Skip to main content

Full text of "The Journal of the College of Science, Imperial University of Tokyo, Japan = Tokyo Teikoku Daigaku kiyo. Rika"

See other formats 


»^ 


$^  W.  ^  M  :^ 

m  :^  m  ftz  m 

^       ^       ^       Pv       W 

THE 

JOURNAL 

OF  THE 

COLLEGE  OF  SCIENCE, 

IMPERIAL  UNIVERSITY  OF  TOKYO, 
Vol.  XXXII. 


^    M    #    ^    i:    ^    fP    If 

PUBLISHED  BY  THE  UNIVERSITY. 
TOKYO,  JAPAN. 

1911-1913. 
MEIJI  44— TAISHO  2. 


9i- 


Publishing  Committee. 


O  »G^ 


Prof.  J.  Sakurai,   LL.  D.,  BUfakuhakiishi,  Director  of  the  College,  (ev  oßcio). 

Prof.  Ï.   Ifima,  Ph.  D.,  Rigakuhakushi. 

Prof.  F.  Omori,  Rigakuhakushi. 

Prof.  S.  Watasé,  Ph.  /).,  Rigakuhakushi. 


-oJ4«- 


All  counuunications  relating  to  this  Journal  should  be  addressed  to  the 
Director  of  the  College  of  Science. 


3^3^ 


CONTENTS. 


Art.  1.— Revisio  Aceracearum  Japonicarum.  (With  33  plates).  By  G.  Koidzumi. 

— Publ.  Augi\st  2ncl,  1911. 

Art.  2.— Beobachtungen  an  einer  Süsswasser  Peridinee.  (Mit  1  Tafel). 

By  N.  Ohno.— Publ.  November  20th,  1911. 
Art.  3.— Observations  and  Experiments  on  the  Ctenophore  Egg:    i.  The 

Structure  of  the  Egg  and  Experiments  on  Cell-division.     By  N.  Yatsu. 
—Publ.  April  29th,  1912. 

Art.  4.— Etudes  Anthropologiques.   Les  Aborigènes  de  Formose.   Fase.  II. 

—By  K.  ToRii.— Publ.  January.  16th,  1912. 
Art.  5.— Climatic  Changes  in  Japan  since  the  Pliocene  Epoch.  (With  i  plate). 

By  M.  Yokohama. —Publ.  Oct.  2nd,  1911. 

Art.  6.— On  Nepheline-basalt  from  Yingé-mên,  Manchuria.   (With  2  plates). 

By  B.  KotA— Publ.  June  7th,  1912. 
Art.  7.— The  Systematic  Motions  of  Sun-spots.  (With  3  plates). 

By  S.  HiRAYAMA.— Publ.  June  7th,  1912. 
Art.  8.— The  Metallogeny  of  the  Japanese  Islands.  (With  1  map). 

By  C.  IwASAKi.— Publ.  November  3rd,  1912. 

Art.  9.— Oogonium  Liberation  and  the  Embryogeny  of  Some  Fucaceous 
Algae.  (With  3  plates  and  5  text-figures).  By  M.  Tahara.— Publ. 
March  31st,  1913. 

Art.  10.— Beiträge  zur  Kenntnis  der  Morphologie  und  Stammesgeschichte 

der  Gorgoniden.  (Mit  13  text-fignres).  By  K.  KiNosHiTA. — Publ.  March 

31st,  1913. 
Art.  11.— Ueber   die   charakteristische   Mannigfaltigkeit   der  Partiellen 

Differentialgleichungen  erster  Ordnung.    By  T.  Y'oshie. — Publ. 

August  25th,  1913. 

Art.  12.— On  Cyathocormus  mirabilis  nov.  gen.,  nov.  sp.,  the  Type  of  a 
New  Family  of  Compound  Ascidians  from  Japan.  (With  3  plates 
and  0  text-figures).  By  A.  Oka.— Publ.  March  30th,  1913. 


PRINTED  BY  THE  TOKYO  PRINTING  CO.,  LTD. 


JOriiXAL  OF  THE  COLLEGE  OF  SCIEKCE,  IMPERIAL  UNIVERSITY   TOKYO 

VOL.  XXXII.,  ARTICLE  1. 


Revisio  Aceracearum  Japonicarum. 

G.  Koidzumi. 

{With  S3  Platen) 

Introduction. 

In  the  year  1902,  Dr.  F.  Pax  published,  in  Engler's  "Das 
Pflanzenreich,"  his  excellent  monograph  of  the  Aceraceae,  in 
which  the  genus  Acer  was  divided  into  13  sections,  comprising 
114  species  in  all,  besides  a  large  number  of  varieties  and  forms. 

Since  that  publication,  new  species  have  been  added  to  the 
genus  from  Eastern  Asia,  especially  from  the  Sinico-Japanese 
region,  augmenting  the  total  number  of  known  species  up  to  127. 

In  1905,  a  new  section  of  the  genus  was  proposed  by  Dr. 
Eehdepv,  based  on  certain  species  found  in  eastern  continental 
Asia,  in  addition  to  those  of  Pax. 

The  present  monograph  deals  with  Acer  found  in  Japan 
proper,  Loochoo  Islands,  Formosa  and  Sachalin.  Based  on  the 
results  of  careful  examinations  and  revisions  of  each  species,  I 
have  attempted  to  present  a  new  systematic  arrangement  of  the 
genus  on  the  principle  of  bringing  out  as  far  as  possible  the 
phylogenetic  relations  of  the  members. 

The  vernacular  names  of  every  species  are  taken  principally 
from  the  works  of  Prof.  J.  Matsumura^  and  Prof.  M.  Shirai.' 

The  present  work  was  taken  up  at  the  suggestion  of  Prof. 
J.  Matsumura,  to  whom  I  beg  herewith  to  express  my  grateful 
thanks  for  the  many  acts  of  kindness  renderd  me  during  my 
studies.  I  am  also  under  great  obligation  to  Prof.  M.  Shirai 
who  helped  me  in  many  matters  regarding  the  literature.  Further 
to  Dr.  Y.  Shirasawa,  Prof.  G.  Yamada,  and  Mr.  T.  Kawakami 
of  the  Formosan  government,  I  am  much  indebted  for  the  valuable 
material  placed  at  my  disposal. 

1-      Bm&^^M-     ed.  3.     (1897J. 

2.  u^mm^mm-  {1908). 


^  Art    1  —  Oi.  Koidz;iiui. 

Aeeraceae.     DC 

Aceraceae,  DC.  Tht'or.  Elem.  Bot.  (1S13)  ;— Lindl.  Nat.  Syst.  ed.  2.  (1838)  p.  81  ;  Veg. 
Kingd.  (1847)  p.  387;— Pax,  in  Engl,  et  Prantl.  Nat.  Pfl.  Fam.  III.  5.  (1893)  p.  263,  et  Pfl. 
Etich,  8  Heft  (1902)  p.  1  ;— Engl.  Svllb.  Nat.  Pfl.  Fam.  (1904].  p.  155. 

Acera,  Juss.     Gen.  PI.  (1789).  50. 

Acerineae,  DC.     Prodr.  I.  (1824).  p.  593  ;— Endl.  Gen.  PI.  (1836—40)  p.  1055. 

Sapindaceae,  subord      Acerineae,  Benih.  et  Hook.     Gen.  PL  I.  (1867)  p.  391. 

Sapindaceae,  series  8.     Acereae,  Bull.     Xat.  Hist.  PL  V.  (1874)  p,  427. 

Sapindaceae,  subord.     Acerinae,  Drude,  in  Schenk,  Handb.  III.  B.  (1887)  p.  3C0. 

Flores  actinomorphi.  Carpella  2.  Ovaria  in  loculis  biovu- 
lata.  Fructus  mericarpiis  sainaratis  2-compositus.  Semina  exaril- 
lata.     Folia  opposita,   simplicia  vel  imparipinnata,  exstipulata. 


Acer.     L. 

Acer,  (Todrnf)  L,  Sp.  PL  ed.  1  (1753)  p.  1034,  et  Gen.  PL  (1754).  p.  1155;— DC.  Prodr.  I. 
•(1824).  p.  593;— Endl.  Gen.  PL  (1836—40).  p.  1056;— Benth.  et  Hook.  Gen.  PL  I.  (1867)  p. 
409  ;— Baill.  Nat.  Hist.  PL  V.  427  ;-Pax,  in  Engl,  et  Prantl.  Nat.  Pfl.  Fam.  III.  5.  p.  269, 
Engl.     Eot.  Jahrb.  VII.  (18S6).  p.  177.  et  Engl.  Pfl.  Eeich.  (IV.  163.)  8.  Heft.  (1902)  p.  6. 

Negimdo,  Ludwig,  Gen.  PL  ed.  3.  (17C0).  309  ;— DC.  Prodr.  I.  596  ;— Benth.  et  Hook.  Gen. 
PL  I.  409  ;— Endl.  Gen.  PL  1056. 

Xegundinm,  Rafin,  in  Desv.  Journ.  Bot.  If.  (1 809).  p.  170. 

JMac,  Adans,  Fam.  II.  (1763).  p.  383. 

Exincer,  Opiz,  Seznam  (1852),  p.  42. 

Flores  lioterochlamydei  rarissime  monochlamydei,  actinomor- 
phi, cyclici  rarius  semicyclici,  pentameri  vel  rarissime  tetrameri, 
hermaphroditi  vel  imisexuales.  Sepala  5,  rarissime  4,  libera  vel 
pi.  in.  connata,  in  aestivatione  imbricata  vel  quincunciales. 
Petala  imbricata  sepalis  alterna,  vel  nulla.  Discus  varius,  extra- 
vel  mtrastaminalis,  annularis  vel  lobatus,  interdum  abortivus  vel 
deficiens.  Stamina  4 — 10,  diplostemoni,  saepissime  8,  hypogyna 
vel  perigyna;  antheris  bilocularibus,  innatis,  longitudinaliter 
dehiscentibus.  Carpella  2  rarissime  3,  connata  (syncarpa  et 
oligoinera).  Ovarium  2-lobum,  2-loculares,  loculis  biovulatis; 
stylis  2  inter  lobos  ovarii  insertis,  intus  stigmatosis;  ovula 
pendula,  collateralia  vel  deinde  superposita,  integumento  duplici, 
raphe  dorsali  cum  micropyle  supera.  Fructus  samarae  2,  indéhis- 
centes, demum  ab  axi  secedentes,  commissura  plus  minus  persis- 
tente. Semina  in  loculis  1,  exalbuminosa,  testa  membranacea. 
Embryo  plerumque  diplecolobeus,  cotyledonibus  plicatis,  radicula 


ReTisio  Aceracearum  Japonicirum.  3 

elongata.  Arbor  vel  frutices.  Folia  opposita,  petiolata,  exstipula- 
ta,  membranacea,  vel  subcoriacea,  simplicia  vel  imparipinnata. 
Inflorescentia  racemosa,  corymbosa  vel  paniculata;  rarias  bractea- 
ta;  andro-polygama,  andro-nionoecia  vel  -clioecia,  ititerdum 
dioecia. 

C7at'is  tliaynostica  Sectiotinm. 

I.  Intrastaminalia  :  (Discus  intrastaminalis.  Stamina  hypogyna 
vel  in  flore  ^  tantum  perigyna.  Flores  cum  foliis  nascentes  vel 
panllo  tantum  praecociores.    Gemmae  perulae  interiores  elongatae.) 

^     Inflorescentia      andro-polygama,     elongato-paniculata,     in 
ramulis    foliatis    terminalis.       Flores  pentameri.       Folia 

palmato-lobata. Pari-iflora. 

'*  *     Inflorescentia   andro-dioecia    vel-monoecia.      Flores  penta- 
meri rarissime  tetrameri. 
A     Inflorescentia  in  ramulis  foliatis  terminalis. 
O     Folia  indivisa. 

D     Folia  utrinque  pluri-costata  ;  flores  5  meri Indivisa. 

DD     Folia  utrinque  multi-costata;  ûores  A-meri.... Carp inifoUa. 
OO     Folia  pi.  m.  palmato-lobata. 

D     Inflorescentia  corymbosa Glabra. 

DD     Inflorescentia  racemosa. 

©     Antherae  ellipticae,  laevae.     Ovaria  puberulento- 
tomentella.     Folia     3 —  (sub  5)  lobata;     lobis 

inaequaliter  serra  ta Macrantha. 

®®     Antherae  ovatae,  apice  apiculatae,  leviter  scabrae. 
Ovaria  glabra.     Folia   5-lobata;   lobis  utrinque 

lobulato-inscisis Palmatoidca. 

A  A     Inflorescentia  in  ramulis  aphyllis  termmaUs. . .  .Lithocarpa. 
*"^     FJores    pentameri,    dioeci.     Inflorescentia  9-  in  ramulis 
foliatis,     ^   in    aphyllis    terminalis.      Folia    palmato- 
lobata Argiita. 

II.     Extrastaminalia :       (Discus    extrastarainalis,    rarius    valde 
abortivus  vel  nuUus.) 

A.     DUyji  valb  abortiva^    vel  deflciens.     Flores  longe  ante 


4 


Art.  1.  — G.  Koidzumi. 


folia    nascentes.      Gemmae    perulae    interiores    non    elongatae. 
ÇAdiscantha.) 

*     Folia  simplicia.     Flores  andro-dioeci.     Sepala  non  connata. 

Discus  dentiformis.     Stamina  5 — 8 Ihihra. 

"^  "^  Folia  ternata  vel  pinnatim  5-foliolata.  Flores  dioeci. 
Sepala  connata.  Stamina  4—6.  Discus  nullus.  Corolla 
deficiens Kcrjundo. 

B.     Discus  bene  evolutus.     Flores  cum  foliis  nascentes  vel  paullo 
tantum  praecociores.     Gemmae  perulae  interiores  elongatae. 
*  Flores  ^  perigyni.     Folia  lobata. 
O     Inflorescentia  andro-polygama. 

-A.     Foliorum  lobi  obtusiusculi,  saepe  integri.     Stamina 
disco     crasso    prope    ejus   marginem    interiorem 

inserta.   Fructus  loculi  duri Cauipcstria . 

-AA     Foliorum     lobi    grosse     sinuato-dentati.      Stamina 
disco  medio  inserta.     Fructus   loculi   planiusculi 

Platanoidca . 

OO     Flores  dioeci.      Sepala    florum  "^  cum  petalis  coalita. 

Diaholica. 

OOO     Inflorescentia   andro-dioecia   vel-monoecia.      Corolla 

nulla.     Sepala  connata Saccliarina . 

*  *     Flores  omnes  hj^DOgyni.     Folia  lobata,  ternata  aut  indivisa. 
O     Flores  dioeci,    tetrameri,    minuti,    elongato-racemosa. 

Folia  ternata Ciasifolia. 

OO     Inflorescentia     andro-polygama.      Folia     lobata    vel 
intégra. 
A     Folia  pi.  m.  palmato-lobata. 

@     Inflorescentia  paniculata,  vel  racemoso-pani- 

culata.     Folia  3 — 5  lobata :Spicata. 

©©     Inflorescentia     corymbosa.       Folia     5-pluri 

lobata .Palniata. 

AA     Folia  intégra Integrifolia . 

OOO     Inflorescentia    andro-dioecia    vel    monoecia.      Folia 
ternatisecta.     Flores  umbellati Trifoliaia. 


EeTisio  Aceracearimi  Japonicarum. 
lèuleoc  sectionnu»  getieris. 

I.  J ntrastaminalia,  Pax,  1885. 

1.  Glabra,  Pax,  1S85. 

2.  Parviflora,  ni. 

3.  Indivisa,  Pax,  1885. 

4.  Carpinifolia,  m. 

5.  Macrantha,  Pax,  1885. 

6.  Palmatoiclea,  m. 

7.  Litliocarpa,  Pax,  1885. 

8.  Arguta,  Rehder,  1905. 

II.  Extrastamimdia,  Pax,  1885. 

9.  Negundo,  Pax,  1885. 

10.  Cissifolia,  rn. 

11.  Rubra,  Pax,  1885. 

12.  Saccharina,  Pax,  1885. 

13.  Spicata,  Pax,  1885. 

14.  Palmata,  Pax,  1885. 

15.  Integrifolia,  Pax,  1885. 

16.  Trifoliata,  Pax,  1885. 

17.  Campestria,  Pax,  1885. 

18.  Platanoidea,  Pax,  1885. 

19.  Diabolica,  m. 


ludeoc  specierunt,  varietafutnj'oftiiai'unique  A.ceraeearuni 

JfapoiticarMitn. 

I.     Iiitmstaiinnalia,    Pax. 

Sect.   1,   Parviflora,  Koidz. 

1 .  Acer  parviflorum,  Fß.  et  Sav Tetsu-Jcaede. 

Sect.  2,  Indivisa,  Pax. 

2.  Acer  distylum,    S.  et  Z Blaniha-kaede. 

3.  Acer  crataegifolium,  S.  et  Z Ko-iwi-kaede, 


6  Art.  1.  — G.  Koidzumi. 

rar.  Veitchii,  Nichols Fw'ri- kmirikacde. 

4.  Acer  insulare,   Mak Shma-wikaede. 

5.  Acer  Kawakamii,  Koidz.  sp.  nov Onaga-hiede. 

6.  Acer  ovatifolium,  Koidz.  sp.  nov Koha-onagahaede. 

7 .  Acer  morrisonensis,   Hay Talcasago-nrikacde, 

Sect.  3.  Carpinifolia,  Koidz. 

8.  Acer  carpinifolium,   S.  et  Z Yamasliiha-kacde. 

vSect.  4.  Macrantha,  Pax. 

9.  Acer  rufinerve,  B.  et  Z Uriliada-kaede. 

forma,  albo-limbaturn,  Hook,  fil Fuiri-urihada. 

10.  Acer  capillipes,   Max Hosoe-iirihada . 

var.  fujisanense,  Koidz,  var.  nov Hiroha-asliiboso-iirinohi . 

1 1 .  Acer  rubescens,  Hay Tahisago-iwihada. 

Sect.  5.  Palmatoidea,  Koidz. 

12.  Acer  micranthum,  S.  et  Z Ko-minchiede. 

13.  Acer  Tschonoski,  Max Minc-lcaedc. 

Sect.  6.  Arguta,  Rehdee. 

14.  Acer  arguturn,  Max   Asanolia-kacde. 

II.      Extnisiaminalia,  Pax. 

Sect.  7.  Cissifolia,  Koidz. 

15.  Acer  cissifoliuni,  Kocir Mitsudc-kaede. 

Sect.  8.  Rubra,  Pax. 
10.      Acer  rubrum,   L Hana-kaedc. 

Sect.  1).  Spicata,  Pax. 

17.     Acer  trifidum,  Hook,  et  Arn    Tû-kacdc. 

forma  integrifoliuni,  (Mak)  ni Bland a-tôkaede. 

var.  ningpoense,    Hanct Tô-kaede. 


Re  Visio  Aceracearum  Japonicarum.  7 

var.  formosaiuim,    Hay Talcasago-tühaede.. 

18.  Acer  ginnala,   Max Karahogi-haede. 

var.  yezoense,  Koidz.  var.  nov Yezo-karalîogikaede.. 

19.  Acer  spicatum,  Lam. 

var.  ukurunduense,  Max     Ogarahana,  Hozalci-haede. 

20.  Acer  Oliverianum,  Pax. 

var.  Nakaharae,  Hay.     var.  nov Shima-vioiniji. 

scar,  formosanum,   Koidz.     svar.  nov 

Itomaki-sliimamomiji,. 

svar.  trilobatum,   Koidz.     svar  nov 

Mitsude-slimamomiji,. 

Sect.  10.  Palmata,  Pax. 

21.  Acer  Sieboldianum,  Miq. 

«•    typicum,    Max Itai/ameigetm.  Kihana-uclimaliaede. 

svar.  albiflorum,  m Shirohaua-itaiiameigctsu. 

svar.  tortuosum,  m. 

scar,  micropbyllum,    ni Hime-ucliiwa'kaede. 

svar.  Moniijigasane,  m Momiji-gasane. 

svar.  Sodenoucbi,    m Sodenouchi. 

ß-    tsusimense,  m Koba-itai/ameigetsu. 

svir.  Kasatoriy ama,  m Kasatoriijama. 

svar.  laxifolium,  m Hina-itayameigetsu.. 

svar.  Ayaigasa,  m Aijaigasa. 

