogota,  1859,  A.  Lindig  ijoo,  1701,  ijij  (listed 
by  Gottsche  in  Ann.  Sci.  Nat.  Bot.  V.  i :  186.     1864). 

Venezuela:  Colonia  Tovar,  Moritz  134  (listed,  but  erro- 
neously ascribed  to  Colombia,  in  Syn.  Hep.  789.  1847;  also  listed 
by  Hampe  in  Linnaea  20:  333.     1847). 

Ecuador:  Tunguragua,  R.  Spruce  (listed  in  Trans.  Bot.  Soc. 
Edinburgh  15:560.     1885). 

As  its  name  implies,  Marchantia  poJymorpha  is  an  exceedingly 
variable  species,  and  a  full  account  of  its  numerous  forms  is 
given  by  Nees  von  Esenbeck.^*'  He  recognizes  two  principal 
varieties,  A.  Communis  and  B.  Alpestris,  and  under  each  variety 
he  describes  a  series  of  sub-varieties  and  groups  of  more  inferior 
rank.  A.  communis  is  prevalent  at  lower  altitudes  but  sometiines 
ascends  to  higher  elevations  in  sheltered  localities;  B.  alpestris 
is  restricted  to  mountainous  regions.     A.  communis  is  charac- 

'"  Naturg.  Europ.  Leberm.  4:65-71.     1838. 


236 


Alexander  W.  Evans, 


terized  by  a  lax  habit  and  by  female  receptacles  having  elongated 
rays  and  long  slender  stalks;  B.  alpestris,  by  a  compact  habit 
and  by  female  receptacles  having  shorter  rays  and  shorter  and 
thicker  stalks. 

Nees  von  Esenbeck's  varieties   and   subordinate   groups   are 
accepted  without  question  in  the  Synopsis  Hepaticarum,  but  later 


Fig.  I.    Marchantia  polymorph  a  L. 

Appendages  of  ventral  scales,  x  100;  F  represents  the  basal  portion  of 
an  appendage ;  the  other  figures  represent  apical  portions.  A.  Opdal, 
Norway,  /.  Hagen,  in  V.  Schiflfner's  Hep.  Europ.  Exsic.  13,  type  of  var. 
mamillata  Hagen.  B-D.  New  Haven,  Connecticut,  A.  W.  Evans.  E,  F. 
Wilbraham,  Massachusetts,  E.  A.  Chapin,  forma  aquatica.  G.  Mount 
Hood,  Oregon,  /.  Roll,  type  of  M.  oregonensis  Steph.  H.  Oaxaca,  E.  W. 
Nelson  2330.    I.    Jamaica,  A.  W.  Evans  176.    J.    Patagonia,  /.  B.  Hatcher. 


writers  have  largely  neglected  them.  There  are  two,  however, 
which  appear  from  time  to  time  in  local  lists  and  taxonomic 
works.  One  of  these  is  A.  communis,  a  aquatica,  usually  quoted 
as  "var.  aquatica  Nees,"  and  the  other  is  B.  alpestris,  quoted 
as  "var.  alpestris  Nees."  The  differences  between  these  two 
so-called  varieties  are  indeed  striking.  In  var.  aquatica  the 
thallus  shows  a  distinct  median  band  on  the  upper  surface  usually 
pigmented  with  purple  and  associated  with  the  absence  of  air- 


American  Species  of  Marchantia.  237 

chambers,  as  Schiffner  has  pointed  out;  in  var.  alpestris  the 
thallus  is  uniformly  green  on  the  upper  surface,  and  air-chambers 
are  everywhere  present.  In  var.  aquatica  the  margin  of  the 
thallus  is  entire  or  nearly  so,  and  the  upper  surface  completely 
lacks  epidermal  papillae  or  bears  them  very  rarely;  in  var. 
alpestris  the  margin  of  the  thallus  is  more  or  less  denticulate  from 
projecting  cells,  and  epidermal  papillae  occur  in  greater  or  less 
abundance  in  the  marginal  portions.  In  var.  aquatica  the 
appendages  of  the  ventral  scales  are  entire  or  nearly  so,  and 
the  spreading  rhizoids  are  smooth;  in  var.  alpestris  the  appen- 
dages are  distinctly  denticulate,  and  the  spreading  rhizoids 
tuberculate. 

About  fifteen  years  ago  another  so-called  variety  was  distin- 
guished by  Hagen  under  the  name  var.  mamillata.  It  was  based 
on  a  supply  of  specimens  collected  by  its  author  at  Opdal  in 
Norway  and  distributed  by  Schiffner  in  Hep.  Europ.  Exsic.  15. 
Apparently  Hagen  himself  did  not  publish  his  variety.  Schiff- 
ner^^ did  so,  however,  and  quoted  Hagen's  original  diagnosis, 
as  follows :  "Cellulae  epidermicae  et  f rondis  dorsalis  et  carpo- 
cephali  acute  mamillosae."  In  commenting  on  this  diagnosis 
Schiffner  showed  that  the  mamillose  appearance,  so  strongly 
emphasized,  was  due  to  epidermal  papillae  and  that  these  were 
restricted  to  the  marginal  portions  of  the  thallus.  He  showed 
further  that  the  female  receptacles  in  all  forms  of  M.  poly- 
morpha  were  mamillose  in  Hagen's  sense.  Var.  mamillata,  there- 
fore, is  based  on  exceedingly  vague  characters  and  has  little  or 
nothing  to  distinguish  it  from  var,  alpestris. 

Although  var.  aquatica  and  var.  alpestris  are  at  first  sight  so 
distinct  from  each  other  they  are  connected  by  intermediate 
forms,  and  their  differences  seem  to  be  associated  with  definite 
dift'erences  in  environmental  conditions.  They  represent,  tliere- 
fore,  modifications  rather  than  varieties  in  the  taxonomic  sense. 
Probably  the  most  logical  disposition  to  make  of  them  is  to  regard 
them  as  forms,  as  Miiller^-  has  done,  and  to  cite  them  as  forma 
aquatica  (Nees)  K.  Miill.  and  forma  alpestris  (Nees)  K.  MiilL, 
respectively.  Other  forms,  less  distinct  than  these,  might  like- 
wise be  distinguished,  but  it  would  hardly  be  a  profitable  task 
to  designate  them  by  names. 

"  Lotos  49 :  93.    1901. 

^' Rabenhorst's  Kryptogamen-FIora  6:306.     1907. 


238  Alexander  W.  Evans, 

Of  the  various  synonyms  quoted  under  M.  polymorpha  the  first 
six  require  no  special  mention,  since  no  question  has  arisen  about 
them  for  many  5^ears.  In  fact  three  of  these  synonyms  were 
never  pubhshed  adequately  by  their  author  at  all,  and  nothing 
would  now  be  known  about  them  if  Nees  von  Esenbeck  had  not 
included  them  among  the  synonyms  of  his  varieties  and  forms  of 
M.  polymorpha.  It  is  perhaps  worthy  of  note,  however,  that 
M.  stellata  and  M.  umhellata  were  based  on  female  and  male 
specimens,  respectively,  showing  how  deep  an  impression  the 
very  different  receptacles  made  on  the  early  observers.  The  last 
three  synonyms  deserve  a  few  words  of  comment. 

The  first,  M.  vittafa,  was  described  from  specimens  collected 
by  its  author  on  the  island  of  Madeira.  It  is  characterized  by 
the  presence  of  a  longitudinal  median  band  on  the  thallus,  deep 
purple  in  color,  and  by  a  female  receptacle  bearing  three  to  ten 
terete  rays.  The  authors  of  the  Synopsis  Hepaticarum  cite 
the  species  but  do  not  number  it,  thus  implying  that  they  doubt 
its  validity.  They  refer  it  with  some  question  to  one  of  the 
varieties  of  M.  polymorpha.  Although  type  specimens  of  ilf. 
vittata  have  not  been  available  for  study  there  can  be  little  doubt 
that  the  species  should  be  referred  to  M.  polymorpha  forma 
aquatica,  on  account  of  its  median  purple  band.  This  conclusion 
is  supported  by  the  fact  that  Schiffner^^  quotes  M.  polymorpha 
var.  aquatica  definitely  from  Madeira,  although  he  makes  no 
mention  of  M.  vittata.  In  fact  the  writer  has  found  no  references 
to  the  species  later  than  the  date  of  the  Synopsis. 

The  next  species,  M.  Syckorae,  was  based  on  female  speci- 
mens collected  by  Syckora  and  by  Corda  in  Bohemia.  Nees  von 
Esenbeck,  without  having  seen  specimens,  gave  a  description  of 
the  species  based  on  Corda's  notes  and  figures.  The  features 
emphasized  are  the  stellate  female  receptacles  with  terete  rays 
and  the  monocarpous  invokicres,  each  consisting  of  two  distinct 
membranes  divided  into  six  lanceolate  acuminate  lobes  with 
serrate-dentate  margins.  In  spite  of  the  peculiarities  in  the 
involucre  Nees  von  Esenbeck  suspected  that  M.  Syckorae  was 
nothing  but  a  form  of  M.  polymorpha,  and  yet  it  is  cited  and 
numbered  in  the  Synopsis  Hepaticarum.  Many  years  later 
Dedecek^*  definitely  included  M.  Syckorae  among  the  synonyms 

^^  Oesterr.  Bot.  Zeitschr.  51:116.     1901. 

"Arch.  Naturw.  Landesdurchf.  Bohmen,  Bot.  5^:20.     1886. 


American  Species  of  Marchantia. 


239 


Fig.  2.     Marchantia  polymorpha  L. 

Anatomical  details.  A,  B.  Epidermal  pores  of  thallus,  surface  view, 
X  225.  C.  Pore  in  cross-section,  x  225.  D-I.  Inner  openings  of  pores, 
X  225.  J.  Marginal  portion  of  thallus,  in  section,  showing  two  epidermal 
papillae,  x  225.  K.  Stalk  of  male  receptacle,  cross-section  near  base, 
X  40.  L.  Female  receptacle,  section  of  part  of  disc,  showing  a  pore  and 
three  epidermal  papillae,  x  225.  M.  Part  of  involucre,  x  50.  N.  Part 
of  cupule,  showing  three  lobes,  x  50.  O,  P.  Epidermal  papillae  from 
cupule.  in  section,  x  225.  A,  D.  Opdal,  Norway,  /.  Hagcn,  in  V.  Schiff- 
ner's  Hep.  Europ.  Exsic.  75,  type  of  var.  mamillata  Hagen.  B,  I,  J. 
Jamaica,  A.  IV.  Evans.  C.  Oaxaca,  E.  W.  Nelson  2330.  E,  L-P.  New 
Haven,  Connecticut,  G.  E.  Nichols,  A.  W.  Evans.  F-H.  Wilbraham, 
Massachusetts,  E.  A.  Chapin,  G  and  H  representing  forma  aquatica.  K. 
Eberswalde,  Germany,  A.  W.  Evans. 


240  Alexander  W.  Evans, 

of  M.  polymorpha  and  Schiffner^^  has  since  followed  the  same 
course. 

The  type  material  of  the  last  synonym,  M.  oregonensis,  con- 
sists of  a  series  of  male  specimens  collected  on  Mount  Hood, 
Oregon,  by  J.  Roll,  in  1888.  A  portion  of  the  type  in  the  Under- 
wood herbarium  has  been  examined  by  the  writer.  In  his 
original  account  of  M.  oregonensis,  Stephani  emphasizes  the 
dentate  and  spinose  appendages  of  the  ventral  scales  and  states 
that  he  knows  no  other  Marchantia  of  temperate  regions  in 
which  similar  appendages  occur.  Howe^^  soon  pointed  out,  how- 
ever, that  the  appendages  in  many  European  and  American 
specimens  of  M.  polymorpha  agreed  with  those  of  M.  oregonensis 
and  expressed  the  opinion  that  the  peculiarity  emphasized  by 
Stephani  had  no  specific  significance.  Fig.  i,  G,  drawn  from 
M.  oregonensis,  fully  supports  Howe's  statements.  As  a  matter 
of  fact  the  appendages  are  not  deeply  enough  toothed  to  be 
called  "spinose"  or  even  "dentate" ;  it  would  be  more  accurate 
to  describe  them  as  denticulate  or  crenulate.  In  his  Species 
Hepaticarum,  published  the  same  year  as  Howe's  observations, 
Stephani^'^  still  maintains  the  validity  of  M.  oregonensis.  He 
describes  the  appendages  as  variously  and  remotely  dentate-spinose 
and  states  that  they  are  composed  of  small  subequal  cells.  Here 
again  Fig.  i,  G  brings  out  a  slight  inaccuracy,  by  showing  that 
the  marginal  cells  are  distinctly  smaller  than  tlie  interior  cells. 
In  his  critical  notes  he  no  longer  emphasizes  the  features  of  the 
appendages  but  calls  attention  to  the  cruciate  internal  openings 
of  the  epidermal  pores,  stating  that  no  other  North  American 
species  has  pores  of  this  character.  In  the  material  studied  by 
the  writer  no  pores  of  a  distinctly  cruciate  type  were  found; 
they  agreed,  rather,  with  the  pores  of  M.  polymorpha  forma 
alpestris,  as  shown  in  Fig.  2,  D.  Since  both  of  the  distinctions 
relied  upon  by  Stephani  thus  break  down  there  seems  to  be  no 
reason  why  M.  oregonensis  should  not  be  considered  a  simple 
synonym  of  M.  polymorpha. 


"  Engler  &  Prantl,  Nat.  Pflanzenfam.  i' :  37.     1893. 
"  Mem.  Torrey  Club  7  :  62.     1899. 
"Bull.  Herb.  Boissier  7:531.     1899. 


American  Species  of  Marchantia.  241 

2.     Marchantia  plicata  Nees  &  Mont. 

Marchantia  (  ?)  plicata  Nees  &  Mont. ;  Montagne,  Ann.  Sci.  Nat. 

Bot.  II.  9 :  43.    1838. 
Marchantia  lamellosa  Hampe  &  Gottsche-;   G.  L.  &  N.  Syn.  Hep. 

527.    1846. 
Marchantia    vtdcanica    Spruce,     Trans.    Bot.     Soc.     Edinburgh 

15  :  559.     1885  (as  synonym). 

Thallus  green,  not  glaucous,  often  more  or  less  pigmented  with 
purple  on  the  lower  surface,  usually  1-1.5  cm.  wide,  often  6-8  cm. 
long  or  even  more,  occasionally  dichotomous,  the  successive  forks 
usually  about  2  cm.  apart,  texture  delicate,  margin  entire  ;•  epi- 
dermis composed  of  thin-walled  cells,  averaging  about  20/x  in 
length  and  13/i  in  width,  papillae  absent;  pores  (with  their  sur- 
rounding cells)  mostly  65-80/i,  long  and  50-60/A  wide,  gradually 
decreasing  in  size  toward  the  margin,  the  smallest  measuring 
about  50  X  40/X,  surrounded  by  three  or  (usually)  four  rows  of 
cells  (two  rows  being  in  the  lower  series),  each  row  being  usually 
composed  of  four  cells,  inner  opening  mostly  four-sided,  with 
slightly  convex  to  distinctly  concave  sides,  somewhat  roughened 
by  a  resinous  deposit ;  air-chambers  low,  isodiametric  or  some- 
what elongated,  their  boundaries  very  indistinct  when  viewed 
through  the  epidermis,  everywhere  present  (except  close  to  the 
margin),  rows  of  photosynthetic  cells  usually  less  than  three 
cells  long;  compact  ventral  tissue  mostly  twenty  to  twenty-five 
cells  thick  in  the  median  portion,  destitute  of  slime  cells  and 
sclerotic  cells,  the  cell-walls  slightly  thickened  and  with  distinct 
pits ;  ventral  scales  in  four  to  six  rows  on  each  side  of  the  thallus, 
median  and  marginal  scales  in  distinct  rows,  laminar  scales  in 
two  to  four  indistinct  rows,  scales  often  pigmented  with  purple, 
marginal  scales  more  or  less  imbricated  and  usually  projecting 
beyond  the  margin ;  appendages  of  median  scales  orbicular-ovate 
to  orbicular,  mostly  0.65-0.9  mm.  long  and  0.65-0.8  mm.  wide, 
somewhat  narrowed  toward  the  rounded  and  sometimes  apiculate 
apex,  margin  minutely  and  irregularly  denticulate  or  crenulate, 
a  tooth  sometimes  consisting  of  an  entire  cell  borne  on  a  slightly 
projecting  stalk  cell,  cells  rapidly  decreasing  in  size  toward  the 
margin,  median  cells  mostly  70-90/x  in  length  and  40-60/x  in 
width,  marginal  cells  only  25-5O1U.  in  length  and  12-20/x  in  width, 
cells  containing  oil-bodies  15-30/*  in  diameter,  usually  about  ten 
on  each  appendage,  restricted  to  submarginal  portions :  male 
receptacle  borne  on  a  stalk  2-3  cm.  long,  with  two  rhizoid-fur- 
rows,  destitute  of  dorsal  air-chambers,  the  disc  mostly  1-1.2  cm. 
broad  (when  well  developed),  deeply  lobed,  the  lobes  or  rays 
mostly  eight  (sometimes  nine  or  ten),  the  two  basal  rays  usually 
separated  by  a  wider  sinus  than  the  others,  2-4  mm.  long,  rounded 
at  the  apex  and  with  thin  wavy  margins,  covered  ventrally  with 


24  2  Alexander  W.  Evans, 

densely  imbricated  scales  in  several  rows:  female  receptacle 
borne  on  a  stalk  6-8  cm.  long  (when  well  developed),  with  two 
rhizoid- furrows  and  a  single  broad  dorsal  band  of  air-chambers, 
the  disc  mostly  1.2-1.6  cm.  broad,  deeply  lobed,  the  lobes  or  rays 
spreading  at  maturity,  mostly  eleven  (sometimes  nine  or  ten), 
5-6  mm.  long,  the  two  basal  rays  usually  shorter  than  the  others 
and  separated  by  a  wider  sinus,  rays  terete,  rounded  at  the  apex, 
destitute  of  surface-papillae ;  involucre  sometimes  pigmented, 
deeply  and  irregularly  lobed,  the  lobes  long-acuminate  and  dentate 
to  ciliate  on  the  sides :  spores  pale  yellow,  12-14/x  in  diameter, 
smooth  or  nearly  so ;  elaters  3-5/^  wide,  bispiral :  cupules  deeply 
lobed,  the  lobes  as  in  M.  polymorpha,  outer  surface  with  epidermal 
papillae.     (Figs.  3,  4.) 


The  species  seems  to  be  confined  to  the  high  mountains  of 
South  America.     The  following  specimens  have  been  examined : 

Colombia:    Boqueron,  Bogota,  W.  Weir  (N.  Y.). 

Ecuador:  Quito,  December,  1847,  W.  Jameson  (N.  Y.,  listed  by 
Mitten  as  M.  Berteroana  in  Jour.  Bot.  &  Kew  Misc.  3:361. 
1851);  Pichincha,  R.  Spruce  (distributed  in  Hepaticae  Spruce- 
anae). 

Peru:  Cuzco,  July,  1911,  H.  IV.  Foote  (Y.,  listed  by  the 
writer  as  M.  lamellosa  in  Trans.  Conn.  Acad.  18 :  299.  1914)  ; 
same  locality,  September,  1914,  Mr.  &  Mrs.  J.  N.  Rose  ipodo 
(N.  Y.,  Y.)  ;  Ollantaytambo,  May,  1915,  Cook  &  Gilbert  672 
(U.  S.,  Y.)  ;  San  Miguel,  Urubamba  Valley,  June,  191 5,  Cook 
&  Gilbert  1162  (U.  S.,  Y.)  ;  Lucumayo  Valley,  June,  1915, 
Cook  &  Gilbert  1^21  (U.  S.,  Y.). 

Bolivia:  between  Chupe  and  Janacache,  province  of  Yungas, 
A.  d'Orbigny  20p  (M.,  type)  ;  Sorata,  February,  1886,  H.  H. 
Rusby  3005  in  part  (N.  Y.,  listed  by  Spruce  as  "M.  plicata 
Nees?"  in  Mem.  Torrey  Club  i:  140.  1890)  ;  Songo,  Novem- 
ber, 1890,  M.  Bang  pio  (N.  Y.,  U.  S.,  Y.,  listed  by  Rusby  as 
M.  polymorpha  in  Mem.  Torrey  Club  4:274.  1895);  Sorata, 
September,  1901,  R.  S.  Williams  2144  (N.  Y.,  Y.). 

The  type  specimen  of  M.  lamellosa  w^as  collected  at  the  fol- 
lowing locality : 

Venezuela  :  Paramo  de  Mucuchies,  Moritz  45  (listed,  but 
erroneously  ascribed  to  Colombia,  in  Syn.  Hep.  527.  1846;  also 
listed  by  Hampe  in  Linnaea  20 :  333.     1847). 


American  Species  of  MarcJiautia. 


243 


The  present  species  was  based  on  a  specimen  without  recep- 
tacles or  gemmae.  When  originally  described  its  generic  posi- 
tion was  considered  doubtful,  but  the  authors  of  the  Synopsis 
saw  clearly  that  it  represented  a  MarcJiantia  and  suggested  its 


Fig.  3.     Marchantia  plicata  Nees  &  Mont. 

Appendages  of  ventral  scales,  x  100.  D  represents  the  basal  portion  of 
an  appendage ;  the  other  figures  represent  apical  portions.  A.  Ecuador, 
PV.  Jameson.  B-D.  Peru,  H.  W.  Foote.  E-G.  Peru,  Cook  &  Gilbert 
672,  1162,  1321.    H.    Bolivia,  A.  d'Orbigny,  type.    I.    Bolivia,  M.  Bang  910. 


relationship  to  M.  polymorpha.  Unfortiuiately  the  absence  of 
receptacles  made  a  positive  conclusion  impossible,  and  they  were 
therefore  obliged  to  place  it  among  the  species  "incertae  sedis." 
A  portion  of  the  type  material  from  the  Montague  herbarium, 
kindly  forwarded  by  M.  Paul  Hariot,  has  been  carefully  com- 


244  Alexander  W.  Evans, 

pared  with  the  other  specimens  cited  above.  The  epidermal  cells 
are  unusually  delicate  and  are  slightly  smaller  than  tlie  averages 
given,  measuring  perhaps  i8x  lo^u,,  but  the  epidermal  pores  and 
ventral  scales  agree  very  closely  with  those  of  the  other  speci- 
mens. It  is  clear,  therefore,  in  the  writer's  opinion,  that  all  the 
specimens  cited  represent  the  same  species. 

The  original  material  of  M.  lamellosa  has  not  been  available 
for  study,  but  the  specimen  from  Colombia,  listed  above,  was 
referred  to  this  species  by  Mitten  and  is  evidently  the  same  as 
the  specimens  from  Ecuador,  Peru,  and  Bolivia.  On  the  basis 
of  this  specimen  and  the  full  description  in  the  Synopsis  Hepati- 
carum,  M.  lamellosa  is  here  reduced  to  synonymy.  It  should  be 
noted,  however,  that  Stephani  considers  it  valid,  so  that  this 
reduction  is  perhaps  unwarranted.  He  ascribes  to  the  species 
cruciate  pores  and  papillate  rays  on  the  female  receptacles.  In 
one  of  the  Peruvian  specimens.  No.  672,  some  of  the  pores  are 
as  cruciate  as  those  of  M.  oregonensis,  but  other  pores  are  not 
cruciate  at  all,  so  that  a  considerable  range  of  variation  is  present. 
The  lack  of  receptacles  in  these  specimens  makes  it  impossible  to 
determine  whether  papillate  rays  are  associated  witli  pores  which 
approach  the  cruciate  condition,  although  the  constant  absence 
of  papillae  in  all  the  fruiting  specimens  studied,  which  are  clearly 
the  same  as  No.  672,  makes  such  an  association  improbable. 
Even  if  papillae  occasionally  occurred  they  would  hardly  afford  a 
basis  for  a  specific  separation.  Stephani  cites  M.  lamellosa  from 
the  type  locality  and  also  from  Ecuador  (Chimborazo  and  Altar, 
Hans  Meyer)  }^  He  cites  M.  plicata  from  the  type  locality,  from 
Ecuador  (Quito,  Ortoneda,  Spruce),  from  Colombia  (Lindig), 
and  from  Venezuela  (Merida,  Morits).  It  is  probable  that  the 
Lindig  specimens  are  tliose  listed  by  Gottsche  under  M.  poly- 
morpha  and  that  the  Moritz  specimens  are  those  doubtfully 
referred  by  Hampe^®  to  M.  Berteroana.  Of  course,  in  the 
absence  of  the  specimens  themselves,  this  matter  can  not  be 
definitely  decided. 

Although  M.  plicata  and  M.  polymorpha  are  closely  related 
species  it  is  usually  easy  to  distinguish  them.  Some  of  the  dif- 
ferential characters,  however,  are  vague  and  subject  to  varia- 
tion.    When  M.  plicata  is  well  developed  the  thallus  and  the 

^*  See  Meyer,  In  den  Hoch-Anden  von  Ecuador  517.    Berlin,  1907. 
"  Linnaea  20  :  333.     1847. 


American  Species  of  Marchantia. 


245 


sexual  receptacles  are  larger  than  in  the  most  robust  forms  of 
M.  polymorpha,  approaching  or  equalling  in  this  respect  the 
more  southern  M.  Berteroana.  The  thallus  also  shows  a  tendency 
to  fork  at  infrequent  intervals,  so  that  it  presents  the  appearance 
of  being  more  elongated  than  in  M.  polymorpha. 

The  lack  of  epidermal  papillae  on  the  rays  of  the   female 
receptacle  seems  also  to  be  a  distinguishing  character.     In  fact 


Fig.  4.    Marchantia  plicata  Nees  &  Mont. 

Anatomical  details.  A-C.  Epidermal  pores  of  thallus,  surface  view, 
X  225.  D.  Pore  in  cross-section,  x  225.  E.  Inner  opening  of  pore,  x  225. 
F,  G.  Two  parts  of  the  same  involucre.  A,  B,  D,  E.  Peru,  Cook  & 
Gilbert  1162,  1321.  C.  Bolivia,  A.  d'Orbigny,  type.  F,  G.  Colombia, 
W.  Weir. 


papillae  of  this  type  are  restricted  to  the  outer  surface  of  the 
cupules.  In  spite  of  the  large  size  of  the  thallus  the  epidermis  of 
M.  plicata  is  unusually  delicate  in  texture  and  the  air-chambers 
unusually  low. 

The  ventral  scales  exhibit  considerable  variation  with  respect 
to  size  and  amount  of  pigmentation.  In  typical  examples  the 
ventral  surface  is  almost  covered  with  purple  scales,  but  this  con- 
dition is  by  no  means  constant;  the  scales  may  only  partially 
cover  over  the  surface  and  the  pigmentation  may  be  very  slight. 