22.  Acer  Shirasawanum,  Koidz.     sp.  nov Oh-itayameigetsu. 

var.  tenuifolium,  in Hina-uchiicakaede.. 

svar.  Murasame,  m Mwasame. 

23.  Acer  japonicum,  Thg, 

«•    typicum,  Ge.  v.  Scnw- Haiiclihra-kaede,  Akahana -ucliiivakaede ^ 
forma  angustilobum,  m. 
forma  macropbyllum,  m. 
forma  tenuilobum,  ni. 
forma  semiovatum,  m. 
forma  ascendens,  m. 
forma  crassifolium,  m. 


^  Art.  l.—  G.  Koidzumi. 

ß'    aureum,  Gr.  v.  Schw. 
T-    Parsonii,  Veitch. 

'^.    Heyliachii,  Matsum.  Mai-hujaku. 

^'    microphyllum,  m Yezo-meigetsii-kaede, 

C-    circumlobatum,  m Oh-meigetsu. 

V-    villosum,  m. 

^-    Kasado,  m Kasado. 

'■    Kokonoye,  m Kokonoe. 

X-    Sayosigure,  m Sayoshigure. 

^-    Matsuyoi,  m Matsuyoi. 

24.     Acer  palmatum,   Thunb Kaede,  Momiji, 

ssp.  Ci.    genuinum,  m. 

(.1.  spectabile,   m Iroha-niomiji. 

forma  Chisiwo,  m Chishio. 

forma  Komonnisiki,  m Komon-nishiki. 

svar.  crispum,  m Okushimo. 

svar.  Higasayama,   m Higasa-yama. 

h.  amabile,  m Iroha-momiji. 

forma  Akajinisiki,  m Akaji-nisliiki . 

/or»?«  Tsuchigumo,  ni Tsucliigumo. 

forma  Hanaizuminisiki,  m — HanaizuminisMki. 

forma  Oridonisiki,  m Orido-nisliiki. 

suar.  Kagiri,   m Kagiri. 

^''•s/^.  ß-    septenlobum,  (Thg. )  m Takao-momiji. 

car.  latilobatum,  ni Hiroha-momiji. 

var,  speciosum,  m Nomura-kaede. 

var,  palmatipartitum  m. 

forma  Senri,  m Senri. 

forma  Ichigioin,  m Ichigyoin. 

forma  Akitsuta,  m Akilsuta. 

forma  Tsukubane,  m Tsiikiihane. 

forma  Ohsakazuki,  m Oh-sakazuki. 

svar.  Tanabata,  m Tanahata. 

scar,  linearilobum,  n:i Shimcnouchi. 

forma  lineare,  (Gr.  v.  Schw.)  m. 
forma  atro-lineare,  (Gr.  v.  Schw.)  m. 


Revisio  Acericearum  Japonicarum,  9 

s^p.  r-    Matsumurae,  m Yama-momiji. 

a.  spontaneum,  m. 

foniia  angustilobnm,  m. 

forma  circumlobatum,  m. 

forma  acutum,  m. 

forma  rectangulare,  m. 

forma  obtusum,  m. 
svar.  elegans,  m. 
svar.  f ormosan urn,    m Takasago-momiji. 

b.  horten  se,  in. 

svar.  pahïiatilobum,  m. 

forma  Sigitatsu,  m Slwjitatsu. 

forma  Nisikikasane,  m NisJiikigascme. 

forma  Asanoha,  m Asanoha. 

svar.  palmatipartitum,  m. 

forma  Monnisiki,  ni MonnisMhi. 

forma  Akitsusima,  m Akitsiishima. 

forma  Tokouatsu,  m Tokonatsu. 

forma  Yugure,  lu Yügure. 

forma  Nokibata,    m Nokihata. 

forma  Kihachijo,  m Kiliacliijö. 

forma  Kageorinisiki,  m Kageorinisliiki. 

forma  Siguresome,  m Shigiirczome. 

forma  Takinogawa,  m Takinogaica. 

forma  Kurabuyama,  m Kurahuyama. 

forma  Awoba,  m Aoha. 

forma  Karukaya,  m Karukaya. 

forma  Murasakitaka,  rn Miirasakitaka. 

forma  Asaji,   m isaji. 

/or»za  Akegarasu,  m Akegarasii. 

forma  Murehibari,  ra Murchihari. 

svar.  heterolobura,  m. 

forma  Wabibito,  m Wahihito. 

forma  Sensunagasi,  m Sensunagashi. 

forma  Hibari,  m. 
svar.  dissectum,  (Tug.)  m. 


10  Art.  I.— G.  KoLdzumi. 

1.  iniiltifîdum,  m. 

forma  Awosidare,    m Aoshidare. 

forma  Tamukeyama,  m Chinmen-momiji. 

2.  palmatisectum,  m. 

forma  Matsukaze,  m Matsiikazè. 

forma  Ohsiusiclare,  m Ohshyiisliidare. 

svar.  sessilifolium,  m. Hagoromo-kaede. 

Sect.  11.  Integrifolia,  Pax. 

25.  Acer  obloiignii),  Wall Kimmoha-lxacàe. 

Sect.  12.  Trifoliata,  Pax. 

26.  Acer  nikoense,  Max C]wja)iol-i\   Ohmitsiule-hiede. 

Sect.  1.3.  Platanoidea,  Pax. 

27.  Acer  Miyabei,  Max Kurohi-  ilaya. 

28.  Acer  pictum,  Thuxb. 

«.    typicuin,  Gr.  v.  Schw. 

uïar.  eupictum,  Pax. Ilaya-htede. 

svar,  mono,  (Max.)  Pax Itaya-lmede. 

srar.  Savatieri,  Pax Itomnld-itaya . 

forma  septenlobum,  m. 
forma  novemlobuni,  ni. 

srar.  Mayri  (G.  v.  Schw. )  ni Itnya-hicdc. 

svar.  Futagoyan^a,  m. 

ß'    Paxii,  Gr.  v.  Schw Oni-itaya. 

/'•   dissectum ,  Wesmael A sahi-kaede. 

svar.  subtrifiduni,  Mak Yafjuruma-lmedc. 

f*.    glaucuiï) ,  m Urajiro-itaya . 

.war.  latilobum,  m Urajiro-itoiiuiki-itaya. 

Sect.  14.  Diabolica,  Koidz. 
20.     Acer  dial)olicurii,  Bl Kaji-kacde,  Oni-momiji. 


Revisio  Aceraceaniui  Japonicarum.  1 1 

Descriptiones  Specrerum  Aceris  Japoniae. 
I.     Intrastaminalia,  Pax. 

Instrastaminalia,  Pax,  iu  Engl.  Eot.  Jahrb.  YI.  (1385)  p.  £27.  (Periayna,  Pax,  1.  c.  pro 
parte.) 

Sect.   1.   Parcißora.  m. 

InfloresceDtia  andro-polygama,  elongato-paniculata.  Flores 
pentameri.  Stamina  8,  in  flore  '^^  perigyni;  antlieris  scabris. 
Discus  intrastaniinalis.  Stylus  bipartitus.  Folia  membranacea, 
palmatilobata.     Species  unica. 

1.     Acer  parviflorum,  Fr.  et  Sa  v.  (Tab.  I.) 

Fr.  et  Sav.  En.  PL  Jap.  II.  (1879)  p.  Îi21  ;— Max.  Mtl.  Picl.  X.  (18S0)  595;- Pax,  in  Engl. 
Bot.  Jahrb.  VU.  (iSCö)  p.  247,  et  Engl.  Pfl.  Riich,  8  Heft  {IV.  163),  (1£02)  p.  t9;-C.  K.  Schn. 
111.  Hacdb.  Laubh.  II.  (1907;  p.  236  ;— Leveil.     Bull.  Soc  Bot.  Fr.  VI.  (1806)  p.  592. 

A.  pennsylvaiUciivi,  var.  iHirvißorum,  Wesmael,  in  Bull.  See.  Bot.  Eelj.iqueXXIX.  (1810)  C2. 

Nom.  JAP.     Tetsu-hiede  (ß-:^^  ^),  Tetm-no-ld  (ibid.) 

Arbusculus  vel  arbor;  ranuilis  novelli?,  foliis  inflorescentibus- 
que  rufo-pubescentibus,  cito  glabriusculis.  Folia  membranacea, 
adulta  ad  axillas  costarum  tantum  rufo-barbarta,  aeciuilonga 
(6 — 20  cm.)  ac  lata;  ambitu  quadrangulata,  sed  in  foliis  trilobatis 
ovata;  basi  cordata,  H-lobata;  lobis  tiiangulari-ovatis  brève 
acuminatis,  duplicato-serratis;  extimis  valde  abbreviatis;  petiolis 
e.longatis  basi  parum  dilatatis.  Inflorecentia  spicato-paniculata; 
Acribus  andro-polygamis,  foliis  coaetaneis;  pedicellis  ebracteolatis, 
minute  puberulentibus.  Calyx  extus  ciliolatus;  limbis  5,  ovatis 
obtusis.  Petala  5,  alba  v.  lutescentia,  oblonga  obtusa,  quam 
sepala  breviora.  Stamina  8;  filamentis  subulatis  glabris;  antheris 
scabris.  Discus  crenatus,  intrastaminalis.  Ovarium  puberulento- 
tomentosum;  stylis  ad  basin  bipartitis;  stigmatibus  revolutis. 
Samarae  loculi  rufo-tomentelli;  alis  angulo  obtuso  divergentibus. 

Obs.     Julio— Augusto  fl.,  Augusto— Octobri  fr. 

Hab.  in  sylvis  montuosis  insulae  Hondo  et  Sikoku;  Hondo: 
Kurikomayama  (Rikuzen),    Ötöge,    Gassan,    Hagurosan,    lidesan, 


i^  Art.  1. — G.  Koidzomi. 

Azumasan  (Uzen),  Simizutôge  (Simotsuke),   Myôkôsan,  Kurobe- 
yama  (Yechigo),  Togakusiyama  (Sinano),  Fujisan  (Suruga),  Sikok: 
Tsurugiyama  (Awa). 
DisTR.     endemica. 


Sect.  2.  Indicim,  Pax. 

IiuUvim,  Pax,  in  Engl.  Bat.  Jahrb,  VI  (18S5)  327.  VTL  (1S86)  210. 

Inflorescentia  racemosa.  Flores  andro-dioeci  vel-monoeci. 
Sepala  petalisque  5.  Stamina  8,  hypogyna,  in  flore  '^  perigyna. 
Petala  rarius  eroso-serrata.  Discus  intrastaminalis.  Stylus  pro- 
funde bilobatus.  Folia  membranacea,  indivisa,  saepe  subtrilobata 
immixta,  utrinque  pauci — pluri  costata. 

Species  ad  6,  Hondoensis,  Sikokuensis,  Kiusiuensis,  et  Form- 
osae  incolae. 


Clavis  speciemm  ttichotonia. 

1.  Folia  elliptica,  glabra,   crenata;  samarae   loculi  ferrugineo-to 

mentosa. A.  disti/Ium,   S.  et  Z. 

1  olia  0  vata  acuminata  ;   samarae  glabrae 2 

2.  Folia  nunquam  trilobata 3 

Folia  saepe  triloba  immixta 4 

3.  Folia  crenulata  nunquam  dentato-lobulata A.  ovatifoliuin,  m. 

Folia    serrata,    saepe    inciso-serrata;    serraturis    subito    acutis 

patentibusque;  petalis  eroso-serratis A.  Kawakamii,  m. 

4.  Folia  inciso-serrata;  serraturis  inaequalibus,  plerumque  incar- 

vatisque ;     petalis    eroso-serratis A.    insulare,    Making. 

Petala  intégra  ;   foliis  serraturis  patentibus 5 

5.  Florum  pedicellis  glabris;  folia  ab  initio  glabra,   subtus  glau- 

cescentes,   saepius  distincte  trilobata;  alae  samarae  horizon- 

taliter  patentes .-( .  cmtaegi/olium,  S.  et  Z. 

Florum    pedicellis    puberulentibus;     folia   juniora    ad   venas 
puberula,  le viter  trilobata A.  morrisonense,  Hayata. 


Eevisio  Aceracearum  Japonicarum.  i--' 

2.     Acer  distylum,  S.  et  Z.     (Tab.  II.) 

SiKBOLD,  et  ZüCCARiNi,  Fl.  Jap.  Farn.  Nat.  in  Abb.  Akad.  Munch.  IV.  2  (1846)  154  ;— Miq. 
Prol.  Fl.  Jap.  (1866).  21  ;— Walp.  Ann.  I  (1846)  960;— Fr.  et  Sav.  En.  PI.  Jap.  I.  (1875)  89  ;— 
Max.  Mél.  Biol.  X  (1880)  595;— Pax,  in  Engl.  Bot.  Jahrb.  VII  (1886)  216,  et  Engl.  Pfl.  Reich. 
8  Heft  (1902)  34;— Leveil.  in  Bull.  Soc.  Bot.  Fran.  VI  (1906)  590  ;—C.  K.  Sohn.  III.  Haiidb. 
Laubh.  II.  (1907)217. 

Nom.  Jap.  Hitotsuha-lcaede  (nom.  vulg.)  Maruha-hjcde,  Itufjo- 
hicde,    Clndori-vo-ki.     (^  «  H  ¥•  15  IS  iXmm-  ^^mm.  1832.) 

Arbor;  ramulis  juvenilibus  ferrugiDeo-tomentosis;  gemmae 
perulis  chartaceis,  ovato-lanceolatis,  extus  ferrugineo-pubescentibus 
praeditae.  Folia  membranacea,  juniora  petiolisque  ferrugineo- 
tomentosa,  mox  glabra,  ovato-elliptica,  crenata,  apice  subito  brève 
acuminata,  basi  cordata.  Inflorescentia  foliis  coaetanea,  sub- 
complicato-racemosa,  ferrugineo-tomentosa;  floribus  andro- 
dioeciis.  Calyx  5-fidus,  laciniis  utrinque  puberulento-tomentosis, 
oblongis,  apice  rotundatis.  Petala  5,  alba  vel  lutescentia,  oblonga 
basi  parum  angustata,  apice  rotundata,  sepalis  aequilonga.  Discus 
bene  annulato-evolutus,  intrastaminalis.  Stamina  plerumque  8, 
in  flore  "^  leviter  exerta,  antheris  laevibus.  Ovarium  dense 
puberulento-tomentosum,  stylis  ad  basin  bipartitis,  laciniis 
subulatis.  Samarae  loculi  ovato-oblongi,  ferrugineo-tomentosi; 
alis  angulo  obtuse  divergentibus. 

Obs.     Maio— Junio  fl.,  Augusto — Septembri  fr. 

Hab.  in  sylvis  insulae  Nippon:  Hayachine,  Sengantöge 
(Rikuchiu),  Kurikomayama  (Rikuzen),  Azumasan  (Uzen),  Aidzu 
(Iwashiro),  Nikkö  (Shimotsuke),  Togakushiyama  (Shinano),  Cbi- 
chibu  (Musashi),  Ohyaraa  (Sagami),  Ibukiyama  (Ohmi). 

DisTR.     endemica. 

3.     Acer  crataegifolium,  S.  et  Z.  (Tab.  III.) 

Sieb,  et  Zucc.  F).  Jap.  Farn.  Xat.  in  Abh.  Akad.  Münch.  IV.  2.  (1846),  p.  155,  et  Fl.  Jap.  II. 
(1870),  84.  t.  147  ;— Miq,  Prol.  Fl.  Jap.  (1S66),  21  ;— Fr.  tt  Sav.  En.  PI.  Jap.  I.  (1875),  89  ;— Max. 
Mél.  Biol.  X.  (1880),  596  ;— Pax,  Engl.  Bot.  Jahrb.  VII.  (1886),  p.  248,  et  Engl.  Pfl.  Eeich.  8 
Heft  (1902).  p.  36;— Leveil.  et  Vnt.  Bull.  So3.  Bot.  Fr.  VI.  (1906),  p.  590  ;— C.  K.  Schn.  111. 
Handb.     Laubh.  II.  (1907),  216. 

A.  cmtae()i folium.  2.  tTjp.cuvi,  Graf.  v.  Schw.  in  Gartenfl.  (1S93)  4c5  ;— Pax.  Pfl.  Eeich.  I.e.  36. 
^    A,  cüciiUobracteatiim,  Leveil.  et  Vnt.  1  c.  59, 


14  Art.  1.— G.  Koidzumi. 

Nom.  Jap.  Shirahashinold  (¥  1*  +  -  ¥■  îI  tî  in  âl  ^.  5^  ^  ^  M  Ä  iS. 
1121)]—Hana-kaede  (!?7T«5¥-  ^W^m^m, -X^^^-  1708;  —  Yanm- 
kaetle  (;^  5fn  tJc  |i)  ; — lion-wi  (fg  !)tl  ,r>  f|)  ;  — Sliira-kaede  (?K  1*  /\  ¥•  laf  H  lg 
Ui  !=■  *E  i^  >^: Sß :£•  ]So7) •,—Ao-iiri  (Indigenis  nikkoensis  et  chichi- 
biieusis  vocatur); — Me-winoki,  Uii-kaede,  Ko-urikaalc  (J.  INIatsu- 
MURA,  ShokubutsLi-]\[eii,  no.  30). 

Arbor;  ramulis  no  vis  ferrugineo-pubescentibus,  mox  gkibris. 
Gemmae  perukie  angaste  vel  spatbulato-oblougae,  parce  pilosae. 
Folia  membranacea,  glabra;  nascentia  praesertim  ad  axillas 
nervorum  ferrugineo-tomentosa;  subtus  glauca  vel  glaucescentia, 
ovata  acuminata,  basi  cordata,  inciso-serrata  saepius  leviter  tri- 
lobata,  utrinque  pauci(4 — 6) — costata;  acuminibus  acutis  vel 
obtusiusculis;  petiolis  gracilibus.  Inflorescentia  racemosa,  foliis 
coaetanea;  pedicellis  glabris;  floribusluteis,  andro-dioeciis.  Calyx 
5-fidus,  extus  glaber;  laciniis  oblongis  apice  rotundatis.  Petala  5 
oblonga  sepalis  parum  longiora.  Stamina  8,  inclusa;  antheris 
rotuudatis  laevibu=5.  Discus  carnosus,  intrastaminalis.  Ovaria 
glabra;  stylis  profouiide  bilobatis,  laciniis  subulatis  revolutis. 
Samara  glabra,  loculis  reticulato-nervosis,  roseo-coloratis,  pulcher- 
rimis;  alis  horizontaliter  patentibus,  0 — 8  mm.  latis.  Maio — Junio 
ti.,  Augusto  fr. 

Hab.  in  Hondo:  Hayacliine  (Rikucbiu),  Aidzu  (Iwasbiro), 
Nikko  (Shimotsuke),  Akagisan  (Ködzuke),  Chichibu,  Tamagawa 
(Musashi),  Norikurasan,  Ontakesan  (Shinano),  Komono  (Ise), 
Köyasan  (Kii),  Prov.  Swö,  Tottori  (Inaba),  Prov.  Yamato. 
Kiusiu:      Hikosan  (l^uzen). 

Var.  Veitchii,  Nichols.  iu  Gard.  Chron.  (ISSl),  II.  75;— Pax,  Engl.  pfl. 
Keich.  1.  c,  37. 

No-M.  Jap.     Fuôii-kourikaedc  (nom.  no  v.) 

Folia  juvenilia  roseo-  dem  urn  albo-marmorata,  inprimis 
secus  marginem. 

Hab.  culta. 

DisïR.  SP.     endemica. 

4.     Acer  insulare,  [Marino.  (Tab.  IV.) 

T.  Making,  ia  Bat.  Mi^.  Tokyo,  XXIV.  (101))  293. 


Revisio  Accracearum  Japonicarum.  15 

A.  rufinerve,  Engl.  Bot.  Jahrb.  VI.  î.9;— Matsdm.  et  Ito,  Tent.  FI.  Lutch.  I.  120;— Forb. 
et  Hemsl.  Joiar.  Lin.  Soc.  XXIII.  142  (non  S.  et  Z.) 

J.  (.viiffZanoH,  Matsüm.  Bot.  Ma j.  Tokyo,  XII.  (1593)  63;— MA.TSUM.  et  Ito,  1.  c.  120  (?i07i 
Wall.) 

Arbor  ramis  glabris,  cortice  lutescenti-viride.  Folia  membra- 
nacea,  adulta  glabra,  juniora  subtus  secus  costas  et  in  axillis 
nervorum  ferrugineo-pubescentia,  ovata  ovali-ovata,  utrinque 
5 — 7  costata,  basi  cordata  vel  subcordata,  apice  longe  acuminata, 
margino  inaequaliter,  vel  interdum  subinciso-serrata;  serraturis 
plerumque  incumbentibus;  7 — 14  cm.  longa,  4 — 14  cm.  lata; 
petiolis  mox  glabris  2,5—9  cm.  longis.  Inflorescentia  elongato- 
racemosa,  cire.  8  cm.  longa,  parce  ferrugineo-pubescentes;  pedicel- 
lis  fîliformibus  3— 8  cm.  longis;  floribus  foliis  coaetaneis,  cire.  10 
mm.  in  diametro.  Calyx  extüs  parce  puberulens;  sepalis  5, 
lineari-oblongis,  obtusis  vel  acutiusculis,  cire  3  mm.  longis. 
Petala  5,  cuneato-oblanceolata,  obtusa  vel  acutiuscula,  eroso- 
serrata,  4,5 — 5,5  mm.  longa.  Discus  crenatus,  intrastaminalis. 
Stamina  8,  ßlamentis  subulatis,  antheris  ellipticis.  Ovaria 
puberulento-tomentosa;  stylis  profunde  bilobatis,  laciniis  revolutis. 
Fructus  glaber,  cire  24  mm.  longa;  alae  angulo  obtuso  vel 
obtusissimo  divergentes. 

Nom.  Jap.      Shima-urihiedc  (T.  Makixo.) 

Hab.  Kiusiu:  Prov.  Ohsumi,  insul.  Yakushima,  insul. 
Amami-Ohslnma. 

5.     Acer  Kawakamii,  Koidz.  (Tab.  V.) 

KoiDZ.  in  Bot.  Mao-.  Tokyo,  XXV.  no.  290  (Mart.  1011)  p.  102. 

A.  CMidaUun,  Matsum.  et  Hat.  En.  PL  Forrnos.  (1905)  96  [non  Wall.) 

A.  canititi folium,  Hatat.  in  Jour.  Coll.  Soi.  Tokyo,  XXX.  1.  (Jun.  1911)  p.  65. 

Arbor;  ramulis  ab  initio  glabris;  gemmae  acutae  pauci- 
perulatae.  Folia  membranacea,  glabra,  juniora  subtus  in  nervis 
parce  et  ad  axillas  costarum  copiose  rufo-puberula,  ovata  vel 
ovato-oblonga,  caudato-acuminata,  basi  aperte  cordata,  subae- 
qualiter  serrata,  sublobulata  immixta;  serraturis  patentibus;  basi 
5-nervata  ceterum  4 — 6  penninervia,  6 — 10  cm.  longa,  3 — 4.5  cm. 
lata;  petiolis  gracilibus  limbo  brevioribus,  3 — 5  cm.  longis. 
Racemus  sat  multiflorus,  glaber  vel  parce  pube  rufescenti 
tectas;  floribus  andro-dioeciis,  foliis  coaetaneis.     Sepala  5,  obovata 


lö  Art.  1. — G.  Koidzumi. 

obtusa.  Petala  5,  subspathulata,  quam  sepala  loiigiora,  circ. 
3 — 3.5  mm.  longa,  eroso-serrata.  Discus  intrastaminalis. 
Ovarium  glabrum;  stylis  quam  stigmata  revoluta  brevioribus. 
Samarae  cum  ioculis  20 — 22  mm.  longae;  alis  angulo  obtuso 
divergentibus. 

Nom.  Jap.      Onaga-liaede  (M.  Shirai) 

Hab.     ia  sylvis  temperatis  Formosae:     Arisan,  Hakkutaisan. 

DisTR.     enderaica. 

Nota,  species  arete  affinis  A.  laxifforo,  Pax,  ab  hoc  diversa 
petalis  spath ulatis  eroso-serratis;  fohis  non  crenulatis. 

6.     Acer  ovatifolium,  sp.  no  v.  (Tab.  VI.) 

Arbor;  cortice  ramulorum  laevi,  luteo-viride.  FoHa  mem- 
branacea,  adulta  utrinque  glabra,  ovata  rarius  ovato-oblonga, 
caudato-acuminata,  basi  rotundata  vel  aperte  cordata,  margine 
minute  crenulato-serrulata  v.  crenulata,  interdum  sublobulata 
immixla,  basi  5-nervata  ceterum  5— G  penninervia;  petiolis 
glabris  1.5 — 2.0  cm.  longis.  Inflorescentia  andro-dioecia,  multi- 
flora,   racemosa,    dense  ferrugineo-tomentella.      Flores 

Fructus  glaber,  circ.  1.5  cm.  longus,  alae  angulo  recto  vel  obtuso 
divergentes;  pedicellis  brevibus  4 — 5  mm.  longis. 

Nom.  Jap.      Koha-oncuja-haede  (nom.  no  v.) 

Hab.     in  sylvis  temperatis  Formosae:      Ako  (Yokusensha). 

Nota.  Species  ab  A.  Hookcri,  MiQ.  quocum  junxit,  distinctis- 
sima  ist  foliis  minoribus  4.5 — 6.5  cm.  longis,  2 — 3  cm.  latis, 
minute  crenulatisque. 

7.     Acer  morrisonense,  Hayata,  sp.  nov.  (Tab.  VII.) 

Arbor;  ramis  glabris  nigro-viridescentibus;  ramulis  juveni- 
liljus  parce  rufo-puberulis.  Gemmae  ovatae  perulis  paucis 
praeditae.  Folia  membranacea,  cito  glabra,  juniora  utrinque 
praesertim  subtus  secus  nervös  rufo-puberula,  leviter  triloba  vel 
trilobatisve,  rarius  indivisa  immixta,  basi  aperte  cordata,  apice 
longe  acuminata,  subduplicato-serrulata,  utrinque  5 — G-costata, 
G — 11  cm.  longa,  3 — 7  cm.  lata;  lobis  lateralibus  multo  minoribus 


Kovisio  Aceracoaruui  Japonicarum.  17 

obtusis;  petiolis  gracilibus  3 — 4  cm.  longis.  Flores  andro-dioeci, 
ill  race  mum  pendulum  parce  puberulum,  ramulo  bifoliato  in- 
sidentem  dispositi;  pedicellis  capilliformibus  ad  8  mm.  longis. 
Sepala  5  ovata,  obtusa,  virescentia.  Petala  5  sepalis  aequilonga, 
rotundato-ovata,  flavescentia.  Stamina  8,  antheris  Jaevibus; 
filamentis  subulatis  glabris.  Discus  lobatus  intrastaminalis. 
Fructus 

Nom.  Jap.      Takasiujo-wihaetlc.  (nom.  nov.) 

Hab.     Formosa:      in  monte  Morrison. 

DiSTE.     endemica. 

Nota.  Species  affine  videtur  A.  crataegifulio,  S.  et  Z.,  sed  ab 
hoc  et  a  reliquis  speciebus  Indivisoriun  foliis  saepius  leviter 
trilobis,  utrinque  viridibus,  subtus  tenuissime  reticulatis  manifeste 
distincta. 


Sect.  3.  Carfiinifolia,  m. 

Inflorescentia  racemosa,  ^  saepe  umbellata  vel  subcorym- 
bosa.  Flores  tetrameri,  andro-dioeci.  Sepala  4.  Petala  4. 
Stamina  5 — 6,  hypogyna.  Discus  intrastaminalis.  Stylus  ad 
basin  partitus,  laciniis  subulato-curvatis.  Folia  membranacea, 
utrinque  multicostata. 

Species  unica  Japonensis. 

<^.     Acer  carpinifolium,  S.  et  Z.  (Tab.  VIII.) 

SxEEOLD  ct  ZuccARiNi,  Fl.  Jap.  Fam.  Xat.  in  Abb.  Akad.  Müuch.  IV.  2.  (1S46)  loi,  et  Fl. 
Jap.  IL  (1870)  81,  t.  142  ;-Fß.  et  Sav.  En.  PL  Jap.  L  (1875)  89  ;-MrQ.  ProL  FL  Jap.  (1866)  21  ;- 
Mas.  Mél.  BioL  X.  (1880)  595  ;-Pax,  in  Engl.  Bot.  Jahrb.  VIL  (1836)  217,  et  Engl.  Pfl.  Reich. 
8  Heft  (1902)  34  ;— Leveil.  Bull.  Soc.  Bot.  Fr.  VI  (1906)  539  j-C.  K.  Schn.  IlL  Handb.  Laubh. 
IL  (1907)  215. 

Nom.  Jap.  Yauiüs]uha-];acd('^  Cliidorinoki,  Shinnsliide  (i^m:^^); 
Shirn-shide  (^i^-h^.  ^fm^mm-^ïâW.:^-^.  imj);—Taniasa  (^i^^^. 
(mm'±^mm-W:^mm.  18SS);—Ara]iagj,  Tsuhanoh  (indigenis  Chichi- 
buensis  vocatur). 

Arbor;    rainulis    novellis   glabris   vel   sparce   villosis,    cortice 


18  Art.  l.—G.  Kfidzumi. 

ramorum  cinereo-fusescente.  Gemmae  perulae  intimae  membra- 
naceae  oblanceolatae  acuminatae  versus  basin  valde  angastatae. 
Polia  membranacea,  supra  mox  glabra,  subtus  praesertim  ad 
venas  adpresse  pubescentia,  oblonga,  acuminata,  basi  rotundata 
saepe  1  éviter  cordata  vel  subauriculata,  utrinque  argute  incisoser- 
rata,  parallele  multi-costulata  ;  petiolis  brevibus  glabris.  Inflores- 
centia  foliis  coaetanea,  laxe  puberula.  Flores  andro-dioeci,  albi 
vel  lutesecntes,  '^  longe  racemosi;  $  umbellati,  corymboso- 
cymosi,  vel  racemosi;  pedicellis  capilliformibus  tenuissimis. 
Calyx  extus  ciliatus  ;  sepalis  4  oblongis  apice  rotundatis.  Petala 
4,  oblonga  apice  obtusa,  saepissime  ciliolata,  sepalis  aequilonga, 
in  floribus  1^  caduca.  Stamina  5 — 6  inclusa;  antheris  laevibus. 
Discus  lobatus  intrastaminalis.  Ovarium  margine  pilosum, 
«eterum  glaberrimum;  stylis  brevissimis  subnullis;  stigmatibus 
■elongatis  subulatis.  Samara  glabra,  loculis  planis  ovatis;  alis 
angulo  obtuso  divergentibus;  pedicellis  elongato-capilliformibus, 
fere  saemper  nutantibus. 

Hab.  Hondo:  Nambu  (Rikuchiu),  Aizu  (Iwashiro),  Nikko 
(Shimotsuke),  Chicliibu,  Ohdake  (Musashi),  Olij^ama,  Hakone 
(Sagami),  Usuitöge,  Ontake,  Toriitöge,  Höfukujitöge,  Togakuslii- 
yama,  (Shinano),  Amagisan,  (Idzu)  Prov.  Kii,  Prov.  Nagato. 
Sikok:     Prov.  Tosa;  Prov.  Avva.  Kiusiu:     Kudshusan. 

DisTK.     endemica. 

Nota.  Folia  ad  16,5  cm.  longa,  8,5  cm.  lata;  samara  ad 
3  cm.  longa,  1,3  cm.  lata. 


Sect.  4.  Macrantha^  Pax. 

Macrantha,  Pax,  ia  Engl.     Bot.  .Jahrl).  VI.  (1885)  328,  et  VII.  (I88ü)  24t. 

Inflorescentia  racemosa.  Flores  andro-dioeci  vel  monoeci, 
pentameri.  Stamina  8  hypogyna,  in  flore  "^  perigyna.  Discus 
intrastaminalis.  Antherae  laevae,  ellipticae.  Ovaria  pl.  m.  pube- 
rulento-tomentosa;  stylis  profunde  bilobatis.  Folia  membranacea, 
3 — 5-palmatilobata,   inaequaliter  serrata. 


Revisio  Acsracearum  Japoniciirum.  !•) 

Clavis  specierttui. 

1.  Folia  trilobata,  rarius  simul  sub  pentaloba  iramixta 2 

Folia  quinquelobata 3 

2.  Folia  ambitu  obovata,   subtus  ad  venas  inflorescentiaque  rufo- 

feiTUginea.     Flores  médiocres,  pedicellis  brevibus.     Fructus 

alae  16 — 20  mm.  longae,  10  (7 — 10)  mm.  latae 

A.  riifinerce,  S.  et  Z. 

Felia  ambitu  ovata  glabra.  Flores  minuti,  pedicellis  gracile 
elongatis,  circ.  10  mm.  longis.  Fructus  alae  circ  5  mm. 
latae,  10  mm .  longae A.  capillipes,  Max. 

3.  Folia   glabra,    5-nervata,    ceterum   G — 11-penninervia;    lobis 

acuminatis.     Fructus  pedicelli  capilliformes   10 — 12   mm. 

longi A.    capillipes^   var.  fujisancnse,   m . 

Folia  juuiora  subtus  rufo-birta,  5-nervata,  ceterum  4—5 
penninervia;       lobis      caudato-acuminatissimis.       Fructus 

pedicelli  7 — 10  mm.  longi;  alis  majoribus  

A.   riihesœns,  Hay  ata. 

9.     Acer  rufinerve,  S.  et  Z.  (Tab.  IX). 

SiEBoLD  et  ZuccARiM,  Fl.  Jap.  Faui.  Nat.  in  Abh.  Akad.  Miiuch.  IV.  2.  (184'3)  155,  et  Fl. 
Jap.  IL  (1S70)  85,  t.  14S  ;— Miq.  Prol.  Fl.  Jap.  (1866)  20  ;— Koch  in  Ann.  Mus.  Lugcl.  Bat.  I. 
(1863—64)  251;  — Walp.  Ann.  I.  961  ;— Max.  Mél.  Biol.  X.  (1880)  593  ;— Pax,  in  Engl.  Bot. 
Jahrb.  VII.  (1886)  247,  et  Engl.  Pfl.  Eeich.  8  Heft  (1902)  69  ;— Fr.  et  Sav.  En.  PL  Jap.  I.  (1875) 
89;— Leveil.  Bull.  Soc.  Bot.  Fr.  VI.  (1906).  593  :—C.  -v.  Schn.  111.  Handb.  Laubh.  II.  (1907) 
237. 

A.  pennsylvaniciim,  var.  rufinerve,  Wesml.  in  Bull.  Soc.  Bot.  Belgique,  XXIX.  (2890)  62. 

A.  rufinerve,  forma  normale,  Gß.  v.  Schw.  in  Gartfl.  (1893)  454. 

Nom.  Jap.  Urlhada-hude  (m %'a ^) ;  — lizulu,  Konjinoki  (%m-h¥- 
^i^^lEm-  H3fellJ^7tc|ffl.  1825)  ;  —Küinuri-kaede,  Ohba-winoki  (X  ^  A  if. 
7K  ^  Äl  5^  ^-  =f^  p^n  m^ik  a.)  182G  ;  —Ao-kacde,  Oh-minekacde  (5?  1*  A  ^.  n^ 
B  22  m  ^-  *E  ii  ^  Sß  iJ;-  1837)  ;  —  (Jrikko  (JtmmmM^);  —Ao-momiji  (?  m 
7Ê  ¥■  /h  fr  M  m  -^-  -^  Sf  tS  ^  iE-  1801)  ;  Uri-i  (nom.  indig.  Chichi- 
buensis);  Oh-urik.mU  (J.  ^Matsumuea,  Shokubutsu-Meii,  no.  45); 
Urinoki  (indig.  Provinciae  Settsu  vocatur). 

Arbor  vel  arbusculus,  ramis  laevibus,  cortice  luteo-virides- 
ceiite  nigricanto-striato  ;  ramulis  novellis  glabris;  gemmae  perulae 
spathulato-oblongae  extus  rufo-tomentosae.     Folia  membranacea, 


20  Art.  1.  — G.  Koidzuuii. 

fiubtus  secus  nervös,  mox  ad  veiiarum  axillas  tantuni  rufo-barbarta, 
ambitu  obovata,  basi  leviter  cordata  vel  i'otuiidata,  triloba,  vel 
quinquelobata  si  lobis  basalibus  adsunt  minimis,  inciso-vel 
subduplicato-serratis  ;  petiolis  denium  glabris.  Racemi  foliis 
coaetanei  rufo-pubescentes;  floribus  sulphureis  glabris,  andro- 
dioeciis;  pedicellis  ebracteatis,  flore  minoribiis  eirc.  3 — 5  mm. 
longis.  Sepala  5  spathulato-oblonga,  apice  rotundata.  Petala  5, 
obovato-oblonga  apice  emarginata  vel  rotundata,  sepalis  aeqni- 
longa.  Stamina  3,  inclusa,  antheris  laevibus.  Discus  intrastami- 
nalis,  margine  crenato,  intus  glaber.  Ovaria  puberulento- 
tomentella;  stylis  bifidis,  laciniis  subulatis  intus  longe  stigmatosis. 
Alae  samarae  5  m.m.  latae,  10  m.m.  longae,  subparallelae  vel 
angulo  acuto  interdum  fere  recto  divergentes,  loculis  rufo- 
tomentellis  multo  longiores.     Maio  fl.,  Augusto  fr.  mat.. 

Hai].  Hando:  Nanshözan,  Himekamiyama,  Iwatesan, 
Goyözan  (Rikuchiu),  Chökaizan  (Ugo),  Azumasan,  Gassan, 
Asahidake,  Ichinenhö,  lidesan  (Uzen),  Aizu  (Iwashiro),  Nikko, 
Shimidzutöge  (Shimotsuke),  Hakone  (Sagami),  Amagisan  (Idzu), 
Fujisan  (Suruga),  Hakusan  (Kaga),  Tateyama  (Yettchiu),  Kasu- 
gasan  (Yamato),  Prov.  Nagato. — Shikok:  Prov.  Awa,  Tosa, 
Sanuki. — Kiusiu:  Iwatake  (Buzen). 

DiSTE.     endemica. 

Xota.  Afiine  Acer  pcjnisijîudnico,  et  A.  üujmcntoso,  sed  a  priori 
pedicellis  florum  rufo-pubescentibus;  fruc^us  loculis  quam  ala 
inulto-brevioiibus,  fructus  pedicellis  baud  ultra  5  mm.  longis; 
ab  altero  foliis  iructibusc[Ue  non  glabris  differt. 

furiiia   albo-limbatum,    Hook.    fil.     in  Bot.  :5[ag.  (isa)),  t.  579  5;-Pax, 

Engl.  Tfl.  Keich.  8  Heft  (1902)  G9. 

A.  Tul'incrve,i.  manjitinium,  et  m  irmora'nm.  P.vx,  i.i  Engl.  Bot.  Jalir'i.  VII.  (183G)  247. 

Folia  albo-pulvurulento-marginata  vel  -maculata  glabriuscula; 
racemis  glabris  vel  ]-arius  parce  puberulis,  i)edicellis  infericribus 
longioribus  cire.  10  mm,  longis;  fructibus  glalnis  alis  angulo 
obtuso  divergentibus. 

Nom.  .Jap.      i'uiri-unhadakacik  (nom.  nov.) 
JJühui/nki-kaalc  (bort,  nom.) 

IIa  15.  culta. 


Rovisio  Aceracearum  Japoniearnm.  21 

10.     Acer  capillipes,    Max.  (Tab.  X.) 

C.  J.  Maxuiowicz,  in  M.l.  Biol.  VI.  (1867)  ?.67.  et  X.  (ISSO)  593  ;— Fr.  et  Sav.  En.  PI.  Jap.  I. 
(1875)  ,88  ;— Pax,  in  Engl.  Bot/Jahrb.  VII.  (188G)  2-46,  et  Pfl.  Reich.  8  Heft  at.02)  67  ;— Leveil. 
in  Bull.  Soc.  Bot.  Franc.  VI.  (1806)  589  ;—C.  K.  Schn.  III.  Hanclb.  Laubh.  IT.  (1907)  238  ;— 
Making,  in  Bot.  Mag.  Tokyo  XXIV  (1910)  p.  292. 