246  Alexander  W.  Evans, 

The  crowded  marginal  scales,  however,  visible  from  above,  seem 
to  be  a  constant  feature,  although  these  scales  may  not  be  any 
more  conspicuous  than  in  M.  polymorpha.  The  appendages  of 
the  ventral  scales,  as  in  other  species,  yield  some  of  the  most 
important  characters  (Fig.  3).  The  small  marginal  cells  are 
exceedingly  irregular,  forming  various  angles  with  the  periphery 
of  the  appendage  and  often  projecting  in  the  form  of  blunt  teeth. 
Frequently  a  projecting  cell  will  be  borne  on  a  broader  basal  cell, 
a  two-celled  tooth  of  a  peculiar  type  being  thus  produced. 
Usually  the  difference  in  size  between  the  marginal  cells  and  the 
interior  cells  is  ver\^  marked,  but  the  difference  is  less  when  the 
appendages  are  poorly  developed  (Fig.  3,  A).  The  apices  of  the 
appendages  are  especially  variable.  In  some  cases  a  distinct 
apical  tooth  two  cells  long  is  present,  making  the  appendage 
apiculate  (Fig.  3,  F)  ;  in  other  cases  the  apical  tooth  is  hardly 
distinguishable  from  the  neighboring  teeth  (Fig.  3,  G)  ;  in  still 
other  cases  there  is  no  indication  whatever  of  an  apical  tooth 
(Fig.  3,  H,  I). 

AMien  the  appendages  are  compared  with  those  of  M.  poly- 
morpha they  are  found  to  have  many  features  in  common.  In 
both  species  they  are  similar  in  form  and  show  a  decrease  in  the 
size  of  the  cells  in  passing  from  the  middle  to  the  margin ;  in 
both  species  the  apex  is  variable  and  the  margin  is  normally 
denticulate.  In  M.  plicata,  however,  the  cells  are  larger  and  the 
decrease  in  size  more  abrupt,  the  denticulation  tends  to  be  more 
pronounced,  owing  largely  to  the  frequency  of  two-celled  teeth, 
and  the  marginal  cells  tend  to  be  more  irregular.  Although  these 
differences  are  of  a  comparative  nature  and  subject  to  variation, 
they  will  usually  be  found  serviceable  in  separating  the  species. 

3.     Marchantia  Berteroana  Lehm.  &  Lindenb. 

M archantia  B ert eroana  Lehm.  &  Lindenb. ;  Lehmann,  Pug.  Plant. 
6:21.     1834. 

Marchantia  tahularis  Nees,  Naturg.  Europ.  Leberm.  4:  71  (foot- 
note).    1838. 

Marchantia  cephaloscypha  Steph.  Hedwigia  22:51.     1883. 

Thallus  green  or  bluish  green,  sometimes  glaucous,  often  more 
or  less  pigmented  with  purple  or  brownish  near  the  margin  and 
on  the  lower  surface,  ustially  1-1.5  cm.  wide  and  6-8  cm.  long. 


American  Species  of  Marchantia.  247 

variously  dichotomous,  the  successive  forks  sometimes  2-3  cm. 
apart  but  often  closer  together,  texture  usually  tough  and 
leathery,  margin  entire  or  minvitely  and  irregularly  denticulate 
or  crenulate,  more  or  less  plicate ;  epidermis  composed  of  cells 
with  thin  or  slightly  thickened  walls,  mostly  20-6o[x  long  (aver- 
aging about  32ju,)  and  16-24/x  wide  (averaging  about  19/a), 
papillae  absent;  pores  (with  their  surrounding  cells)  mostly 
60-80/X  long  and  50-6o;ia  wide,  surrounded  usually  by  six  rows 
of  cells  (three  in  each  series),  each  row  being  usually  composed 
of  fovir  cells,  inner  opening  cruciate,  the  bounding  cells  commonly 
four  (rarely  three  or  five),  slightly  roughened;  air-chambers  of 
medium  height,  usually  a  little  longer  than  broad,  their  boundaries 
indistinct  when  viewed  through  the  epidermis,  present  every- 
where, rows  of  photosynthetic  cells  usually  three  or  four  cells 
long;  compact  ventral  tissue  about  twenty-five  cells  thick  in  the 
median  portion,  destitute  of  slime  cells  and  sclerotic  cells,  usually 
thin-walled  and  with  indistinct  pits ;  ventral  scales  in  two  rows, 
median  and  laminar,  no  marginal  scales  being  present,  scarcely 
imbricated,  pale  or  brownish ;  appendages  of  median  scales 
orbicular-ovate  to  broadly  orbicular,  usually  somewhat  narrowed 
toward  the  rounded  apex,  mostly  0.6-1  mm.  in  length  and  about 
the  same  in  width,  margin  minutely  and  often  regularly  crenulate 
or  denticulate  from  projecting  cells,  cells  showing  an  abrupt 
decrease  in  size  toward  the  margin,  median  cells  mostly  60-85/x 
long  and  20-40/*  wide,  marginal  cells  (in  one,  two,  or  three 
rows)  mostly  12-2O/U,  long  and  8-i2/x  wide,  cells  containing  oil- 
bodies  about  20fx.  in  diameter,  about  five  on  each  appendage, 
restricted  to  submarginal  portions :  male  receptacle  borne  on  a 
stalk  1-5  cm.  long  with  two  rhizoid-furrows,  destitute  of  dorsal 
air-chambers,  the  disc  about  i  cm.  in  diameter,  more  or  less 
deeply  lobed  when  well  developed,  the  lobes  or  rays  mostly  eight, 
usually  2-3  mm.  long,  rounded  at  the  apex  and  with  thin  wavy 
margins,  covered  ventrally  (except  in  the  marginal  portions)  with 
imbricated  scales  in  several  rows,  sinuses  usually  subequal  in 
width :  female  receptacle  borne  on  a  stalk  mostly  3-8  cm.  in 
length,  with  two  rhizoid-furrows  and  a  broad  dorsal  band  of  air- 
chambers,  the  disc  mostly  0.8-1  cm.  broad,  deeply  lobed,  the 
lobes  or  rays  spreading  at  maturity,  mostly  nine,  2-3  mm.  long, 
separated  by  subequal  sinuses,  terete,  rounded,  destitute  of  epi- 
dermal papillae;  involucre  much  as  in  M.  polymorpha:  spores 
brownish  yellow,  mostly  8-io/a  in  diameter,  smooth;  elaters  about 
5/x  wide,  bispiral :  cupules  deeply  lobed,  the  lobes  as  in  M.  poly- 
morpha, outer  surface  with  epidermal  papillae.       (Fig.  5.) 

A  widely  distributed  species,  restricted  (according  to  our 
present  knowledge)  to  the  Southern  Hemisphere.  The  follow- 
ing South  American  specimens  have  been  examined : 


248  Alexander  W.  Evans, 

Argentina:  Buenos  Aires,  without  date,  Twiedie  (H.)  ;  La 
Plata,  collector  and  date  unknown  (H.). 

Chile:  Concepcion,  November,  1905,  R.  Thaxter,  16,  66 
(H.,  Y.)  ;  Port  Corral,  January,  1906,  R.  Thaxter,  62,  ^4,  Q2 
(H.,  Y.)  ;  Hermite  Island,  Cape  Horn,  1843,  ^-  ^-  Hooker  (H.). 

Juan  Fernandez:  without  definite  localities,  1830,  C.  Bert  era 
(H.,  N.  Y.),  type)  ;  H.  N.  Moseley  (N.  Y.,  Challenger  Expedi- 
tion) ;  1901,  G.  T.  Hastings  218  (N.  Y.,  U.  S.). 

Falkland  Islands:  without  definite  locality,  1843,  J-  •^• 
Hooker  (H.). 

The  following  specimens  from  other  regions  have  likewise 
been  examined: 

St.  Helena  :  without  definite  locality  or  date,  /.  Melliss 
(N.  Y.)  ;  1844,  /.  D.  Hooker  (N.  Y.)  ;  February,  1890,  W.  H. 
&  A.  H.  Broimi  264  (U.  S.,  United  States  Eclipse  Expedition  to 
Western  Africa,  listed  as  M.  tahularis  by  Stephani  in  Bull.  Herb. 
Boissier  7:  1899). 

Cape  Colony:  Table  Mountain,  F.  Krauss  (N.  Y.)  ;  Mon- 
tague Pass,  /.  C.  Bruetel  (N.  Y.)  ;  without  definite  localities  or 
dates,  W.  H.  Harvey  (N.  Y.),  Capt.  Rabenhorst  (Y.). 

Australia:  Victoria,  Robertson  (5jj  (N.  Y.)  ;  Swan  River, 
1846,  /.  Drummond  (H.)  ;  without  definite  locality,  F.  von 
Miiller  (H.,  Y.). 

Tasmania:  without  definite  localities,  1823,  Lawrence 
(N.  Y.)  ;  1838,  R.  Gunn  (N.  Y.)  ;  no  date,  W.  Archer  (N.  Y.)  ; 
western  mountains,  no  date,  Lazvrefice  (N.  Y.)  ;  Tasman  Penin- 
sula, roadside  from  Long  Bay  to  Tarrand,  February,  1899,  1956b 
(H.,  distributed  by  E.  Levier  as  M.  cephaloscypha). 

New  Zealand:  without  definite  localities,  no  date,  A.  Sinclair 
(N.  Y.)  ;  1881,  E.  Craig  (Y.)  ;  no  date,  /.  Remy  (N.  Y.,  dis- 
tributed by  C.  Roumeguere  as  M.  nitida)  ;  Waikehi,  A.  Sinclair 
(H.)  ;  Raipara,  1850,  S.  Mossmann  (H.)  ;  Point  Cooper,  col- 
lector and  date  unknown  (N.  Y.)  ;  North  Island,  1904,  W.  A. 
Setchell  32  (Y.)  ;  Lord  Auckland's  Group,  November,  1840, 
/.  D.  Hooker  (H.)  ;  Campbell  Island,  November,  1840,  /.  D. 
Hooker  (H.)  ;  without  definite  locality,  date,  or  collector's  name 
(B.,  type  of  M.  cephaloscypha). 

Other  South  American  stations  of  interest,  cited  in  the  litera- 
ture, are  the  following: 


American  Species  of  Marchantia.  249 

Argentina:  Staten  Island,  Tierra  del  Fuego,  C.  Spegazsini 
(listed  by  Massalongo  in  Nuovo  Gior.  Bot.  Ital.  17:  258.     1885). 

Chile  :  San  Augustin  and  San  Francisco  del  Monte,  F.  J.  F. 
Meyen  (listed  in  Syn.  Hep.)  ;  Otway  Gulf,  P.  Savatier  (listed 
and  figured  as  M.  tabularis  by  Bescherelle  and  Massalongo  in 
Compt.  Rend.  Miss.  Sci.  Cap  Horn  5 :  247.    pi.  4,  f.  ly.    1889), 

The  identity  of  M.  Berteroana  and  M.  tabularis  was  established 
by  Schiffner,  although  Nees  von  Esenbeck,  in  proposing  M. 
tabularis,  recognized  its  close  relationship  with  the  older  species. 
M.  Berteroana  was  based  on  specimens  collected  by  Bertero  on 
the  island  of  Juan  Fernandez,  while  the  type  material  of  M. 
tabularis  came  from  Table  Mountain  in  Cape  Colony.  The 
Synopsis  Hepaticarum  recognizes  both  species;  it  cites  M.  Ber- 
teroana from  Chile  and  St.  Helena,  as  well  as  from  the  type 
locality,  and  gives  Devil's  Peak  in  Cape  Colony  as  a  second  sta- 
tion for  M.  tabularis.  Under  M.  Berteroana  three  varieties  are 
recognized :  a,  from  Juan  Fernandez,  /?,  biflora,  from  Chile ;  and 
y,  anactis,  from  Juan  Fernandez  and  St.  Helena.  In  a,  according 
to  the  description,  the  rays  of  the  female  receptacle  are  one  third 
longer  than  the  involucres,  the  latter  enclosing  three  to  five 
flowers;  in  ji,  the  rays  are  the  same  as  in  a  but  are  fibrillose, 
while  the  involucres  usually  contain  only  two  flowers;  in  y,  the 
rays  do  not  project  beyond  the  involucre  at  all. 

In  discussing  M.  Berteroana,  Schiffner-"  points  out  that  the 
differences  relied  upon  by  Nees  von  Esenbeck  in  separating  M. 
tabularis  are  of  no  significance  and  that  the  same  thing  is  true 
of  the  differential  characters  assigned  to  the  three  varieties  of 
M.  Berteroana.  He  points  out  further  certain  mistakes  in  the 
original  description  of  tliis  species  and  also  in  the  description 
given  in  the  Synopsis.  He  based  his  conclusions  on  a  large  series 
of  original  and  authentic  specimens,  several  of  which  have  been 
studied  by  the  present  writer,  and  there  seems  to  be  no  reason 
for  doubting  the  accuracy  of  his  observations. 

Schiffner  was  apparently  the  first  to  give  a  satisfactory  descrip- 
tion of  tlie  scale  appendages  in  M.  Berteroana.  He  calls  attention 
to  the  finely  crenulate  margin,  to  the  border  of  very  small  cells 
in  one  or  two  rows,  and  to  the  sharp  distinction  in  size  between 

^"Oesterr.  Bot.  Zeitschr.  46:41-44,  100-103.    1896. 

Trans.  Conn.  Acad.,  Vol.  XXI  17  1917 


250 


Alexander  IV.  Evans, 


the  marginal  cells  and  the  cells  which  adjoin  them.  To  M. 
polymorpJia  he  assigns  appendages  which  are  minutely  but 
sharply  denticulate,  and  adds  that  the  cells  gradually  increase  in 


Fig. 


Marchantia  Berteroana  Lehm.  &  Lindenb. 


Appendages  of  ventral  scales  and  other  anatomical  details.  A-D. 
Appendages  of  ventral  scales,  x  100:  D  represents  the  basal  portion  of 
an  appendage ;  the  other  figures  represent  apical  portions.  E,  F.  Epi- 
dermal pores  of  thallus,  surface  view,  x  225.  G.  Pore  in  cross-section, 
X  225.  H-J.  Inner  openings  of  pores,  x  225.  K.  Stalk  of  female 
receptacle,  cross-section  near  base,  x  40.  A.  Juan  Fernandez,  G.  T. 
Hastings  218.  B,  E,  G,  H.  Chile,  R.  Thaxter  66.  C,  D.  Cape  Colony, 
Capt.  Rabcnhorst.    F,  I-K.    Australia,  F.  von  Miiller. 


size  in  passing  from  the  margin  to  the  median  portion.  In  most 
cases  the  distinctions  given  by  Schiffner  apply  very  definitely,  but 
in  the  forma  aquatica  of  M.  polymorpha  (Fig.  i,  E,  F)  the 
appendages  are  usually  destitute  of   distinct  denticulations,  the 


American  Species  of  Marchantia.  251 

entire  or  crenulate  margin  closely  simulating  that  of  M.  Ber- 
teroana.  Even  here,  however,  the  gradual  decrease  in  the  size  of 
the  cells  as  the  margin  is  approached  is  in  contrast  to  the  abrupt 
decrease  found  in  M.  Bertcroana. 

In  habit  and  in  general  appearance  M.  Berteroana  resembles 
M.  polymorpha  very  closely,  and  it  is  not  surprising  that  the  early 
observers  failed  to  distinguish  it  as  a  species.  The  thallus,  to 
be  sure,  is  more  robust,  it  tends  to  be  thicker  and  more  leathery 
than  in  the  northern  species,  and  the  rays  of  the  female  receptacle 
tend  to  be  shorter,  but  these  differences  are  not  always  pro- 
nounced. There  are,  however,  several  distinctive  features,  in 
addition  to  the  scale-appendages,  which  deserve  to  be  emphasized. 
In  the  first  place  M.  Berteroana  seems  to  lack  marginal  scales 
altogether.  Sometimes  the  laminar  scales  form  a  vague  double 
row,  some  of  the  scales  being  nearer  the  margin  than  the  others, 
but  even  under  these  conditions  there  is  quite  a  little  space 
between  the  outermost  scales  and  the  margin.  In  most  cases 
the  row  of  laminar  scales  is  more  definite  and  the  region  without 
scales  is  consequently  wider.  In  M.  polymorpha  the  marginal 
scales  are  apparently  always  present,  although  they  do  not  always 
attain  the  same  degree  of  development.  M.  Berteroana  is  further 
distinguished  by  its  cruciate  epidermal  pores,  and  by  its  lack 
of  epidermal  papillae  on  both  thallus  and  female  receptacle.  The 
lack  of  marginal  scales,  the  crenulate  scale  appendages,  and  the 
cruciate  pores  will  serve  also  to  distinguish  M.  Berteroana  from 
^1/.  plicata,  although  there  is  no  evidence  as  yet  that  their  ranges 
overlap. 

The  writer  has  fortunately  been  able  to  examine  an  original 
specimen  of  M.  ceplialoscypha.  The  species  was  based  on  gem- 
miparous  material  with  young  female  receptacles  collected  some- 
where in  New  Zealand,  neither  the  date  nor  the  collector's  name 
being  given.  A  few  years  later  Stephani-^  referred  to  his  species 
a  series  of  specimens  from  various  parts  of  Australia  and  was 
able  to  add  the  characters  derived  from  mature  female  receptacles 
and  ripe  capsules.  In  his  Species  Hepaticarum,--  in  addition  to 
New  Zealand  and  Australia,  he  cites  Tasmania,  Fuegia,  Pata- 
gonia and  Chile  as  localities  for  the  plant  and  notes  that  it  is  not 

-'  Hedwigia  28 :  265.     1889. 

"Bull.  Herb.  Boissier  7:391.     1899. 


252  Alexander  W.  Evans, 

rare.  Still  later  he^^  quotes  stations  on  Juan  Fernandez,  the 
Chilean  island  of  Chiloe,  and  the  Falkland  Islands.  He  tlierefore 
gives  the  species  a  very  extensive  distribution  in  the  Southern 
Hemisphere.  It  has  already  been  noted  that  he  does  not  credit 
M.  Berteroana  to  America  at  all,  the  only  specimens  which  he  cites 
being  from  Cape  Colony,  the  Transvaal,  and  the  island  of  St. 
Helena. 

In  his  descriptions  of  M.  cephaloscypha  Stephani  emphasizes 
the  cruciate  pores,  the  large  scale-appendages  bordered  with  very 
small  cells,  the  nine-rayed  female  receptacles  with  smooth  and 
terete  rays,  the  eight-lobed  male  receptacles,  and  the  spinose 
cupules.  It  will  at  once  be  noted  that  all  of  these  features  are 
found  in  M.  Berteroana.  There  are,  however,  certain  discrep- 
ancies between  Stephani's  descriptions  and  the  account  of  M. 
Berteroana  giY en  above.  He  states,  for  example,  in  his  original 
description  that  the  cupules  are  contracted  at  base  and  apex  and 
that  the  ventral  scales  are  in  three  rows  on  each  side  of  the 
thallus,  one  row  of  tongue-shaped  scales  being  close  to  the 
margin.  In  his  last  description  he  still  emphasizes  the  contracted 
apices  of  the  cupules  but  makes  no  allusion  to  the  three  rows 
of  ventral  scales,  perhaps  because  he  has  already  given  a  triseriate 
arrangement  of  the  scales  as  a  generic  character. 

The  type  specimen  shows  that  some  of  the  specific  characters 
emphasized  by  Stephani  are  based  on  misconceptions.  The  single 
cupule  present,  for  example,  is  contracted  at  the  throat  but  flares 
widely  at  the  mouth.  Even  if  the  mouth  itself  were  contracted 
this  condition  might  easily  be  due  to  immaturity  and  figures  of 
a  young  cupule  of  M.  polymorpha  by  Mirbel,^*  in  which  the 
mouth  is  distinctly  contracted,  fully  support  this  view.  There 
are,  moreover,  no  marginal  ventral  scales,  although  tlie  margin, 
being  irregularly  crispate,  produces  the  effect  of  scales.  The 
appendages  of  the  median  scales  are  slightly  crenulate  and  show 
one  or  two  rows  of  marginal  cells,  the  rays  number  nine  in  the 
female  receptacle  and  are  destitute  of  papillae,  the  surface,  of 
the  cupule  bears  numerous  papillae,  and  the  pores  are  of  the  cru- 
ciate type.  The  writer  therefore  feels  justified  in  considering  M. 
cephaloscypha  a  simple  synonym  of  M.  Berteroana. 


'Kungl.  Svensk.  Vetensk.-Akad.  Handl.  46°:  5.     ign- 
'  Mem.  Acad.  Sci.  13 :  pi.  4,  f.  31,  32.     1835. 


American  Species  of  Marchantia.  253 

Section    II.     Chlamidium 
4.     AIarchantia  paleacea  Bertol. 

Marchantia  paleacea  Bertol.    Opus.  Sci.  Bologna  i :  242.    1817. 
Marchantia  papillafa  /?  italica  Raddi,  Mem.  Soc,  Ital.  Modena 

19:44.     1823. 
Fimhriaria  paleacea  Corda;  Opiz,  Beitr.  zur  Naturg.  648.    1828. 
Fegatella  Michelii  Corda,  /.  c.  649.    1828  (according  to  Nees  von 

Esenbeck). 
Marchantia  nepalensis  Lehm.  &  Lindenb. ;  Lehmann,  Pug.  Plant 

4:  10.     1832. 
Marchantia  nitida  Lehm.  &  Lindenb.  /.  c.  11.     1832. 
Marchantia  squamosa  Raddi;    Lehm.  &  Lindenb.  /.  c.  12.     1832 

(as  to  the  East  Indian  plant). 
Marchantia  tholophora  Bisch.     Nova  Acta  Acad.  Leop.-Carol. 

17:989.     1835. 
Marchantia  calcarata  Steph.  Bull.  Herb.  Boissier  5  :  98.    1897. 
Marchantia  planipora  Steph.  /.  c.  98.    1897. 

Thallus  pale  green,  often  glaucous,  sometimes  more  or  less 
pigmented  with  purple,  especially  near  the  margin  and  on  the 
lower  surface,  usually  0.5-0.8  cm.  wide  and  2-4  cm.  long, 
repeatedly  dichotomous,  the  successive  forks  usually  i  cm.  or 
less  apart;  texture  firm  but  scarcely  leathery,  margin  entire; 
epidermis  composed  of  cells  with  more  or  less  thickened  walls, 
sometimes  in  two  layers,  mostly  35-70/x  long  (averaging  about 
43)Li)  and  20-40/^  wide  (averaging  about  30ju,),  papillae  absent; 
pores  (with  their  surrounding  cells)  mostly  70-90/^  long  and 
65-8511!,  wide,  sometimes  measuring  as  much  as  140  x  lOOjx,  sur- 
rounded usually  by  six  (or  seven)  rows  of  cells  (three  in  the 
upper  and  three  or  four  in  the  lower  series),  each  row  being 
usually  composed  of  four  cells  or  the  innermost  row  of  the  upper 
series  of  from  four  to  eight  cells,  inner  opening  cruciate,  the 
bounding  cells  smooth ;  air-chambers  usually  high,  isodiametric 
or  slightly  elongated,  their  boundaries  indistinct  when  viewed 
through  the  epidermis,  present  everywhere,  rows  of  photosyn- 
thetic  cells  often  six  or  seven  cells  long  but  sometimes  shorter; 
compact  ventral  tissue  mostly  twenty  to  thirty  cells  thick  in  the 
median  portion,  the  walls  sometimes  pigmented,  more  or  less 
thickened  and  showing  distinct  pits,  sclerotic  cells  usually  dis- 
tinct, scattered,  ten  to  twenty  in  a  cross-section  of  thallus,  more 
abundant  in  median  region  but  not  confined  to  this,  slime-cells 
sometimes  lacking,  sometimes  more  or  less  abundant,  especially 
toward  the  margin;  ventral  scales  in  two  distinct  rows,  the 
laminar  scales  alternating  with  the  median  and  only  a  little  nearer 


254  Alexander  W.  Evans, 

the  margin ;  appendages  of  median  scales  oblong,  ovate,  or  ovate- 
orbicular,  mostly  0.6-0.75  "^"^-  lo"g  ^'^d  0.45-0.6  mm.  wide, 
usually  narrow^ed  toward  the  rounded,  obtuse  or  acute  apex, 
margin  entire  or  vaguely  and  irregularly  denticulate  or  dentate, 
rarely  with  a  basal  lobe,  cells  showing  a  slight  and  gradual 
decrease  in  size  toward  the  margin,  median  cells  isodiametric  to 
distinctly  longer  than  broad,  mostly  25-60/x  long  and  20-30/A  wide, 
marginal  cells  mostly  30-40/^  long  and  i2-20/a  wide,  very  irregular, 
the  long  axis  sometimes  parallel  with  the  margin  and  sometimes 
at  an  angle  with  it,  cells  containing  oil-bodies  sometimes  absent 
altogether,  when  present  about  20jU  in  diameter,  one  to  three  or 
more  in  number  and  indefinite  in  position :  male  receptacle  borne 
on  a  stalk  5-7  mm.  high,  with  two  rhizoid-furrows,  destitute  of 
dorsal  air-chambers,  the  disc  5-6  mm.  broad,  very  shortly  or 
sometimes  (according  to  Schiffner)  more  deeply  eight-  (to 
twelve-)  lobed,  the  lobes  or  rays  rounded  and  with  a  thin  wavy 
margin,  ventral  scales  restricted  to  middle  portion  of  disc :  female 
receptacle  borne  on  a  stalk  2-4  cm.  high,  with  two  rhizoid-furrows 
and  a  single  broad  dorsal  band  of  air-chambers,  the  disc  about 
0.5  cm.  broad,  usually  nine-lobed.  the  lobes  or  rays  spreading  at 
maturity,  0.8-1.2  mm.  long,  separated  by  subequal  sinuses  or  with 
the  deep  sinus  between  the  basal  ray  broader  than  the  others, 
fiat,  dilated  at  the  truncate  or  emarginate  apex,  disc  with  a  median 
hemispherical  or  papilliform  protuberance  about  0.5  mm.  in 
diameter  and  nine  distinct  ridges  corresponding  with  the  rays ; 
involucre  much  as  in  M.  polymorpha:  spores  brownish  yellow, 
about  34/x  in  diameter,  with  a  narrow  hyaline  margin  about  2/x 
wide,  outer  face  bearing  a  series  of  low  lamellae  sometimes  form- 
ing an  indistinct  reticulum ;  elaters  mostly  6-8/a  wide,  bispiral ; 
cupules  with  toothed  lobes  much  as  in  M.  polymorpha,  but  lacking 
epidermal  papillae.     (Figs.  6-8.) 

A  widely  distributed  species  in  tropical  and  subtropical  regions. 
The  following  North  American  specimens  have  been  examined : 

Arizona:  Huachuca  Mountains,  1910,  L.  N.  Gooding  824 
(N.  Y.). 

Puebla:  Puebla,  1906,  Frcre  Arscne  (N.  Y.)  ;  Honey  Station, 
October,  1908,  Barnes  &  Land  ^oj  (Y.)  ;  banks  along  Avenida 
Hidalgo  and  path  to  barranca,  Tezuitlan,  Barnes  &  Land  544 
(Y.)  ;  Santa  Barbara,  near  Puebla,  November,  1909,  Frcre 
Nicolas  s  (Y.). 