A.  p>"i.mylraiiku)n,  var.  capdlipc",  VVesmael  in  Bull.  Soc.  Bot.  Belgique,  XXIX.  (1890)  62. 

Nom.  Jap.  Oh-karahana  (-^i&-h¥-  ^^^  lE^.  U  3fe  lü  :?: ?l^ ü. 
l&2ô);~Hosoe-l-aedc  (R.  Yatabe  n:^m^^m.  vul.  I.  (1900)  p. 
411.) ;—Äshiboso-iiriNol-i  ( M .  Shirai  B:^mm  ^  mm  1 908. ) ;  —Hosoc- 
un'hada  (nom.  nov.) 

Arbor  glaber  10  metralis.  Rarnuli  cortice  laeve,  luteo-viridis- 
cente  iiigricanto-striato.  Folia  membranacea  ambitu  ovata,  tri- 
lol:»ata  rarius  subquinquelobata.  basi  rotundata,  lobis  lateralibus 
vakle  minoribus,  omnibus  acuminatis  subduplicato-serratis 
serraturis  ovatis  nmcronatis;  petiolis  quam  limbo  brevioribus  basi 
parum  incrassatis.  Flores  andro-dioeci,  foliis  coaetanei,  minuti,  in 
racemum  multiflorum  pendentem,  elongatum,  pedunculo  3 — 5  cm. 
longo  insidentem  dispositi;  pedicellis  capilliformibus  circ.  10  mm. 
longis.  Sepala  oblonga  obtusa  virescentia.  Petala  oblonga  sepalis 
aequilonga  lutescentia.  Stamina  8  antheris  laevibus.  Discus 
intrastaminalis,  mai-gine  crenato.  Ovaria  parce  puberula,  st}dis 
bifidis,  stigmatibus  snbulatis  exertis.  Samara  cum  loculis  13-14 
mm.  longa,  ad  4 — 5  mm.  lata;  alis  angulo  obtuso  divergentibus; 
pedicellis  10-12  mm.  longis. 

Hai;,     in  sylvis  Hondo:  Nikkö,  Ontakesaii,  Fujisan. 

DiSTR.      endemica. 

var.  fujisanense,    m. 

Folia  ambitu  subquadrangulata,  di.-^tincte  quinqaelobata  lobis 
extimis  multo  minoribus. 

Nom.  Jap.      Hirolia-asliihoso-urinnlii. 

Hae.  Fujiyama. 

11.     Acer  rubescens,  Hayata,   sp.  nov.  (Tab.  XI.) 

Ai'boi-,  rami  purpureo-viridi  glabri  ;  gemmae  magnae  ovoideae 
pauci-perulatae.  Folia  membranacea,  elongato-petiolata,  juniora  ad 
basin  costaruin  subtus  rufo-I)arbarta,  mox  glabi-a,  ambitu  rotundato 


22  Art.  1.— G.  Koidzumi. 

vel  subqnadraiigulato-ovata,  quinquelobata,  basi  cordata  pentaner- 
via,  ceterum  4 — 5  penninervia,  6 — 9  cm.  longa  4 — 7  cm.  lata; 
lobis  inaequaliter  serratis  caudato-acuminatis  ;  petiolis  gracile 
elongatis,  laminae  aequantibns  vel  paulo  brevioribus.  Inflorescentia 
in  ramulis  2  foliatis  terminalis,  fructifera  glabra  elongato-racemosa. 
Flores  ut  videtur  andro-dioeci,  graciliter  pedicellati.  Fnictus  in 
racemum  pendulum  circiter  ad  8  cm.  longum  dispositi.  Samarae 
plerumque  1,3  cm.  longae  (raro  ad  1,7  cm.  longae),  loculis 
ovoideis  3( — 5)  mm.  longis,  convexis,  parce  venosis;  alls  horizon- 
talibus  vel  angulo  obtusissimo  divergentibus,  medio  4 — 5  mm. 
latis,  basi  contractis;  pedicellis  capilliformibus  7 — 10  mm. 
longis. 

Nom.  Jap.     Tahisago-iinliachuaede  (nom.  nov.) 

Hab.  in  sylvis  temperatis  Formosae:  in  monte  Morrison  (ad 
7000 — 7500  ft.  alt.),  Hokkutaisan,  Arisan,  Kanmutsusha. 

DisTR.     endemica. 

Not.  Species  affinis  Acer  capillipes,  var.  fitjisaaensi,  Koidz., 
distinguitur  foliis  junioribus  parce  hirtis,  utrinque  5 — 6  penni- 
nervis,  lobis  caudato-acuminatissimis;  fructibus  pedicellis  brevio- 
ribus. 

Sect.  5.  Falniatoidca,  m. 

Inflorescentia  racemosa.  Flores  andro-dioeci  vel  monoeci. 
Sepala  petalisque  quinque.  Stamina  8  liypogjaia,  in  fl.  ^  perigyna  ; 
antheris  leviter  scabris,  ovoideis  apiculatis.  Ovaria  glabra  ;  stylis 
apice  bilobatis.  Discus  intrastaminalis.  Folia  membranacea  pro- 
funde 5-lobata,  lobis  incisis  vel  lobulatis.  Species  ad  3  Japoniae 
et  Sinensis. 


Viavis  »pecierum  dichofotna. 

*  Hacemus  15 — 22-floratus,  floribus  minutis,  sepalis  suborbicu- 
latis,  petalis  ellipticis;  fructus  minor  13 — 20  mm.  longus, 
(3—)  4 — 6  mm.   latus A.   ntiminthwii,   S.  et  Z. 

**      Racemus  6— 10-floratiis,  floribus  mediocribus,  sepala  petalis- 


Eevisio  Aceracearum  Japonicarimi.  23 

que  auguste  spathulata;   fructus  major  20 — 25  mm.  longus, 
G — ]  2  m  m .   latus A.    Tschonoskii^   jMax. 

12.     Acer  micranthum,  S.  et  Z.  (Tal).  XII.) 

SiEEoLD  et  ZoccARiNi,  Fl.  Jap.  Farn.  Nat.  in  Abh.  Akad.  Münch.  IV.  2.  (184'3)  155,  et  Fl. 
Jap.  II.  (1870)  80,  t.  141  ;— MiQ.  Prol.  Fl.  Jap.  (I860)  21  ;— Fr.  et  Sav.  En.  PL  Jap.  I.  (1875)  89  ;  — 
Max.  Mél.  Biol.  X.  (1880)  598  ;-Pax,  Engl.  Bot.  Jahrb.  VII.  (18StJ),  24S,  et  Engl.  Pfl.  Reich.  8 
Heft  (1902)  70  ;— C.  K.  Schn.  Ilh  Handb.  Lanbh.  II.  (1902)  233;  — Leveil.  in  Bull.  Soc.  Bot.  Fr. 
VI.  (1906).,  591. 

A.  Tschonoflai,  Komakj.  («;)»  Max.)  Fl.  I\lan.  II.  73-3  (vidi  specim  )  :— Xakai,  FI.  Korea.  I. 
131. 

Nom.  Jap.  Ko-minc-kacde  (J.  Matsümüea,  Shokubutsu-Meii, 
no.  35.) 

Arbuscula,  minis  glabris  viridescentibus;  gemmae  tegmentis 
interioribus  obovato-oblongis  vel  late  spatliulatis  apice  rotundatis, 
utrinque  glabris,  marginibus  ciliolatis  praeditae.  Folia  membran- 
acea  glabra  vel  subtus  in  axillis  costarum  rarius  secus  nervös  rufo- 
barbulata,  palmato-quinquelobata,  basi  cordata  ;  lobis  extimis  ab- 
breviatis  ceterum  lanceolato-ellipticis  vel  ovato-lanceolatis,  caudato- 
acuminatis  argute  serratis,  omnibus  utrinque  incisis;  petiolis  graei- 
libus  quam  lamina  brevioribus.  Raceinus  glaber  circiter  15—22 
florus;  floribus  minutis  pallide  roseis,  foliis  coaetaneis,  andro- 
dioeciis.  Sepala  minuta  ovata  vel  rotundata,  quam  petala  2 — 3- 
plo  breviora.  Petala  obovata  vel  obovato-elliptica  1,5 — 3,0  mm. 
longa.  Stamina  <S  in  sinubus  disci  inserta,  exerta;  antheris  vix 
scabris  apiculatis.  Discus  crenatus  intrastaminalis.  Ovaria  glabra, 
stylis  apice  bilobatis.  Samara  glabra  13 — 20  mm.  longa  (3 — ) 
4—6  mm.  lata;  alis  cum  loculis  horizontaliter  vel  obtusissime 
patentibus. 

Hab.  in  sylva.  Hondo:  Sengantöge,  Himekamiyama,  Omyö- 
jin  (Rikuchiu),  Gassan,  Adzumasan  (Uzen),  Aidzu,  Bandaisan 
(Iwashiro),  Nikko  (Shiinotsuke),  Mitsuminesan  (Musashi),  Hakone 
(Sagami),  Amagisan  (Idzu),  Obara  (Yettchiü),  Tateyama  (Kaga), 
Kiso  (Shinano),  Shikok:  Tsurugisan  (Awa),  Yahazuyama,  Nano- 
gawamura  (Tosa) ; — Kiusiu:  Hikosan  (Buzen),  H(')manzan. 

DisTR.     Korea,  Manshuria. 

Not.      Affine  A.  Maxinwn-iczii^    Fax,   et  Acer  Tsclionoskii,  Max. 


24  Art.  1.— G.  Koidzuini. 

sed  a  priori  foliis  lobis  inter  se  fei-e  aequantibus  ;  ab  altero  floribus 
minutis;  foliis  lobis  caudato-acuminatissimis;  fructilnis  niinoribus, 
i\lis  horizontaliter  patentibus  differt. 

13.     Acer  Tschonoskii,  Max.  (Tab.  XIIT.) 

C.    J.    Maximowicz,    in    Mél.     Biol.    XII.    (188G)    432  ;— Pax,    in    E.s-ol.    Prl.    Eoieh.    8 
Heft  (1902)  70; 

A.  pellucidohracleatum,  Leveil.  et  Vnt.  in  Bull.  Soc.  Bot.  Fr,  VI.  (190G),  592. 

Nom.  Jap.  Mine-kaede,  Hime-ogarabana  (J.  ]\Iatsumura,  Shoku- 
butsu-Meii,  no.  49)]—Hahisa}i-iiwnriji(Jtm^n). 

Arbusculus  dumosus,  ramuli  viridescentes  glabri,  gemmae 
tegmentis  interioribus  ut  in  praecedente.  Folia  membranacea 
supra  glabra  subtus  ad  nervös,  vel  rarius  adulta  tantum  in  axillis 
costarum  rufo-barbarta,  palmato  quinquelobata,  basi  cordata;  lobis 
ovalibus  subito  brève  acuminatis,  argute  serratis  utrinque  sublobu- 
lato-incisis;  petiolis  gracilibus  quam  lamina  brevioribus.  Flores 
andro-monoeci,  5-meri,  in  racemum  circiter  G — 10  ( — 13)  florum 
erectum,  pedunculo  2 — 3  cm.  longo  insidentem  dispositi;  pedi- 
celli  glabri;  sepala  5  lineari-spathulata,  obtusa;  petala  5,  sepalis 
conformia,  ]mllide  rosea,  sed  parant  superantia,  4 — 4,5  mm.  longa; 
stamina  8  petalis  subaequilonga  ;  discus  leviter  crenatus  intra- 
staminalis;  ovaria  glabra  stylo  stigmatibus  revolutis  vix  longiore. 
Samara  loculis  liorizontalibus;  alis  in  angulo  obtuso  divergentibus; 
20 — 2;")  Dim.  longae,  G — 12  mm.  latae. 

Hab.  in  subalpinis  Japoniae  mediae  et  borealis.  Yezo: 
Raidentôge,  Makkarinupuri,  Prov  Tokachi,  Yunosawa, — Hontô: 
Iwakisan  Hakködasan  (Mutsu),  Iwateyama,  Goyôzon,  llayaeliine 
(Rikuchiû),  Kurikomayama  (Rikuzen),  Cbôkaisan  (Ugo),  Gassan, 
Asahidake,  Zaôsan,  lidesan,  Azumasan  (Uzen),  Komagatake, 
INIyokôzan,  Niôzan  (Yechigo),  Osenodaira,  Nikko  (Shimotsuke), 
Tcgakusliiyama,  Önogawa,  Ariakcyama,  Norikurayama.  Ontake 
(Shinano),  Tateyama  (Yettcbin),  Hakusan  (Kaga). 
DisTB.     endemica. 

Nota.  Affine  Acer  Maximowiczii,  Pax,  et  Acßr  micrantlio,  S,  et 
Z.,  sed  a  priori  foliis  lobis  inter  se  fere  subacqualibus;  ab  altero 
floribus  majoril)Us,  petalis  sepalisque  lineari-spatludatis  4 — 4,;")  mm. 


Kevisio  Acerri.ceciruLu  Japonicaruui.  25 

loDgis,  in  raceinuin  (> — 10  florum  dispositis;  foliis  subtiis  plerumque 
secus  venas  rufu-barbartis,  subito  breve  acuminatis;  alis  samame 
obtuse  divergentibus,  latioribusque;  fructibus  uiajoribus  20 — 25 
mm.  loDgis.  (i — 12  mm.  latis,  differt. 


'&>' 


Sect.    C).    Anjvta^    Rehdee. 

RtHDER,  ia  Saegknt,  Trees  and  Shr.  I.  (1935)  131  ;-C.  K.  S.-hn.  111.  Haadb.  II.  2.  (1909/  24k 

Infîorescentia  mascula  secus  ramulos  elongotos  e  gemmis 
lateralibus  coryml)oso-fasciculata  ;  foemina  semper  in  rainulo 
bifoliato  terminalis,  racemosa.  Flores  dioeci,  tetrameri  (sep.  4, 
pet.  4,  stani.  4,  carpel.  2).  Discus  intrastaminalis  4 — lobatus. 
Stamina  hypogyna.  Stylus  profunde  bilobatus.  Folia  mem- 
branacea  palmatilobata. 

14.     Acer  argutum,   Max.   (Tab.  XIV.) 

C.  J.  Maximowicz,  in  Mel.  Biol.  VI.  (1S67)  36S,  et  X.  (ISSö)  594;  — Fr.  et  Siv.  En.  PI.  Jap.  I. 
(1875)  91  ;— Pax,  in  Engl.  Bjt.  Jahrb.  VIF.  il58Ü)  252,  et  Pfl  Reich.  8  Heft  (1902)  72  ;— Levkil. 
in  Ball.  Soc.  Bot.  Fr.  VI.  (1936)  589  ;— C.  K.  S.^hn.  Hl.  H.indb.  Laubh.  II.  (1909)  244. 

A   diahoUcum,  subsp.  arputum,  Wksm.\el,  in  Bull  Soc  Bot.  Belgique,  XXIX.  (1S?0)  C3 

Nom.  Jap.  Jsanolia-kaedi',  Mii/iiiiia-iiioin/ji  (,J.  Matsümura, 
Shokubutsu-Meii,  no.  2G). 

Arbor  ramulis  novellis  incano-puberulis;  gemmae  tegmentis 
interioribus  oblongis  vel  ellipticis  apice  plerumque  obtusis  utrinque 
puberulis.  Folia  membranacea  juvenilia  subtus  incano-pubes- 
centia  supra  glabriuscula  vel  ad  venas  parce  puberula,  adulta 
tantum  subtus  secus  costas  pilosiuscula,  circumscriptione  orb- 
icularia,  quinquelobata  basi  aperte  cordata,  lobis  ovatis  acuminatis 
utrinque  arguteinciso-serratis;  petiolis  gracilibus  quam  lamina 
brevioribus  superne  parce  pilosiusculis.  Inflorescentia  primum 
corymbo.sa  tum  racemosa;  floribus  lutescentibus  vel  albis  dioeciis 
foliis  coaetaneis;  pedicellis  giabris  gracile  elongatis;  bracteolis 
minulis.  Calyx  glaber,  sepalis  4  auguste  oblongis  vel  oblongo- 
lanceolatis  oljtusis.  Petala  4  sepalis  breviora.  Stamina  4  (in  fl.  ^ 
abortivi)  exerta,  filamentis  ontherisc{ue  giabris  in  faciebus 
exterioribus    disci   insertis.      Di-^cus   carnosus  4-lobatu-,    in    flore 


2i3  Art.  1. — G.  Koidzuini. 

^  rudimentarius  ad  dentés  reductus.  Ovaria  glabra,  st3dis  ad 
basin  fere  bipartitis.  Fructus  glaber  in  racemuin  pendentem 
dispositi,  locnlis  nlisque  horizontaliter  patentibus;  pedicellis  gracile 
elongatis. 

Haiî.  Mundo:  Nikko,  Ashio  (Shimotsuke),  Chichibu  (Musa- 
sbi),  Ariakeyania,  Togakushiyama,  Kiso  (Shinano),  Taira-no-ya 
(Hida)., — Shikok:  Tsurugisan  (Awa). 

DiSTR.     endernica. 


II.     Extrastaminalia,   Pax. 

h.rirastavüiialici.  Pax,  in  Engl.  Bot.  Jahrb.  VI  (18S5),  326. 
Adiscantlin,  Pax,  1.  c,  327. 
l'erigyna.  Pax,  1.  c.  327  (pro  parte) 


Sect.  7.   Cissîfolia,  m. 

Inflorescentia  elongato-racemosa.  Flores  dioeci  niinuti. 
Sepala  petalisque  4.  Stamina  4  ( — 5)  bypogyna.  Stylus  profunde 
bilobatus.  Discus  extrastaminalis.  Folia  membranacea  ternata, 
foliolis  petiolulatis.     Species  unica. 

15.     Acer  cissifolium  (S.  et  Z.)  Kocii.  (Tab.  XY.) 

C.  Koch,  in  Acn.  Mus.  Liigtl-Batav.  I.  (1864)  252  ;— Fr.  et  Sav.  En.  PI.  Jap.  I.  (1875)  90  .:-Max. 
Mél,  Biol.  X.  (18bO)  610  ;-Pax,  in  Engl.  Bot.  Jahrb.  VII.  (1886)  204.  et  Pfl.  Etich.  8  Heft  (1902) 
29;— Leveil.  in  Bull.  Soc.  Bot.  Fr.  VI.  (ISCC)  589  :-C.  K.  Schn.  111.  Handb.  I  aubh.  II.  (1907) 
210. 

Xet/iiiido  c iy." ijoliit III,  S  1KB.  et  Zacc.  Fl.  Jap.  I  am.  Nat.  in  Abb.  Akad.  Münch.  IV.  2.(1846) 
159  ;— MiQ.  Prol.  Fl.  Jap.  (1866)  22  (?) 

Xeiiundo  (?)  nihoeniii',  Miq.  Prol.  FI.  Jap.  22.  (1866). 

Negimdo  nikocnfe,  Miq.  Archiv.  Neel.  t.  II.  (lSö7). 

Nom.  Jap.  Mitsiule-kaede  {%ms.¥-  m^n^Tî^m.  W^mum- 
]>^-2-2);—Mitsudc-momiji\  Auniialio-lardc  (%îAA¥-  7K^m:^.M-  ^tJj  nT.  i^ 
«  fô  îfi-  ]  825)  ;—Anialwgi  (^W^^Pkmm);  —Amahild  (-  *  fö  !j^  ^  m  m). 

Ar))or  innovatione  pubescente;  gemmae  perulis  interioribus 
oblongis  apice  rotundatis  dense  rufo-tomentosis.  Folia  mem- 
branacea trifoliolata,  nascentia  supra  dense  subtus  ad  costas  petio- 
lisque  pubescentia,  adulta  utrinqiie  glabra,  aut  supra  pilosa  et  tunc 


Eevisio  Aceracearum  Japonicarum.  27 

subtus  in  axillis  venarum  tomentella;  petiolis  communis  elongato- 
gracilibus  mox  glaberrimis;  foliolis  petiolulatis  ellipticis  vel  oblon- 
gis  saepe  ovato  vel  obovato-oblongis  interdum  versus  basin 
cuneatis  aut  subrhombeo-oblongis,  apice  in  acumen  longum 
acutum  integerrimum  productis,  sursus  pauci  grosse  dentato- 
serratis  rarius  incisis  vel  integerrimis,  serraturis  cuspidatis.  Inflore- 
scentia  longe  racemosa,  pedunculis  pedicellisque  pubescentibus, 
floribus  minutis  lutescentibus  dioeciis.  Calyx  4-fidus  extus 
puberulus,  laciniis  ovatis  vel  lanceolato-ovatis  acutiusculis.  Petala 
4,  auguste  spathulata  quam  sepala  duplo  longiora.  Discus  extra- 
staminalis  lobatus.  Stamina  4  ( — 5),  filamentis  exertis  auguste 
fusiformibus.  Ovaria  glabra,  stylis  brevissimis,  stigmatibus 
recurvatis.     Fructus  glaber,  alis  angulo  acuto  divergentibus. 

Hab.  Yezo:  Niikapp  (Hidaka). — Hondo:  Hirosaki  (Mutsu), 
Hiraidzumi,  Kukaitoge,  Tsunatoriyama,  Kadoma  (RikuchiQ), 
Aidzu,  lidesan  (Iwashiro),  Nikko  (Shimotsuke),  Hakone  (Saga- 
mi),  Ontake,  Norikuradake  (Shinano),  Sbimokura  (Bittchiu), — 
Sikok:    Nanogawa  (Tosa),  Tsurugisan  (Awa). 

DiSTR.     endemica. 

Sect.   8.  Piuhra,   Pax. 

F.  Pax,  in  Engl.  Bot.  Jahrb.  VI  (1S85)  32Ö,  et  VII.  (1836)  179. 

Inflorescentiae  secus  ramulos  fasciculatae.  Flores  andro- 
dioeci  longe  folia  ante  nascentes.  Sepala  petalisque  5.  Stamina 
5 — 8  hypogyna.  Discus  valde  abortivus.  Stylus  profunde  bilo- 
batus.  Folia  membrauacea  palmatilobata.  Gemmae  perulae  in- 
teriores  non  elongatae.     Species  uuica. 

IG.     Acer  rubrum,   L.  (Tab.  XVI.) 

LiNN.  Sp.  PI.  ed.  1.(1753),  1055;— DC.  Prodr.  I.  (1824)  595  ;— Eobinson,  in  Geat,  Syn.  Fl.  I. 
(1897)  437;— Pax,  in  Engl.  Bot.  Jahrb.  VII.  (1886)  181,  et  Pfl.  Eeich.  8  Heft  (1902)  37;— T. 
Nuttal,  in  North  Am.  Fl.  (1835)  II.  34  ;-Makino,  in  Bot.  Mag.  Tokyo.'XVI.  (1902)  93. 

A.  pycnanthum,  C.  Koch,  in  Miq,  Ann.  Mus.  Lugd.  Batav.  I.  (1863-64)  250;— S.  et  Z.  Fl. 
Jap.  IL  (1870)  86,  t.  143,  fig.  I.  et  1—4  ;— Miq,  Prol.  Fl.  Jap.  (I860)  21  (p.p.)  ;— Fb.  et  Sav.  En. 
PI.  Jap.  I.  (1875),  II.  (1879)  322  ;— Max.  Mél.  Biol.  X.  (1880)  591  ;— Pax,  in  Engl.  Bot.  Jahrb. 
VII.  (1886)  254. 

A.  semiorbinduium,  Pax,  1.  c.  (1886)  181. 


—  J  Art.  1. — G^.  Koi'lzuuù. 

Nom.  Jap.  Hunanoki.  (X  ft  p^  ¥■  A  g:  Sj /;  ^.  i^S^m^-  189(');  — 
Hana-h.iedc  (.J.  Maïsumura,  Shokubut.-u-Meii,  iiu.  44); — M((jnsiiri- 
noki  {^"^.^^m-  B  *  ^  4^  ^.  i£  ö:  SR.  fol.  28.  fig.) 

Ai'bui"  iimgiia,  cortice  ramulorum  glaucesceiite  ;  geinmae 
pcrulae  coriaceae  ovatae  vel  ovato-rotundatae  obtusae  intus 
tomentosae;  innovationibus  foliis  junioribusque  rufo-pubescentibus 
mox  glabris.  Folia  chartacea,  subtu.s  intense  glauca,  trinervia  vel 
subpentanervia,  trilobata,  indivisa  immixta,  basi  rotundata  vel 
leviter  cordata  interdum  rotundato-obtusa,  obtusiter  acaminata, 
daplicato-inciso-serrata,  serratnris  obtusis,  circ.  .3 — 7  cm.  lata  5 — 9 
cm.  longa;  petiolis  gracilibus  ad  G  cm.  longis.  Flores  andro-dioeci, 
rubri,  secus  i-amos  fasciculato-umbellati,  foliis  praecociores  ; 
pedicellis  glabris  filiforniibus;  perulis  rubris  vel  coccineis.  Bepala 
lanceolata  acuta,  petal  is  spathulatis  obtusis  subaequantia.  Stamina 
8,  antheris  rubris  oblongis.  Discus  abortivus.  Stigmata  2  ovario 
glabro  inserta  elongata  undique  papulosa.  Fructus  glaber,  alis 
rectis  medio  lati.ssimis  in  angulo  acuto  divergentibus. 

Hab.      Hondo.  Prov.  Mino  et  C)hmi. 

DisTR.     in  America  boreali  atlantica,  a  Canada  ad  Floridam. 


Sect.  1).   bpic-ita,   Pax. 

F.  Pax,  in  Engl.  Bot.  Jahrb.  VI.  (1335)  32o,  et  VII.  (18S6)  132. 

Intlorescentia  paniculata  vel  racemosa,  interdum  cor3'mboso- 
paniculata.  Flores  andro-polygami.  Sepala  peatalisque  5. 
Stamina  8  hj^pogyna.  Discus  carnosus  extrastaminalis.  Folia 
simplica  o  vel  5 — 7  lobata,  memln-anacea  raro  coriacea. 

Vitivis  Mpeciernut  ffic/iofoma. 

1.  Antherae  scabrae;  folia  r)-lul>ata 

A.   sjnrjiuiii^  viw.  iikiiriDiJihiisr^  Max. 

Antherae  laevae 2. 

2.  Folia  palmato-ö-lobata  argute  serrulatn.  ...--i.  Olivrrnuiwii,  Pax. 
Folia  trilobata 3. 


Eevisio  Acerace:irnm  Japonicarum.  2j 

3.     Folia  adultii  integerrima A.  irifiduui.  Hk.  et  Arn. 

Folia  irregularitev  serrata ...A.  (iinnala,  Max. 

Folia  aequaliter  argute   serrulata 

A.    Oliccritinuiii^    var.    Nakaliarac,   svav.   triloliatuiii,   m. 

17.     Acer  trifidum,   Hk.  et  Akn.  (Tab.  XVII.) 

Hooker  et  Arndt,  ia  Bot.  Beech.  Voj.  (ISil)  17-i  {nee  thg)  ;— Sieb,  et  Zocc.  Fl.  Jap.  Farn.  Xat 
in  Abb.  Akad.  Münca.  IV.  2.  (134G)  157  ;.— Walp.  Ann.  I.  961  ;— C.  Koch,  in  Miq.  Ann.  Mns. 
Lugd.  Bativ.  I.  (1363—61)  251;— Miq.  Prol.  Fl.  Jap.  (1833)  19  ;— Fr.  et  Sav.  En.  PI.  Jap.  I 
(1875)87,11.  (1879)  186;— Fosb.  et  Hemsl.  Jour.  Lian.  Soc.  XXIIf.  (1836-88)  142  ;— Pax  in 
Engl.  Bot.  Jahrb.  VII.  (1883)  186,  et  Pfl.  Eeich.  8  Heft  (1902)  10  ;— Max.  Mél.  Biol.  X.  (1880)  603  ; 
— Mak.  Bot.  Mas.  Tokyo,  XV.  lli  ;— Leveil.  Bull.  Soc.  Bot.  Fr.  VI  (19J6)  593  ;—C.  K.  Sohn. 
111.  Handb.  Laubh.  II.  (1907)  197 

A.  Buergerianuvi,  Miq.  Prol.  Fl.  J.ip.  (18J6)  20  ;  — Fr.  et  Sav.  En.  PL  Jap.  I.  (1875)  88. 

A.  palmatuin,  var.  trilohum,  C.  Koch,  in  Miq.  Ann.  Mus.  Li\gd.  Batav.  I.  (1834)  251. 

A.  trincrve,  Dippel,  Laubh.  IL  428  (1892)  ;— Pax,  in  Engl.  Pfl.  Eeich,  8  Heft  (1902)  12  ;— 
Mak.  Bot.  Mag.  Tokyo,  XIV.  p.  136. 

Nom.  Jap.  Tölutcde  (nom.  viilg.); — Kahtduinn  (yci^Mir  :^^-  ■^li 
±^m-  mn:^:^^m  1 820)  ;  —Hanakazara  (X  ï*  A  ¥■  7K  ^  Kl  y\  M-  ^f^  «*!.  m 
=Sfèît-  1825). 

Arbor  magna,  ramis  glabris,  innovationibus  tomentosis. 
Folia  nascentia  pannosa,  demnm  glabra,  raembranacea  vel  tenuiter 
chartacea,  supra  nitida  subtus  pallide  viridia  vel  glaucescentia, 
triner  via  trilobata,  indivisa  immixta,  basi  cuneata  vel  rotundata, 
lobis  jolerumque  aequilongis  plus  minus  acutis  integerrimis  rarius 
pauci-serratis  ;  petiolis  glabris  laminae  aequantibus  vel  brevioribus. 
Flores  andro-polygami  lutescentes  foliis  coaetanei  pentameri,  in 
paniculam  compositam  dispositi;  pedicello  tomentoso.  Sepala 
ovata  acuta  glabra.  Petala  sepalis  longiora  lineari-lanceolata. 
Discus  extrastaminalis  pl.m.  carnosus.  Stamina  8  inclusa;  an- 
theris  innatis  laevibus.  Ovarium  bilobum  pilosum;  stigmata 
fîliformia  circinato-revoluta.  Samarae  glabrae  ad  2  cm.  longae; 
alis  parallelis  erecti-;  5  —  8  mm.  latis,  saejoissime  sese  invicem 
obtegentibus. 

Hab.     Formosa. 

DisTPv.     in  China  jiu.-trali. 

Not.     Species  valde  affinis  Acer  l'axü,  Feaxch.     distinguitur 
foliis  membranaceis,  fructus  alis  parallele  erectis. 


30  Art.  1.— G.  Koidzumi. 

var.  ningpoense,   Hange,  in  Jor.  Bot.  xi.  (i873)  i63;-PÄX,m  engl.  Bot. 

Jahrb.  VIL  (1886)  187  ;-C.  K.  S:hn.  111.  Handb.  Laubh.  II.  (1907)  198. 
A.  Paxil,  var.  ninfjpo3n<ie,  Pax,  in  Engl.  Pfl.  Reich.  8  Heft  (1902)  11. 
Â.  Baergeriamim,  var.  ningpoense,  Rehder,  in  Sargent  Trees  and  Shr.  I.  (1905)  179. 

Differt  a  typo  praecipne  fructus  ali.s  angulo  circiter  55° 
divergentibus. 

Nom.  Jap.     Tokaede. 
Hab.   ill  hortis  culta. 

forma  integrifolium,  m. 

A.  trifidiivi,  var.  integrifolium,  Making,     in  Bot.  Mag.  Tokyo,  XV.  112  ;  — Leveil.  Bull.  Soc. 
Eot.  Fria.  VI.  (1906)  593. 

Folia  subrhombeo-oblonga  integerrima,  raria.s  biloba  triloba 
immixta,  basi  rotundata  vei  obtusa.  apice  plerumque  obtusiter 
acuminata;   fructus  alis  erecto-patentibus. 

Nom.  Jap.     Maniha-tokaede. 

Hab.     in  hortis  culta. 

var.  formOSanum,  HaYATA,  Levell.  in  BuII.  Soc.  Bot.  Fran.  VI.  (1906) 
593  ;-C.  K.  S:;hn.  III.  Handb.  Laubh.  II.  (1937)  198. 

A.  irifidum,  var.  ?  Hatata,  in  Matsum.  et  Hatata,  En.  PI.  Formos.  (1905)  97. 

Folia  basi  cordata,   leviter  trilobata  lobis  obtusissimis.     Alae 
samarae  horizontaliter  patentes. 
Nom.   Jap.      Takasago-tökaede. 
Hab.     Formosa. 


18.     Acer  Ginnala,  Max.  (Tab.  XVHL) 

C.  .T.  Maximowicz  in  Mél.  Biol.  IL  (1857)]4.15  ;— Rupr.  ibid.  522  ;— Pax,  in  Engl.  Bot.  Jahrb.  VIL 
(1836)  18'),  et  Pfl.  Reich.  S  Heft  (1902)  12  ;— Fretn,  Oesber.  Bot.  Zeitsch.  (1902)  17  ;— Komaro. 
Fl.  Mansh.  IL  719  ;— C.  K.  Schn.  111.  Handb.  Laubh.  IL  (1907)  196  ;-Nakai,  Fl.  Korea.  I.  (1909) 
134. 

A.  tataricum,  var.  Ginnala,  Max.  Prim.  Fl.  Amur.  (i859)  67,  Mél.  Biol.  X  (1830)  604,  et  Fl. 
Mongol.  13}  ;— Regel,  Tent.  Fl.  Uss.  no.  106  ;— Forb.  et  Hemsl.  Jour.  Linn.  Soc.  XXIII.  142  ;— 
Palib.  Consp.  Fl.  Kor.  I.  59;— Korsh.  ia  Act.  Hort.  Petr.  XII.  318  ;— Schmidt,  11.  Amur.  no. 
79  ;  — Regel,  in  Gart.  fl.  (1377)  338  ;— Leveil.  Ball.  Soc.  Bot.  Fr.  VI.  (1906)  593. 

A.  tataricum,  b.  la'Mniatum,  Regel,  in  Bull.  Ph.  Math.  Akad.  Petrop.  XV.  p.  213. 

A.  tatari-um,  Fr.  et  Sav.  En.  pl.  J.ip.  I.  (1875)  89,  II.  (1879)  32  5  ;— Leveil.  l.c  593, 

A.  tataricum,  var.  amminatuin,  Franch.  PI.  David.  I.  76. 

A.  tataricum,  var.  aidznensc,  Franch.  in  Bull.  Soc.  Bot.  Fr.  XXVL  (1380)  84  ;— Pax,  in  Engl. 
1.  c  VIL  185  (1886). 

A.  titariaim,  a.  euginwili,  Pax,  in  Engl.  1  o.  VII.  135,  et  P.l.  Reich.  8  Heft  (1902)  12. 


Eevisio  Aceracearum  Japonicirum.  31 

Nom.   Jap.   Karahogi-kaede,  Kanokogi  (3Ç 1*  h  ¥•  ^  ?S  B  ^  ^.  ^  7f^  la 

ta-    1 832)  ;  —Mochincsso   (X  Êic  A  ¥•   7}^nmi^m-   ^^^m^^)&-    1 825)  ; 

Yacliiitai/a,     Xanai/e-uaga    (àt  M  M  lÊi  ^  it^  m)  ; — Hanakaedc    (J.    Matsu- 
MURA,  Shokubutsu-Meii,  no.  33.) 

Arbor,  ramuli  cortice  griseo-  vel  fasco-brunneo;  niinulis 
juvenilibus  leviter  pilosis.  Folia  juvenilia  utrinque  ad  nervös 
plus  minus  dense  rnolliter  pilosa;  adulta  membranacea  supra  glabra 
subtus  ad  venas  pilosa,  ovato-oblonga  vel  ovata,  acuta  vel  acumi- 
nata, basi  leviter  cordata  vel  rotundata  interdum  subtruncata, 
trilobata  rarius  indivisa  vel  subquinquelobata,  inciso-serrata  vel 
irregulariter  subduplicato-serrata  interdum  crenato-serrata,  versus 
basin  integerrima;  lobis  lateralibus  multo  minoribus;  petiolis 
gracile  elongatis  supra  sulcatis.  Panicula  dense  pubescens;  flori- 
bus  andro-polygamis.  Calyx  glabriusculus  5-fidus,  sepalis  ovatis 
obtusis.  Petala  alba  oblanceolata  quam  sepala  longiora.  Stamina 
8  antheris  laevibus.  Discus  bene  evolutus  extrastaminalis.  Ovari- 
um dense  visllosura,  stylis  bilobatis.  Samara  glabriuscula,  loculis 
reticulato-venosis,  alls  ad  18  mm.  longis  apice  rotundatis  angulo 
acuto  divergentibus  rarius  sese  invisem  obtegentibus. 

Hab.  in  sylvis  montuosis  per  totam  Japoniam.  Yezo: 
Shibetsu  (Nemuro),  Sapporo  (Isbikari),  Prov.  Iburi; — Honto: 
Nambu  (Rikuchiu)  Yonezawa  (Uzen),  Aidzu  (Ewashiro),  Nikko 
(Shimotsuke)  Togakushi,  Ontake  (Shinano),  Ohnogôri  (Hida);  — 
Kishu:  Hikosan  (Buzen). 

DiSTR.  Asia  orientali  (Mongolia,  China,  Manshuria,  Korea, 
Amuria). 

var.  yezoense,   m. 

Alae  samarae  parallelae  sese  invisem  obtegentes  ceterum  ut 
in  typicum. 

Nom.  Jap.      Yezo-karakogi-kasde. 

Hae.     Yezo:    Prov.  Ishikari. 

19.     Acer  spicatum,    Lam. 
var.  ukurunduense.  Max.  (Tab.  XIX.) 

Acer  spicatum.,  L  vm.  Pncycl.  Metho  1.  IÎ.  (I7i6)  331  ;  — DC.  Pro  Ir.  I  (1824)  593  ;-Pax,  in  Engl. 


32  Art.  1.  — G.  Koidzumi 

Bot.  Jahrb.  YII.  (ISÜC.)  188,  it  Fä.  reich.  8  Heft  (1902)  16  ;— Eobinsox,  in  Gra.t,  Synopt.  Fl.  Nor. 
Am.  I.  (18'j7j  435  ;-K.  Koch,  in  Dendrolog.  T.  (1SG9)  522  ;-C.  K.  Schn.  111.  Handb.  Laubh.  II. 
(1907)  199. 

Acer  gpicatitm,  Lam.  var.  iiIiuriDKluensc,  Maxim.  Piim.  Î1.  Amur.  (1859)  65  ;  et  Mtl.  Biol.  X. 
(1880)  504;-Fr.  et  Sav.  Ph  Jap.  I.  (1&75)  88  ;-Leveil.  Bull.  Soc.  Bot.  Fr.  VI.  (K06)  593  ;-Pax, 
1.  c.  VIL  133,  et  Pfl.  reich.  I.e.  16  ;— Fb.  Schmidt,  Eeis.  Amur.  Sachal.  p.p.  3j,  119;— Korsch. 
in  Act.  Hort.  Petr.  XII.  317  ;-Koidz.     PI.  Sachal.  Xakah.  (1910)  89. 

A.  uhiintmluenfe,  Tradtv.  et  Met.  FI.  Ochot.  no.  78  ;— Eupr.  Mt  1.  Biol.  IT.  520  ;— Komafo. 
Fl.  Mansh.  II.  722  ;— Xakai,  Fl.  Korea.  I.  iS-i  ;— C.  K.  Schn.  Ill,  Handb.  Laubh.  IL  199. 

A.  dedyle.  Max.,  Eopr.  in  Mél.  Biol.  II.  520. 

A.  siiicatuvi,  var.  ussuricnse,  Bdd.  (ex  Komaro.) 

A.  caudatum,  var.  itkuriinduense,  Eehder  (ex.  C.  K.  Schneider). 

A.  lasiocarpnm,  Leveil.  et  Vnt.  in  BulL  Soc.  Bot.  Fr.  VI.  (1906).  591. 

Nom.  Jap.  (hjarahana,  Araliana  (^Cife-b^-  ^nt^^IE^.  03felU^ 
?I^Iil-  1825; — Hozahi-haede  (jtmMm^^il^.m); — Amluuja  (^^nx^-  ^Mf 
sa  m  ^-  -^  I?  mmtZ.  l soi )  ;  —  Yama-asagani  (it  Ä  â  ^ ^  tt). 