Vera  Cruz:  Orizaba,  1855,  F.  Milller  2245  in  part  (N.  Y.)  ; 
walls  of  Lost  River  sink,  Orizaba,  November,  1908,  Barnes  & 
Land  668  (Y.). 

Guatemala:  Coban,  Alta  Verapaz,  1310  m.  alt.,  1892,  H.  von 
Tuerckheim  4960  (N.  Y.). 


American  Species  of  MarcJiautia.  255 

Cuba:  without  definite  locality,  C.  Wright  (H.,  N.  Y.,  Y., 
distributed  in  Hep.  Cubenses  as  M.  domingensis)  ;  La  Peria, 
Oriente,  600-660  m.  alt.,  February,  191 1,  /.  A.  Shafer  pog6 
(N.  Y.,  Y.)  ;  Monte  Verde,  Oriente,  on  walls  of  the  ruined  man- 
sion of  Lescaille,  where  Wright  lived,  August,  191 3,  Brother 
Leon  4089  (N.  Y.). 

Jamaica:  Whitfield  Hall  Plantation,  December,  1896,  W. 
Harris  iiodj  in  part  (N.  Y.)  ;  along  path  from  Cinchona  to 
Clyde  River,  July,  1903,  A.  W.  Evans  14  (Y.)  ;  vicinity  of  Cin- 
chona, February,  1905,  C.  E.  Cummings  29,  ji  (N.  Y.,  Y.)  ; 
Mabess  Road,  May,  1906,  D.  S.  Johnson  46  (Y.)  ;  Morce's  Gap 
and  vicinity,  August,  1906,  A.  IV.  Evans  462  (Y.). 

The  following  specimens  from  Europe,  the  Azores,  and  Asia 
have  likewise  been  examined : 

France:  Mentone,  November,  1864,  /.  T.  Moggridge  (N.  Y.). 

Italy:  without  definite  locality,  G.  Raddi  (N.  Y.,  labeled  M. 
papillata  (3  italica)  ;  Monte  Oliveto,  near  Pegli  in  Liguria,  May, 
1851,  L.  Caldesi  (N.  Y.)  ;  near  Genoa,  May,  1855,  G.  de  Notaris 
(N.  Y.,  also  distributed  in  Rabenhorst's  Hep.  Europ.  2"/)  ;  Ripoli, 
near  Florence,  May,  1899,  E.  Levier  (Y.,  Micheli's  locality)  ; 
Trezzo  sull'  Adda,  province  of  Milan,  July,  1899,  F.  A.  Artaria 
(C.  C.  H. ;  also  distributed  in  Schiffner's  Hep.  Europ.  Exsic.  /j). 

Azores:  San  Miguel  1865,  F.  D.  Godman  (N.  Y.)  ;  x^ugust, 
1894,  C.  S.  Brown,  365  (N.  Y.),  jdd  (N.  Y.)  ;  August,  1894, 
W.  Trelease  1320  (N.  Y.),  132 1  (N.  Y.)  ;  May  and  June,  1898, 
B.  Carriero  /13,  /18  (Y.).  The  specimens  collected  by  Brown 
and  Trelease  have  been  listed  by  Trelease  in  Rept.  Missouri  Bot. 
Gard.  8 :  187.  1897.  The  following  specimens,  however, 
although  listed  under  M.  paleacea,  should  be  referred  to  Cono- 
cephalum  conicum  (L.)  Dumort. :  C.  S.  Brown  36/  (from  San 
Miguel),  368  (from  Fayal),  jdp  (from  Pico)  and  W.  Trelease 
131/  (from  Terceira).  A  specimen  collected  by  B.  Carriero  at 
Furnas,  in  1888,  has  been  reported  by  Schiffner  in  Oesterr.  Bot. 
Zeitschr.  51 :  116.     1901. 

China:   Szechwan,  no  date,  E.  Faber  mo  (N.  Y.). 

India  (including  Nepal)  :  Northwest  Himalayas,  no  date. 
H.  Falconer  10/3  (N.  Y.,  U.  S.,  Y.)  ;  /.  F.  Royle  (N.  Y.)  ;  Nepal, 
N.  Wallich  (N.  Y.,  types  of  M.  nepalensis,  M.  nitida,  and  M. 
squamosa)  ;    Shagak  Valley,  1847,  T.  Thomson  166 1,  1663  (N. 


256  Alexander  W.  Evans, 

Y.)  ;  Kumaon,  Strachey  &  IVinterbottom  (N.  Y.,  listed  by 
Strachey,  as  M.  paleacea  and  M.  nitida,  in  Cat.  PI.  Kumaon,  234. 
1906)  ;  Musooric  (Northwest  Himalayas),  Arnigadh,  December, 
1895,  W.  Gollan  210  (N.  Y.,  Y.,  distributed  by  E.  Levier  as  M. 
nepalensis). 

Japan:  Nagasaki,  April-May,  1875,  R.  Oldham  (N.  Y.,  listed 
by  Mitten,  as  M.  nitida,  in  Trans.  Linn.  Soc.  II.  Bot.  3 :  205. 
1891)  ;  Kigo,  August,  1877,  Ahlberg  (B.  type  of  M.  planipora)  ; 
Yamakita,  Spidzuoka,  May,  1899,  Abbe  Faurie  (Y,,  distributed 
as  M.  nitida  in  Hep.  du  Japon  82)  ;  Tokyo,  May,  1897,  K.  Miyake 
40  (C.  C.  H.,  Y.,  determined  as  M.  diptera)  ;  Kuzunmura, 
August,  1898,  K.  Okudaira  26  (C.  C.  H)  ;  Kyoto,  August,  1900, 
K.  Miyake  (Y.,  determined  as  M.  planipora)  ;  Mt.  Futatabisan, 
near  Kobe,  April,  1903,  Abbe  Faurie  (Y.,  distributed  as  M. 
diptera  in  Hep  du  Japon  1268)  ;  Jigokudani,  May,  1903,  vS'. 
Okamura  (C.  C.  H.)  ;  Kochi,  May,  1904,  S.  Okamura  (C.  C.  H.)  ; 
Kanagawa,  Sagami,  no  date,  ilf.  Maeda  41  (Y.)  ;  Osaka,  May, 
1905,  vS*.  Okamura  (C.  C.  H.)  ;  Ikku,  May,  1908,  S.  Okamura 
(C.  C  H.). 

Java:  Mt.  Pangerango,  April,  1894,  V.  Schiffner  (Y.,  dis- 
tributed as  M.  nitida  in  Iter  Ind.  59). 


Within  recent  years  M.  paleacea  has  been  recorded  from  the 
Caucasus,  from  Dalmatia,  Spain  and  Portugal,  and  from 
Morocco.  Although  the  writer  has  seen  no  specimens  from  any 
of  these  countries,  there  can  be  little  doubt  regarding  the  cor- 
rectness of  the  determinations.  Some  of  the  records  for  M. 
nitida,  however,  are  open  to  suspicion.  This  species  has  been 
reported  from  the  Philippines,  Tahiti,  Samoa  and  New  Zealand, 
as  well  as  from  the  Fiji  and  Hawaiian  Islands.  Specimens  from 
the  first  four  of  these  localities,  which  have  been  determined  as 
M.  nitida,  have  been  examined  and  are,  in  the  writer's  opinion, 
referable  to  other  species.  No  specimens  from  either  the  Fiji  or 
the  Hawaiian  Islands  have  been  available  for  study. 

Although  M.  paleacea  is  here  reported  from  a  number  of  North 
American  localities  it  is  remarkable  that  there  are  no  earlier 
records  for  the  species  from  America.  In  fact  the  only  trust- 
worthy records  for  M.  tholophora,  here  considered  a  synonym  of 
M.  paleacea,  are  the  following : 


American  Species  of  MarcJiaiitia. 


257 


Oaxaca:  near  Oaxaca,  Sommerschu  (the  type-locality); 
Chinantla,  F.  Liehmann  (listed  by  Gottsche  in  Mex.  Leverm. 
1863). 

It  has  already  been  noted  that  M.  paleacea  was  distinguished 
by  the  Florentine  botanist  Micheli  as  long  ago  as   1729.     His 


Fig.  6.     Marchantia  paleacea  Bertol. 

Ventral  scales,  x  27.  A-G  represent  median  scales  (with  appendages)  ; 
H-0,  laminar  scales  (without  appendages).  A-D,  H-J.  Italy,  F.  A. 
Arteria,  in  V.  Schiffner's  Hep.  Europ.  Exsic.  13.  E,  K.  Vera  Cruz, 
Barnes  &  Land  668.  F,  L.  Cuba,  C.  Wright,  in  Hep.  Cubenses,  as  M. 
domingcnsis.    G,  M-0.    Java,  V.  Schiffner,  in  Iter  Ind.  59,  as  M.  nitida. 


figure  brings  out  clearly  the  general  habit  of  the  plant  and  many 
of  the  features  of  the  cupules  and  female  receptacles.  The  rays 
of  the  latter  are  shown  to  be  flat  at  their  extremities,  and  the 
center  of  the  disc  is  marked  by  a  distinct  rounded  elevation.  In 
the  only  receptacles  where  the  rays  can  be  counted  ten  or  eleven 
are  represented,  but  one  receptacle  in  profile  shows  only  four 
rays,  indicating  that  sometimes  at  least  fewer  than  ten  rays  were 
present. in  the  material  figured.     It  is  now   admitted   that  the 


258  Alexander  IV.  Evans, 

normal  number  of  rays  is  nine.  IMicheli,  in  his  description, 
emphasizes  the  glaucous  color  of  the  thallus.  The  species  is  still 
abundant  in  the  vicinity  of  Florence,  where  it  was  originally 
collected,  and  specimens  from  this  region  are  among  those  cited 
above. 

Bertolini's  description  is  drawn  from  female  plants,  and  he 
distinctly  states  that  both  male  receptacles  and  cupules  were 
unknown  to  him.  His  material  came  from  the  vicinity  of  Chia- 
vari  in  Liguria.  He  adds  very  little  to  !Micheli's  account,  but 
gives  the  number  of  rays  definitely  as  ten  and  notes  that  their 
extremities  are  obtuse  or  almost  truncate.  Apparently  his  species 
was  not  very  widely  known  at  first  because  neither  Raddi  nor 
Lehmann  and  Lindenberg  make  any  allusion  to  it.  Raddi's  M. 
papUlata  (3  ifalica  was  based  on  ]\Iicheli's  description  and  figure 
and  on  specimens  collected  at  Micheli's  original  locality.  M. 
nepalensis  and  M.  nitida  were  based  on  material  collected  by 
Wallich  in  Nepal  and  M.  squamosa  on  two  specimens,  one  col- 
lected by  Wallich  in  Xepal  and  the  other  by  Raddi  in  Brazil. 

In  1835  Taylor-^  accepted  M.  paleacea  as  a  species  and  referred 
to  it  not  only  the  Italian  specimens  originally  cited  but  also 
specimens  from  Xepal  collected  by  ^^'allich.  Although  there 
seem  to  be  no  specimens  in  the  Taylor  herbarium  labeled  "M. 
paleacea",  there  are  two  with  a  manuscript  name  of  Taylor's 
from  the  Wallich  collection.  In  one  case  M.  n'ltida  is  given  as 
a  synonym  and  M.  squamosa  as  a  doubtful  synonym,  so  that 
these  specimens  probably  -represent  the  M.  paleacea  of  Taylor's 
paper.  Unfortunately  his  figures  and  description  do  not  corre- 
spond in  all  respects  with  authentic  specimens  of  the  species  in 
question  and  have  therefore  g-iven  rise  to  considerable  con- 
fusion. The  most  marked  discrepancy  is  in  his  account  of  the 
female  receptacles,  where  the  number  of  involucres  is  given  as 
four  to  six,  instead  of  eight  (corresponding  with  nine  rays), 
but  an  error  of  this  sort  might  easily  be  made  if  poor  material 
was  examined.  Taylor's  specimens  are,  indeed,  imperfectly 
developed,  but  they  show  the  thallus  characters  of  M.  paleacea 
very  clearly,  and  his  determination  may  therefore  be  considered 
correct. 

In  proposing  M.  tholopliora  as  a  species  Bischoff  makes 
no  mention  of  M.  paleacea  or  of  the  various  species  based  on 

■'Trans.  Linn.  Soc.  17:378.  pi.  12,  f.  5.     1835. 


American  Species  of  Mar  chant  ia. 


259 


Wallich's  specimens,  so  that  the  works  of  Bertolini  and  of 
Lehmann  and  Lindenberg  may  have  been  unknown  to  him.  His 
description  was   drawn   from   Sommerschu's  material,   collected 


Fig.  7.     Marchantia  paleacea  Bertol. 

Portions  of  ventral  scales.  A.  Margin  of  basal  portion  of  median 
scale,  X  100.  B-L.  Appendages  of  median  scales,  x  100.  A-C.  Italy, 
F.  A.  Artaria,  in  V.  Schiflfner's  Hep.  Europ.  Exsic.  13.  D.  Azores,  F.  D. 
Godman.  E,  F.  Vera  Cruz,  Barnes  &  Land  668.  G,  H.  Cuba,  C.  Wright, 
in  Hep.  Cubenses,  as  .1/.  domingensis.  I-K.  Nepal,  N.  JVallich;  I,  type 
of  M.  ncpalcnsis;  J,  type  of  M.  nitida;  K,  type  of  M.  squamosa;  L, 
Java,  V.  Schiffncr,  in  Iter  Ind.  59,  as  M.  nitida. 


near  Oaxaca,  Mexico,  and  is  unusually  detailed.  The  species 
is  recognized  as  valid  by  subsequent  writers,  the  authors  of  the 
Synopsis  Hepaticarum  placing  it  next  to  M.  nitida  and  Stephani 
following  their  example.  The  original  material  has  not  been 
available   for   study.      Bischoff's   description,  however,   and   the 


26o  Alexander  W.  Evans. 

figure  which  he^'^  afterwards  published  show  almost  beyond  a 
doubt  that  the  species  represents  a  synonym  of  M.  paleacea. 
This  conclusion  seems  further  warranted  by  the  fact  that  the 
true  M.  paleacea  is  now  known  from  several  Mexican  localities. 

Nees  von  Esenbeck,  in  describing  M.  paleacea,  places  it  in  the 
section  Chlamidium  and  calls  especial  attention  to  the  features 
of  the  female  receptacle.  He  gives  the  normal  number  of  rays 
as  nine  and  mentions  the  median  protuberance  of  the  disc  and 
the  dilated  apices  of  the  rays.  Among  the  synonyms  of  the 
species  he  includes  M.  nitida  without  question,  basing  his  opinion 
on  specimens  received  directly  from  Lindenberg.  He  quotes 
a  statement  of  the  latter  author  to  the  effect  that  M.  nitida  is 
very  close  to  the  Italian  M.  paleacea  and  perhaps  identical  with 
it,  accompanied  by  the  remark  that  M.  paleacea  had  not  been  seen 
by  him  when  he  published  M.  nitida  as  a  new  species.  Nees  von 
Esenbeck  hesitates  somewhat  in  the  case  of  Taylor's  M.  paleacea 
but  inclines  toward  the  opinion  that  this  plant  also  must  be  the 
same  as  Bertolini's  species. 

In  spite  of  these  statements  M.  nitida  is  reinstated  as  a  valid 
species  in  the  Synopsis  Hepaticarum  and  Taylor's  M.  paleacea, 
so  far  as  the  Nepal  specimens  are  concerned,  is  given  as  a 
synonym  under  it,  the  range  of  M.  paleacea  being  again  restricted 
to  Italy.  Both  M.  nepalensis  and  M.  squamosa  are  likewise 
accepted  as  valid  and  these  two  species,  together  with  M.  paleacea 
and  M.  nitida,  are  included  under  the  section  Chlamidium.  Many 
years  later,  in  1899,  Stephani,^'^  in  recognizing  these  four  species, 
placed  M.  nitida  and  M.  nepalensis  in  his  section  with  unsym- 
metrical  receptacles,  while  he  placed  M.  paleacea  and  M. 
squamosa  in  the  section  with  symmetrical  receptacles. 

In  1898  doubt  was  again  thrown  on  the  validity  of  M.  nitida 
by  Schiffner,^^  who  stated  that  it  was  probably  synonymous  with 
M.  paleacea.  Two  years  later  he  reaffirmed  this  idea  and  added 
that  M.  calcarata  Steph.,  according  to  a  specimen  in  his  herba- 
rium, was  surely  identical  with  M.  nitida}^    In  the  following  year 

^"Handb.  Bot.  Term,  und  Systemk.  2:  pi.  55,  f.  2727.    1842. 

"  Bull.  Herb.  Bossier  7 :  402,  522.     1899. 

"^  Conspect.  Hepat.  Archip.  Indici  50.     Batavia,  1898. 

"°F1.  de  Buitenzorg  4:31.  1900.  A  specimen  in  the  writer's  herbarium, 
collected  by  the  Abbe  Faurie  at  Tokyo,  Japan,  and  distributed  (Hep.  du 
Japon  2360)  under  the  name  M.  calcarata,  represents  M.  polymorpha. 
Stephani  himself  now  regards  the  true  M.  calcarata  as  a  synonym  of  M. 
dipfera  Mont.,  a  species  which  evidently  requires  further  study. 


American  Species  of  Marchantia.  261 

he  maintained  still  more  definitely  that  M.  paleacea  and  M.  nitida 
were  identical  and  stated  further  that  in  his  opinion  M.  nepalensis 
also  would  have  to  be  considered  a  synonym.^"  He  criticised 
Stephani  for  placing  M.  paleacea  in  one  section  of  the  genus, 
while  he  placed  M.  nepalensis  and  M.  nitida  in  another,  thus 
implying  that  their  relationship  to  the  Italian  species  was  at  best 
remote. 

The  writer  would  agree  with  Schiffner  in  his  reductions.  Type 
specimens  of  M.  papillata  j8  italica,  of  M.  nepalensis,  and  of  M. 
nitida  have  all  been  available  for  study.  The  first  and  third  are 
in  good  condition  and  show  close  agreement  with  each  other 
and  with  the  abundant  material  oi  M.  paleacea  from  other  locali- 
ties. The  type  specimen  of  M.  nepalensis  bears  very  immature 
female  receptacles,  but  the  thallus  characters  are  those  of  M. 
paleacea  and  there  seems  to  be  no  reason  for  attempting  to  main- 
tain the  species  as  valid.  Two  other  species  quoted  above  among 
the  synonyms  of  M.  paleacea  remain  to  be  considered.  The  first 
of  these  is  il/.  squamosa.  Wallich's  specimens  of  tliis  species 
are  clearly  the  same  as  il/.  paleacea,  and  since  these  specimens 
are  the  ones  first  quoted  by  Lehmann  and  Lindenberg,  they  might 
logically  be  considered  the  type.  It  is  probable,  however,  that 
Raddi  originally  gave  the  name  M.  squamosa  to  his  own  Bra- 
zilian specimens,  and  this  is  apparently  the  view  held  by  Stephani 
who  quotes  only  the  specimens  from  Brazil.  Raddi's  specimens 
have  not  been  seen  by  the  writer.  If  they  should  prove  distinct 
from  M.  paleacea  it  might  still  be  possible  to  maintain  ilf. 
squamosa  as  valid.  It  is  imfortunate  that  Raddi  published 
nothing  on  his  species  himself.  The  second  species  to  be  con- 
sidered is  M.  planipora,  which  the  writer  knows  from  a  portion 
of  the  type  material  and  from  specimens  sent  by  Professor 
Miyake.  These  specimens  agree  with  M.  paleacea,  and  the 
descriptions  given  by  Stephani  bring  out  no  essential  differences. 

There  is  usually  little  difficulty  in  distinguishing  M.  paleacea 
even  in  the  absence  of  receptacles.  At  the  present  time  it  is  the 
only  known  North  American  species  in  which  the  epidermal  pores 
constantly  conform  to  the  cruciate  type.  In  this  respect  it  agrees 
with  M.  Berteroana  of  the  Southern  Hemisphere,  a  much  larger 
plant  with  very  different  scale-appendages  and  terete  rays  on  the 
female  receptacle.     Aside  from  the  cruciate  pores  M.  paleacea 


Lotos  49: 92.    1 90 1. 


262 


Alexander  IV.  Evans, 


Fig.  8.     Marchantia  paleacea  Bertol. 

Anatomical  details.  A,  B.  Epidermal  pores  of  thallus,  surface  view, 
X  225.  C-F.  Pores  in  cross-section ;  C,  D,  F,  x  225 ;  E,  x  300.  G,  H. 
Pores,  inner  view,  x  225.  I.  Stalk  of  male  receptacle,  cross-section,  x  40. 
J.  Stalk  of  female  receptacle,  cross-section,  x  40.  K.  Part  of  involucre, 
x  50.  A,  C,  D,  G,  J,  K.  Italy,  F.  A.  Artaria,  in  V.  Schiffner's  Hep. 
Europ.  Exsic.  /_?.  B,  E,  H.  Vera  Cruz,  Barnes  &  Land  66S.  F.  Java, 
V.  Schiffncr,  in  Iter  Ind.  39,  as  M.  nitida.     I.     Puebla,  Frcrc  Arscnc. 


American  Species  of  Marchantia.  263 

is  distinguished  from  M.  polymorpha  by  its  somewhat  smaller 
size,  by  its  total  lack  of  epidermal  papillae,  by  having  the  pores 
bounded  by  six  or  seven  rows  of  cells,  by  the  presence  of 
s'clerotic  cells,  by  the  flattened  rays  of  the  female  receptacles,  by 
the  lack  of  marginal  scales  and  by  marked  differences  in  the 
appendages  of  the"  median  scales.  In  M.  polymorpha  epidermal 
papillae  are  always  present  on  cupules  and  female  receptacles, 
the  pores  are  surrounded  by  only  four  or  five  rows  of  cells,  there 
are  no  sclerotic  cells,  the  rays  of  the  female  receptacle  are  terete, 
and  marginal  scales  are  always  present. 

The  features  of  the  appendages  in  AI.  paleacea  deserve  par- 
ticular attention  (Fig.  7,  B-L).  When  a  long  series  is 
examined,  it  will  be  seen  that  they  exhibit  marked  differences 
in  their  apices  and  margins,  although  they  are  almost  constantly 
longer  than  broad  and  maintain  an  oblong  or  ovate  form.  The 
apex  is  sometimes  rounded,  sometimes  truncate,  and  sometimes 
apiculate  or  even  acute,  while  the  margin  may  be  entire  through- 
out, variously  toothed,  or  even  provided  with  a  basal  lobe.  A 
tooth,  on  its  part,  may  be  the  slightest  and  vaguest  projection 
of  a  marginal  cell,  it  may  be  a  distinctly  projecting  cell,  or  it  may 
consist  of  a  cell  borne  on  a  stalk-cell;  in  some  of  the  broader 
teeth  two  adjoining  cells  may  even  be  involved.  It  must  be 
admitted,  however,  that  large  and  complicated  teeth  are  the  excep- 
tion. In  commenting  on  the  type  specimen  of  M.  nitida  Schiff- 
ner  states  that  the  appendages  are  broadly  ovate,  less  pointed  at 
the  apex  and  scarcely  toothed,  those  of  his  Javan  material  being 
broadly  cordate,  abruptly  pointed  and  with  irregular  and  distant 
marginal  teeth.  These  differences,  which  he  considers  of  little 
importance,  are  shown  by  Fig.  7,  J,  L,  and  at  first  sight  are 
somewhat  striking.  Since,  however,  equally  extreme  conditions 
are  sometimes  found  on  a  single  specimen,  as  shown  by  Fig. 
7,  G,  H,  it  is  evident  that  Schiffner  did  not  underrate  their 
value.  The  appendages  also  vary  in  the  number  of  cells  with 
oil-bodies  which  they  show.  In  some  cases  no  such  cells  are 
present ;  in  one  case  as  many  as  ten  were  counted ;  in  the 
majority  of  cases  there  are  from  one  to  three.  The  gradual 
decrease  in  size  between  the  median  cells  and  the  marginal  cells 
is  usually  evident,  although  the  actual  measurements  are  not  very 
different,  and  a  distinct  margin  is  never  apparent  as  in  M. 
Berteroana.    It  may  be  noted  that  Stephani  assigns  smaller  mar- 


264  Alexander  W.  Evans, 

ginal  cells  to  M.  nepalensis,  M.  nitida,  and  M.  tholophora,  while 
he  states  that  the  cells  of  M.  paleacea  are  subequal  in  size,  a 
specific  difference  which  is  not  supported  by  actual  comparisons. 
When  contrasted  with  the  appendages  of  M.  polymorpha,  those 
of  M.  paleacea  are  seen  to  be  narrower,  usually  less  toothed,  and 
composed  of  larger  cells,  which  show  a  less  marked  decrease 
in  size  between  the  median  and  marginal  regions. 

The  male  receptacles  of  M.  paleacea  seem  to  be  infrequent. 
In  the  few  cases  seen  the  receptacles  have  been  remarkably  like 
those  of  M.  polymorpha,  although  borne  on  shorter  stalks.  In 
other  words  the  disc  has  been  shortly  eight-lobed  with  rounded 
rays  and  narrow  sinuses.  According  to  Schiffner  the  rays  as 
they  grow  older  become  longer  and  give  the  disc  a  palmate 
appearance.  If  this  is  true  the  male  receptacles  exhibit  a  con- 
siderable range  of  variation.  The  cupules  of  M.  paleacea,  with 
their  dentate,  sharp-pointed  lobes,  likewise  agree  with  those  of 
M.  polymorpha  and  its  allies,  except  that  the  outer  surface  is 
free  from  epidermal  papillae. 

When  well  developed  the  female  receptacle  of  M.  paleacea 
consists  of  a  disc  with  nine  horizontal  flat  rays  borne  on  an 
elongated  stalk.  In  many  cases  the  rays  are  about  twenty  degrees 
apart  and  present  the  appearance  of  being  symmetrically 
disposed.  Even  here,  however,  the  single  plane  of  symmetry 
is  marked  by  the  sinus  between  the  two  basal  rays,-  which  is  much 
deeper  than  the  others.  When  the  basal  rays  are  separated  by 
a  sinus  more  than  twenty  degrees  wide  the  plane  of  symmetry  is 
more  apparent  and  the  disc  does  not  show  a  radial  appearance. 
Since  the  width  of  the  sinus  between  the  basal  rays  varies 
markedly  it  should  not  be  made  the  basis  for  specific  separations, 
although  this  has  evidently  been  done  in  the  past.  The  extremi- 
ties of  the  rays  are  variously  dilated  and  are  truncate  or  even 
emarginate  at  the  apex.  In  the  center  of  the  disc  the  hemi- 
spherical or  bluntly  conical  protuberance  is  usually  distinct,  and 
the  same  thing  is  true  of  the  nine  rounded  ridges  extending  from 
the  protuberance  to  the  beginnings  of  the  rays.  When  the  recep- 
tacle is  young  or,  in  some  cases,  when  fertilization  has  not  taken 
place,  the  rays  do  not  spread  horizontally  but  extend  downward, 
and  usually,  under  these  circumstances,  the  median  protuberance 
and  the  radiating  ridges  are  only  slightly  developed.  In  fact 
they  are  not  always  distinct  even  when  the  rays  have  assumed 


American  Species  of  Marchantia.  265 

a  horizontal  position.  Although  the  normal  number  of  rays  in 
M.  paleacea  is  nine,  just  as  in  M.  polymorpha,  deviations  from 
this  number  sometimes  occur,  a  reduced  number  being  associated 
with  poor  development. 

According  to  Prescher  scattered  slime  cells  of  small  size  are 
present  in  the  walls  of  the  cupules  of  M.  paleacea  but  are  lack- 
ing altogether  in  the  thallus  and  in  the  female  receptacles. 
Although  the  thallus  sometimes  agrees  with  his  account  it  does 
not  always  do  so.  Cases  have  been  observed  in  which  slime 
cells  formed  a  rather  conspicuous  feature  of  the  ventral  tissue 
of  the  wings,  an  occasional  cell  of  this  character  being  present 
even  in  the  thickened  median  region.  In  other  cases,  tlie  slime 
cells  were  less  abundant.  In  the  male  receptacles,  which  Prescher 
did  not  examine,  slime  cells  are  fairly  numerous.  It  is  evident 
from  these  observations  that  the  presence  or  absence  of  slime 
cells  can  not  be  utilized  in  distinguishing  M.  paleacea. 

5.     Marchantia  breviloba  sp.  nov. 

Thallus  pale  green,  more  or  less  glaucous,  sometimes  a  little 
pigmented  with  purple,  especially  near  the  margin  and  on  the 
lower  surface,  mostly  0.5-0.8  cm.  wide  and  2-5  cm.  long,  repeat- 
edly dichotomous,  the  successive  forks  usually  1-2  cm.  apart, 
texture  firm  but  not  leathery,  margin  entire ;  epidermis  composed 
of  cells  with  somewhat  thickened  walls,  sometimes  in  two  layers 
mostly  45-90/1,  long  (averaging  about  65]^,)  and  20-4OJU.  wide 
(averaging  about  28/x),  papillae  absent;  pores  (with  their  sur- 
rounding cells)  mostly  125-150/*  long  and  ioo-120/x  wide,  sur- 
rounded usually  by  six  (or  seven)  rows  of  cells  (three  in  each 
series  or  sometimes  four  in  the  outer  series),  innermost  row 
of  upper  series  usually  composed  of  four  cells,  second  row  of 
four  to  eight  cells  and  third  row  of  eight  or  more  cells,  each  row 
of  lower  series  usually  composed  of  four  cells,  inner  opening 
usually  four-sided  (sometimes  three-,  five-,  or  six-sided),  the 
sides  being  concave  and  forming  acute  angles  with  one  another, 
bounding  cells  of  pore  more  or  less  roughened  with  a  resinous 
deposit;  air-chambers  usually  high,  more  or  less  elongated,  their 
outlines  very  indistinct  when  viewed  through  the  epidermis, 
present  everywhere,  rows  of  photosynthetic  cells  often  four  or 
five  cells  long;  compact  ventral  tissue  mostly  twenty  or  twenty- 
five  cells  thick  in  the  median  portion,  the  walls  sometimes  pig- 
mented, more  or  less  thickened  and  showing  distinct  pits,  scle- 
rotic cells  distinct,  scattered,  about  forty  in  a  cross-section, 
largely  confined  to  median  region,  sometimes  as  much  as  0.7  mm. 
in  length,  slime  cells  about  o.i  mm.  in  diameter,  usually  con- 

Trans.  Conn.  Acad.,  Vol.  XXI  18  iqt? 


266  Alexander  W.  Evans, 

spicuous  (often  three  or  four  in  a  cross-section  of  thallus),  scat- 
tered but  more  abundant  toward  the  margin,  sometimes  present 
in  the  walls  between  air-chambers ;  ventral  scales  in  two  distinct 
rows,  the  laminar  scales  alternating  with  the  median  and  only 
a  little  nearer  the  margin  (a  large  portion  of  the  ventral  surface 
being  free  from  scales)  ;  appendages  of  median  scales  ovate  to 
orbicular,  mostly  0.5-0.65  mm.  long  and  0.45-0.55  mm.  wide, 
narrowed  toward  the  rounded,  obtuse,  or  apiculate  apex,  margin 
sinuate,  sparingly  and  irregularly  crenulate  or  denticulate  from 
projecting  cells,  cells  showing  a  gradual  decrease  in  size  toward 
the  margin,  median  cells  usually  distinctly  longer  than  broad, 
mostly  60-120JU,  long  and  28-40/i,  wide,  marginal  cells  mostly 
30-50/X  long  and  18-25/x  wide,  irregular,  the  long  axis  usually 
forming  an  angle  with  the  margin,  rarely  parallel  with  it,  cells 
containing  oil-bodies  lacking:  male  receptacle  borne  on  a  stalk 
1,5-2  cm.  high,  with  two  to  four  rhizoid-furrows  and  a  single 
narrow  dorsal  band  of  air-chambers,  the  disc  mostly  1-1.5  cm. 
broad,  deeply  six-  (or  seven-)  lobed,  the  lobes  or  rays  palmately 
disposed  (the  basal  sinus  being  almost  a  straight  line),  mostly  2-5 
mm.  long  and  1.5  mm.  wide,  rounded  and  with  a  thin  wavy 
margin,  ventral  scales  imbricated,  mostly  in  two  rows:  female 
receptacle  borne  on  a  stalk  6-8  cm.  long,  with  four  rhizoid-fur- 
rows (except  close  to  the  base)  and  a  single  broad  dorsal  band 
of  air-chambers,  the  disc  mostly  0.6-0.8  cm.  broad,  usually 
eleven-lobed,  sometimes  seven-  to  nine-lobed,  the  lobes  or  rays 
short,  I  mm.  long  or  less,  flat,  scarcely  or  not  at  all  dilated  at 
the  truncate  apex,  basal  sinus  considerably  broader  than  the 
others,  upper  surface  of  disc  plane  or  with  low  ridges  corre- 
sponding with  the  lobes ;  involucre  ciliate,  not  lobed ;  spores  yel- 
lowish brown,  about  34/A  in  diameter,  with  a  hyaline  margin 
about  4iti  wide,  outer  face  bearing  a  few  low  lamellae,  sometimes 
forming  a  very  indistinct  reticulum ;  elaters  about  8/x  wide, 
bispiral :  cupules  shortly  and  irregularly  ciliate-dentate,  the 
teeth  sometimes  adjoining  and  sometimes  separated  by  sinuses 
of  varying  width,  mostly  two  or  three  cells  long  and  one  or  two 
cells  wide  at  the  base,  epidermal  papillae  lacking.     (Fig.  9.) 

The  following  specimens  of  this  species,  which  seems  to  be 
very  local,  have  been  examined : 

Jamaica  :  without  definite  localities  or  dates.  Wilds  (N.  Y., 
four  specimens)  ;  Hardware  Gap  and  vicinity,  April,  1903, 
W._  R.  Maxon  1115  (U,  S.,  Y.)  ;  July,  1903,  A.  W.  Evans  175, 
203  (Y.)  ;  Chestervale,  July,  1903,  A.  IV.  Evans  211  (Y.)  ; 
vicinity  of  Cinchona,  March,  1905,  C.  E.  Cummings  28  (N,  Y., 
Y.)  ;  St.  Catherine's  Peak  and  vicinity,  August,  1906,  A.  W. 
Evans  441  (Y.).    The  specimens  collected  by  Wilds  include  both 


American  Species  of  Marchantia. 


267 


Fig.  9.  Marchantia  breviloba  Evans 
Plants,  natural  size,  and  various  anatomical  details.  A.  Male  plant,  x  i. 
B  Female  plant,  x  i.  C-E.  Appendages  of  ventral  scales,  apical  por- 
tions X  100.  F.  Epidermal  pore  of  thallus,  surface  view,  x  225.  U 
Pore  in  cross-section,  x  225.  H.  Inner  opening  of  pore,  x  225.  I,  J. 
Stalk  of  male  receptacle,  cross-sections,  x  40;  I  was  cut  near  base;  J, 
near  apex.  K.  Stalk  of  female  receptacle,  cross-section,  cut  near  base. 
A.  Jamaica,  W.  R.  Max  on  11 15.  B-D,  F-K.  Jamaica,  A.  W.  Evans  175. 
type.    E.    Jamaica,  A.  IV.  Evans,  203. 


268  Alexander  W.  Evans, 

female  and  gemmiparous  material.  Two,  bearing  the  numbers 
5  and  6,  are  labeled  "Marchantia  conica" ;  the  other  two  bear 
no  name.  No.  175,  collected  by  the  writer,  may  be  designated 
the  type. 

The  thallus  of  M.  breviloba  bears  a  strong  resemblance  to 
that  of  M.  paleacea,  being  of  about  the  same  size  and  similarly 
subject  to  pigmentation.  The  ventral  scales  and  the  appendages 
of  tlie  median  scales  in  these  two  species  likewise  have  certain 
features  in  common.  The  arrangement  of  the  scales,  for  exam- 
ple, is  very  similar  and  the  appendages  agree  in  form,  in  the  grad- 
ual decrease  in  the  size  of  the  cells  in  passing  toward  the  margin, 
and  in  some  of  the  peculiarities  of  the  margin  itself.  Even  the 
pores  look  a  good  deal  alike  when  examined  through  a  lens.  A 
detailed  examination,  however,  quickly  brings  out  points  of  dif- 
ference. In  M.  breviloba  the  pores  are  not  of  the  cruciate  type, 
the  inner  opening  (so  far  as  observed)  being  surrounded  by  evenly 
bulging  cells  and  thus  usually  exhibiting  a  four-sided  outHne 
with  concave  sides  and  narrow  angles ;  in  M.  paleacea  the  pores 
are  distinctly  cruciate.  In  M.  breviloba  the  cells  of  the  appen- 
dages are  markedly  larger  than  in  M.  paleacea  and  oil-containing 
cells  seem  to  be  constantly  absent;  in  M.  paleacea  oil-contain- 
ing cells  can  often  be  detected.  In  M.  breviloba  slime  cells  seem 
always  to  be  numerous  and  conspicuous;  in  M.  paleacea  they 
are  less  frequent  and  may  be  absent  altogether:  this  last  dif- 
ference, unfortunately,  is  one  to  be  used  with  caution. 

The  differential  characters  yielded  by  the  receptacles  and 
cupules,  in  separating  M.  breviloba  from  M.  paleacea,  are  even 
more  marked  than  those  derived  from  the  thallus.  In  M.  brevi- 
loba the  male  receptacle  is  borne  on  a  long  stalk  with  a  distinct 
band  of  air-chambers  and  usually  witli  four  rhizoid-f urrows ; 
and  the  disc  is  imsymmetrically  divided  into  six  or  seven 
elongated  lobes,  the  basal  sinus  being  much  broader  tlian  the 
odiers  and  often  approximating  a  straight  line.  In  M.  paleacea 
the  male  receptacle  is  borne  on  a  very  short  stalk  without  air- 
chambers  and  with  only  two  rhizoid-f  urrows ;  while  the  disc  is 
very  shortly  and  apparently  radially  divided  into  eight  (or  more) 
often  indistinct  lobes,  the  basal  sinus  being  of  about  tlie  same 
width  as  the  others. 

In  M.  breviloba  tlie  stalk  of  the  female  receptacle  shows  four 
rhizoid-furrows,  and  the  disc  is  very  shortly  lobed,  the  lobes 


American  Species  of  Marchantia.  269 

being  usually  more  than  nine  and  scarcely  if  at  all  dilated  at  the 
apex.  In  M.  paleacea  the  stalk  of  the  female  receptacle  shows 
only  two  rhizoid-furrows,  and  the  disc  is  more  deeply  divided, 
the  lobes  being  usually  nine  (or  fewer)  and  more  or  less 
distinctly  dilated  at  the  apex;  the  upper  surface  of  the  disc, 
moreover,  shows  a  median  protuberance  and  nine  ridges  cor- 
responding with  the  rays,  these  structures  being  very  indistinct 
in  M.  brez'iloba  or  absent  altogether.  In  M.  breviloba  the  cupula 
is  simply  short-ciliate ;  in  M.  paleacea  it  shows  a  series  of  ciliate 
or  dentate  lobes. 

There  is  no  difficulty  in  separating  AI.  breviloba  from  M.  poly- 
morpha.  The  new  species  is  smaller,  it  lacks  marginal  ventral 
scales,  the  appendages  of  the  median  scales  have  larger  cells 
and  usually  fewer  marginal  teeth,  the  epidermal  cells  are  more 
complex  being  surrounded  by  more  rows  of  cells,  there  are  no 
epidermal  papillae,  the  stalk  of  the  male  receptacle  has  air- 
chambers  and  usually  more  rhizoid-furrows,  the  disc  is  more 
deeply  and  more  unsymmetrically  lobed,  the  stalk  of  the  female 
receptacle  has  more  rhizoid-furrows,  the  disc  is  less  deeply  lobed, 
and  the  lobes  are  flattened  instead  of  being  terete.  The  difference 
in  the  cupules,  already  noted  in  connection  with  M.  paleacea, 
would  apply  equally  well  in  separating  the  species  from  M.  poly- 
morpJia. 

6.     Marchantia  domingensis  Lehm.  &  Lindenb. 

Marchantia  domingensis   Lehm.    &   Lindenb. ;     Lehmann,    Pug. 

Plant.  6  :  22.     1834. 
Marchantia  inflexa  Nees  &  Mont.;    Montague,  Ann.  Sci.  Nat. 

Bot.  II.  9  :  43.     1838. 
Marchantia  disjuncta  Sulliv.  Am.  Jour.  Sci.  II.  i :  74.     1846. 
Marchantia  linearis  G.  L.  &  N.  Syn.  Hep.  529.     1847  (in  part). 

Not  Lehm.  &  Lindenb. 
Marchantia  martinicensis  Spreng. ;   G.  L.  N.  /.  c.  531.     1847  (^s 

synonym). 
Marchantia  Elliottii  Steph.  Bull.  Herb.  Boissier  7  :  400.    1899. 
Marchantia  caracensis  Steph.  /.  c.  526.     1899. 

Thallus  pale  to  dark  green,  not  glaucous,  slightly  or  not  at 
all  pigmented  with  purple,  usually  4-6  mm.  wide  and  2-3  cm. 
long,  dichotomous,  the  successive  forks  usually  1-1.5  cm.  apart, 
texture  delicate,  margin  entire ;  epidermis  composed  of  cells  with 


270  Alexander  W.  Evans, 

slightly  thickened  walls,  sometimes  in  two  layers,  mostly  30-60/A 
long  (averaging  about  45iu.)  and  15-30/1.  wide  (averaging  about 
23/i),  papillae  absent;  pores  (with  their  surrounding  cells) 
mostly  90-I30/A  long  and  70-80/*  wide,  surrounded  usually  by  six 
(or  seven)  rows  of  cells  (three  or  four  in  the  upper  and  three 
in  the  lower  series),  the  two  lower  rows  of  the  upper  series 
usually  composed  of  eight  (six  to  ten)  cells  apiece  (more  rarely 
of  only  three  to  five  cells),  the  other  rows  of  four  (three  to 
five)  cells  apiece,  inner  opening  usually  four-sided,  more  rarely 
three-  or  five-sided,  with  the  sides  straight  or  nearly  so,  the 
bounding  cells  more  or  less  obscured  by  a  resinous  deposit ;  air- 
chambers  of  medium  height,  isodiametric  or  somewhat  elongated, 
their  boundaries  sometimes  distinct  and  sometimes  vague  when 
viewed  through  the  epidermis,  present  everywhere,  rows  of 
photosynthetic  cells  often  four  or  five  cells  long  but  sometimes 
shorter;  compact  ventral  tissue  about  twenty  cells  thick  in  the 
median  portion,  the  walls  sometimes  pigmented,  more  or  less 
thickened  and  showing  distinct  pits,  sclerotic  cells  usually  dis- 
tinct, scattered,  mostly  five  to  thirty  in  a  cross-section,  more 
abundant  in  the  median  portion  but  sometimes  present  in  the 
wings,  in  the  latter  case  often  distinctly  visible  without  section- 
ing, slime-cells  lacking;  ventral  scales  in  two  distinct  rows,  the 
laminar  scales  alternating  with  the  median  scales  and  not  much 
nearer  the  margin ;  appendages  of  the  median  scales  broadly 
lanceolate  to  ovate,  when  well  developed  mostly  0.35-0.6  mm. 
long  and  0.27-0.45  mm.  wide  but  sometimes  considerably  smaller, 
apex  apiculate,  acute,  or  cuspidate,  margin  more  or  less  densely 
denticulate  or  dentate,  the  teeth  usually  one  or  two  cells  long, 
cells  showing  a  gradual  and  slight  decrease  in  size  toward  the 
margin,  median  cells  usually  longer  than  broad,  mostly  40-80/x 
long  and  20-40/1  wide,  marginal  cells  mostly  20-40/x  long  and 
15-20/*  wide,  irregular  but  the  long  axis  usually  at  right  angles 
or  nearly  so  to  the  margin,  cells  containing  oil-bodies  apparently 
always  lacking:  male  receptacle  borne  on  a  stalk  5  mm.  long 
or  less,  with  two  to  four  rhizoid-f urrows  and  a  single  broad  dorsal 
band  of  air-chambers,  the  disc  variable  in  size  but  mostly  6-8  mm. 
broad,  deeply  lobed,  the  lobes  or  rays  usually  four  to  six  but 
sometimes  two,  three,  seven  or  eight,  palmately  disposed,  the 
basal  sinus  a  very  broad  angle  or  a  straight  line,  mostly  3-6  mm. 
long  and  2-3  mm.  wide,  rounded,  with  a  thin  wavy  margin  extend- 
ing across  the  basal  sinus,  ventral  scales  imbricated,  in  two  or 
more  rows;  female  receptacle  borne  on  a  stalk  1.5-2  cm.  high, 
with  two  to  four  rhizoid-furrows  and  a  single  broad  dorsal 
band  of  air-chambers,  the  disc  mostly  5-7  mm.  wide,  the  lobes 
or  rays  spreading  at  maturity,  extending  about  half  way  from 
the  margin  to  the  center,  normally  seven  but  often  only  five  or  six, 
more  rarely  eight  to  eleven,  slightly  or  not  at  all  dilated  at  the 
truncate,  irregularly  crenate  or  slightly  emarginate  apex,  upper 


American  Species  of  MarcJiantia.  271 

surface  of  disc  and  rays  usually  plane  but  sometimes  more  or  less 
convex,  basal  sinus  broader  than  the  others  and  sometimes  form- 
ing a  very  obtuse  angle ;  involucre  very  delicate,  the  margin 
minutely  and  often  irregularly  crenulate  to  short-ciliate,  the  teeth 
usually  varying  from  one  to  three  cells  in  length ;  spores  brown- 
ish yellow,  about  28yu  in  diameter,  the  outer  face  bearing  a  series 
of  low  irregular  ridges  not  forming  a  network ;  elaters  about  6/x 
wide,  bispiral :  cupules  closely  short-ciliate,  the  cilia  mostly  one 
to  four  cells  long,  outer  surface  without  papillae.      (Figs.  10-12.) 

A  widely  distributed  species  in  the  southern  United  States,  the 
West  Indies,  Mexico,  Central  America,  and  Venezuela.  The  fol- 
lowing specimens  have  been  examined : 

Tennessee:  Etowah,  June,  1909,  F.  McCormick  (C.  C.  H., 
Y.,  listed  as  M.  disjuncta  by  tlie  writer  in  Bryologist  13 :  33. 
1910). 

Georgia:  west  bank  of  Ocmulgee  River,  Hawkinsville,  June, 
1902,  R.  M.  Harper  1^8 2a  (N.  Y.,  U.  S.)  ;  Samochechobee  Creek, 
near  Killen,  Clay  County,  October,  1902,  R.  M.  Harper  lygih 
(N.  Y.,  U.  S.)  ;  Flint  River  swamp,  below  Albany,  Dougherty 
County,  August,  1903,  R.  M.  Harper  ipjie  (N.  Y.,  U.  S.). 

Florida  :  Devil's  Millhopper,  Alachua  County,  February,  1909, 
R.  M.  Harper  14  (N.  Y.)  ;  April,  1915,  N.  L.  T.  Nelson  45,  46, 
47,51  (Y.). 

Alabama:  banks  of  the  Alabama  River  near  Claiborne, 
JV.  S.  Sullivant  (H.,  type  locality  of  M.  disjuncta,  specimens 
distributed  in  Muse.  Alleg.  286  and  Hep.  Bor.-Amer.  128)  ; 
Auburn,  May,  1896,  L.  M.  Underwood  (N.  Y.,  Y.,  distributed,  as 
M.  disjuncta,  in  Hep.  Amer.  182)  ;  June,  1897,  Earle  &  Baker 
52  (N.  Y.). 

Texas:  Fort  Worth,  1887,  G.  S.  Thompson  10  (N.  Y.,  U.  S., 
listed  as  M.  disjuncta  by  Underwood  in  Bot.  Gaz.  20:  69.  1895)  ; 
Hallettsville,  May,  1892,  G.  C.  Nealley  59  (C.  C.  H.,  U.  S.)  ; 
Austin,  March,  1909,  F.  D.  Heald  (Y.).  * 

Arkansas:  Fort  Harvey,  no  date,  F.  L.  Harvey  2  (N.  Y., 
listed  as  M.  disjuncta  by  Underwood,  I.  c). 

Mexico  :  without  definite  locality  or  date,  C.  H.  Schidtz  1229 
(B.,  listed  as  M.  papillata,  by  Stephani  in  Bull.  Herb.  Boissier 
7:397.     1899). 

Hidalgo:  Tula,  C.  G.  Pringle  io6/j  (Y.,  distributed  in  PI. 
Mex.  under  a  manuscript  name  of  Stephani). 


272  Alexander  W.  Evans, 

Puebla:  banks  along  Avenida  Hidalgo  and  path  to  barranca, 
Tezuitlan,  1908,  Barnes  &  Land  jjj  (Y.), 

Tamaulipas  :  near  Victoria,  June,  1907,  E.  Palmer  446 
(N.  Y.,  Y.). 

Vera  Cruz  :  Orizaba  and  vicinity,  1855,  F.  Miiller  2^yj  (N.  Y., 
listed  as  M.  disjimcfa  by  Underwood,  /.  c.)  ;  1857,  C.  Mohr 
(N.  Y.,  Y.)  ;  March,  1890,  W.  Stone  114,  115  (N.  Y.,  listed  by 
Underwood,  /.  c.)  ;  1892,  /.  G.  Smith  (N.  Y.,  Y.)  ;  Cordova, 
1885,  VV.  G.  Farlow  18  (N.  Y.,  listed  as  M.  tholophora  by  Under- 
wood, /.  c.  70). 

Guatemala:  Black  River,  ^S.  Watson  2pjb  (H.,  N.  Y.,  listed  as 
M.  linearis  by  Underwood,  /.  c.  69)  ;  near  the  Finca  Sepacuite, 
Alta  Verapaz,  March  and  April,  1902,  Cook  &  Griggs  82,  255, 
^05  (U.S.,  Y.). 

Bahamas:  Fort  Charlotte,  New  Providence,  April,  1905, 
L.  J.  K.  Brace  3pi6  (N.  Y.,  Y.,  listed  as  M.  chenopoda  by  the 
writer  in  Bull.  Torrey  Club  38:206.  191 1)  ;  New  Providence, 
October,  1904,  L.  J.  K.  Brace  8/3  (N.  Y.). 

Cuba:  without  definite  localities  or  dates,  C.  Wright  (dis- 
tributed as  M.  disjuncta  and  M.  linearis  in  Hep.  Cubenses)  ;  San 
Andre,  April  14,  1865,  C.  Wright  (H.,  Y.)  ;  valley  of  the  San 
Juan  River,  near  Matanzas,  March,  1903,  Britton,  Brit  ton  & 
Shafer  326  (N.  Y.,  Y.)  ;  Guines,  Havana,  March,  1905,  M.  T. 
Cook  (N.  Y.,  Y.)  ;  Almendares  River,  near  Puentas  Grandes, 
Havana,  April,  1908,  Brother  Leon  ^23  (N.  Y.,  Y.)  ;  falls  of  the 
Habanilla  and  near  Siguanea,  Trinidad  Mountains,  Santa  Clara, 
March,  1910,  E.  G.  Britton  4855,  jo/6  (N.  Y.,  Y.)  ;  vicinity  of 
Guane,  Pinar  del  Rio,  March,  191 1,  Britton,  Britton  &  Cowell 
97/0  (N.  Y.,  Y.)  ;  vicinity  of  Pinar  del  Rio,  March,  191 1,  E.  G. 
Britton  looij  (N.  Y.,  Y.)  ;  Finca  Guerrero,  Rio  Yayabo,  St. 
Spiritus,  Santa  Clara,  December,  191 1,  Brother  Clement  44  (N. 
Y.,  Y.)  ;  Banaos  Hills,  Santa  Clara,  August,  191 3,  Brother  Leon 
4036  (N.  Y.,  Y.)  ;  Ensenada  de  Mora,  Oriente,  March,  191 2, 
Britton,  Cozvell  &  Shafer  13005  (N.  Y.,  Y.). 

Santo  Domingo  :  Azui,  province  of  Seibo,  November,  1909, 
N.  Taylor  281  (N.  Y.,  Y.). 

Jamaica  :  without  definite  locality  or  date,  N.  Wilson  jpj, 
611  (N.  Y.,  listed  as  M.  disjuncta  by  Underwood,  /.  c.)  ;  Hart- 
ford  and   adjoining   properties,   near   Priestman's   River,   June, 


American  Species  of  Marchantia.  273 

1904,  W.  R.  Maxon  2535  (U.  S.,  Y.)  ;  road  from  Holly  Mount 
to  Resource,  February,  1905,  W.  Harris  88/6  (N.  Y.,  Y.)  ; 
vicinity  of  Mandeville,  September,  1907,  E.  G.  Britton  429 
(N.  Y.,  Y.)  ;  February,  1910,  i'.  Broum  280,  282,  283  (N.  Y., 
Y.)  ;  Kempsport  and  Roaring  River  Falls,  March,  1908,  E.  G. 
Britton  668,  708  (N.  Y.,  Y.)  ;  Porus  to  Clarendon  Park,  Sep- 
tember, 1908,  A^.  L.  Britton  3/8^  (N.  Y.,  Y.)  ;  Cokely,  near 
Castleton,  and  Hope  Gardens,  January,  1914,  W.  Harris  (N.  Y.). 