Arbnscula  ramis  viridibus  vel  fuscis,  juvenilibiis  pilosis.  Folia 
membraiiacea  supra  glabra  juniora  subtus  piloso-tomentosa,  aclulta 
praesertim  ad  costas  pubescentia  pallidiora  vel  subcanescentia, 
rarissime  utrinque  perfecte  glabra,  amibtu  rotundata  vel  ovalia 
(5 — )  7  ( — 9)-niervia,  5 — 7-lobata,  basi  cordata  vel  nperte  eordata 
interdum  cordato-rotundata;  lol)is  acuminatis  inciso-dentatis  vel 
irregulariter  subduplicato-dentata,dentibus  mucronatis  vel  aristatis; 
petiolis  elongatis  mox  glabriuscnlis.  Racemus  niultitlorus  niolliter 
pilosus;  fioribus  andro-polygamis  viridi-flavescentibus.  Sepala 
(4 — )  5  ovata  vel  ovato-lanceolata  acuta  extus  ininute  pilosa. 
Petala  oblanceolata  vel  subspathulata  quam  sepala  lougiora  caduca 
fugacea.  Discus  lobatus  extrastaminalis.  Stamina  8,  filamentis 
antherisque  minutissime  scabriusculis.  Ovaria  villoso-tomentosa, 
stylis  profunde  bilobatis  laciniis  revolutis  intus  stigmatosis. 
Fructus  minute  pilosus,  alae  angulo  obtuso  divergentes. 

ITaiî.  in  subalpinis  Japoniae  mediae  et  septemtrionalis. 
Sacbalin:  Trctya-jiadj  ;  Kuriel  ;  Yezo:  Teinesan,  Sapporodake, 
Jözankei,  Muiwayama  (Isbikari),  ]Makkarinu])uri  (Shiribeshi), 
Konbumuri  (Hidaka),  Titose  (Iburi), — Hondo:  Iwakisan,  Hakkn- 
dasan  (Mutsu),  Iwatesan,  Hayachine  (Hikucliiü),  Chokaizan 
(Ugo),  Asabidake,  Gassau,  Zaösan,  Azumasan,  lidesau  (Uzen), 
Bandaisan,  Aizu,  lliucbigatake  (Iwashiro),  Xikko  (Shimotsuke). 
Togakusliiyaina,  Ilakuljasan.     Yatsugatake,  Komagatake,  Ariake- 


Eevisio  Aceracearum  Japonicarnm.  àS 

yama  (Sliinano);    Hakusaii  (Kaga);  Tairaiioyu  (Hida);  Myokösaii 
(Yechigo). 

DisTR.  VAR.     China,  Manshuria,  Korea,  Amuria. 

DiSTE.  SP.     America  boreali,  Asia  orientali. 

Not.     a  typo  differt  foliis  5 — 7  lobatis,  inciso  grandi-dentatis, 

20.     Acer  Oliverianum,  Pax. 

F.  Pax  in  Hooker  Icon.  PI.  XIX  (1889)  sub.  t.  1897,  et  Engl.  Pfl.  Eeich  8  Heft  (1902)  21  ;— C.  K. 
ScHN.  lU.  Handb.  Lanbh.  II.  (1907).  206. 

var.  Nakaharae,  Hayata.  (Tab.  XX.) 

a.  serrnïatum,  Hayata.  in  sched.  (Sut cuius  storilis  !) 

Arbor,  ramuli  glabri  pnrpureo-brunnei  vel  olivaceo-virides. 
Folia  rnembranacea  quinquenervia,  juniora  subtus  praesertim  ad 
nervorum  basin  parce  molliter  pilosa,  adulta  gkibra  palmato- 
quinqnelobata,  basi  aperte  cordata  raro  truncata,  argute  serrata, 
serraturis  incumbentibus,  lobis  triangularibus  vel  ovato-lanceolatis 
acuminatissimis;  petiolis  gracilibus  quam  lamina  paullo  bieviori- 
bus.  Panicula  corymbosa  multiflora  glabra,  floribus  albis  foliis 
coaetaneis  andro-polygamis.  Sepala  5  ovata  obtusa  extus 
puberula.  Petala  5  quam  sepala  paullo  longiora  rotundato-obovata 
apice  undulata.  Stamina  8  antheris  laevibus.  Ovarium  pilosum 
mox  glabrum,  stylis  gracilibus  2 — 2  J  m.m.  Ion  gis,  stigmatibus 
revolutis  brevioribus.  Samara  glabra  loculis  ovoideis  alis  angulo 
obtuso  divergentibus. 

Nom.  Jap.     Shima-momiji  (T.  Kawakami). 

Hab.     Formosa:    in  sylvis  temperatis,   Taitö,  Kierun,  Nanô, 
Sintek . 

DisTK.  Sp.     China  australi. 

Not.     Differt  a  typo  alis  saniarae  obtuse  divergentibus. 

subvar.     formosanum,  m.  (fig.  1.) 

Foha    ambitu    subquadrangulata   basi   cordata,    lobis    5   late 
triangularibus  subito  acuminatisque. 
Nom.  Jap.     liomald-sliimamomiji. 


M 


Art.  1. — G.  Koiclzumi. 


Hab.     Formosa:  Chösökei. 


Hab.,     For 


svar.  ;9,  trilobatum,  m.  (fig.  2.) 

Acer  TuUcheri,  var.  Shinrichte,   Hayata. 
in  SchecL. 

Folia  glabra,  trilobata  rarius 
tetraloba  intermixta,  lobis 
ovatis  longe  acuminatis  regulari- 
ter  serratis,  serraturis  leviter 
incunibentibns;  petiolis  glabris 
cum  lamina  usque  15  cm.  longis. 
Alae  samarao  obtusissime  diver- 
gentes. Folia  basi  rotundata, 
lobis  intermediis  saepe  longiori- 
bus. 

Nom.    Jap.,     Milsucle-slu'ma- 
momiji. 
mosa:  Hakkutaizan,  Sintek. 


Revisio  Aceracearum  Japonicaruui.  oO 

Sect.   10.     Palmata,   Pax. 

Pax,  ia  Exgl.  Bat.  Jahrb.  VI.  (1S85)  326,  VII.  (1836)  198. 

Iiiflorescentia  coiyinboso-paniculata  vel  corymbosa,  bracteata 
vel  nulla.  Flores  andro-polygami.  Sepala  petalisque  5.  Stamina  8 
hyi^ogyna.  Discus  extrastaminalis.  Stylus  apice  bilobatus.  Folia 
membranacea  simplicia  (5 — )  7 — pluri  palmatilobata. 

Species  -i. 

Clavis  speeicriitti.. 

■*  Antherae  scabrae.  Folia  plerumque  9-lobata  subtus  secus 
venas  petiolisque  pubescentia.  Inflorescentia  bracteolata, 
floribus  ochro-leiicis.     Ovaria  villosa.     Alae  saraarae  hori- 

zontaliter  patentes. A.  Sieholdianum,  Miq. 

*"       Antherae  laevae,  inflorescentia  ebracteolata. 

zs  Folia  plerumque  11-lobata,  secus  costas  pedicellisque 
in  juventate  tomcntosis.  Flores  purpurei.  Ovaria 
villoso-tomentosa.      Alae    samarae    in    angulo  obtuso 

divergentes A.  japomcwn,  Thg. 

z^A      Folia  plerumque   11-lobata,   petiolisque  ab  initio  glabr- 
iuscula.       Flores    ochro-leuci.       Ovaria    pubescentia. 

Alae  samarae  obtusissiine  divergentes 

A.  Sh imsaiüanum,  m . 

AAA      Folia  plerumque  7-lobata,  nascentia  subtus  secus  costas 
parce  hirta  vel  glabra.     Flores  plerumque  purpurei. 

Ovaria  glabra.     Alae  samarae  obtuse  divergentes 

A.  palmatum,  Thg. 


21.     Acer  Sieboldianum,  Miq.  (Tab.  XXI.) 

MiQQEL,  Prol.  Fl.  .Jap.  (1S;5)  19  ;-Pß  et  S  av.  Eu.  PL  Jap.  I.  (1S75)  87  ;-M.vx.  M.l.  Biol.  X. 
(1830)  603,  XII.  (1833)  4î3;— Pax,  ia  Engl.  Pfl.  Reich.  8  Heft  (1903)  23,  et  Bjt.  Jahrb.  VII. 
(1836)  200  ;— C.  K.  Sohn.  111.  Handb.  Laubh.  II.  (1907)  209. 

A.  japoiiimm,  var.  Sieboldianum,  Fr.  et  S^v.  En.  PI.  Jap.  II.  (1870)  317  ;— Leveil.  Bull.  Soc. 
Bot.  Fr.  VI.  (1906)  591. 

Ä.Siehüdianum,  a.  tjpi:a,  Mas.  Mil,  Biol.  XII.  (1333)  433  ;— Pax  iu  Engl.  Pfl.  lieich.  S 
Heft  (1902)  2'. 


36  Art.  1.— G.  Koidzimü. 

Arbor  rami  glabri,  cortice  badio-brunneo  vel  purpuras- 
cente,  ramuli  juveniles  cinereo-tomentosi.  Folia  niembranacea 
juniora  supra  sparce  puberula  subtus  presertim  in  nervis 
petiolisque  cinereo-tomentosa,  adulta  supra  glabra  subtus  secus 
costas  petiolisque  pubescentia,  in  axillis  venarum  villoso-barbarta, 
ambitu  rotundata  vel  rotundato-ovalia,  5 — 8  cm.  longa,  G — 9  cm. 
lata,  9  lobata  rarius  7 — 11  loba  immixta,  basi  aj^erte  cordata  in- 
terdum  subtruncata;  lobis  ovatis  acutis  vel  breve  acuminatis  argute 
serratis  vel  irregulariter  subduplicato-serratis;  petiolis  limbo  brevio- 
ribus  vel  aequantibus  dense  pubescentibus,  basi  gemmas  occultan- 
tibus.  Flores  andropolygami  pentameri  ochracei,  in  corymbum 
pluriflorum  pendentem,  pedunculo  dense  niveo-velutino  1 — 3  cm. 
longo  insidentem  dispositi  ;  bracteolis  linearibus  vel  lineari-lanceo- 
latis.  Sepala  ovata  vel  lanceolata  acuta  utrinque  dense  minute 
puberula.  Petala  ovata  vel  rotundato-ovata  obtusa,  calycis  lobis 
breviora,  margine  plicato.  Discus  extrastaminalis  pauci-lobatus. 
Stamina  8  sub  anthesin  exerta;  antberis  scabris.  Ovarium  dense 
villosum  stylis  apice  bilobatis.  Samara  pubescens  deinde  sub- 
glabra,  alae  oblongae  horizontaliter  patentes  ad  13  m.m.  longae 
5  m.m.  latae. 

Hae.  in  sylvis  montuosis  llondûensis,  Sikokuensis  et 
Kiusiuensis. 

DiSTR.     Korea,  Manshuria. 

a.  typicum,  Max.  inaiél.  Biol.  XII.  (1SS8)433;-Pax.  inENGi.  pa.  Eeich 
8  Heft.  ^1002)  25. 

Nom.  Jap.  Iiaijaiueiijetsu  (J.  Matsumuka,  Shokubutsu-Meii. 
no.  4G);  KihaiKt-ncltiira-l.aedc  (E.  Yatabe,  Nippon-Shokubutsu- 
hen.  I.  p.  417.) 

Folia  0-lobata,  rarius  7-loba  immixta,  basi  aperte  cordata, 
5 — 7  ( — 8)  cm.  lata;  floribus  flavescenlilms. 

Hab.  Hondo:  Zaôsan,  Nikkôsan  Komuno,  Ontake,  AVada- 
tôge.  Shikok:  Tsuiaigisan,  Nanogawamura,  Yabazuj^ama.  Kiushiu: 
Homanzan,  Inutake,  Ilikosan,  Amami-ohslnma  (Prov.  Obsumi). 

subvar.     albiflorum,  m. 
Floribus  albis,  ccterum  ut  in  typicum. 


Eevisio  Aceraceanim  Jtiponicarum.  37 

Nom.  Jap.      Sli irohcDia-itaijame'ujctm . 
Ha]î.      Hondo:   Yonezawa,  Nikk(3. 

subvar.     tortuosum,  (Max.) 

A.  SiehohUanum,-^.  tortuofmm.  Max.  Mel.  Biol.  XIL  (1838)  433; -Pax  Pfl.  Reich.  1.  c.  3'». 

Habitus    tortuoso-ramosissimus;     foliis    lobis    9    pleramquo 
brevissime  acutis  vel  obtusis. 
Nom.  Jap. 
Hab.     culta. 

subvar.     microphyllum,  (Max.) 

A.  Sieboldianum.  ß.  microphyllum,  Max.  1.  c  ;— Pax  1.  c.  25. 

Folia  minora,  5 — G  cm.  longa. 

Nom.  Jap.     Hime-ucliiwa-haede . 
Hab.     culta. 

subvar.     Momijigasane,  m. 

Foliis  lobis  anguste  oblongis,  versus  basin  subito  constrictis. 
Nom.  Jap.     Momiji-gasane  (UMM)  {W" M  ^  ^  ^,  M^i&M  i^) 
Hab.     culta. 

subvar.     Sodenouchi,  m. 

Folia  minora  semper  9  lobata. 

Nom.  Jap.     Sode-no-iichi  (liU  /  t^)  (K7ic-b¥,   iP'm^^<m,  i^mi&i^ 
g>.  1710) 

Hab.     culta. 

ß.     tsusimense,  m. 

Folia  7-lobata  basi  aperte  cordata  vel  subtruncata;   lobis  oblongis 
inciso-serratis,  sinubus  profundioribus. 

Nom.  Jap.     Kolia-ücujamcigetsu . 

Hab.     Kiushiu:  Insl.  Tsushima. 

subvar.     Kasatoriyama,  m. 
Folia  7-lobata,  basi  cordata,  lobis  ellipticis  acutis. 


38  Art.  1.— G.  Ivoidzn.mi. 

Nom .  Jap.     Kasatoriyama  (^MÜi)  (ß.n'.-h¥.   ^ ü ^  ^  tf.    if  fS  üü 
ISO?.  1710) 

Hab.     culta. 

subvar.     laxifolium,  m. 

Folia   ambitu   rotuDdata,    T-loljata,    basi   piofunde   cordata,    Jobis 
inciso-serratis. 

XoM.  Jap.     Hina-ilai/aiiieigctsu. 

Hap.     culla. 

.^ul:)Yar.     Ayaigasa,  m. 

Folia  minora  7-lobata. 

Nom.  Jap.     Ayaigam  (m  m  ^)  (fMiP^^  if.  m  ff  iü  1^,  ti^) 
Hap.     culta. 

22.     Acer  Shirasawanum.  .^p.  nov.  (Tab.  NXII.) 

Arbor  glaber,  rami  cortice  cinerasceiite.  Folia  membrauacea, 
juniora  supra  vel  utrinque  secus  nervös  et  in  axillis  venarum  parce 
pilosa,  adulta  supra  semper  glabra  subtus  pilis  persistentibus, 
ambitu  lotuudata  vel  rotundato-subreniformia,  basi  cordata, 
palmato-11-lobata,  lobis  acuminatis  duplicato-serratis;  petiolis 
elongatis  ab  initio  vel  mox  glabiis.  Inflorescentia  umbellato- 
corymbosa,  glabra  vel  parce  pul)erula  ebracteolata  pedunculata; 
floribus  albis  vel  ochro-leucis  andro-polygamis  (vel  andro-dioecis 
interdum  andi'o-monoecis  ?).  Sepala  5  ovata  acuta  e.xtus  raro 
puberula,  atropurpurea  vel  purpurea.  Petala  5  ovata  apice 
rotundata  quam  sepala  breviora.  Discus  carnosus  extrastaminalis. 
Stamina  8  exerta,  antheris  laevibus.  Ovaria  villosa  stjdis  glabris 
stigmatibus  leviter  revolutis.  Samarae  glabrae  loculis  borizon- 
talibus,  alis  obovato-oblongis  leviter  arcuato-ascendenti]>us. 

Nom.  Jap.      (ih-itayamci(jctsu. 

Hap.     Hondo:  Nikkn,  Usuitüge,  Hakoneyama. 

DiSTP.     endemica. 

Not.  Affme  Acer  Sicholdiano,  Miq.  et  Acer  jiahiuilo,  Tiig.  sed 
a   priori    foliis  11-lubatis  petiolisque    glabris;    antheris  laevibus; 


Eevisio  Aceracearum  Japonicarum.  ov 

inflorescentibus  glabris  ebracteolatis;  sepalis  extus  purpurascen- 
tibus;  ab  altero  foliis  11-lobatis  lobis  duplicato-serratis.  subtus  ad 
nervös  et  in  axillis  venaruin  parce  pilosis;  alis  samarae  obtusissime 
divergentibus;  ovariis  villosis;  petalis  albis  differt. 

Var.     tenuifolium,  m.  (Tab.  XXIIT.) 

Folia  tenuiora  basi  profunde  cordata,  ambitu  rotundata,  9  rarius 
11  circumlobata,  lobis  incisis;  saniaris  ad  apicem  ramidorurn 
solitariis. 

Nom.  Jap.     Hina-iichiicahaedc. 

Hab.     Hondo:   Nikko,  Fujiyama,  Ontake. 


sub  var.     Murasame, 


m, 


Folia  rotundata  usque  5  cm.  longa  1 1-circumlobata  basi  profunde 
cordata,  lobis  incisis  acuminatis  subtus  secus  costas  adpresse  pilosis. 

Nom.  Jap.     Murasame  (HM) 

Hab.     in  hortis  culta. 

23.     Acer  japonicum,  Thuxb.  (Tab.  XXIV.— XXV.) 

Thünbekg,  Fl.  Jap.  (17S4)  IG.  et  Icon  Fl.  Jap.  clec.  2.  t.  19  (1794)  ;— Sieb,  et  Zucr.  Fl.  Jap.  Farn. 
Xat.  in  Abh.  Akad.  Münch.  IV.  2  (1846)  156,  et  Fl.  Jap.  II.  (1870)  82.  t.  144  :-  Miq.  Prol.  Fl.  Jap. 
(186  ■)  18  ;— Fe.  et  Sav.  Eu.  PI.  Jap.  I.  (1875)  87  ;— Max.  Mel.  Biol.  X.  (IfeSO)  C05  ;-Pax,  in  Engl. 
Bot.  Jahrb.  VII.  1 1886)  109,  et  Pfl.  Eeich,  8  Heft  (1902)  24  :-DC.  Proclr.  I.  (1824)  595  -—CK.  Schn. 
111.  Hanclb.  Laubh.  I.  (19G7)  203  ;— Leveil.  Bull.  See.  Bot.  Fr.  VI.  (1206)  591  :— A.  Grat  in  Peert 
Expecl.  Jap.  309;— C.  Koch  in  Miq.  Ann.  Mus.  Lugd.  Bitar.  I.  251  ;— Xakai,  Fl.  Korea.  I.  p.  135. 

Arbor  rami  cortice  atro-purpurascente  vel  rufo-fusescente, 
ramulis  novellis  glabriusculis.  Gemmae  joerulae  caducae,  ex- 
teriores  coriaceae  rotundatae,  interiores  omnes  dense  villosao 
oblanceolatae  vel  lineari-spathulatae  apice  rotundatae  vel  obtusae. 
Folia  membranacea  juvenilia  utrinque  dense  villosa,  cito  supra, 
glabra  subtus  in  axillis  venarum  tantum  barbarta,  ambitu  rotun- 
data, 11-lobata  rarius  7— lo  lobata,  basi  profunde  cordata;  lobis 
ovato-ellipticis  acuminatis  iiiciso-serratis;  petiolis  limbo  multo 
brevioribus  glabrescentibus  bnsi  plerumque  dilatatis,  junioribus 
pl.  m.  villosis.  Flores  andro-polygami  pentameri  purpurei  spec- 
tabiles,  in  corymbum  laxiflorum  pendentem  dispositi;  pedunculis 


40  Art.  l.-G.  KoicUuiiû. 

pedicellisque  glabris  vel  villoso-tonientellis  ebracteolatis.  Sepala 
purpurea,  late  elliptica  obtusa,  ampla  5 — 7  mm.  longa,  demum 
reüexa.  Petala  ovato-orbiculata  obtusa  vel  rotundata,  alba  vel 
purpurascentia,  sepalis  fere  duplo  breviora.  Discus  carnosus 
extrastaminalis.  Stamina  8  sub  anthesiii  exerta,  filamentis 
antherisque  laevibus.  Ovarium  dense  villosum  stylis  elongatis 
apice  bilobatis.  Samara  pubescens,  alae  angulo  obtuso  dive- 
rgentes. 