Porto  Rico :  Las  Marias  Road  and  vicinity  of  Mayaguez, 
March,  1906,  Britton  &  Marble  5^4,  613  (N.  Y.,  Y.)  ;  Lares  to 
San  Sebastiano,  April,  1913,  Britton  &  Marble  2/^7,  2803  (N.  Y., 
Y.)  ;  Rio  de  Maricao,  April,  191 3,  E.  G.  Britton  2494  (N.  Y., 
Y.)  ;  Ciales,  August,  1913,  /.  R.  Johnston  P40  (N.  Y.,  Y.)  ; 
between  Arecibo  and  Utuado,  July,  1901,  Underzvood  &  Griggs 
836  (U.  S.,  Y.)  ;  March,  1914,  E.  G.  Britton  2074  (N.  Y.,  Y.)  ; 
Lares,  June,  1901,  Underzvood  &  Griggs  36  (U.  S.,  Y.)  ; 
June,  1914,  /.  R.  Johnston  2070  (N.  Y.,  Y.)  ;  Monte  Mon- 
toso,  February,  1915,  Britton  &  Coiuell  4177  (N.  Y.,  Y.)  ; 
La  Juanita,  near  Las  Marias,  February,  191 5,  E.  G.  Britton 
3965  (N.  Y.,  Y.)  ;  vicinity  of  Utuado,  March,  191 5,  E.  G. 
Britton  5168,  51 12  (N.  Y.,  Y.). 

St.  Kitts  :  Bethesda,  /.  C.  Brentcl  (N.  Y.,  listed  as  M.  linearis 
in  Syn.  Hep.  529  and  as  M.  domingensis  by  Stephani  in  Bull.  Herb. 
Boissier  7 :  399.  1899)  ;  without  definite  locality,  1853,  Walwyn 
(N.  Y.)  ;  Old  Road  and  Lambert  Estate,  September  and  Octo- 
ber, 1901,  Britton  &  Cowell  481,  626  (N.  Y.,  U.  S.,  Y.). 

Montserrat:  mountain  pass  to  Roches  and  Tar  River,  Feb- 
ruary, 1907,  /.  A.  Shafer  867,  871  (N.  Y.,  Y.). 

Guadeloupe:  Morne  Rouge,  Gombeyre  (Basse-Terre)  and 
Morne  Papillon,  1897-1900,  Pere  Duss  250,  231,  312,  392  (N.  Y., 
Basse-Terre  specimens  listed  as  M.  disjuncta  by  Schiffner  in 
Oesterr.  Bot.  Zeitschr.  57:  51.     1907). 

Dominica:  without  definite  locality  or  date,  W.  R.  Elliott  1292 
(B.,  type  of  M.  Elliott ii). 

Martinique:  without  definite  locality,  date,  or  collector's 
name  (M.,  received  from  Merat,  type  of  M.  inflexa)  ;  without 
definite  locality  or  date,  Sieber  378  (N.  Y.,  probably  type  of  M. 
martinicensis)  ;  St.  Pierre  and  between  Deux-Choux  and  Gros- 
Morne,  1899,  1900,  Pere  Duss  342,  393  (N.  Y.)  ;   Morne  Rouge, 


2  74  Alexander  W.  Evans, 

August,  1 90 1,  Pere  Duss  578,  580,  581  (N.  Y.)  ;  Deux-Choux, 
September,  1903,  Pere  Duss  (N.  Y.,  determined  as  M.  caracensis 
by  Stephani). 

St.  Vincent:  without  definite  localities  or  dates,  L.  Guilding 
(H.,  N.  Y.,  specimens  in  the  Taylor  and  Mitten  herbaria,  labeled 
"March,  linearis — chenopoda.")  : 

Grenada:  Annandale,  St,  George's,  March,  1906,  W.  E. 
Broadway  (N.  Y.). 

Trinidad:  without  definite  locality,  1878-80,  A.  Fendler 
(N.  Y.,  U.  S.,  Y.,  distributed  as  M.  chenopoda)  ;  Mareval  Valley, 
1913,  R.  Thaxter  (H.,  Y.)  ;  La  Lenia  Valley,  1913,  R.  Thaxter 
(H.,  Y.). 

Venezuela:  Rio  Cartude,  Caracas,  1856,  Gollmer  (B.,  type 
of  M.  caracensis)  ;  Caracas,  August,  1902,  A.  F.  Blakeslee. 
(H.,Y.). 

The  following  stations,  cited  in  literature,  should  also  be  noted : 
Santo  Domingo:   without  definite  locality,  date  or  collector's 

name  (type). 

Guadeloupe:    without  definite  locality  or  date,  L'Herminier 

6p  (listed,  as  M.  linearis,  by  Bescherelle  in  Jour,  de  Bot.  7 :  193. 

1893). 
Martinique:   without  definite  localities  or  dates,  C.  Bclanger 

124  (listed,  as  M.  linearis,  by  Bescherelle,  I.  c.)  ',  K.  von  Martins; 

C.  Bclanger  24  in  part,  ^74;  Hahn  774  (the  last  three  listed,  as 

M.  inflexa,  by  Bescherelle,  /.  c). 

The  type  specimen  of  M.  domingensis  was  collected  in  Santo 
Domingo,  neither  the  date  nor  the  collector's  name  being  men- 
tioned in  the  original  publication.  Unfortunately  this  specimen 
has  not  been  available  for  study.  In  its  absence  the  writer 
has  been  obliged  to  rely  upon  West  Indian  material  determined 
as  M.  linearis  and  upon  the  type  specimen  of  M.  inflexa.  These 
are  referred  to  M.  domingensis  without  question  by  Stephani  and 
agree  in  all  essential  respects  'with  the  other  specimens  listed. 
The  type  specimens  of  M.  disjuncta,  M.  Elliottii  and  M.  caracensis 
have  likewise  been  examined  and  show  a  similar  agreement. 
The  writer  feels  convinced,  therefore,  that  the  synonymy  given 
above  is  correct. 

The  type  specimen  of  A4.  inflexa,  received  from  the  ]\Iontagne 
herbarium,  is  very   fragmentary  but  bears  two   female  recep- 


American  Species  of  Marchantia. 


275 


Fig.  10.     Marchantia  domingensis  Lehm.  &  Lindenb. 

Appendages  of  ventral  scales,  x  100.  A,  B.  Alabama,  Earle  &  Baker 
^2.  C.  Cuba,  C.  Wright,  distributed  in  Hep.  Cubenses,  as  M.  disjuncta. 
D,  E.  Hope  Gardens,  Jamaica,  W.  Harris.  F.  St.  Kitts,  /.  C.  Breutel, 
specimen  in  the  Mitten  herbarium,  labeled  M.  linearis  and  M.  chenopoda. 
G.  St.  Kitts,  Britton  &  Cowell,  626.  H.  Martinique,  type  of  M.  inflexa. 
I,  J.    Mareval  Valley,  Trinidad,  R.  Thaxter. 


276  Alexander  IV.  Evans, 

tacles :  the  first  shows  nine  truncate  rays  clearly,  the  basal  sinus 
being  broader  than  the  others ;  the  second  shows  ten  less  clearly 
marked  rays.  The  type  of  M.  disjuncta  agrees  closely  with  the 
specimens  distributed  by  Sullivant,  by  Austin,  and  by  Under- 
wood. They  are  well  represented  in  the  beautiful  figures  pub- 
lished by  Sullivant,^^  and  the  female  receptacles  show  a  variable 
number  of  rays.  The  type  of  M.  Elliottii  bears  numerous  female 
receptacles,  which  show  from  five  to  nine  rays  apiece,  the  rays 
being  plane  or  slightly  convex  and  blunt.  The  type  of  M. 
caracensis  bears  both  cupules  and  female  receptacles,  the  latter 
showing  five  rays  apiece.  All  of  these  specimens  show  the 
dentate  scale  appendages  and  other  features  characteristic  of 
the  species. 

The  thallus  of  M.  domingefisis  is  smaller  and  usually  more 
delicate  in  texture  than  in  any  of  the  preceding  species.  In  wet 
localities  the  photosynthetic  layer  is  especially  thin  and  the  epi- 
dermis is  rarely  more  than  a  single  cell  in  thickness,  so  that 
plants  growing  under  these  conditions  present  an  unusually 
fragile  aspect.  The  epidermal  pores,  however,  maintain  their 
complex  structure  and  form  conspicuous  whitish  dots  on  the 
upper  surface.  Sometimes  the  boundaries  of  the  air-chambers 
show  distinctly  through  the  epidermis,  but  they  are  usually  indis- 
tinct. The  structure  of  the  pores  is  much  the  same  as  in  M. 
breviloha,  although  the  inner  opening  is  bounded  by  straighter 
lines. 

The  sclerotic  cells  in  the  thallus  exhibit  a  great  deal  of  varia- 
bility. When  abundantly  developed  they  occur  both  in  the 
thickened  median  portion  of  the  thallus  and  in  the  wings,  those 
in  the  latter  position  showing  distinctly  as  elongated  brown  spots 
when  examined  from  underneath.  The  sclerotic  cells  appear  to 
be  separated  from  one  another  by  parenchyma  when  a  cross  sec- 
tion of  a  thallus  is  examined.  As  a  matter  of  fact,  in  the  median 
portion  of  the  thallus  at  least,  they  often  form  elongated  strands 
running  for  a  considerable  distance,  the  acute  ends  of  the  cells 
slightly  overlapping.  There  are  many  cases,  however,  where  the 
sclerotic  cells  are  very  scantily  developed.  Sometimes  there  are 
none  at  all  present  in  the  wings  although  the  median  portion 
still  retains  them;   sometimes  even  the  median  portion  seems  to 


Mem.  Am.  Acad.  II.  3 :  pi.  3.    1846. 


American  Species  of  Marchantia. 


277 


lack  them  completely,  although  no  specimens  have  yet  been  seen 
in  which  a  careful  examination  failed  to  show  traces  of  sclerotic 
cells  in  this  region. 

The  appendages  of  the  ventral  scales  (Fig.  10)  have  better 
developed  teeth  than  in  any  other  North  American  species  of 
Marchantia,  although  the  South  American  M.  papillata  is  a  close 


Fig.  II.     Marchantia  domingensis  Lehm.  &  Lindenb. 

Epidermal  pores  of  thallus,  x  225.  A,  B.  In  surface  view.  C,  D.  In 
cross-section.  E,  F.  Inner  openings.  A,  C,  E.  Texas,  F.  D.  Heald. 
B,  D,  F.    Cokely,  near  Castleton,  Jamaica,  W.  Harris. 


rival  in  this  respect.  The  teeth  are  very  irregular,  the  simplest 
being  single  cells  which  project  as  rounded  or  bluntly  pointed  pro- 
cesses. Between  these  simple  teeth  and  irregular  lobes,  several 
cells  long  and  wide,  are  all  possible  gradations.  The  apical  tooth 
tends  to  be  longer  than  the  others,  although  this  tendency 
is  not  always  apparent.  The  median  cells  of  the  appendages  are 
often  longer  than  broad  and  a  decrease  in  the  size  of  the  cells 


278  Alexander  W.  Evans, 

between  the  middle  portion  and  the  margin  is  distinctly  evident. 
There  is  no  definite  border,  however,  as  in  M.  Berteroana. 
Sometimes,  in  wet  situations,  the  appendages  remain  very  small, 
although  even  under  these  circumstances  the  marginal  teeth  form 
a  conspicuous  feature.  The  more  sharply  pointed  apices,  the 
larger  and  more  irregular  teeth,  the  larger  marginal  cells  and  the 
lack  of  cells  containing  oil-bodies  will  at  once  distinguish  the 
appendages  of  M.  domingensis  from  those  of  M.  polymorpha. 

According  to  the  original  description  of  M.  domingensis  the 
female  receptacle  is  semicircular  and  ten-lobed  to  the  middle, 
the  lobes  being  dilated  and  emarginate-crenate  at  the  apex.  The 
Synopsis  gives  the  number  of  lobes  as  seven  to  ten,  while  Stephani 
states  that  nine  lobes  are  present.  The  original  description  of 
M.  inflexa  assigns  nine  to  eleven  lobes  to  the  receptacle,  while 
Sullivant  gives  seven  to  nine  as  the  number  of  lobes  in  M. 
disjuncta;  here  again  Stephani  places  the  number  of  lobes  defi- 
nitely at  nine,  and  gives  the  same  number  for  M.  Elliottii.  In 
M.  caracensis,  however,  he  states  that  only  five  or  six  lobes  are 
present.  As  a  matter  of  fact  the  receptacles  are  exceedingly 
variable  and  it  is  not  easy  to  decide  what  the  typical  or  normal 
number  of  lobes  really  is.  In  the  material  from  the  mainland 
seven  is  perhaps  the  usual  nimiber  but  five  lobes  often  occur 
and  more  than  seven  have  been  observed  in  several  instances.  In 
the  West  Indian  material  nine  lobes  are  present  more  frequently, 
but  seven  or  even  only  five  lobes  are  not  unusual.  Sometimes  the 
lobes  seem  to  be  subdivided,  so  that  it  is  not  always  easy  to  count 
them  except  by  means  of  the  involucres  which  alternate  with  them» 
The  lobes  vary  not  only  in  number  but  in  thickness.  In  some 
cases  they  are  very  thin  and  flat,  in  other  cases  thicker  and 
convex.  When  the  lobes  are  fleshy  the  center  of  the  disc  some- 
times shows  a  low  swelling,  but  it  is  usually  plane,  and  the  recep- 
tacle never  shows  the  conspicuous  median  protuberance  and 
radiating  ridges  which  are  so  characteristic  of  M.  paleacea. 

The  study  of  the  involucre  is  beset  with  considerable  difficulty 
on  account  of  its  extreme  delicacy.  This  has  apparently  been 
the  cause  of  considerable  confusion  in  the  published  descrip- 
tions. In  M.  domingensis,  for  example,  the  involucre  is  said  to 
be  laciniate-ciliate  or  shortly  fimbriate ;  in  M.  inflexa,  laciniate ; 
in  M.  disjuncta,  sparingly  dentate  or  subentire ;  in  M.  caracensis, 
shortly  fimbriate.    Fig.  12,  E-J,  brings  out  the  range  of  variation 


American  Species  of  Marchantia. 


279 


observed  by  the  writer.  The  nearest  approach  to  an  entire  con- 
dition is  seen  in  Fig.  12,  E,  although  even  here  the  margin  is 
distinctly  and  closely  crenulate;  in  Fig.  12,  F,  the  crenulate  con- 
dition is  more  pronounced,  some  of  the  teeth  being  two  cells  long ; 
in  Fig.  12,  G,  a  crenulate  portion  directly  adjoins  a  short-ciliate 
portion,  in  which  the  cilia  are  two  or  three  cells  long;    in  Fig. 


Fig.  12.     Marchantia  DOMiNGENsisLehm.  &  Lindenb. 

Anatomical  details.  A.  Stalk  of  male  receptacle,  cross-section,  x  40. 
B-D.  Stalks  of  female  receptacles,  cross-sections,  x  40,  C  having  been 
cut  near  base.  E-J.  Parts  of  involucres,  x  100:  H  and  I  were  dissected 
from  a  single  involucre;  J,  from  another  involucre  on  the  same  recep- 
tacle. K.  Part  of  cupule,  x  100.  A.  Florida,  A''.  L.  T.  Nelson  47.  B,  F. 
Texas,  F.  D.  Heald.  C,  H-J.  Jamaica,  E.  G.  Britton  668.  D.  St.  Kitts, 
Britton  &  Cowell  626.  E.  Alabama,  Earle  &  Baker  52.  G.  Guatemala, 
Cook  &  Griggs  82.    K.    Cokely,  Jamaica,  W.  Harris. 


12,  H-J,  taken  from  a  single  receptacle,  the  variation  to  be 
expected  is  shown  with  especial  clearness,  some  of  the  marginal 
teeth  or  cilia  being  straight  and  some  curved.  In  all  probabihty 
the  laciniate  and  fimbriate  involucres  of  the  descriptions  have 
been  the  result  of  irregular  tears  in  old  material. 

In  the  case  of  the  male  receptacle,  Sullivant  states  that  the 
number  of  lobes  in  M.  disjuncta  is  normally  seven,  although  some 


2  8o  Alexander  W.  Evans, 

of  his  figures  show  six,  four,  or  only  three  lobes.  Stephani 
places  the  number  at  about  eight  in  his  account  of  M.  disjuncta 
but  makes  no  mention  of  the  male  receptacles  in  M.  domingensis, 
M.  Elliottii  or  M.  caracensis.  Apparently  a  good  deal  of  varia- 
tion is  to  be  expected  from  varying  environmental  conditions,  as 
GoebeP-  has  recently  noted,  poorly  developed  material  showing  a 
reduced  number  of  lobes.  Even  when  only  two  rays  are  present 
the  upper  part  of  the  stalk  shows  two  rhizoid-furrows,  although 
the  lower  part  shows  but  one.  The  deeply  lobed  male  receptacle 
will  distinguish  M.  domingensis  from  all  the  preceding  species 
except  M.  breviloha.  In  addition  to  its  greater  size  this  species 
differs  from  M.  domingensis  in  the  appendages  of  the  ventral 
scales,  which  are  larger  and  much  less  toothed ;  in  the  slime  cells 
of  the  thallus,  these  structures  being  apparently  never  found  in 
M.  domingensis;  in  the  less  deeply  lobed  female  receptacles ;  and 
in  the  distinctly  ciliate  involucres, 

7.     Marchantia  papillata  Raddi 

Marchantia  papillata  a  brasiliensis  Raddi,  Mem.  Soc.  Ital.  Modena 

19 :  44.     1823 ;   20 :  pi.  6a,  f.  3,  4.     1829. 
Marchantia  platycnemos  Schwaegr. ;    Gaudichaud,  Freyc.  Voy. 

Bot.  218.     1827  (as  to  the  Brazihan  plant). 
Marchantia  androgyna  Nees ;   Martius,  Fl.  Brasil.  i :  308.     1833. 

Not  L. 
Marchantia    subandina    Spruce,    Trans.    Bot.    Soc.    Edinburgh 

15:561.     1885. 

Thallus  dull  green,  not  glaucous,  sometimes  more  or  less  pig- 
mented with  purple,  usually  1.5-3  mm.  wide  and  0.8-1. 15  cm. 
long,  dichotomous,  the  forks  usually  only  1.5-3  mm.  apart,  texture 
firm,  margin  entire ;  epidermis  composed  of  cells  with  more 
or  less  thickened  walls,  usually  in  a  single  layer,  mostly  20-50/i 
long  (averaging  about  30^11)  and  i2-20;a  wide  (averaging  about 
i5/x),  papillae  absent;  pores  (with  their  surrounding  cells) 
mostly  ^o-yoiM  long  and  40-45/i  wide,  surrounded  usually  by  five 
(or  six)  rows  of  cells  (two  or  three  in  the  upper  series  and  three 
in  the  lower  series),  each  row  usually  composed  of  four  cells 
(rarely  of  three  or  five),  the  lowest  row  of  the  upper  series 
sometimes  with  from  five  to  eight  cells,  inner  opening  usually 
four-sided  (rarely  with  three  or  five  sides),  the  sides  concave, 

'*  Organographie  der  Pflanzen,  2d  ed.  699.     1915. 


American  Species  of  Marchantia.  281 

bounding  cells  more  or  less  obscured  by  a  resinous  deposit;  air- 
chambers  low,  isodiametric  or  somewhat  elongated,  their  outlines 
very  indistinct  when  viewed  through  the  epidermis,  present 
everywhere,  rows  of  photosynthetic  cells  usually  two  or  three 
cells  long;  compact  ventral  tissue  about  fifteen  cells  thick  in 
the  median  portion ;  the  walls  sometimes  pigmented,  more  or  less 
thickened  and  showing  distinct  pits,  sclerotic  cells  scattered, 
mostly  fifteen  to  twenty  in  a  cross-section,  more  abundant  in  the 
median  portion  but  often  present  in  the  wings,  sometimes  clearly 
visible  without  sectioning,  slime  cells  lacking;  ventral  scales  in 
two  rows,  the  row  of  laminar  scales  more  or  less  irregular  but 
tending  to  alternate  with  the  median  scales  and  not  much  nearer 
the  margin ;  appendages  of  the  median  scales  ovate,  when  well 
developed  mostly  0.3-0.45  mm.  long  and  0.25-0.3  mm.  wide  but 
sometimes  considerably  smaller,  apex  apiculate,  acute,  or  cuspi- 
date, margin  subentire  or  usually  more  or  less  closely  denticulate 
or  dentate,  the  teeth  irregular,  mostly  one  or  two  cells  long, 
rarely  larger  and  more  lobe-like,  cells  showing  a  gradual  and 
slight  decrease  in  size  toward  the  margin,  median  cells  usually 
longer  than  broad,  mostly  49-60/*  long  and  25-30/x  wide,  marginal 
cells  mostly  30-45/*  long  and  15-25/*  wide,  irregular  but  usually 
with  the  long  axis  at  right  angles  or  nearly  so  to  the  margin, 
cells  containing  oil-bodies  apparently  always  lacking,  male  recep- 
tacle borne  on  a  stalk  about  3  mm.  long  with  two  rhizoid-fur- 
rows,  the  disc  about  0.8  cm.  wide,  deeply  four-  to  eight-lobed,  the 
lobes  or  rays  palmately  disposed  (the  basal  sinus  being  very  broad), 
about  3  mm.  long  and  i  mm.  wide,  rounded  at  the  apex  and  with 
a  thin  wavy  margin,  ventral  scales  imbricated :  female  receptacle 
borne  on  a  stalk  1.5-2  cm.  long,  with  two  rhizoid-furrows  and 
a  single  broad  dorsal  band  of  air-chambers,  the  disc  mostly 
3-4  mm.  broad,  normally  nine-lobed  (but  sometimes  with  five  to 
eight  lobes),  the  lobes  or  rays  1-1.5  mm.  long  and  about  i  mm. 
wide,  distinctly  dilated  at  the  truncate  to  emarginate  apex,  strongly 
convex  on  upper  surface,  basal  sinus  broader  than  the  others, 
upper  svirf ace  of  disc  with  a  low  median  protuberance ;  involucre 
very  delicate,  irregularly  lobed  and  crispate,  otherwise  entire  or 
slightly  and  irregularly  crenulate ;  pores  yellowish  brown,  about 
26/*  in  diameter,  outer  face  bearing  a  series  of  very  low  ridges 
not  forming  a  network,  margin  narrow  and  often  indistinct,  less 
than  2/*  broad ;  elaters  about  8/*  broad,  bispiral :  cupules  spar- 
ingly and  irregularly  denticulate  to  short-ciliate,  the  teeth  being 
projections  of  marginal  cells  or  from  one  to  four  cells  long, 
epidermal  papillae  lacking.       (Fig.  13.) 

The  following  specimens  have  been  examined: 

Brazil  :   Rio  de  Janeiro,  without  date,  G.  Raddi  (N.  Y.,  type)  ; 

without  date,  /.  Milne  (N.  Y.)  ;   without  definite  locality  or  date, 

W.J.Bnrchell  1857  (N.  Y.). 

Trans.  Conn.  Acad.,  Vol.  XXI  19  1917 


282  Alexander  W.  Evans, 

Paraguay:  Paraguari,  August,  1883,  B.  Balansa  (N.  Y., 
U.  S.,  distributed  in  PI.  du  Paraguay  4006,  and  listed  by  Stephani 
in  Rev.  Bryol.  14:58.     1887). 

Peru:  near  the  Rio  Huallaga,  R.  Spruce  (type  of  M.  sub- 
andina,  distributed  in  Hepaticae  Spruceanae). 

Bolivia:  Isapuri,  October,  1901,  R.  S.  Williams  214=,  (N.  Y., 
Y.). 

Although  the  writer  has  seen  no  specimens  of  M.  papillata 
from  other  localities,  the  following  records  may  be  cited  from  the 
literature : 

Cuba  :  "ad  terram  in  locis  humidis  prope  S.  Marcos,"  Ramon 
de  la  Sagra,  gemmiparous  specimens  (listed  by  Montague  in 
Ramon  de  la.  Sagra,  Hist.  Fis.  Pol.  y  Natur.  Cuba  9 :  290.    1845). 

Martinique:  without  definite  locaHty  or  collector's  name 
(listed  by  Underwood  in  Bot.  Gaz.  20:70.  1895);  without 
definite  locality  or  date,  A.  Plee  18 21  (listed  by  Bescherelle  in 
Jour,  de  Bot.  7 :  193.     1893). 

Brazil:  Rio  Janeiro,  C.  Gaudichaud  (listed,  as  M.  platycne- 
inos,  in  Freyc.  Voy.  Bot.  218.  1827)  ;  Minas  Geraes,  K.  von 
Martins  (listed  by  Nees  von  Esenbeck  in  Naturg.  Europ.  Leberm. 
4:109.  1838);  "ad  muros  humidos  aquaeductus,  loco  Corco- 
vado  dicto"  near  Rio  de  Janeiro,  and  "ad  terram  juxta  flumen  Rio 
Negro",  A.  d'Orbigny  (listed  by  Montagne  in  D'Orbigny,  Voy. 
dans  I'Amer.  Merid.  f  :  397.     1839) . 

Falkland  Islands  :  without  definite  locality  or  date,  C.  Gaudi- 
chaud (listed,  as  M.  platycnemos,  in  Freyc.  Voy.  Bot.  218.    1827). 

The  original  M.  papillata  included  two  varieties,  a.  brasiliensis 
and  ^.  italica.  Nees  von  Esenbeck^^  soon  showed,  however,  that 
the  second  variety  was  a  synonym  of  M.  paleacea;  he  therefore 
reserved  the  name  M.  papillata  for  the  first  variety,  a  course 
which  has  been  followed  by  subsequent  writers.  He  was  also 
the  first  to  recognize  the  fact  that  his  Brazilian  M.  androgyna 
belonged  to  M.  papillata  and  to  include  M.  platycnemos  among 
the  synonyms  of  the  same  species.  It  is  possible,  however,  that 
M.  platycnemos  ought  still  to  be  maintained  as  a  species,  at  least 
in  part.     It  was  based  on  three  specimens,  the  first  from  the 


Naturg.  Europ.  Leberm.  4:101.     li 


American  Species  of  Marchantia.  283 

Marianne  Islands  in  the  Pacific,  the  second  from  the  Falkland 
Islands,  and  the  third  from  Brazil.  The  first  specimen  may  be 
regarded  as  the  type  of  the  species  since  it  is  mentioned  first. 
A  portion  of  this  specimen  in  the  herbarium  of  the  New  York 
Botanical  Garden  shows  that  the  type  is  distinct  from  M.  papil- 
lata,  as  here  understood,  but  that  it  is  very  close  to  M.  emarginata 
and  perhaps  synonymous  with  it.  Nees  von  Esenbeck's  inclu- 
sion of  M.  platycnemos  among  the  synonyms  of  M.  papillata 
must  therefore  be  considered  as  applying  to  a  part  only  of 
Schwaegrichen's  species  as  originally  described. 

Of  Raddi's  original  figures,  /.  5  is  said  to  represent  male  plants 
and  /.  4,  female  plants.  This  is  obviously  an  error,  the  recep- 
tacles shown  under  /.  j  being  clearly  female.  In  /.  ja  six 
receptacles  are  drawn,  two  showing  six  lobes  apiece  and  one 
seven  lobes,  the  number  being  doubtful  in  the  other  three.  In 
/.  56  an  enlarged  receptacle  with  nine  lobes  is  represented;  the 
lobes  show  clearly  the  enlarged  apices  with  more  or  less  distinct 
emarginations,  and  no  difiference  is  brought  out  between  the  basal 
sinus  and  the  others.  The  receptacles  shown  in  /.  4  are  very 
doubtful  and  bear  a  disc  which  is  scarcely  lobed  at  all.  Unless 
drawn  from  very  immature  material  they  probably  belong  to 
some  other  species  than  M.  papillata.  It  should  be  noted,  how- 
ever, that  the  Synopsis  describes  the  disc  of  the  male  receptacle 
as  "subdimidiato  excentrico  marginibus  repando-lobatis,"  thus 
evidently  recognizing  a  male  receptacle  in  Raddi's  so-called 
female  receptacle ;  but  Stephani  apparently  discards  this  view, 
since  he  does  not  mention  the  male  receptacles  at  all. 

The  specimen  of  M.  papillata,  quoted  above  as  the  type,  is  in 
the  ]\Iitten  herbarium  and  was  received  from  Hooker.  It  is 
very  fragmentary  but  includes  three  female  receptacles,  two 
showing  eight  lobes  apiece  and  the  third,  seven  lobes.  The 
dilated  apices  of  the  lobes  and  the  broader  basal  sinus  are  clearly 
apparent.  This  specimen  has  been  carefully  compared  with  the 
other  specimens  cited  and  found  to  agree  with  them  in  all  essen- 
tial respects.  Spruce  compares  M.  subandina  with  both  M.  papil- 
lata and  M.  Berteroana,  which  he  knew  from  description  only. 
He  ascribes  to  the  species,  however,  a  polyoicous  inflorescence 
and  monospiral  elaters.  Unfortunately  the  specimens  which  he 
distributed,  although  agreeing  with  his  description  in  other 
respects,    show    a    strictly    dioicous    inflorescence    and    bispiral 


284 


Alexander  W.  Evans, 


elaters,  so  tliat  his  statements  about  the  inflorescence  and  elaters 
must  have  been  based  on  a  misconception. 

The  thallus  of  M.  papillata  is  even  smaller  than  in  M.  domin- 
gensis;  it  is,  in  fact,  the  smallest  American  species  known  at  the 


Fig.  13.     Marchantia  papillata  Raddi 

Appendages  of  ventral  scales  and  other  anatomical  details.  A-E. 
Appendages  of  ventral  scales,  B  and  C  having  been  dissected  from  the 
same  thallus,  x  100.  F.  Epidermal  pore  of  thallus,  surface  view,  x  225. 
G,  H.  Pores  in  cross-section,  x  225.  I.  Inner  opening  of  pore,  x  225. 
J.  Stalk  of  female  receptacle,  cross-section,  x  40.  K-M.  Parts  of 
involucres,  x  100.  A.  Brazil,  G.  Raddi,  type.  B,  C,  G,  J,  K.  Paraguay, 
B.  Balansa  4006.  D-F,  H,  I.  Bolivia,  R.  S.  Williams  2145.  L,  M.  Peru, 
R.  Spruce,  type  of  M.  subandina. 

present  time.  So  far  as  the  structure  of  the  thallus  is  concerned 
the  agreement  with  M.  domingensis  is  very  close.  Slime  cells 
seem  to  be  always  lacking  in  both  species  and  sclerotic  cells  are 
usually  abundant  and  equally  conspicuous.  The  appendages  of 
the  median  scales,  moreover,  are  essentially  the  same,  although 
the  marginal  teeth  in  M.  papillata  exhibit  a  slightly  wider  range 


American  Species  of  Marchantia.  285 

of  variability.  There  are,  however,  certain  differences  in  the 
epidermal  cells  and  pores  which  deserve  some  emphasis.  The 
cells  and  pores  are  not  only  distinctly  smaller  than  in  M.  domin- 
gensis,  but  the  pores  are  constructed  on  a  simpler  plan,  the 
opening  being  surrounded  by  fewer  rows  of  cells  and  the  number 
of  cells  in  each  row  being  less  subject  to  variation.  The  cells 
bounding  the  inner  opening  tend  to  be  more  convex.  It  must 
be  admitted  that  these  differences  are  very  slight  and  might 
not  deserve  much  attention  if  they  were  not  supported  by  other 
dift'erences  derived  from  the  female  receptacle. 

The  variability  of  the  receptacles  wnth  respect  to  the  number 
of  lobes  has  already  been  commented  upon  in  connection  with 
the  type  specimen  and  becomes  still  more  apparent  from  a  study 
of  the  published  descriptions.  According  to  Nees  von  Esenbeck 
the  usual  number  of  lobes  is  seven,  eight  to  ten  being  sometimes 
present ;  according  to  Spruce  nine  lobes  are  present  in  M.  sub- 
andina;  according  to  Schiffner,  who  studied  Raddi's  specimen 
in  the  Lindenberg*  herbarium,  the  normal  number  of  lobes  in 
M.  papiUata  is  six,  a  larger  number  being  unusual ;  according 
to  Stephani  both  M.  papiUata  and  M.  suhandina  have  nine  lobes 
apiece.  In  the  writer's  opinion  nine  may  be  regarded  as  the 
normal  number  of  lobes,  although  a  smaller  number  frequently 
occurs.  In  the  number  of  lobes,  therefore,  the  species  agrees  on 
the  whole  with  M.  domingensis.  The  receptacle,  however,  is 
smaller;  the  lobes  are  more  dilated  and  more  frequently  emargi- 
nate  at  the  apex;  the  medium  protuberance  of  the  disc  and  the 
convexity  of  the  lobes  are  more  pronounced ;  and  the  involucre  is 
less  variable,  being  entire  or  nearly  so  and  apparently  never  bear- 
ing elongated  teeth  or  cilia.  Just  how  constant  these  differences 
are  can  only  be  established  by  the  study  of  more  material.  If  they 
should  be  found  to  intergrade  it  might  become  necessary  to 
reduce  M.  domingensis  to  synonymy  under  M.  papiUata,  but  tlie 
differences  seem  sufficient  at  the  present  time  to  justify  the 
maintenance  of  both  species  as  valid. 

The  group  of  species  to  which  M.  domingensis  and  M.  papil- 
lata  belong  is  well  represented  in  paleotropic  regions.  Among 
the  species  which  are  referable  to  this  group  M.  emarginata 
R.  Bl.  &  N.,3*  M.  linearis  Lehm.  &  Lindenb.,^^  and  M.  Schaden- 

'*  Nova  Acta  Acad.  Caes.-Leop.  Carol.  12 :  192.     1824. 
''Lehmann,  Pug.  Plant.  4:8.     1832. 


2  86  Alexander  W.  Evans, 

bergii  Steph.^^  may  be  especially  mentioned.  The  first  of  these 
has  a  very  extensive  distribution  and  is  reported  by  Stephani 
from  China,  the  Himalayas,  Japan,  Java  and  the  Philippine 
Islands ;  the  second  is  known  from  various  parts  of  India ;  the 
third,  from  the  Philippine  Islands  only.  These  three  species  are 
closely  related,  and  the  differences  brought  out  by  Stephani  are 
not  very  convincing.  As  indicated  in  the  synonymy  the  authors 
of  the  Synopsis  referred  specimens  of  M.  doniingcnsis  to  M. 
linearis,  an  error  first  pointed  out  by  Stephani.  Since,  however, 
M.  linearis  is  so  closely  related  to  M.  emarginata  it  will  be 
sufficient  to  compare  the  two  American  plants  with  this  latter 
species,  a  full  description  of  which  has  been  published  by 
Schiffner.^' 

In  size  M.  emarginata  is  comparable  with  M.  domingensis;  in 
the  structure  of  the  female  receptacle,  with  M.  papillata.  It 
agrees  with  both  species  in  the  possession  of  sclerotic  cells  in 
the  thallus ;  in  the  general  features  of  the  ventral  scales ;  in  the 
closely  toothed  appendages  of  the  median  scales ;  and  in  the 
structure  of  the  involucre.  Schiffner  describes  the  latter  as  lobed 
and  almost  entire,  but  it  is  sometimes  possible  to  detect  a  few 
short  teeth,  especially  toward  the  outer  extremities.  A  few  dif- 
ferences in  the  structure  of  the  thallus  may  be  mentioned.  In 
M.  emarginata,  for  example,  although  sclerotic  cells  are  present 
they  are  never  so  abundant  or  so  conspicuous  as  they  sometimes 
are  in  M.  domingensis.  The  thallus  is  further  distinguished  by 
the  possession  of  slime  cells  and,  according  to  Schift'ner,  by  the 
occasional  presence  of  epidermal  papillae.  It  would  be  unwise, 
however,  to  lay  much  stress  on  any  of  these  differences,  since 
the  structures  on  which  they  are  based  are  so  very  variable. 

The  female  receptacle  of  M.  emarginata  shows  the  features 
described  under  M.  papillata  in  an  intensified  form.  The  median 
protuberance  is  not  only  more  pronounced,  but  the  lobes  them- 
selves might  almost  be  described  as  costate,  while  their  apices  are 
more  markedly  dilated  and  emarginate.  These  features  are  of 
course  subject  to  variation.  The  male  receptacles  are  distin- 
guished by  their  long  and  slender  stalks.  The  cupules  are  much 
the  same  as  in  two  American  species. 

'"Bull.  Herb.  Boissier  7:524.     1899. 
"  FI.  de  Buitenzorg  4:31.     1900. 


American  Species  of  Marchantia.  287 

8.     Marchantia  Bescherellei  Steph. 

MarcJiantia  Bescherellei  Steph.  Rev.  Bryol.  15:86.     1888. 

Thallus  dull  green,  not  glaucous,  slightly  or  not  at  all  pig- 
mented with  purple,  mostly  5-7  mm.  wide  and  2-3  cm.  long, 
dichotomous,  the  forks  1-1.5  cm.  apart,  texture  very  delicate, 
margin  entire,  sometimes  vaguely  and  irregularly  plicate ;  epi- 
dermis composed  of  cells  wath  very  thin  walls,  in  a  single  layer, 
mostly  40-70/^  long  (averaging  about  50/*)  and  20-35JU.  wide 
(averaging  about  25^11),  papillae  absent;  pores  (with  their  sur- 
rounding cells)  mostly  160-200/A  long  and  i20-i60ja  wide, 
surrounded  usually  by  seven  or  eight  rows  of  celk,  three  or  four 
in  the  upper  series  and  four  in  the  lower  series,  the  two  lower 
rows  of  the  upper  series  usually  composed  of  ten  to  twelve  or 
even  more  cells  apiece,  the  other  rows  of  five  or  six  cells,  rarely 
of  only  four  cells  apiece,  inner  opening  usually  five-  or  six- 
sided,  rarely  only  four-sided,  with  the  sides  strongly  concave, 
resinous  deposit  slight ;  air-chambers  low,  somewhat  elongated, 
their  boundaries  indistinct  when  viewed  through  the  epidermis, 
present  everywhere  except  close  to  the  margin,  rows  of  photo- 
synthetic  cells  usually  two  or  three  cells  long ;  compact  ventral 
tissue  mostly  twelve  to  fifteen  cells  thick  in  the  median  portion, 
abruptly  thinning  out  in  the  wings,  the  walls  somewhat  pig- 
mented, slightly  or  not  at  all  thickened,  sclerotic  cells  scattered, 
sparingly  developed,  slime  cells  lacking;  ventral  scales  in  two 
rows,  the  laminar  scales  alternating  with  the  median  scales  and 
not  much  nearer  the  margin ;  appendages  of  the  median  scales 
ovate,  when  well  developed  mostly  0.5-0.6  mm.  long  and  0.35- 
0.45  mm.  wide  but  often  smaller,  apex  acute,  margin  sparingly 
and  irregularly  dentate,  the  teeth  in  the  basal  portion  often 
larger  and  sharper  and  sometimes  lobe-like,  cells  showing  a 
gradual  and  slight  decrease  in  size  toward  the  margin,  median 
cells  usually  longer  than  broad,  mostly  50-80jli  long  and  30-40/A 
wide,  marginal  cells  mostly  30-50/x  long  and  15-25/x  wade,  irregu- 
lar but  usually  perpendicular  or  nearly  so  to  the  margin,  cells 
containing  oil-bodies  lacking:  male  receptacle  borne  on  a  stalk 
about  I  cm.  long  (in  the  only  example  studied),  with  two  rhizoid- 
furrows  and  (apparently)  with  a  single  broad  dorsal  band  of 
air-chambers,  the  disc  7  mm.  broad,  with  four  short  and  rounded 
lobes  or  rays  with  thin  margins,  basal  sinus  more  than  180  degrees, 
the  other  sinuses  narrow,  ventral  scales  apparently  in  two  rows : 
female  receptacle  borne  on  a  stalk  about  2  cm.  long  (in  the  only 
example  studied),  with  two  rhizoid-furrows  and  dorsal  air- 
chambers  apparently  in  two  distinct  bands,  the  disc  about  7  mm. 
wide,  with  five  short  and  rounded  lobes  or  rays,  upper  surface  of 
receptacle  plane,  basal  sinus  a  straight  line  or  nearly  so ;  involucre 
delicate,  closely  and  irregularly  dentate,  some  of  the  teeth  three 
or  four  cells  long  and  two  to  four  cells  wide  at  the  base,  other 


2  88  Alexander  W.  Evans, 

teeth  smaller:  spores  (according  to  Stephani)  yellowish  brown, 
23)11  in  diameter,  erose  along  the  ridges,  otherwise  smooth. 
(Fig.  14.) 

Known  only  from  the  following  specimens : 
Brazil  :  Rio  Janeiro,  A.  Glazioii  6^48  (B.,  N.  Y.,  type)  ;  E.  Vie 
12^  (B.,  listed  by  Stephani  in  Bull.  Herb.  Boissier  7  :  406.    1899). 

The  material  of  M.  Bescherellei  in  the  Boissier  herbarium,  por- 
tions of  which  have  been  examined  by  the  writer,  includes  the 
female  type  specimens  and  the  male  specimens  collected  by  Ule. 
A  sterile  specimen  of  the  type  material,  in  the  Mitten  herbarium, 
has  likewise  been  examined.  It  will  be  noted  that  Spruce,^^  who 
published  a  list  of  Glaziou's  specimens,  makes  no  mention  of 
No.  6348.  According  to  the  label  on  the  specimen  in  the  Mitten 
herbarium,  Spruce  thought  that  this  number  might  perhaps 
represent  a  new  genus  of  the  Marchantiaceae,  but  he  evidently 
reached  no  definite  conclusion  about  it.  Probably  he  had  only 
sterile  material  at  his  disposal,  because  the  female  receptacle 
shows  at  once  that  Stephani  was  correct  in  referring  the  plant 
to  the  genus  Marchantia.  Unfortunately  the  specimens  studied 
by  the  writer  were  very  fragmentary  and  remained  shriveled  after 
long  soaking  in  water.  It  was  therefore  impossible  to  gain  from 
them  an  adequate  idea  of  the  species,  and  some  of  the  statements 
made  about  the  structure  must  be  regarded  as  more  or  less 
tentative. 

The  texture  of  M.  Bescherellei  is  exceedingly  delicate  and  the 
thallus  thins  out  abruptly  in  passing  from  the  midrib  to  the 
wings.  In  the  latter  the  ventral  tissue  becomes  reduced,  accord- 
ing to  Stephani,  to  a  single  cell  in  thickness,  and  the  marginal 
portion,  where  the  entire  thallus  is  only  one  cell  thick,  is  four 
cells  broad.  Although  the  air-chambers  are  low  the  photosyn- 
thetic  tissue  is  well  developed  and  characteristic  and  the  pores 
are  large  and  complex. 

The  appendages  of  the  ventral  scales  are  composed  of  cells 
which  show  a  gradual  decrease  in  size  in  passing  from  the  median 
portion  toward  the  margin,  resembling  in  this  respect  the  appen- 
dages of  M.  domingensis  and  M.  papillata,  but  the  margin  itself  is 

*"  Rev.  Bryol  15 :  33,  34.     1888. 


American  Species  of  Marchantia. 


289 


very  different  in  being  much  more  sparingly  toothed.  Among  the 
preceding  species  the  appendages  find  their  closest  counterparts  in 
M.  paleacea  and  M.  breviloha.  Their  apices,  however,  are  more 
uniformly  sharp-pointed,  their  teeth  tend  to  be  sharper,  and  their 
marginal  cells  are  more  frequently  placed  at  right  angles  to  the 
margin. 

With  regard  to  the  female  receptacle  there  are  marked  dis- 
crepancies between  Stephani's  original  description  and  the  later 
description  of  his  Species  Hepaticarum.     According  to  the  orig- 


FiG.  14.     Marchantia  Bescherellei  Steph. 

Anatomical   details.     A-C.     Appendages   of  ventral   scales,   x    100. 
Part  of  an  involucre,  x  100.    Drawn  from  the  type  specimen. 


D. 


inal  account  the  disc  is  green,  convex  in  the  middle,  five-lobed  for 
one  third  the  distance  from  margin  to  center,  the  lobes  being 
rounded  and  shortly  incised  at  the  apex,  plane  and  horizontal, 
delicate  and  beautifully  reticulated.  Doubt  is  thrown,  however, 
upon  the  constancy  of  the  five-lobed  condition.  The  involucres 
are  described  as  reddish,  firm  in  texture,  and  shortly  ciliate. 
According  to  the  later  account  the  disc  is  brownish  green,  delicate 
and  veiny,  plano-convex  in  the  center,  and  nine-lobed,  the  lobes 
being  plane,  connate  almost  to  the  apex,  rounded  and  very  shortly 
incised.  The  involucre  is  said  to  be  hyaline,  small-lobed,  irregu- 
larly and  shortly  fimbriate.  In  the  only  receptacle  seen  by  the 
writer  the  disc  is  five-lobed,  the  two  basal  lobes  being  only  about 


290  Alexander  W.  Evans, 

half  as  broad  as  the  other  three  lobes  and  the  sinus  being 
practically  a  straight  line.  The  three  broad  lobes  appear  to  be 
very  shortly  incised  at  the  apex  but  they  are  actually  only  emar- 
ginate,  the  apparent  incision  being  really  filled  by  an  extension 
of  the  membranous  margin  of  the  lobes.  In  other  respects  the 
lobes  agree  with  Stephani's  accounts.  The  involucre  is  distinctly 
toothed,  but  the  teeth  are  scarcely  long  enough  or  sharp  enough 
to  be  described  as  cilia  or  fimbriations.  The  discrepancy  in  the 
number  of  lobes  which  Stephani's  accounts  bring  out  might  of 
course  be  due  to  a  variability  of  the  species,  which  could  only 
become  evident  through  the  study  of  more  extensive  material. 

It  is  unfortunate  that  the  structure  of  the  stalks  of  the  recep- 
tacles must  be  left  in  doubt.  In  Stephani's  original  description 
the  stalk  of  the  female  receptacle  is  said  to  bear  two  dilated 
dorsal  lamellae  but  no  mention  is  made  of  these  in  his  later 
account,  and  nothing  whatever  is  said  about  the  structure  of  the 
stalk  of  the  male  receptacle.  From  the  scanty  supply  of  material 
available,  it  has  been  impracticable  to  prepare  cross-sections  of 
stalks,  so  that  the  writer  is  unable  to  confirm  Stephani's  state- 
ment or  to  add  further  details.  If  the  stalk  of  the  male  recep- 
tacle bears  a  band  of  air-chambers  the  relationship  might  be  with 
M.  domingensis.  If  two  bands  of  air-chambers  are  present  in 
the  stalk  of  the  female  receptacle,  a  relationship  with  M.  clieno- 
poda  would  be  indicated,  and  it  is  worthy  of  note  that  Mitten 
referred  Glaziou's  type  specimen  to  M.  brasiliensis  Lehm.  & 
Lindenb.,  a  species  which  is  now  included  among  the  synonyms 
of  M.  chenopoda  L.  According  to  our  present  knowledge,  how- 
ever, the  systematic  position  of  the  species  can  hardly  be 
determined. 

9.     Marchantia  chenopoda  L. 

MarchanUa  chenopoda  L.  Sp.  Plant.  1137.     I753- 

Marchantia  androgyna  L.     /.  c.  1138.       1753  (in  part)  ;   Swartz, 

Fl.  Ind.  Occ.  1882.    1806. 
Chlamidium  indicum  Corda ;   Opiz,  Beitr.  zur  Naturg.  647.    1828 

(nomen  nudum). 
Marchantia  Szvartdi  Lehm.  &  Lindenb. ;    Lehmann,  Pug.  Plant. 

4:9.     1832. 
Marchantia  cartilaginea  Lehm.  &  Lindenb.  /.  c.  4:  31.     1832. 
Marchantia  brasiliensis  Lehm.  &  Lindenb.  /.  c.  4:  32.     1832. 


American  Species  of  Marchantia.  291 

Grimaldia  peruviana   Nees  &  Mont. ;    Montagne,   Fl.   Boliv.   in 

D'Orbigny,  Voy.  dans  TAmer.  Merid.  7- :  53.     1839. 
Marchantia  peruviana  Nees ;  G.  L.  &  N.  Syn.  Hep.  538.    1847. 
Marchantia  Dillenii  Lindb.  Krit.  Gransk.  Hist.  Muse.  47.     1883. 

Thalltis  pale  or  yellowish  green,  sometimes  more  or  less 
glaucous,  usually  tinged  with  purple  or  brownish  on  the  lower 
surface,  usually  4-7  mm.  wide  and  2-3  cm.  long,  dichotomous, 
the  successive  forks  averaging  about  i  cm.  apart,  texture  varying 
from  firm  and  often  leathery  to  delicate,  margin  entire  ;  epidermis 
composed  of  cells  with  slightly  thickened  walls,  often  in  two 
layers,  mostly  30-60/A  long  (averaging  about  40/^)  and  15-30/t 
wide  (averaging  about  22/x),  papillae  absent,  slime  cells  often 
present,  averaging  about  6o/x  in  diameter ;  pores  variable  in  size, 
usually  (with  their  surrounding  cells)  measuring  ioo-i70jU,  in 
length  and  80-130/A  in  width  but  sometimes  considerably  smaller, 
usually  surrounded  by  seven  rows  of  cells  (four  in  the  upper 
series  and  three  in  the  lower),  more  rarely  by  six,  eight  or  even 
nine  rows,  the  two  upper  rows  of  the  upper  series  and  the  two 
lower  rows  of  the  lower  series  composed  of  four  to  six  cells 
apiece,  the  third  row  of  each  series  usually  of  twice  as  many, 
and  the  fourth  row  of  the  upper  series  usually  of  a  much  larger 
number,  sometimes  of  as  many  as  eighteen,  inner  opening  four- 
to  six-sided  with  the  sides  straight  or  more  or  less  concave,  cell- 
walls  mostly  smooth  throughout;  air-chambers  of  medium 
height,  isodiametric  or  somewhat  elongated,  their  boundaries 
usually  distinct  but  sometimes  obscure  when  viewed  through 
the  epidermis,  present  everywhere,  cells  of  partition  walls  some- 
times including  slime  cells,  rows  of  photosynthetic  cells  usually 
three  or  four  cells  long;  compact  ventral  tissue  mostly  twenty 
to  twenty-five  cells  thick  in  the  median  portion,  the  w^alls  some- 
times pigmented,  more  or  less  thickened  and  showing  distinct 
pits,  sclerotic  cells  usually  distinct,  scattered,  mostly  twenty  to 
fifty  in  a  cross-section,  usually  abundant  in  both  median  portion 
and  wings,  slime  cells  usually  present,  scattered,  tending  to  be 
more  abundant  in  the  wings,  rarely  more  than  six  or  eight  in  a 
cross-section ;  ventral  scales  in  two  distinct  rows,  the  laminar 
scales  alternating  with  the  median  scales  and  not  much  nearer 
the  margin ;  appendages  of  the  median  scales  very  variable, 
lanceolate  to  broadly  ovate,  when  well  developed  mostly  0.45- 
0.65  mm.  long  and  0.3-0.4  mm.  wude  but  sometimes  considerably 
smaller,  apex  acuminate,  acute  or  apiculate,  margin  entire  or 
variously  and  irregularly  toothed,  the  teeth'  rarely  numerous  and 
often  restricted  to  the  basal  portion,  cells  of  about  the  same  size 
throughout  or  showing  a  slight  and  gradual  decrease  in  size 
toward  the  margin,  median  cells  usually  longer  than  broad, 
mostly  60-90/*  long  and  25-40/x  wide,  marginal  cells  mostly  30-70jtt 


292  Alexander  W.  Evans, 

long  and  20-35/*  wide,  irregular,  the  long  axis  varying  from 
parallel  to  perpendicular  to  the  margin,  cells  containing  oil- 
bodies  usually  absent,  rarely  one  or  two  present :  male  receptacle 
borne  on  a  stalk  mostly  1-2  cm.  long,  with  two  rhizoid-furrows 
and  no  air-chambers,  the  disc  mostly  0.8-1  cm.  broad,  deeply  lobed, 
the  lobes  or  rays  mostly  four,  rarely  five  or  six,  palmately  dis- 
posed, the  basal  sinus  sometimes  more  than  180  degrees  broad, 
the  lobes  mostly  5-7  mm.  long  and  1.5-2  mm.  wide,  rounded,  with 
a  thin  wavy  margin  extending  across  the  basal  sinus,  ventral 
scales  imbricated,  mostly  in  two  rows :  female  receptacle  borne 
on  a  stalk  2-4  cm.  high,  with  two  rhizoid-furrows  and  two  narrow 
dorsal  bands  of  air-cliambers,  the  disc  convex,  mostly  6-8  mm. 
wide,  shortly  five-lobed,  the  lobes  or  rays  convex,  rounded  and 
separated  by  shallow  sinuses,  the  basal  sinus  much  broader  than 
the  others  and  approximating  a  straight  line ;  involucre  firm,  the 
margin  sparingly  dentate  to  closely  ciliate  or  laciniate,  the  teeth  or 
cilia  varying  from  one  to  five  cells  in  length,  sometimes  forking ; 
spores  brownish  yellow,  about  2611  in  diameter,  narrowly  mar- 
gined, the  oviter  face  bearing  a  few  low  ridges  not  forming  a 
network ;  elaters  about  6/a  wide,  bispiral :  cupules  closely  short- 
ciliate,  the  cilia  mostly  two  to  four  cells  long,  outer  surface 
without  papillae.     (Figs.  15-20.) 

A  widely  distributed  species  is  tropical  America.  The  follow- 
ing specimens  have  been  examined : 

Puebla:  banks  of  Avenida  Hidalgo  and  path  to  barranca, 
Tezuitlan,  October,  1908,  Barnes  &  Land  §^/. 

Vera  Cruz  :  Jalapa  and  vicinity,  September,  1906,  Barnes  & 
Land,  no  number  (Y.)  ;  July,  1908,  C.  G.  Pringle  15326  (Y., 
distributed  in  PI.  Mex.  under  a  manuscript  name  of  Stephani)  ; 
November,  1908,  Barnes  &  Land  556,  614,  626a  (Y.)  ;  vicinity 
of  Orizaba,  November,  1908,  Barnes  &  Land  631,  6/0  (Y.). 