Hab.     in  sylvis  montuosis  Yezoensis  et  Hond(3ensis. 

DisTR.     Korea.  (Manshuria?) 

«_  typicum,  Gr.  V.  ScHW.  in  Gattn.  A.  (ISDS/  XLII.  p.  TO^;— PAxin 
Pfl.  Reich  I.e.  24  (19U2). 

Folia  aequilonga  ac  lata,  lobis  ovato-ellipticis  acuminatis  vel 
•caudato-acuminatis ;  alae  samarae  angulo  obtuso  divergentes. 

Nom.  Jap.  Ha-iichiwa-kaede  (J.  Matsumüra,  Shokubutsu- 
Meii,  110.  34);  Ahahana-uclmva-kaede  (R.  Yataee.  B:^^à^lisM.  1^  —  ^. 
p.  417);  Oguraijaina,  Jünihitoyc,  Meigclsii-hacde  (p- M  IP"  ^ 'es  SS: .  ^W^^ 
#.  110.  4). 

Hab.  Yezo:  Moiwadake,  Teinesan,  Saj^poro,  Jôzankei  (Ishi- 
kari);  Makkarinupuri  (Shiribeshi);  Sbikifu  (Iburi).  Hondo: 
Tokiwano,  Iwakij^ama,  Hakködasan  (Mutsu);  Iwateyama,  Haya- 
cliine,  Nanshözan  (Rikucliiu);  Kurikomayama  (Rikuzen);  Chö- 
kaizan  (Ugo);  Gassan,  Asahidake,  Zaösan,  Adzumasan,  lidesan 
(Uzen);  Bandaisan,  Aidzu  (Iwasbiro);  Nikkösan  (Shimotsuke) 
Harunasan  (Kotsuke);  Togakuslii^^ama,  Kiso,  Onogawa,  Nori- 
kuradake,  Ontake,  Yatsugatake  (Shinano);  Prov.  Hida. 

foniiK  angustilobum,  m. 
Foliis  lobis  angustioribus  oblongo-lanceolatis  acuminatis. 

foniKi  macrophyllum,  m. 

A.  jnpoiiiciiiii,  var.  iiiacrop]iyllitm,  Gk.  v.  Sghvv.  I.e.  709.  (1893) 

Folia  valde  ampla  14  cm.  longa  ac  lata,  lobis  caadato-acuminatis. 


Eivisio  Aceracearum  Japonicarum.  41 

forma  tenuilobum,  m. 

Folia  8 — 10  cm.  longa,  11 — 13  cm.  lata,  lobis  ovatis  breve 
acuminatis,  infimis  sese  invisera  obtegentibus;  alae  samarae  semi- 
ovatae. 

forma  semiovatum,  m. 

Folia  subtus  secus  costas  petiolisque  dense  villosa,  basi  profunde 
cordata,  lobis  ovatis  acuminatis;  alae  samarae  semi-obovatae. 

forma  ascendens,  m. 
Alae  samarae  arcuato-ascendentes. 

forma  crassifolium,  m. 

Folia  crassiora  subchartacea,  subtus  petiolis  ramulisque  plus  minus 
canescentia. 

ß.     aureum,  Gr.  v.  Schw.   ic  709;-pax1.c.  2i. 

Folia  aureo-tincta. 
Nom.  Jap. 
Hab.     culta. 

r.     Parsonii,  Veitch,    ex  gr.  v.  schw.  i.e.  709  ;-pax  i.e.  25. 

Folia  profunde  inciso-lobata. 
Nom.  Jap. 
Hab.     culta. 

o.     Heyhachii,  Maïsum.,    MAKiNoiaBot.  Ma- Tokyo  (i'jo4)  115.   (Tab. 
XXV.) 

Acer  circumlohatum,  Max.  var.  IL'ijliacliii,  Mak.  ibid.  XXIV.  (1910)  74. 

Folia  11 — 13  palmatipartita,  lobis  ambitu  oblanceolatis  longe 
acuminatis,  sursum  utrinque  profunde  lobulatis,  deorsum  integer- 
rimis  versus  basin  cuneato-attenuatis. 

Nom.  Jap.     Mai-hijahi. 

Hab.     culta. 


42  Alt.  1.— G.  Koidzumi. 

ç.     microphyllum,  m.    (Tab.  XXV.) 

Folia  minora  7^ — 8^  cm.  lata,  11 — 13  lobata,  lobis  ovatis  acutis 
inciso-serratis. 

Nom.  Jap.      Yezo-mc'Kjclsulîacdc . 

IIaiî.     Yezo. 

•-;.     circumlobatum,  (Max.)  ni.     (Tab.  XXV.) 

Acer  circvmlohatinn,  Max.  Mt'l.  Biol.  \I.  (18G7)  p.  3GS,  et  X.  (1880)  €08  ;— Fr.  et  Sav.  Ed.  PI. 
Jap.  I.  (1875)  8S  ;— Pax  Engl.  Bet.  Jahrb.  TH.  (1886)  190,  ot  Pfl.  Eeich.  I.e.  25  ;— Leveil.  Bull. 
Soc.  Bot.  Fr.  VI.  (1906)  589. 

A.  cimimloiatum ,  a.  insular,',  Pax  1.  c.  pp.  25.  200. 

Folia  chartacea,  in  venannn  axillis  albo  villosa,  nee  rufo-barbarta, 
11-lobata,  lobi.s  basalibus  sese  invisem  saepe  obtegentibus.  Alae 
samarae  horizontaliter  patentes. 

Nom.  Jap.     Oh-mcigctsu  (J.  Matsumura,  1.  c.  no.  28) 

Hab.     Nikko,  Sbinano. 

C.     villosum,  in. 

Folia  9-lubata  ll-lo],>a  intermixta,  supra  parce  subtus  praesertim 
secus  costas  et  in  axillis  venaruin  A'illosu-tomentosa;  petiolis 
villoso-tomentellis. 

Nom.  Jap. 

Hab.     Yezo,  Nikko. 

foinuif  macrophyllum,  ni. 

Folia  ampla  14  cm.  longa  ac  lata. 
Nom.  Jap. 
Hab.     Nikkn. 

0.     Kasado,  m. 

Acer  j(i])oiii(ti III,  a.  ii/jucum,  forma  Kasudo,  ui.  IMss. 

Folia  tenuiora  8  cm.  longa  ac  lata,  lo-lobata,  lobis  acuminatis 
inciso-serratis  serraturis  argutissimis. 

Nom.  Jap.     Kasado  (3?^). 

Hab.     culta. 


Eevisio  Aceracearnm  Japonicarum.  43 

V.     Kokonoe,  m. 

A.  japonkinn,  a.  forma,  Kohoiioe,  m.  Mss. 

Folia  11-lobata,  lobis  ovatis  acutis. 

Nom .  Ja p .     Kolzonoe  {%  S)  (e»-  ü  fi*  Ä-  §f  Ä .  if  ffi  üü  IS  ib .  W.n^). 
Hab.     culta. 

X.     Sayosigure,  m. 

A.  jajwnicuvi,  a.  forma,  Sayosigurc,  m.  Wss. 

Folia  minora  9-circumlobata. 

Nom.  J  at.      Sai/osliigure  (/h  ^  a|  nf)  (p-M^  ^^i  R.  IM^M  If.  i^). 
Hab,     in  hoitis  culta. 

?•.     Matsuyoi,  m. 

A.  jaxoiiicum,  a.  forma,  Mittmyoi,  m.  Mss. 

Folia  minora  0-lobata. 

Nom .  Jap .     Mahnnoi  (^  W)  (^  Hi  ^  ^  if  Ä .  FfUl  ifi  IS  tj?) . 
Hab.     culta. 

24.     Acer  palmatum,  Thukb.  (Tab.  XXYI— XXVIH.) 

Thunbeeg,  Fl.  Jap.  (17S4)  161  .:-BC.  Prodr.  I.  (1824.)  GLE.  ;— K.  Koch,  in  Miq.  Arn.  Mv,s.  Liigcl. 
Batav.  I.  (1863—64)  251,  et  Dtndrol.  I.  (1869)  224  ;— KiQ.  Prcl.  II.  Jap.  (1866-67)  20 .— Fieb.  et 
Zucc.  Fl.  Jap.  n.  (1870)  83  ;— Fb.  (t  Sav.  En.  PI.  Jsp.  I.  (IhVC))  88  :-Max.  MCI.  Bid.  X.  (1S8() 
eC7  ;-Pax  in  E>GL.  Bot.  Jahrb.  YII.  (1886)  101,  et  Pfl.  Ecich  S  Beit  (1^(2)  L5  ;— Ieveil.  Eull. 
Sec.  Bot.  Fran.  VI.  (19(6)  C91  .:-C.  K.  Schn.  111.  Handb.  lanl  h  11.  (1Î07)  £C7  ;-Biels,  11.  Cen- 
tralchin.  in  Engl.  Bot.  Jahrb.  XXIX.  p.  448. 

Acer  lohjmcri  luw,  S.  (t  Z.  El.  Jap.  lam.  Xat.  in  Alh.  Atfd.  Mirch.  lY.  2.  (ie4()  ir8. 

Aibcr  rami  glabri  cortice  cinerascenti-viride  vel  fufco-puipuias- 
cente,  ramulis  novellis  glabis  vel  puberulis.  Gemmae  perulae 
interiores  spathulatae  feirugineo-tcmentosae.  Folia  maxime  varia- 
bilia,  juvenilia  dense  moUiter  pilosa,  adulta  glabra,  membranacea, 
palmato-pluiilobata,  basi  aperte  cordata,  lobis  acuminatis  vel  caud- 
ato-cuspidatis,  serratis  vel  inciso-serratis;  petiolis  gracile  elongatis 
Inflorescentia  umbellato-corymbosa  vel-paniculata,  pedicellis 
filii'oimibus  glaberrimis;  floribus  coaetaneis  andro-polygamis. 
Sepala  5  ovato-oblonga  rotundata  vel  obtusa  intense  purpurea 
extus   saepe   parce  ciliolata.     Petala  5  sepalis  aequilonga  oblonga 


44  Art.  1.  — G.  Koidzumi. 

obtusa  aut  rotundata  basi  parani  angustata  lutescentia  vel  ochro- 
leuca.  Stamina  8  exerta,  antberis  laevibus.  Ovaria  glabra  rarius 
parce  ciliolata  stylis  bilobatis,  stigma tibus  revolutis.  Samara 
glabra  alis  angulo  vario  divergentibus. 

Nom.  Jap.  Kaede  (%7Tc2¥.  M;^.  ^$f#.  'Xm?^^.  1708);  — 
Kaeclenold,   Kaerudenold  (U  M  WM  ^  ii>)    Moniijl   (*  W^^^M);  — Iwatohcni 

Hab.     per  totarn  japoniam  spontanea. 
DiSTR.     Hupeb  (Cliina  centrali).    (Korea?) 

Snh>;i>.  «.     genuinum,  (S.  etZ.)  m.  (Tab.  XXVI.— XXVIII.) 

A.  palmatum,  x.w.  j^almatum,  Koch.  Dendrol.  I.  ^IStîO)  525. 

A.  palmatum,  var.  ficnuinum,  S  et  Z.  Fl  Jap.  IT.  (l'STO)  84  ipp.)  (excl.  A.  t^rptenlohtim,  Thg)  t. 
145  (excl.  fig.  9.) 

A.  palmatum,  var  Thunhergi,  Pax,  svar.  eupalmatuia,  Gr  v.  Schw.  Gartu.  fl.  (1833)  G52  ;  — 
Pax  in  Engl.  Pfl.  Reich  8  Heft  (1902)  2o. 

A.  palmatum.  var.  Thunhergi,  Pax  in  Engl.  Bot.  .Jahrb.  VII.  (I8S6)  232  (p.  p.) 

Gemmae  perulae  intimae  circ.  15  mm.  longae,  glabrae  vel 
ciliolatae.  Folia  5 — 7  palinatilobata  basi  cordata  usque  4,5  cm. 
longa  ac  lata;  lobis  lanceolato-oblongis  acuminatissimis  serrulatis 
aut  inciso-serratis.  Flores  pedicellis  capilliformibus.  Samarae 
minores  alis  circ.  1 — 1,2  cm.  longis  in  angulo  obtuso,  rarius  acuto 
V.  obtusissimo  divergentibus. 

a.     spectabile,  m.    (Tab.  XXVIIl.) 

A.  palmatum,  Thg.  Fl.  Jap.  IGl,  et  Icon.  Fl.  Jap.  (1800)  t  4i. 

.1.  palmatum,  var.  genuinum,  S.  et  Z.  Fl.  Jap.  II.  84  (p.  p.)  t  145  (p.  p.) 

Folia  aequaliter  argute  serrulata. 
Nom.  Jap.     Iwhamomiji. 
PIab.      llondö:    Yugasbima  (Idzu). 

J'onua  Chisio,  m. 

Folia  5 — 7  lol)ata  serrata,   nascentia  intense  et  etiam  in  aestate 
vix  roseo-colorata. 

Nom.  Jap.     Chishio  (ff-'^),  Mösen  (^'îà). 

Hab.     in  bortis  culta. 


Eevisio  Aceracearum  Japonicarum.  4o 

fonna  Komonnisiki,  m. 

Folia  in  varus  inodis  aureo-maculata. 
Nom .  Jap .     Komon-nishild  (/h  WL 1^) . 
Hab.     in  liortis  culta. 

subvar.     crispum,  (Andhe)  m. 

Ä.  imlmatum,   a  TJmnheriji,  svar.   eiipnlnuitum,  i.  crisjjuin  Andre,  in  Illustr.  Hortic.  XVIIl. 
(1870)  241.  t.  43  ;— Pax  I.e.  p.  L'G. 

Folia  profunde  palmatifida;   laciniis  anguste  lanceolatis,  niargine 
leviter  involuto. 

Nom.  Jap.      Okiishimo  (Eft). 

Hap.     in  liortis  culta, 

subvar.     Hikasayama,  m. 

Folia  palmatifida,    lobis   oblongo-lanceolatis    eroso-serratis,    secus 
costas  et  margincs  viridia  ceterum  albo-maculata. 

Nom.  Jap.      Hil-a^aijama  (e  ^Uj). 

Hap.     in  hortis  culta. 

h.     amabilie,  m.  (Tab.  XXVI.) 

A.  imlmatum,  var.  Thunhergi,  Pax,  svar.  eiipalmatum,  Gr.  v.  Schw.  I.e.  (p.  p.) 

Folia  inciso-serrata. 

Nom.  Jap.      Iroha-momiji. 

Hap.     Hondo:     Nikko,     Chichibu,     Kasugasan,    Takiyama 
(Swô);  Sikok:  Kotohirayama;  Kiushiu:  Mitake  (Tsushima). 

forma  Akajinisiki,  m. 

Folia  enascentia  rosea,  inciso-serrata,  lobis  anguste  olilongis  acumi- 
natis. 

Nom.  Jap.     Ahaji-nishihi  (ßiüi^). 

Hab.     culta. 

forma  Tsuchigumo,  m. 
Foliis  lobis  apice  circinatis  pulverulentibus. 


46  Art.  1.— G.  Koiclzumi. 

Nom.  Jap.     Tsucliiijumo  (^  fiH  g  IS  Ä). 
Hab.     culta. 

forma  Hanaidzuminisiki,  m. 

Folia  inciso-serrata,  albo-maculata. 

Nom.  Jap.     Haua-idzumi-iiUlnlü  {^^î%). 
Hab.     in  hortis  culta. 

forma  Oridonisiki,  ni. 

Folia  inciso-serrata  aureo-guttata. 

Nom.  Jap.     Orido-no-nUldhl  (^  ^  l?,). 
Hab.     in  liortis  culta. 

subvar.     Kagiri,  m. 

Folia  leviter  contorta  argute  serrata  i'oseo-niarginata  aut  maculata 
saepe  viridia  immixta. 

Nom.  Jap,     Kaijiri. 

Hab.     culta. 

Suhsp.     septenlobum,  (ïnc)  m.  (Tab.  XXVI.— XXVII.) 

A.  jHiliiiatum,  ß.  HL'ptcnlohum,  (Tho.)  Koch,  in  Ann.  Mus.  I.u^d.  Bativ.  I.  (1SG4)  251,  et 
Dendrol.  I.  (1869)  525. 

A.  septenlobim,  Th3.  Fl.  Jip.  (17St)  161  ;  — DG.  Prodr.  I.  (1321)  595. 

A.  vieikots,  Sieb,  in  Jaahrb.  Maatsch.  Tuinb.  (1344)  t.  2.  fi-j.  5. 

A.  palmatum.  var.  septenlobum,  Miq.  Prol.  Fl.  Jap.  (18ö7)  20. 

A.  paimatum,  rM.palmitlß.lum,  S.  et  Z.  Fl.  Jap.  II.  (1870)  84.  (p.  p.) 

A. palmuttim,  vxr.  Thnnbn-ni,  sv.ir.  .<i:ib.!;pt.'iihb  im  Ga.  v.  Smiw.  Givt.  Fl.  (lidl).  G7S;— Pax 
Pfl.  Reich.  I.e.  26.  (p.  p.) 

A.  palmatum,  rar.  Tknnbji-ji,  svar.  septenlobum,  (Koch)  C.  K.  Sch.v.  I.e.  ilJ07)  207. 

Gemmae  perulao  iiitimae  anguste  spathulatae,  ad  3  cm.  longae 
tomentosae.  Folia  nascentia  dense  pilosa,  adulta  glabra  usque 
6,5  cm.  longa  palmato-7-lobata,  basi  cordata;  lobis  elliptico-  vel 
oblongo-lanceolatis  caudato-acuminatis  vel  subito  cuspidatis, 
regulariter  serratis.  Inflorescentia  corymboso-paniculata.  Alae 
samarae  2 — 2,4  cm.  longae  in  angulo  obtusissimo,  vel  .subhori- 
zontaliter  divergentes. 


Eevisio  Aceracearum  Japonicarnm  47 

XoM.  Jap.      Takawo-inomiji,  Oh-moiuiji. 

Hab.  Hondo:  Morioka,  Tokizawayama,  Omyojin,  Uba- 
yashiki,  Kadoma,  Sengantôge  (Rikuchiu);  Azumasaii  (Uzen), 
\Vakamatsu(Iwashiro);  Shiwobara,  Nikkö  (Shimotsuke);  Chichibu 
(Musashi);  Fujiyama  (Suruga). 

var.     latilobatum,  m.  (Tal.  XXVI). 

Foliis  lobis  ovatis  vel  kite  triangukitis  subito  acuminatis.  Akie 
samarae  liorizontales. 

Xo3r.  Jap.     Hiroha-momiji. 

Hab.  Shiwobara  (Shimotsuke),  Ontake  (Kiso  Provinciae 
Shinauo);  Kasugami,  Nanogawa  (Tosa), 

var.     speciosum,  m. 

Folis  lobis  ovato-lanceolatis  acuminatis  utrinque  purpurascen- 
tibus,  coloribus  autumnalibus  aureo-brunnescentibus. 

Nom.  Jap.     Nomiu-a-hacde,  2Iiisasliino  (ß,M^). 

Hab.     in  hortis  culta,  sed  saepe  sj^ontanea. 


var.     palmatipartitum, 


m. 


Folia    palmato-7-partita    lobis    auguste    oblougis   acuminatis   vel 
subito  acuminatis  versus  basin  cuneato-augustatis  vel  constrictis. 

forma  Senri,  m. 

Folia  nascentia  intense  rosea  deinde   purpureo-rosea,    in   aestate 
viridescentes. 

Nom.  Jap.     «Sc/hy  (-f- m),  Cliisato. 

Hab.     in  hortis  culta. 

forma  Ichigyoin,  m . 

Folia  nascentia  latescentia,  auctumna  profunde  purpurea. 
Nom.   Jap.      Tchigijü-in.  (— 'ff^). 
Hab.     culta. 


48  Art.  1.— G   KoiclziiHi'. 

forma  Akitsuta.  ni. 

Foliis  lobis  basi  subito  vakle  angustaii.^. 
Nom.  Jap.     Akitsuta. 
Hae.     in  hortis  culta. 

forma  Tsukubane,     m. 

Foliis  lobis  anguste  lanceolatis. 

Nom.  Jap.      Tsvluhanc  (^'i^m). 
Hau.     culta. 

forma  Ohsakazuki,  m. 