Guatemala:  without  definite  locality  or  date,  Godman  & 
Sahin  (N.  Y.)  ;  Santa  Rosa,  September,  1894,  Heyde  &  Lux 
62()3  (N.  Y.)  ;  near  the  Finca  Sepacuite,  Alta  Verapaz,  March 
and  April,  1902,  Cook  &  Griggs  83,  141,  3P4  (U.  S.,  Y.)  ;  trail 
from  Pangos  to  Sepacuite,  Alta  Verapaz,  January,  1908,  Maxon 
&  Hay  31 II  (U.  S.,  Y.)  ;  Coban,  Alta  Verapaz,  H.  von  Tuerck- 
heim  60/4  (N.  Y.). 

Costa  Rica  :  Bagnar,  Angostura,  June,  1874,  O.  Kuntse  2102 
(N.  Y.)  ;  la  Verbena,  Alajuelita,  August,  1894,  A.  Tondiiz 
15562  (N.  Y.,  Y.,  distributed  by  E.  Levier  under  a  manuscript 
name  of  Stephani)  ;  Rio  Turrialba,  March,  1896,  /.  D.  Smith 
(N.  Y.)  ;    Cuesta  de  la  Vieja,  road  to  San  Carbos,  April,  1903, 


American  Species  of  Marchantia.  293 

Cook  &  Doyle  iii  (U.  S,,  Y.)  ;  Juan  Vinas,  April,  1903,  Cook 
&  Doyle  ^01  (U.  S.,  Y.)  ;  vicinity  of  La  Palma,  May,  1906, 
W.  R.  Maxon  489  (U.  S.,  Y.). 

Panama:  without  definite  locality  or  date,  B.  Seemann  (N. 
Y.)  ;  Darien,  April  and  June,  1908,  R.  S.  Williams  108 j,  1084 
(N.  Y.,  Y.). 

Cuba:  without  definite  localities  or  dates,  C.  Wright  (dis- 
tributed in  Hep.  Cubenses). 

Jamaica:  without  definite  localities  or  dates,  Fordyce,  W. 
Wright;  Whitfield  Hall,  December,  1896,  W.  Harris  1106^  in  part 
(N.  Y.)  ;  Moody's  Gap,  March,  1895,  W.  Harris  5671  (N.  Y., 
U.  S.,  Y.)  ;  vicinity  of  Cinchona,  November,  1902,  F.  S.  Earle 
3pya  (N.  Y.,  Y.)  ;  July,  1903,  A.  W.  Evans  248  (Y.)  ;  Mount 
Airy,  trail  to  Tweedside,  April,  1903,  W.  R.  Maxon  864  (U.  S., 
Y.)  ;  Second  Breakfast  Spring,  near  Tweedside,  April,  1903, 
W.  R.  Maxon  880  (U.  S.,  Y.)  ;  Morce's  Gap,  August,  1906, 
A.  W.  Evans  405  (Y.)  ;  Ctma  Cuna  Gap,  September,  1908, 
E.  G.  Britton  ggo  (N.  Y.,  Y.)  ;  March,  1909,  Britton  &  Howe 
4032  (N.  Y.,  Y.). 

Porto  Rico:  Adjuntas,  March,  1886,  P.  Sintenis  51  (N.  Y., 
U.  S.,  Y.,  Hsted  by  Stephani  in  Hedwigia  27  :  294.  1888)  ;  road 
from  Ponce  to  Adjuntas,  July,  1901,  Underwood  &  Griggs  'J32 
(N.  Y.,  U.  S.,  Y.)  ;  Military  Road,  north  of  Cayey,  June,  1901, 
Underwood  &  Griggs  2y8  (U.  S.,  Y.)  ;  road  from  Utuado  to 
Arecibo,  July,  1901,  Underwood  &  Griggs  839  (U.  S.,  Y.)  ; 
near  Cayey,  July,  i9CH3,  A.  W.  Evans  p5  (Y.)  ;  between  Ponce 
and  Utuado,  March,  1906,  Britton  &  Marble  778  (N.  Y.,  Y.)  ; 
Mount  Morales,  near  Utuado,  March,  1906,  M.  A.  Hozve  iop8 
(C.  C.  H.,  N.  Y.,  Y.)  ;  summit  of  Loma  la  Mina,  Sierra  de 
Naguabo,  July,  1914,  /.  A.  Shafer  3337  (N.  Y.,  Y.)  ;  La  Juanita, 
near  Las  Marias,  February,  1915,  E.  G.  Britton  3964  (N.  Y.,  Y.)  ; 
La  Chiquita,  near  Maricao,  February,  191 5,  E.  G.  Britton  4099 
(N.  Y.,  Y.)  ;  Britton  &  Cowell  4296  (N.  Y.,  Y.)  ;  between 
Adjuntas  and  Ponce,  March,  1915,  E.  G.  Britton  5367  (N.  Y., 
Y.)  :  Maricao,  July,  1915,  F.  L.  Stevens  1844  (N.  Y.,  Y.)  ; 
Gigante,  July,  1915,  F.  L.  Stevens  1797  (N.  Y.,  Y.). 

Guadeloupe:  without  definite  locality,  1874,  T.  Husnot  (dis- 
tributed in  PI.  des  Antilles  196)  ;  Gombeyre,  1897-1900,  Pcre 
Duss  391  (N.  Y.)  ;  Basse  Terre,  1898  Pere  Duss  233  (N.  Y., 
determined  as  M.  brasiliensis  by  Stephani). 


294 


Alexander  W.  Evans, 


Dominica:  Laudat,  1903,  F.  E.  Lloyd  yd,  78,  285  (N.  Y.,  Y.). 
Martinique:    St.  Pierre,  1899-1900,  Pcre  Duss  spo  (N.  Y., 


Fig.  15.     Marchantia  chenopoda  L. 

Appendages  of  ventral  scales,  x  100 ;  K  represents  the  basal  portion 
of  an  appendage.  A.  Vera  Cruz,  1906,  Barnes  &  Land.  B-G.  Vera 
Cruz,  Barnes  &  Land  614.  H,  I.  Vera  Cruz,  Barnes  &  Land  626a.  J,  K. 
Guatemala,  Cook  &  Griggs  83. 


determined  as  M.  brasiliensis  by  Stephani)  ;  Carbet,  1899,  Pere 
Duss  342  his  (N.  Y.)  ;  Morne  Rouge,  August,  1901,  Pcre  Duss 
581  (N.  Y.). 


American  Species  of  Marchantia.  2,95 

Grenada:   Grand  Etang,  1913,  R.  Thaxter  (H.,  Y.). 

Venezuela:  near  Caripe,  A.  von  Humboldt  (Y.,  specimen 
from  the  Hooker  herbarium,  labeled  "Humboldt  2_^j,"  pre- 
sumably the  basis  for  the  record  in  Nov.  Gen.  Sp.  Plant.  7 :  99. 
1825)  ;  without  definite  locality  or  date,  Funck  &  Schlim  557 
(N.  Y.). 

Colombia:  Andes  Bogotenses,  W.  Weir  (N.  Y.). 

Brazil:  Orgaos  Mountains,  C.  Gaudichaud  (N.  Y.,  specimen 
from  the  Montague  herbarium,  labeled  simply  "Brasilia,"  pre- 
sumably the  basis  for  the  record  in  Voy.  Corv.  la  Bonite,  Bot. 
1:209.  1844-46);  Morro  Velho,  no  date,  G.  Gardner  i^i  (N. 
Y.)  ;  Rio  de  Janeiro,  no  date,  /.  Milne  (N.  Y.)  ;  A.  Gldciou  17992 
(N.  Y.,  listed  by  Spruce  in  Rev.  Bryol.  20  :  60.  1893)  ;  Jacobina, 
Mattogrosso,  October,  1872,  O.  Kuntse  (N.  Y.,  sterile  and  some- 
what doubtful)  ;   near  Sao  Paulo,  April,  1905,  A.  Usteri  i  (Y.). 

Ecuador:  Baiios,  R.  Spruce  (listed  in  Trans.  Bot.  Soc.  Edin- 
burgh 15:562.     1885,  and  distributed  in  Hep.  Spruceanae). 

Peru:  near  Tarapoto,  R.  Spruce  (distributed  in  Hep. 
Spruceanae)  ;  Ollantavtambo,  May,  1915,  Cook  &  Gilbert  /^^ 
(U.  S.,Y.). 

Bolivia:  near  Irupana,  A.  d'Orbigny  226  (M.,  type  of 
Grimaldia  peruviana)  ;  Yungas,  1885,  H.  H.  Rusby,  3001,  3002, 
3003,  3004  (N.  Y.,  U.  S.,  listed  by  Spruce  in  Mem.  Torrey  Club 
1:140.  1890);  1892,  M.  Bang  1545  (N.  Y.)  ;  July,  1893,  P. 
Jay  /I  (N.  Y.,  Y.)  ;  Tumupasa,  December,  1901,  R.  S.  Wil- 
liams 2143  (N.  Y,,  Y.). 

Galapagos  Islands:  Albemarle  Island,  August,  1906,  A. 
Stezvart  68/6. 

The  following  additional  stations,  recorded  in  the  literature, 
are  likewise  of  interest : 

Oaxaca:  Mirador  and  Comaltepec,  F.  Liebmann  (listed  by 
Gottsche  in  Mex.  Leverm.  268.     1863). 

Vera  Cruz  :  near  Orizaba  and  at  Cordoba,  1855,  F.  Midler 
(listed  by  Gottsche,  /.  c). 

Costa  Rica  :  near  San  Jose,  H.  Pittier  6004,  6049;  Marais  de 
la  Palma,  H.  Pittier  6018,  6024  (both  listed  by  Stephani  in  Bull. 
Soc.  Bot.  Belgique  31 :  180.     1892). 

Jamaica:  without  definite  locality  or  date,  P.  Collinson  (type 
of  M.  Dillenii). 


296  Alexander  W.  Evans, 

Martinique:  Morne  de  la  Calabasse,  without  date  or  col- 
lector's name  (type,  cited  by  Plumier)  ;  without  definite  locality, 
date,  or  collector's  name  (t3^e  of  M.  cartilaginea)  ;  without 
definite  locality  or  date,  Hahn  134/;  T.  Hiisnot  igy,  ip8  (the 
last  three  listed  by  Bescherelle  in  Jour,  de  Bot.  7:193.     1893). 

French  Guiana:  near  Cayenne,  1835-49,  Leprieur  1^86 
(listed  by  Montagne  in  Ann.  Sci.  Nat.  Bot.  IV.  3:320.     1855). 

Venezuela:  Merida,  K.  Goebel  (figured  in  Organographie 
der  Pflanzen  258.  /.  13/.    1898). 

Colombia:  Muzo,  Fusagasuga  and  Puripi,  1859,  A.  Lindig 
lyiS,  1718,  1719,  1722,  172^  (listed  by  Gottsche  in  Ann.  Sci. 
Nat.  Bot.  V.  1 :  186.     1864). 

Peru  :  Rio  Huallaga,  November,  1902,  £.  Vie  527  (Hsted  by 
Stephani  in  Hedwigia  44:223.     1905). 

Brazil:  ''Montagne  d'Estrella,"  G.  Raddi  (cited  by  Raddi, 
see  below)  ;  without  definite  locality  or  date,  F.  Sellow  (type  of 
M.  brasiliensis). 

The  specimens  recorded  by  Schiffner  from  tlie  Fiji  Islands 
(Leberm.  Forschungsr.  S.  M.  S.  "Gazelle"  43.  1890)  are 
described  as  having  ciliate-dentate  ventral  scales  and  would 
probably  now  be  referred  to  some  other  species. 

The  interpretation  of  ilf.  chenopoda  is  beset  with  difficulties, 
and  a  history  of  the  species  may  therefore  be  in  place.  The 
Linnaean  description  or  diagnosis  is  very  short  and  reads, 
"Marchantia  calyce  communi  dimidiato  palmato  quadrifido."  If 
the  term  "calyx"  signifies  the  female  receptacle  this  descrip- 
tion would  not  apply  accurately  to  any  of  the  known  American 
species,  where  a  four-parted  receptacle  occurs  only  as  an  abnor- 
mality. If  the  term  signifies  the  male  receptacle  there  are  sev- 
eral species  to  which  the  description  might  perhaps  apply.  In 
any  case  it  would  be  quite  impossible  to  identify  a  definite  species 
by  means  of  the  Linnaean  description  alone. 

Unfortunately  the  only  synonym  which  Linnaeus  quotes,  the 
"Lichen  anapodocarpos"  of  Plumier,  is  likewise  insufficient  to 
lead  to  a  positive  conclusion.  Plumier^®  described  his  plant  from 
material  collected  on  the  Morne  de  la  Calabasse  in  Martinique. 
Linnaeus  cites  the  original  description  and  figure  and  also  the 

■*  Traite  des  Fougeres  de  I'Amer.  143.  pi.  142.    Paris,  1705. 


American  Species  of  Marchantia.  297 

later  description  and  figure  published  by  Dillenius.**'  Plumier's 
figure  represents  certain  reproductive  parts  in  detail  and  a 
thallus  with  a  sinuate  or  vaguely  lobed  margin,  branching  occa- 
sionally by  forking  and  apparently  also  by  ventral  outgrowths. 
From  the  tips  of  some  of  the  branches  the  four-lobed  receptacles 
on  short  stalks  take  their  origin  and  clearly  bring  the  growth 
of  the  branches  to  an  end.  In  two  cases — possibly  in  three — five- 
lobed  receptacles  are  shown.  In  his  text  Plumier  compares  the 
appearance  of  the  plant  with  that  of  the  Indian  fig  and  says  that 
the  upper  surface  is  of  a  pale  green  color  and  roughened  by 
minute  elevated  points.  He  compares  the  entire  receptacle  with 
a  mushroom  and  states  that  one  side  of  the  disc  is  rounded, 
while  the  other  shows  four  semicircular  lobes,  the  whole  resem- 
bling an  inverted  goose  foot.  He  adds  that  each  lobe  opens 
longitudinally,  and  shows  minute  white  "flowers"  in  the  form 
of  tubes.  Each  tube  divides  at  the  apex  after  a  while  into 
four  parts  which  roll  back  and  disclose  an  oval  fruit  filled  with 
"seeds"  like  flour.  It  is  clear  from  this  account  that  he  had 
female  receptacles  before  him  and  that  he  saw  the  involucre, 
the  pseudoperianth,  the  capsule,  and  the  spores. 

Dillenius  took  his  figure  directly  from  Plumier  and  did  not 
know  the  plant  itself.  He  tried  to  improve  the  figure,  however, 
by  indicating  that  the  upper  surface  of  the  thallus  was  covered 
over  with  minute  polygons  as  in  related  species.  Lindberg" 
criticises  the  figure  of  Dillenius  (and  consec^uently  that  of 
Plumier)  by  stating  that  an  autoicous  inflorescence  is  shown, 
both  male  and  female  receptacles  being  represented  on  the 
thallus.  This  criticism  is  undeserved.  The  receptacles  shown 
are  all  female,  the  dorsal  surface  being  represented  in  some 
cases  and  the  ventral  in  others.  In  his  text  Dillenius  brought 
out  the  fact  that  the  receptacles  were  all  the  same  kind,  although 
he  incorrectly  interpreted  the  fruit  of  Plumier  as  an  anther  and 
the  flour-like  seeds  as  pollen,  a  well-known  error  which  he 
repeats  in  his  interpretation  of  the  reproductive  parts  in  other 
bryophytes. 

On  the  basis  of  Plumier's  description  and  figures  it  becomes 
evident  that  the  term   "calyx"    in  the  Linnaean  diagnosis  of  M. 

^'Hist.  Muse.  531.  pi.  75,  f.  5.     Oxford,  1741. 

*"  Krit.  Gransk.  Dillen.  Hist.  Muse.  45.    Helsingfors,  1883. 

Trans.  Conn.  Acad.,  Vol.  XXI  20  1917 


298 


Alexander  W.  Evans, 


chenopoda  refers  to  the  female  receptacle,  and  it  has  already 
been  pointed  out  that  a  quadrifid  female  receptacle  is  not  found 


Fig.  16.     Marchantia  chenopoda  L. 

Appendages  of  ventral  scales,  x  100;  G  represents  the  median  portion 
of  an  appendage  with  two  cells  containing  oil-bodies.  A-C.  Cuba,  C. 
Wright,  in  Hep.  Cubenses.  D,  E.  Jamaica,  A.  JV.  Evans  258.  F,  G. 
Jamaica,  W.  R.  Maxon  405.  H,  I.  Porto  Rico,  F.  L.  Stevens  1844;  J. 
Guadeloupe,  T.  Husnot,  in  Pi.  des  Antilles  796. 

normally  in  any  known  American  species.  Plumier's  work  indi- 
cates further,  that  the  involucres  and  sporophytes  of  his  plant 
are  situated  underneath  the  lobes,  a  condition  which  is  likewise 


American  Species  of  Marchantia.  299 

unknown  among  American  species,  where  tlie  involucres  and 
sporophytes  invariably  alternate  with  the  lobes.  Since  there  are 
apparently  no  specimens  of  Plumier's  plant  in  herbaria,  it  is 
clear  that  M.  chenopoda  L.  represents  an  unidentifiable  plant, 
and  the  logical  course  would  be  to  give  up  the  species  altogether. 

In  the  literature  of  the  Hepaticae,  however,  M.  chenopoda 
has  an  established  place,  and  it  seems  justifiable  to  interpret  it 
according  to  the  descriptions  of  later  waiters.  Even  Plumier's 
figure  gives  us  a  little  help  because  it  shows  that  he  occasionally 
observed  a  five-lobed  receptacle,  although  he  makes  no  mention 
of  such  a  structure  in  his  text.  Since  most  subsequent  vv^riters 
ascribe  to  the  species  definitely  a  five-lobed  receptacle,  and  since 
the  species  to  which  they  assign  the  name  is  abundant  in 
Martinique,  it  is  quite  probable  that  their  M.  chenopoda  is  the 
same  as  Plumier's  plant.  Unfortunately  their  descriptions  and 
figures  are  not  without  discrepancies,  and  it  becomes  evident  that 
Taylor  at  least  did  not  distinguish  between  what  is  here  called 
M.  chenopoda  and  M.  domingensis. 

Apparently  Swartz*-  was  the  first  to  describe  the  male  recep- 
tacles. He  states  that  they  are  subpeltate,  unsymmetrical,  pal- 
mate-quadrifid,  plane  and  verruculose  above  (like  the  thallus), 
and  convex  below,  the  rays  or  lobes  being  linear,  obtuse,  and 
often  unequal,  with  membranous,  undulate  margins.  He  cites  no 
stations  for  the  species  although  he  implies  that  it  occurs  in 
Jamaica.  Ouadrifid  receptacles  are  sometimes  found  in  M. 
domingensis,  but  it  is  probable  that  Swartz  had  the  true  M.  cheno- 
poda before  him,  and  his  description  is  definitely  cited  in  the 
Synopsis  Hepaticarum. 

Schwaegrichen,^^  in  18 14,  quotes  M.  chenopoda  from  Africa 
as  well  as  from  America,  and  F.  Weber,**  the  following  year, 
notes  a  similar  extension  of  range.  Neither  writer  adds  anything 
significant  to  our  knowledge  of  the  species,  and  it  is  probable 
that  their  citation  of  African  stations  is  based  on  incorrect  deter- 
minations, since  all  subsequent  writers  restrict  the  range  of 
AI.  chenopoda  to  America. 

A  few  years  later  Raddi*^  extended  the  known  range  of  the 

"  Fl.  Ind.  Occid.  1880.     Erlangen,  1806. 

^  Hist.  Muse.  Hepat.  Prodr.  32.     Leipzig,  1814. 

*^Hist.  Muse.  Hepat.  Prodr.  103.    Kiel,  1815. 

*°  Mem.  Soe.  Ital.  Modena  19  :  44.     1823 ;    20 :  pi.  6a,  f.  i,  2.     1829. 

Trans.  Conn.  Acad.,  Vol.  XXI  21  1917 


300  Alexander  IV.  Evans, 

species  into  Brazil  and  stated  that  it  occurred  abundantly  at  the 
bottom  of  moist  and  mossy  rocks  on  the  "Montagne  d'Estrella." 
According  to  his  account  M.  chenopoda  is  distinguished  from 
all  the  other  species  of  Marchantia  by  its  receptacles,  which  are 
truncate  on  one  side.  He  adds  that  in  the  male  receptacle  the 
upper  surface  is  plane  and  that  the  four  parts  or  lobes  are  unequal 
in  length,  and  he  criticises  Plumier  for  comparing  this  receptacle 
with  a  goose's  foot;  in  his  opinion  it  is  more  like  the  foot  of  a 
pigeon.  Of  course  this  criticism  has  no  weight,  since  Plumier 
drew  his  account  entirely  from  female  receptacles.  According 
to  Raddi  the  disc  of  the  female  receptacle  is  strongly  convex 
and  either  entire  or  very  shortly  divided,  bearing  on  the  lower 
surface  four  fleshy  or  rib-like  swellings,  between  which  are  borne 
the  capsules,  much  as  in  M.  polymorpha.  He  notes  further  that 
the  upper  surface  of  the  thallus  is  areolate  and  perforated  by 
white  vesicles  and  that  the  lower  surface  is  violet  except  along 
the  margin,  where  it  is  green.  Although  Raddi's  specimens  have 
not  been  available  for  study  it  is  evident  that  he  had  the  true 
M.  chenopoda  before  him.  Not  only  is  his  description  unusually 
clear,  but  the  species  has  since  been  collected  in  other  Brazilian 
localities. 

In  1835  Taylor  published  an  account  of  the  Marchantiaceae 
which  had  come  under  his  observation.  In  his  description  of 
M.  chenopoda'^^  he  comments  on  the  inaccuracies  of  Plumier's 
figures  and  quotes  them  doubtfully,  although  it  was  upon 
Plumier's  work  that  the  species  was  primarily  based.  Accord- 
ing to  Taylor  the  female  receptacle  is  hemispherical  and  divided 
into  from  eight  to  ten  truncate  laciniae,  each  bearing  underneath 
a  single  involucre  with  ciliate  or  serrulate  margins.  He  notes 
further  that  the  stalk  of  the  receptacle  has  two  rhizoid-furrows 
and  adds  interesting  statements  about  the  scales  on  the  vegetative 
thallus  and  about  the  cupules.  The  scales,  in  his  words,  have  an 
entire  and  broadly  ovate  base,  then  a  deep  constriction  at  about 
the  middle,  and  then  a  broadly  ovate  and  ciliate  expansion  (the 
latter  being  what  is  now  known  as  the  appendage).  In  the 
cupules  he  speaks  particularly  of  the  serrate  margin.  It  will 
be  seen  at  once  that  Taylor's  account  of  the  female  receptacle 
is  very  different  from  that  of  his  predecessors,  and  the  speci- 
mens in  his  herbarium  show  that  it  was  drawn  from  M.  domin- 

«  Trans.  Linn.  Soc.  17 :  379-  pl-  ^2,  f.  2.     1835. 


American  Species  of  Metre Jiantia. 


301 


gensis.     His  account  of  the  male  receptacle,  however,  agrees 
esentially  with  that  of  Swartz.    Taylor  cites  Ad.  chenopoda  from 


Fig.  17.    Marchantia  chenopoda  L. 

Appendages  of  ventral  scales,  x  100.  A,  B.  Venezuela,  Funck  &  Schlim. 
C-E.  Brazil,  A.  Glasiou  17992.  F,  G.  Brazil,  1905,  A.  Usteri  i.  H-J. 
Galapagos  Islands,  A.  Stewart  6876. 

Guadeloupe    and    St.    Vincent   as    well    as    from   Jamaica   and 
Martinique. 


302  Alexander  W.  Evans, 

In  the  Synopsis  Hepaticarum  emphasis  is  again  laid  on  the 
receptacles.  The  female  receptacle  is  said  to  be  excentric,  hemi- 
spherical, and  about  five-lobed,  the  lobes  being  obtuse,  subcrenate 
and  soon  obsolete,  with  hyaline,  denticulate  involucres.  The  male 
receptacles  are  said  to  be  unsymmetrical  and  palmately  four-  to 
five-parted.  This  account  of  the  female  receptacle  agrees  closely 
with  that  of  Raddi  and  differs  from  that  of  Taylor.  Both  writers, 
however,  are  cited  under  the  species.  From  Brazil  several 
localities  are  enumerated,  but  the  only  West  Indian  stations  given 
without  question  are  on  the  island  of  Martinique.  The  Synopsis 
quotes  three  synonyms:  M.  androgyna  (female  plants  only), 
M.  Swartzii,  and  Chlamidium  indicum.  These  may  be  considered 
in  order. 

Swartz  apparently  supposed  that  what  he  called  M.  androgyna 
was  the  same  as  M.  androgyna  L.,*'^  a  species  based  on  two  cita- 
tions, the  first  from  Dillenius*^  and  the  second  from  Micheli.** 
According  to  Swartz,  who  treats  the  plant  very  briefly,  M. 
androgyna  is  related  to  Rehoulia  hemisphaerica  (L.)  Raddi;  he 
describes  the  thallus  as  narrower  than  in  that  species  and  states 
that  the  male  receptacles  are  perhaps  sessile  and  that  the  female 
receptacles  are  subentire.  Although  he  cites  no  actual  material 
it  is  probable  that  he  drew  his  description  from  Jamaican  speci- 
mens collected  by  himself,  these  being  definitely  referred  to  by 
later  writers. 

Now  the  Dillenian  species  quoted  by  Linnaeus  under  M. 
androgyna  has  been  the  cause  of  a  great  deal  of  confusion.  It 
was  based  on  two  entirely  different  plants,  a  fact  which  was  first 
pointed  out  by  Lehmann  and  Lindenberg  in  their  discussion  of 
the  Asiatic  M.  linearis  Lehm.  &  Lindenb.^°  They  show  clearly 
that  the  Dillenian  /.  ^B,  which,  as  they  state,  is  essentially  the 
same  as  the  figure  by  Micheli,  represents  Grimaldia  dichotoma 
Raddi,  a  common  species  of  the  Mediterranean  region.  They 
show  further  that  the  Dillenian  /.  ^A  and  /.  jC  represent  a  species 
of  Marchantia,  and  they  suppose  that  this  species  is  the  same 
as  the  M.  androgyna  of  Swartz.  The  two  figures  in  question 
were  drawn  from  specimens  collected  in  Jamaica  by  P.  Collinson ; 

*^Sp.  Plant.  1138.     1753. 

^  Hist.  Muse.  520.  pi.  75,  f.  3.     1741. 

^  Gen.  Nov.  Plant.  3.  pi.  2,  f.  3.     1729. 

■^Lehmann,  Pug.  Plant.  4:9.    Hamburg,  1832. 


American  Species  of  Alarchantia. 


303 


/.  ^A  shows  a  plant  with  numerous  cupules  and  female  recep- 
tacles, while  /.  jC  shows  a  small  forking  fragment  with  cupules 


Fig.  18.    Marchantia  chenopoda 


Appendages  of  ventral  scales,  x  100.  A-C.  Peru,  Cook  &  Gilbert  755. 
D-F.  Bolivia,  A.  d'Orbigny,  type  of  M.  peruviana.  G,  H.  Bolivia,  H.  H. 
Rusby  3004.    I,  J.    Bolivia,  P.  Jay  71. 

only.  The  receptacles  are  so  strongly  convex  that  they  appear 
conical  and  resemble  those  of  Conocephalum  conicum  (L.) 
Dumort.     In  fact,  according  to  Lehmann  and  Lindenberg,  the 


304  Alexander  W.  Evans, 

M.  androgyna  of  Weber^^  is  actually  Conocephalum  conicum,  and 
the  Linnaean  name  has  been  applied  by  other  writers  to  such 
distinct  species  as  Preissia  quadrata  (Scop.)  Corda  and  Reboulia 
hemisphaerica  (L.)  Raddi. 

Lehmann  and  Lindenberg's  conception  of  M.  androgyna  Sw. 
was  based  on  specimens  collected  by  Swartz  in  Jamaica. 
Although  they  considered  these  specimens  identical  with  those 
collected  by  Collinson  they  did  not  take  up  the  name  M. 
androgyna  for  the  species,  probably  because  the  original  M. 
androgyna  L.  was  an  aggregate.  They  described  it  instead  under 
the  new  name  M.  Swartzii.  The  female  receptacle,  according 
to  their  account,  is  unsymmetrical,  hemispherical,  and  subentire 
or  obsoletely  lobed,  the  lower  surface  and  the  stalk  being  villous. 
They  state  further  that  the  upper  surface  of  the  thallus  is  green 
with  many  large  pores  bordered  with  white,  and  that  the  lower 
surface  is  brown  with  scales  in  the  median  portion;  and  they 
suggest  that  the  male  receptacles  of  Swartz's  description  may 
have  been  cupules  only.  So  far  as  the  descriptions  go  M. 
Swartzii  and,  consequently,  M.  androgyna  Sw.  do  not  differ  in 
any  essential  respects  from  M.  chenopoda,  and  the  authors  of  the 
Synopsis  are  probably  correct  in  citing  these  two  species  as 
synonyms  of  M.  chenopoda.  This  view  is  supported  by  a  frag- 
mentary specimen  in  the  Taylor  herbarium,  labeled  M.  Sxvartzii 
by  Lehmann,  which  apparently  represents  M.  chenopoda,  although 
a  positive  conclusion  can  hardly  be  reached  without  sectioning 
the  material. 

A  further  difficulty  in  disentangling  the  synonymy  is,  how- 
ever, encountered.  Although  Lehmann  and  Lindenberg  con- 
sidered Swartz's  and  Collinson's  plants  identical,  this  opinion 
was  not  shared  by  the  authors  of  the  Synopsis  Hepaticarum. 
In  quoting  M.  Szvartzii  as  a  synonym  of  M.  chenopoda  they 
take  pains  to  exclude  the  Dillenian  /.  j  altogether,  although 
/.  ^A  and  /.  jC  are  definitely  quoted  by  the  authors  of  M. 
Swartzii  in  citing  M.  androgyna  Sw.  as  a  synonym  of  their  spe- 
cies. Fortunately  Collinson's  material  is  preserved  in  the  Dil- 
lenian herbarium  and  throws  a  little  light  on  the  subject.  It 
was  studied  by  Lindberg,  who  reached  the  conclusion  that  it 
represented  a  distinct  and  undescribed  species.  This  he  pro- 
posed as  new  under  the  name  M.  Dillenii  Lindb,    He  assigns  to 

"  Spic.  Fl.  Goettingen.  168.     Gotha,  1778. 


American  Species  of  Marchantia. 


305 


the  species  a  delicate  pellucid  thallus  with  indistinct  areolae  but 
with  large  pores,  the  thallus  in  M.  chenopoda  being  thick  and 


Fig.  19.     Marchantia  chenopoda  L. 

Epidermal  structures  of  thallus,  x  225.  A,  B.  Pores,  surface  view. 
C-G.  Pores,  cross-section.  H-K.  Pores,  inner  view.  L.  Slime  cell, 
cross-section.  A,  C,  H-J.  Guatemala,  Cook  &  Griggs  83.  B,  D,  E,  K,  L. 
Jamaica,  W.  R.  Maxon  405.    F,  G.    Peru,  Cook  &  Gilbert  755. 

opaque  with  distinct  areolae  and  small  pores.  He  states  further 
that  the  female  receptacles  are  depressed-semiglobose,  excentric 
and  almost  entire,  the  five  lobes  present  being  very  short,  thick, 


3o6  Alexander  W.  Evans, 

semicircular  in  outline  and  slightly  incurved.  The  receptacles 
described  are  immature  and  do  not  therefore  yield  very  satis- 
factory characters,  but  Lindberg's  description,  so  far  as  it  goes, 
would  clearly  apply  to  M.  chenopoda.  Even  the  characters 
drawn  from  the  thallus  easily  come  within  the  range  of  variation 
to  be  expected  in  so  multiform  a  species,  where  both  the  texture 
and  the  size  of  the  pores  differ  widely  in  different  plants.  The 
writer  would  therefore  follow  Stephani  in  reducing  M.  Dillenii 
to  synonymy,  even  in  the  absence  of  Lindberg's  type  material. 

The  third  synonym  given  in  the  Synopsis,  Chlamidium  indicum, 
is  nothing  but  a  nomen  nudum.  According  to  Corda  it  was  based 
on  No.  375  of  Sieber's  Flora  Martinicensis.  The  Synopsis,  how- 
ever, in  citing  it  as  a  synonym  under  M.  chenopoda,  states  that 
it  was  based  on  No.  378.  In  the  Mitten  herbarium  a  specimen 
of  No.  378  is  preserved  under  the  name  M.  martinicensis.  This 
plant,  which  probably  represents  the  type  of  the  manuscript  spe- 
cies M.  martinicensis  Spreng.,  is  clearly  referable  to  M.  domin- 
gensis,  as  the  authors  of  the  Synopsis  have  already  shown. 
Their  citation  of  No.  378  under  Chlamidium  indicum,  therefore, 
was  probably  an  error  or  due  to  the  fact  that  this  number  was 
a  mixture ;  in  any  case  Corda's  species,  in  the  absence  of  adequate 
publication,  deserves  no  further  attention. 

If  the  work  of  Taylor  is  excepted  it  will  be  seen  that  writers 
up  to  the  time  of  the  Synopsis  Hepaticarum  (1847)  were  prac- 
tically unanimous  in  assigning  to  M.  chenopoda  a  subentire  or 
shortly  five-lobed  female  receptacle  and  a  deeply  four-cleft  male 
receptacle.  The  same  thing  may  be  said  of  subsequent  writers. 
Unfortunately  identical  or  similar  characters  have  been  assigned 
to  other  species.  Aside  from  M.  Dillenii,  which  has  already 
been  alluded  to,  M.  cartilaginea,  M.  brasiliensis,  and  M.  peruvi- 
ana may  be  mentioned  in  this  connection.  The  first  was  based 
on  material  collected  on  the  island  of  St.  Vincent,  no  collector 
being  named ;  the  second  on  Brazilian  material  collected  by 
Sellow ;  the  third  on  Bolivian  material  collected  by  D'Orbigny. 

In  M.  cartilaginea  the  male  receptacles  are  said  by  the  authors 
of  the  species  to  be  slightly  convex  and  borne  on  very  short 
stalks,  while  the  female  receptacles  are  said  to  be  minute  and 
entire  or  obsoletely  crenulate.  Schiffner,  who  studied  the  type 
material,  found  that  the  female  receptacles  were  immature  and 
that    the    so-called    male    receptacles    were    nothing    more    than 


American  Species  of  MarcJiantia.  307 

extremely  young  female  receptacles.  The  distinctive  characters 
of  the  species  thus  break  down,  and  he  reduced  it  to  synonymy, 
retaining  it  as  a  var.  cartilaginea  (Lehm.  &  Lindenb.)  Schiffn. 
under  M.  chenopoda.  Stephani  quotes  it  as  a  simple  synonym. 
Their  views  are  supported  by  the  work  of  Prescher,  who  found 
the  distribution  of  the  slime  cells  the  same  in  M.  cartilaginea  as 
in  M.  chenopoda. 

In  M.  brasiliensis  the  male  receptacle  is  described  as  peltate, 
angled  and  convex,  the  central  portion  being  thickened  and  the 
margin  plane  and  hyaline ;  the  female  receptacle  is  said  to  be 
hemispherical,  symmetrical  and  entire.  Here  again  Schiffner 
showed  that  the  receptacles  in  the  type  specimen  were  immature 
and  that  the  distinctive  characters  drawn  from  the  male  recep- 
tacles could  be  duplicated  by  young  male  receptacles  of  M. 
chenopoda.  He  therefore  regards  M.  brasiliensis  as  synonymous 
with  ilf,  chenopoda,  a  view  which  the  writer  is  disposed  to  share. 
Stephani,  in  maintaining  the  validity  of  the  Brazilian  plant,  dwells 
on  the  symmetry  of  the  female  receptacle  and  describes  it  as 
strongly  convex  and  very  shortly  four-  to  six-lobed.  He  adds 
that  the  entire  appendages  of  the  ventral  scales  can  easily  be 
distinguished  from  the  dentate  appendages  of  M.  chenopoda. 
Since,  however,  he  assigns  both  entire  and  toothed  appendages 
to  il/.  chenopoda  in  his  detailed  description  of  that  species,  and 
since  the  receptacles  on  some  of  the  West  Indian  specimens 
referred  by  him  to  M.  brasiliensis  are  distinctly  unsymmetrical, 
his  differential  characters  can  not  be  regarded  as  having  much 
significance. 

In  the  original  description  of  Grimaldia  peruviana  the  female 
receptacle  is  said  to  be  subglobose  and  crenate  while  the  male 
receptacle  is  said  to  be  discoid  and  sessile.  Apparently  on  account 
of  the  characters  of  the  so-called  male  receptacles  Montagne 
continued  to  regard  the  species  as  a  Grimaldia  even  after  the 
authors  of  the  Synopsis  had  correctly  transferred  it  to  Mar- 
chantia.^-  Probably  the  sessile  structures  which  Montagne 
observed  were  immature  female  receptacles,  but  unfortunately 
the  type  specimen  in  his  herbarium,  a  portion  of  which  the  writer 
has  been  able  to  examine,  is  sterile,  so  that  these  problematical 
organs  could  not  be  studied.     The  compound  pores,  however, 

°'  See  Montagne,  Sylloge  91.    Paris,  1856. 


3o8  Alexander  W.  Evans, 

and  the  cupiiles  show  conclusively  that  the  species  is  a  Mar- 
chantia,  and  the  writer  would  go  even  further  than  Stephani  did 
and  reduce  it  to  a  synonym  of  M.  chenopoda.  This  conclusion  is 
strengthened  by  the  fact  that  numerous  specimens  of  M.  cheno- 
poda have  been  collected  in  Bolivia  by  subsequent  explorers. 

Although  a  wide  range  of  variability  is  assigned  to  M.  cheno- 
poda, according  to  the  writer's  conception  of  the  species,  "an 
equally  wide  range  is  assigned  to  M.  domingensis  and  an  even 
wider  range  to  M.  polymorpha.  The  structures  which  are  per- 
haps most  subject  to  variation  are  the  epidermal  pores,  the  slime 
cells,  tlie  appendages  of  the  ventral  scales  and  the  involucres. 
The  male  receptacles  and  the  female  receptacles  in  most  respects 
exhibit  features  of  a  more  constant  character. 

In  normal  and  well-developed  specimens  the  pores  are  unusu- 
ally large  in  the  middle  of  the  thallus  and  are  only  slightly 
smaller  near  the  margin.  In  other  cases  the  contrast  in  size 
between  the  median  and  marginal  pores  is  much  more  marked; 
in  still  other  cases  even  the  median  pores  may  be  small  or  medium 
sized.  Corresponding  with  these  differences  in  size  there  are 
differences  in  the  number  of  cells  in  the  concentric  rows  around 
the  opening,  although  the  number  of  such  rows  is  usually  seven. 
The  differences  in  number  are  found  especially  in  the  third  and 
fourth  rows  of  the  upper  series  and  in  the  third  row  of  the  inner 
series.  In  the  fourth  row  of  the  outer  series  the  variation  is 
especially  great.  In  small  pores  as  few  as  four  cells  may  be 
present,  in  large  pores  as  many  as  eighteen  cells,  and  all  grada- 
tions between  these  extremes  are  to  be  expected.  In  the  third 
row  of  each  series  similar  but  less  marked  differences  are  encoun- 
tered. In  the  first  and  second  row  of  each  series  four  cells  are 
normally  present  although  three,  five,  six,  or  even  seven  cells 
sometimes  occur. 

The  slime  cells  vary  greatly  in  number  and  in  distribution.  In 
typical  West  Indian  material  they  occur  abundantly  in  the  epi- 
dermis, in  the  walls  of  the  air-chambers,  and  in  the  compact 
ventral  tissue  of  the  thallus.  In  other  specimens  they  are  rare  in 
the  epidermis  or  even  absent  altogetlier,  although  still  persistent 
in  the  walls  of  the  air-chamber  and  in  the  compact  tissue ;  in  still 
other  specimens,  and  this  seems  to  be  especially  true  of  mate- 
rial from  Mexico,  Central  America  and  South  America,  they 
are  restricted  to  the  compact  tissue,  where  indeed  they  may  be 


American  Species  of  Marchantia. 


309 


Fig.  20.     Marchantia  chenopoda  L. 

Anatomical  details.  A.  Cells  from  compact  ventral  tissue  in  cross- 
section,  including  two  sclerotic  cells  and  a  cell  containing  oil-bodies,  x  100. 
B.  Cells  from  same  tissue  in  longitudinal  section,  including  a  sclerotic 
cell,  X  100.  C-E.  Cells  from  basal  portions  of  median  scales,  x  100. 
F.  Stalk  of  male  receptacle,  cross-section,  x  50.  G-I.  Stalks  of  female 
receptacles,  cross-sections,  x  50 :  G,  showing  a  stalk  of  average  size ;  H, 
a  slender  stalk  near  the  middle ;  and  I,  the  same  slender  stalk  near  the 
base.  J-M.  Portions  of  involucres :  J,  x  40 ;  L-M,  x  100.  A-G,  J. 
Jamaica,  A.  W.  Evans  405,  W.  R.  Max  on  880.  H,  I,  Panama,  R.  S. 
Williams  1084.  K.  Vera  Cruz,  Barnes  &  Land  631.  L.  Costa  Rica, 
Cook  &  Doyle  301.    M.    Bolivia,  R.  S.  Williams  2143. 


3IO  Alexander  W.  Evans, 

very  scarce.  Since  these  different  conditions  grade  into  one 
another,  it  seems  impossible  to  use  them  as  a  basis  for  the  segre- 
gation of  M.  chenopoda. 

Very  striking  variations  are  to  be  observed  in  the  appendages 
of  the  median  scales.  If  the  series  represented  in  Figs.  15-19 
is  examined  it  will  hardly  seem  possible  at  first  that  all  can  have 
been  taken  from  a  single  species.  The  appendages  shown  exhibit 
four  more  or  less  distinct  types,  varying  in  shape,  in  the  char- 
acter of  the  margin,  and  in  the  size  of  the  component  cells.  In 
the  first  type,  shown  by  Figs.  15,  A-H,  and  16,  the  appendages 
are  narrowly  ovate  to  lanceolate,  tapering  gradually  to  an 
acute  or  acuminate  apex ;  the  margin  is  either  entire  or 
provided  with  one  or  more  vaguely  defined  and  irregular  teeth; 
and  the  cells  are  large,  showing  no  marked  differences  in  size 
between  the  median  and  marginal  portions.  In  the  second  type, 
shown  in  Figs  15,  I,  and  17,  A,  the  appendages  are  larger 
than  in  the  first  type  and  tend  to  be  more  acuminate;  the  mar- 
gin is  more  distinctly  dentate,  although  the  teeth  are  still  irregu- 
lar; and  the  cells  are  much  the  same  as  in  the  first  type,  except 
for  the  fact  that  the  marginal  cells  in  the  basal  portion  tend  to 
be  smaller.  In  the  third  type,  shown  in  Fig.  18,  A-C,  the 
appendages  are  broadly  ovate  and  apiculate  to  abruptly  acute; 
the  margin  is  entire  or  vaguely  and  sparingly  dentate  or  crenate 
toward  the  base;  and  the  cells  are  everywhere  large,  much  as  in 
the  first  type.  In  the  fourth  type,  shown  in  Fig.  18,  D-J,  the 
appendages  have  about  the  same  form  as  in  the  third  type, 
although  they  sometimes  taper  more  gradually ;  but  the  margin 
is  more  irregular,  varying  from  entire  to  distinctly  and  rather 
closely  crenate,  dentate,  or  even  lobed  in  the  basal  portion;  and 
the  cells  are  distinctly  smaller,  often  showing  a  definite  decrease 
in  size  between  the  median  and  marginal  portions.  Cells  con- 
taining oil-bodies  are  usually  absent  altogether,  but  one  or  two 
sometimes  occur,  as  shown  in  Fig.  16,  G,  H,  J.  These  have  not 
been  observed  except  in  the  first  type  of  appendage. 

Since  the  various  types  of  appendage  are  more  or  less  char- 
acteristic of  definite  regions,  the  first  type,  for  example,  being 
prevalent  in  the  West  Indies  and  the  fourth  in  South  America, 
the  writer  at  first  thought  that  distinct  varieties  with  definite 
geographical  ranges  might  be  distinguished,  using  the  appen- 
dicular differences  as  a  basis.    It  soon  became  evident,  however, 


American  Species  of  Marchantia.  311 

that  this  was  hardly  possible.  Many  instances  were  noted  where 
the  range  of  one  type  overlapped  that  of  another,  and  a  few 
cases  were  observed  in  which  appendages  of  two  distinct  types 
occurred  on  an  individual  thallus  (Figs.  15,  H,  I;  17,  A,  B). 
It  was  impossible,  moreover,  to  associate  the  differences  in  the 
appendages  with  other  differences  showing"  any  degree  of  con- 
stancy. In  the  writer's  opinion,  therefore,  the  numerous  types  of 
appendage  are  to  be  regarded  as  a  further  evidence  of  the  great 
variability  of  the  species. 

In  the  case  of  the  involucres  there  is  again  great  variability, 
although  the  extremes  are  perhaps  less  marked  than  in  M.  domin- 
gensis.  Fig.  20,  L,  shows  an  involucre  in  which  the  teeth  are 
scattered,  short,  and  blunt;  while  in  Fig.  20,  J,  K,  M,  the 
involucres  shown  bear  crowded,  long  and  slender  teeth.  It  will 
be  noted  that  some  of  the  teeth  are  simple  while  others  are  more 
or  less  complex.  Bifid  teeth  are  especially  common  and  often 
show  widely  divergent  divisions.  The  involucres  are  firmer  than 
in  M.  domingensis,  the  cell  walls  being  sometimes  distinctly 
thickened  and  pigmented  with  yellowish  brown. 

In  North  America  the  only  species  with  which  M.  chenopoda 
is  likely  to  be  confused  is  M.  domingensis.  The  two  species  are 
of  about  the  same  size,  the  structure  of  the  epidermal  pores  is 
much  the  same  in  both,  the  sclerotic  cells  in  the  ventral  portion 
of  the  thallus  show  a  similar  distribution  and  the  male  recep- 
tacles are  very  much  alike  in  general  appearance.  There  are, 
however,  striking  differences  which  usually  make  it  possible  to 
distinguish  specimens  even  in  the  absence  of  female  receptacles. 
In  M.  chenopoda  slime  cells  can  almost  always  be  observed  in 
the  thallus  and  often  occur  in  great  abundance;  the  appendages 
of  the  ventral  scales  are  often  entire  and  are  never  very  closely 
toothed ;  and  the  stalk  of  the  male  receptacle  is  destitute  of  air- 
chambers.  In  M.  domingensis  there  are  no  slime  cells  in  the 
thallus  ;  the  appendages  of  the  ventral  scales  are  closely  toothed ; 
and  the  stalk  of  the  male  receptacle  bears  a  band  of  air-chambers. 

If  female  receptacles  are  present  other  important  differences 
may  be  observed.  In  M.  chenopoda,  the  stalk  bears  two  bands 
of  air-chambers ;  there  are  normally  only  five  lobes,  these  being 
very  short ;  and  the  involucre  is  firm  in  texture,  the  margin 
varying  from  dentate  to  ciliate  or  laciniate.  In  M.  domingensis 
the  stalk  bears  a  single  band  of  air-chambers,  there  are  usually 


312  Alexander  W.  Evans, 

more  than  five  lobes,  these  being  more  or  less  elongated ;  and 
tlie  involucre  is  very  delicate  in  texture,  the  margin  varying  from 
crenulate  to  short-ciliate. 

The  features  which  distinguish  M.  chenopoda  from  M.  domin- 
gensis  will  distinguish  it  also  from  the  South  American  M. 
papillata.  Another  South  American  species  to  which  it  may  be 
related  is  M.  Bescherellei,  the  appendages  and  involucres  of 
which  might  easily  come  within  the  range  of  variability  exhibited 
by  M.  chenopoda.  According  to  our  present  knowledge,  M. 
Bescherellei  is  a  more  delicate  species  with  lower  air-chambers 
and  thinner  ventral  tissue.  There  is  no  danger  of  confusing  M. 
chenopoda  with  any  of  the  other  species  recognized  in  the  present 
paper. 

Doubtful  Species 

1.  Marchantia  squamosa  Raddi ;   Lehm.  &  Lindenb.  in  Leh- 

mann.  Pug.  Plant.  4:  12.    1832  (as  to  the  Brazilian  plant). 

Brazil  :  without  definite  locality  or  date,  Raddi. 

Attention  has  already  been  called  to  this  species  and  to  its 
possible  aggregate  nature  (see  p.  261).  Stephani's  description 
agrees  in  most  respects  with  M.  paleacea,  and  it  is  possible  that 
Raddi's  specimens  would  now  be  referred  to  that  species.  If 
this  should  prove  true  it  would  mark  an  interesting  extension  of 
range. 

2.  Marchantia  quinqueloba  Nees,  Naturg.  Europ.  Leberm. 

4:98-    1838. 

West  Indies:   without  locality,  date,  or  collector's  name. 

According  to  the  full  description  given  by  Nees  von  Esenbeck 
this  species  is  probably  a  form  of  M.  domingensis.  In  any  case 
there  seems  to  be  no  reason  for  attempting  to  maintain  it,  since 
the  original  specimens  (according  to  Stephani)  are  poorly 
developed  and  valueless. 

3.  Marchantia  pusilla  Nees  &  Mont. ;   G.  L.  &  N.  Syn.  Hep. 

526.    1847. 

Chile  :  without  locaHty,  date,  or  collector's  name. 

This  species  was  based  on  a  single  very  immature  specimen  and 
is  not  represented  in  the  Montagne  herbarium.  The  original 
description  throws  little  light  on  its  affinities,  and  Montagne  him- 


American  Species  of  MarcJiantia.  313 

self  apparently  had  no  faith  in  its  validity  since  he  does  not 
mention  it  in  his  Sylloge  (1856).  Stephani  therefore  seems 
justified  in  repudiating  it  altogether. 

4.  Marchantia  flabellata  Hampe,  Linnaea  20:  235.    1847. 
Venezuela:    Galipan,  without  date,  Moritz  4/b  (erroneously 

ascribed  by  Hampe  to  Colombia). 

According  to  the  brief  original  account  the  species  is  monoi- 
cous,  the  female  receptacles  are  four-parted,  and  the  male  recep- 
tacles seven-parted.  Although  no  specimens  have  been  available 
the  writer  svispects  tliat  M.  flabellata  may  represent  a  synonym 
of  M.  domingensis.  Should  this  be  established  it  would  show 
that  Hampe  confused  tlie  male  and  female  receptacles  and  incor- 
rectly assigned  a  monoicous  inflorescence  to  his  species.  Unfor- 
tunately the  question  must  be  left  in  doubt. 

5.  Marchantia  Notarisii  Lehm.  Pug.  Plant.  10:22.    1857. 
Chile:   near  Valparaiso,  without  date,  W.  Lehmann. 
Although  Stephani  at  first  threw  doubt  on  the  validity  of  this 

species,  suggesting  that  it  was  probably  synonymous  with  M. 
chenopoda,  he  afterwards  listed  it  without  question  from  the 
Chilean  island  of  Chiloe,  citing  specimens  collected  by  C.  Skotts- 
berg.^^  The  original  description  of  M.  Notarisii  is  very  full  but 
is  justly  criticised  by  Gottsche^*  on  account  of  its  many  ambi- 
guities. It  certainly  seems  to  point  to  M.  chenopoda,  and  the 
writer  would  refer  it  provisionally  to  that  species.  Unfortunately 
no  specimens  of  M.  chenopoda  from  Chile  have  been  available  for 
study. 

Sheffield  Scientific  School, 
Yale  University. 

"Kungl.  Sven.  Vet.-Akad.  Handl.  46':  5.    1911. 

"Bot.  Zeit.  16   (Beil.):28.     1858. 


WELLESLEY  COLLEGE  LIBRARY 


3  5002  03199  3517 


For  particulars  regarding  the  previous  Transactions  of 
the  Connecticut    Academy, 

Address  the  Librarian, 

Andrew  Keogh,  Yale  Station, 

New  Haven,  Conn.,   U.  S.  A. 


CONTENTS  OF  VOLUME  19  OF  THE  TRANSACTIONS 

PAGES ,  PRICE 

i-iio    Fossil  Birds  in  the  Marsh  Collection  of  Yale  Uni- 
versity, by  R.  W.  Shufeldt  (Feb.  191 5)   $1.10 

1 1 1-446     Middlemen  in  English  Business,  1660-1760,  by 

R.  B.  Westerfield  (Feb.  1915) 3.40 

CONTENTS  OF  VOLUME  20  OF  THE  TRANSACTIONS 

PAGES  PRICE 

1-131     The  Materials  for  the  History  of  Dor,  by  George 

Dahl  (May,  1915)   $1.40 

133-144    New  Spiders  from  New  England,  XI,  by  J.  H. 

Emerton 15 

145-160     Canadian  Spiders,  II,  by  J.  H.  Emerton 15 

161-240    The  Historical  Background  of  Chaucer's  Knight, 

by  Professor  Albert  S.  Cook  (Feb.  1916) i.oo 

241-399  Rural  Economy  in  New  England  at  the  Begin- 
ning of  the  Nineteenth  Century,  by  Percy  W. 
Bidwell  (April,  1916)   , 2.00 

CONTENTS  OF  VOLUME  21  OF  THE  TRANSACTIONS 

PAGES  PRICE 

1-144    The  Last  Months  of  Chaucer's  Earliest  Patron, 

by  Professor  Albert  S.  Cook  (Dec.  1916) $1.60 

145-200    The  Relationship  of  the  Tetracoralla  to  the  Hexa- 

coralla,  by  W.  I.  Robinson  (Feb.  191 7) . .      .40 

201-313  The  American  Species  of  Marchantia,  by  Alex- 
ander W.  Evans  (March,  1917) 1.20 


% 


■\