Foliis  lobis  lauceolato-oblougis. 
Nom.  Jap.      Ohsahcuki  (-k^). 
Hau.     culta 

subvar.     Tanabata,  m. 

Foliis  lobis   lanceolato-linearibus   utrinque   attenuatis   serratis  vel 
integerrimis. 

Nom.  Jap.      Tanalmta  (-h  ^). 

Hab.     culta. 

subvar.     linearilobum,  (Miq)  rn. 

A.  jminidlinii,  var.  liiwi rilobuiii,  Miq.  in  Archiv.  Xeerlandeis  II.  (1867)  4'j9  ;  — Îiax.  Mol.  Biol. 
X.  (1880)  C03  ;-Pax  in  Engl.  Bot.  Jahrb.  VII.  (ISSÜ)  202. 

A.  palmalum,  f.  linearilobum,  S.  et  Z.  Fl.  Jap.  II.  (1870)  8i,  t.  14Ü. 

A.  2>ii  Intal  lim,  -{.  liiicarilnhum,  S.  et  Z.,  Pax  in  Engl.  Pfl.  Reich.  I.e.  27  ;— C.  K.  Schk.  I.e.  -07. 

Nom.  Jap.      Slirmenouchi  (ä  0  "h). 

Hap.     in  hortis  culta. 
Folia  palniatisecta   vel  scctisve,    laciniis  linearibus  acutis  integris 
aut  remote  obscuriter  serratis. 

forma  lineare,  m.  ((Ir.  v.  Srnw.)  1-<^-o82;-Pax1.o. 27. 
Folia  viridia. 

forma  atro-lineare,  m.   (Gii.  v.  Sciiw.)  J.c.  gs2;-pax.  i.e.  27. 
Folia  atro-purpuroa. 


Eevisio  Aceracearum  Japonicarum.  49 

Suhsp.    y.     Matsumurae,  ni.  (Tab.  XXVIII.) 

A.  jialmatum,  f.  2)<i/)«a/(;Wi(?»,  S.  et  Z.  Fl.  Jap.  II.  (1870)  84,  (pro  parte  ?) 

A.  pnlmatam,  var.  Thiaihergii,  svar.  auhseptenlohum,  Gb..  \.  Schw.  I.e.  678  ; — Pax,  Pfl.  Eeich. 
I.e.  26  (pi-o  parte). 

Folia  7 — 0-lobata  ad  7 — 8  cm.  longa;  lobis  ellipticis  ovatis 
vel  oblongis,  candato-acuminatis  incisis  vel  inciso-serratis  rarius 
inaequaliter  sabduplicato-serratis.  Inflorescentia  coiymljoso- 
paniculata.  Alae  samarae  angulo  acuto,  obtuso  v.  obtusissiino 
divergentes,  cire.  17 — 25  mm.  longae. 

Nom.  Jap.  Yama-momiji  (J,  Matsumura,  Shokubutsu-Meii, 
no.  40.) 

Hab.     Yezo,  Hondo,  Shikok,  Kiusbiu,  Formosa. 

DiSTE.     China  centrali  ?. 

«.     spontaneum,  m. 

forma     angustilobum,  m. 

Foliis  lobis  angnste  oblongis. 
Nom.  Jap. 
Hab.     ill  Yezo. 

forma     circumlobatum,  m. 

Folia  9  ( — 7)  circumlobata  basi  profunde  cordata,  lobis  basalibus 
conniventibus. 

Nom.  Jap. 

Hab.  Matsumine,  Yudonosan  (Uzen);  Shimidzutöge  (Shimo- 
tsuke). 

forma     acutum,  m. 
Alae  samarae  angulo  acuto  divergentes. 

forma     rectangulare,  m. 
Alae  samarae  in  angulo  fere  recto  divergentes. 

forma     obtusum,  m. 
Alae  samarae  angulo  obtuso  v.  obtusissimo  imtentes. 


-50  Art.  1 — G.  Koidzumi. 

sultvar.     elegans,  m. 
Alae   saniarae   latiores,    circ.    2    cm.    longae    1    cm.     latae    apice 
rotundatae. 

XoM.  Jap. 

IIab,     in  monte  Gassan  et  Ohtôge. 

subvar.     formosanum,  m. 

A.  dupUcato-serratuin,  Hatata,  in  sclied. 

Folia  7-lobata  lobis  lineari-oblongis  aciitis  inciso-serratis. 
Nom.  Jap.      Takasago-momiji. 
Hab.     Formosa. 

I),     hortense,  m. 

subvar.     palmatilobum,  m. 
forma     Sigitatsu,  m. 

Folia  7 — l)-lobata  incisa,  secus  costas  utrinque  am-eo-maculata. 
Nom.  Jap.     Slrirjitatsu. 
Hab.     in  liortis  culta. 

forma     Nisikigasane,  m. 

Folia  am-eo-guttata. 

Nom.  Jap.     Niskihlgasane  (UM). 
Hab.     culta. 

forma     Asanoha,  m. 

Folia  9-circumlobata  lobis  ovatis  acuminatis  inciso-serratis. 
Nom.  Jap.     Asa-no-lia. 
Hab.     in  hortis  culta. 

subvar.     palmatipartitum,  m. 

Folia  palmatipartita  inciso-serrata  v.  lobulata,  lobis  omnibus  fere 
aequiformibus. 

forma     Monnisiki,  m. 

Foliis  lobis  5 — 7,  profunde  incisis  vel  pinnatifidisve,  versus  basin 
cuncato-angustatis. 


Revisio  Acer  ice  irum  Japonicarum.  51 

Nom.  Jap.     .^[on-nisliihi.  (^if,). 
Hae.     in  hortis  culta. 

foniHi     Akitsusima,  m. 

Foliis  lobis  7 — 9,  anguste  oblongis  incisis  acuminatis. 

Nom.  Jap.     Akitsushima.  i 

fonna     Tokonatsu,  m.  ■  ! 

I 

Foliis  lobis  7,  oblongis  breve  acuminatis  incisis.  i 

Nom.  Jap.     Tohmatsu.  f 


forma     Yugure,  \n. 

Foliis  lobis    7   oblongis  breve   acuminatis  incisis,   basalibus  con- 
nive ntibus. 

Nom.  Jap.      Y ü (jure. 

forma     Nokibata,  m. 

Folia  nascentia  intense,  in  aestate  pallide  rosea. 
Nom.  Jap.     Nohibata  ($f  ^) . 
Hab.     culta. 

forma     Kihatsijö,  m. 

Folia  nascentia   viridia,    incisa,    autumnales  aurea  deinde  roseo- 
colorata. 

No  if.  Jap.     Ki-hacltijö  (^A^). 

FTab.     culta. 

forma     Kageorinisiki,  m. 

Folia  brunneo-aurantiaca  mox  atro-bruunea,  incisa. 
Nom.  Jap.     Kageorinishiki  (bi^^U). 
Hab.     in  hortis  culta. 

forma     Siguresome,  m. 

Foliis  lobis  7  oblongis  acuminatis  incisis  basi  pauUo  angustatis. 


52  Art.  1.--G.  Koidzumi. 

Nom  .  Ja r .     Sh igurezomc  (a^  rß  îiè) . 
Hae.     in  lioi'tis  culta. 

foniia     Takinogawa,  u\. 

Folia    7-lobata    basi    Iruncata,     lobis    anguste   oblongis  caudato- 
acumiiiatis  versus  basin  cuneato-angustatis. 
Nom.  Jap.      Tahi-no-gcnra  (îhI  /  jil). 

fonna     Kurabuyama,  m. 

Folia  basi  subtruncata,  lobis  7  ol;)longis  acuminatis  versus  basin 
angustatis. 

Nom.  Jai».     Kwahwjama. 

forma     Aoba,  ni. 

Folia  palniatipartita  incisa,  intense  viridia. 
Nom.  Jap.     Aoha  (#^). 
Hai^     culta. 

forma     Karukaya,  m. 

Foliis  lobis  lineari-lanceolatis  versus  basin  cuneatis  caudato- 
acuminatis  argutissime  inciso-seiratis;  saniarae  alis  loculisque 
horizontaliter  patentibus. 

Nom.   Jap.      Karvhaya  (^ij^). 

Haiî,     in  liortis  culta. 

forma     Murasakitaka,  m. 

Folia  subcliartacea  5-G-7-lobata,  pu]i)urascentia;  lobis  incisis. 
Nom.   Jap.      j\lurasal;itala. 
Haiî.     in  liortis  culta. 

forum     Asaji,  m. 

Folia  7-lobata,  lobis  incisis  subtus  secus  nervo,-;  jiaice  adpresse 
pilosis. 

Nom.  Jap.     ./.sc,// (i^S&). 

Hat;,     culta. 


Ri'visio  Ac^ricearnoi  J.T.ponicArum.  53 

forma     Akegarasu,  m. 

Foliis  lobis  7  ol)longis  incisis  acaminatis. 
Nom.  Jap.     Ahc-garasu. 

fonna     Murehibari. 

Foliis  lobis  5,  oblongis  acutis,  suberoso-incisis. 
Nom.  Jap.     Murc-liihan . 

subvar.     heterolobum,  m. 

Folia  palmatipavtita;  laciniis  valde  polyinorphis,  in  eodem  folio 
una  oblonga  acuminata  inaequaliter  serrata,  alteris  lanceolato- 
lineariV)U3  incisis;  ceteris  pinnatifidis  vel  incisis. 

forma     Wabibito,  m. 

Folia  margine  roseo-guttata. 
Nom .  Jap .      Wahibito  (it  A) . 
Hab.     culta. 

forma     Sensunagasi,  m. 

Foliis  lobis  brunneo  vel  roseo-marginatis. 
Nom.  Jap.      Seiisu-nagaslii  (M^-^). 

forma     Hibari,  m. 

Folia  viridia,  lobis  angustioribus. 
Nom.  Jap.      Hlhari(m^M). 

sLibvar.     dissectum,  (Tiro.)  m. 

A.  lU.^sectum,  Thg.  Fl.  Jap.  (178 i)  16,  et  lorn.  PJ.  Jap.  t.  45. 

A .  p  ilmatum,  var.  multifidum,  Koch,  in  Ann.  Mug.  Liigd.  B.xtav.  I.  (1303—04)  -51. 

A.  dccompositinn,  Miq.  in  Catal.  Hort.  Amsteld.  275,  (1866). 

A.  iKilmatum,  var.  dissectum,  f.  ruhrifoUum,  Miq.  Prol.  Fl.  Jap.  (1867)  20. 

A.  palmatum,  var.  dissectum,  (Thg.)  Koch,  Dendrol.  I.  (1SG9).  224  ;— Pax,  I.e.  27  (1902). 

A.  palmatum,  f.  dissectum,  S.  et  Z.  Fi.  Jap.  II.  (1870)  84. 

A.  palmatum,  f.  decomposltum,  S.  et  Z.  I.e.  84. 

A.  palmatum,  var.  dissectum,  (Thg.)  Max.  McI.  Biol.  X.  (ISSO)  6:'8. 


54  Art.  l.  —  G.  KoiJzuuii. 

Folia  5-7-'J-,    rarius   ll-secta,  lobi   auguste  lanceolati  piunatifidi, 
inciso-serrati. 

1.  multifidum,  m. 

A.  dissecliDii,  Thg.  ].c. 

Folia  palmatisecta  lobis  piniiatipartitis  acuminatis,  laciiiiis  lineari- 
lanceolatis  argute  serratis. 

fonna     Aosidare,  m. 

Folia  viridia. 

Nom.  Jap.     Aosltidare  (#fö^). 

forma     Tamukeyama,  m. 

Folia  purpurea. 

No3i.  Jap.      Tamukeijcuna  (^  lA]  ilj).  Cliiriiucn-hiede. 
Hab.     in  hortis  frequens. 

2.  palmatisectum,  m. 

Folia  palmatisecta,  laciniis  auguste  oblougis  utiinque  acuiuiuatis 
incisis. 

foDiut,     Matsukaze,  m. 

Folia  purpurascentia. 

Nom.  Jap.      Matmkuzc  (f$  E). 
Hab.     culta. 

forma ^     Ohsiusidare,  m. 

Folia  purpurascentia;  ramis  pendulis. 
Nom  .  Ja  p.     a  h  sh  iushidare  (^  i'tl  fè  HE) . 

suljvar.     sessilifolium,  (S.  et  Z.)  Max.    m  Ma.  Bioi.  x.  (isso)  cos;- 

Pax,  1.  c.  23. 

A.  sessilifolhm,  S.  et  Z.  I<1.  Jap.  Faiu.  Xat.  in  Abh.  AkaJ.  Müuch.  (1840)  158. 
Ncffiivdo  scssiU/oliinn,  ISIiq.  Prol.  Fl.  Jap,  (1860)  21. 


Kevisio  Aceracearum  J'aponic.imm. 


55 


Petiolis  brevissimis;  foliis  ternatis  vel  quinatis;  foliolis 
ambitu  lanceolato-oblongis  acuminatis  ntrinque  pinnatifidis  vel 
partitis  interdum  pinnatisectis;  lirnbis  inaequaliter  vel  subduplica- 
to-serratis. 

Nom.  jAr.      Hagoromo-hacde  (^^W.),  Kahtircmiiin, 

Hab.     in  hortis  culta. 

Sect.  11.     Iniegrifolia,  Pax. 

F.  Pax,  in  Engl.  Bot.  Jahrb.  VII.  (1SS6)  207,  VI.  (18S5)  327. 

Inflorescentia  paniculata.  Flores  andro-polygaip.i  pentameri. 
Stamina  8  hypogyiia,  in  flore  "^  perigyna.  Discus  extiastaminalis. 
Folia  coriacea  siniplicia  intégra  vel  integerrima.      Species  unica. 

25.     Acer  oblongum,  Wall.  (Tab.  XXIX.) 

Wall,  in  DC.  Prodr.  I.  (1824)  503  ;— Benth.  Fl.  Hongkon.  47  ;  — Hiern  in  Hook.  Fl.  Br.  Ind.  I. 
693  ;-Max.  Mel.  Biol.  X.  (1880)  599  ; -Pax  in  Engl  Bot.  Jahr.  VII.  (1886)  208,  et  Pfl.  Reich.  8 
Heft  (1902)  31  ;— Fokbes  et  Hemsl.  in  Jour.  Linn.  Soc.  XXIII.  141  ;— Ire  et  Matsum.  Tent.  Fl. 
Lutch.  I.  p.  120  ;— Matsum.  et  Hat  at.  lîn.  PI.  Formos.  16  ;— C.  K.  Schx.  111.  H:,ndb.  Laubh.  II. 
(1907)  214  ;— DiELs  in  E.\gl.  Bot.  Jahrb.  XXIX.  499. 

A.  laurifolium,  D.  Don,  Prodr.  Fl.  Nepal.  (1820)  240. 

A.  huzimbala,  Buch.  Ham.  ex  Pas.  1.  c. 

A.  Itoanum,  Hat  ata,  in  sched. 

A.  oblovrjitvi,  var.  Itoanum,  Hay  ata,  in  sched. 
?  A.  alh^-purpure^cenit,  Hatata,  in  sched.     (Ranul.  i  a^m.  cum  4  foL). 

Arbor  rainis  vetustioribus  cinerascentibns  rugulosis;  rainulis 
hornotinis  brunnescentibns  lenticellis  minutis  conspicne  tectis. 
Folia  coriacea  utrinqne  laevia  ininute  reticulato-venosa,  oblonga 
rarius  ovata  vel  ovato-elliptica  interdura  obovato-oblonga,  obtuse 
acuminata,  intégra,  basi  saepissime  rotundata  trinervata,  subtus 
glauca  vel  glaucescentia,  25-55  mm.  lata  5-11  cm.  longa; 
petiolis  gracile  elongatis.  Inflorescentia  dense  corymboso- 
paniculata.  Flores  minuti  andro-polygami;  pedicellis  dense 
ruf  o-puberulis  ;  sepala  oblonga  apice  rotundata  1-1,5  mm. 
longa  extus  dense  rufo-puberula;  petala  oblongo-linearia  2 — 2,5 
mm.  longa  alba  v.  lutescentia;  stamina  8  exerta,  antlieris 
scabriusculis  ;  discus  extrastaminalis;  ovarium  puberulento- 
tomentosum,     stylis    profunde    bilobatis.     Samara    glabra    loculis 


^6  Art.  1.— G.  Koidzumi. 

plerumque  horizontalibus,  alis  1-1.5  cm.  longa  4-10  mm.  lata, 
aiigulo  acuto  vel  obtuso  divergentibus  interdum  subhorizontaliter 
patentibus. 

Nom.  Jap.     Kum-no-lia-kaecle. 

Hab.     Liukiu  et  Formosa. 

DiSTR.     China  australi,  Hongkong,  Hima]a3^a. 

Sect.   12.     Trifoliata,  Pax. 

r.  Pax.  in  Engl.  Bot.  Jahrb.  VI.  (1885)  SI.',,  VII.  (1336)  203  ;— C.  K.  Schn.  111.  H-i-mlb.  Laubh. 
II.  (1907)  193. 

Inflorescentia  umbellata.  Flores  andro-monoeci  vel  dioeci, 
pentameri.  Stamina  10  (8 — 12)  hypogyna.  Discus  extras- 
taminalis.  Stylus  apice  bilobatus.  Folia  membranacea  trifolio- 
lata.     Species  unica. 

20.     Acer  nikoense,  Maxim.  (Tab.  XXX.) 

c.  J.  Maximowicz,  in  Mél.  Biol.  VI.  (ISi7i  370  ;-Fr.  et  Sw.  En.  PI.  Jrp.  I.  (1875)  90  ;— Pax,  in 
Engl.  Bot.  Jahrb.   VIL  (1S8Ö)  2o'>,  et  Pfl.  Eeich.  8  Heit  (1902)  29  ;— Leveil.  in  Bull.  Soc.  Bot. 
Pr.  VI.  (190G)  591  ;— C.  K.  Sch.v.  111.  Ha^nclb.  Laubh.  IL  (1907)  211.  (non.  Miq.). 
A.  Maximo  ciczianHvi,  Miq.  in  Arch.  Xeerlan.  IL  (1867)  472,  47(3  {nec  Fax.) 

Nom.  Jap.  Clulja-iio-lä,  Meguro  (itm^^^;  B  3^  Ui  ^  ?<^  H)  ;  Kocliù- 
no-lii]  Scminoki,  Ohmitsade-haede  (nov.) 

Arbor  ramis  validis  cortice  cinerascente;  ramulis  annotiiiis, 
foliis  subtus,  petiolis  pedicellisque  canescento-tomentosis.  Folia 
ternata  membranacea  supra  sparce  pilosa  subtus  secus  costas 
praesertim  petiolisque  liir.suto-tomentosa;  petiolis  erecto-patentibus; 
foliolis  mediis  oblongis  petiolulatis,  Jateralibus  oblique  oblongis 
subsessilibus,  omnibu-^  integerrimis  obtusis,  rarius  obscuriter  vel 
inaeqiialiler  remote  obtuse  dentatis.  Inflorescentia  triflora, 
floi'ibus  andro-dioeciis  albis  coaetaneis;  bracteolis  linearibus 
cadacis.  Sepala  5  glabra  v.  extus  parce  puberula,  obovata  v. 
ovata  basi  vix  unguiculata.  Petala  obovata  basi  angustata  sepalum 
aequilonga.  Stamina  cire.  10-12  exerta,  antheris  laevibus. 
Discus  crenatus  extrastaminalis.  Ovarium  puberulento-toment- 
osum    stigmatibus   revolutis.       Samara   3.5-5    cm.  longa,     loculis 


lîevisio  Aceracearum  Ji^ponicarnm.  57 

pube^cento-tornentosis;  alis  circiter  13-14  nini.  Jatis  anguloobtuso 
divergentibus. 

Hau.  Hondo:  Azumasan  (Uzen);  Aidzu  (Iwasbiro);  Nikko; 
Togaknsbi^^ama;  Prov.  Kozuke;  O^-ama  (Sagami);  Cbicbibu 
(Musasbi).     Shikok:  Prov.  Awa;  Kiusbiu;  Prov.  Hizen,  Higo. 

DiSTR.     Cliiiia  centraUs:  Hupeb  (sec.  CK.  Schneider). 


Sect.  13.      P!atanoidea,  Pax. 

F.  Pax  in  Exgl  Bot.  Jahrb.  Vi.  (1S85)  327,  VII.  (ISSü)  2  i3  ;-C.  K.  Sch.v.  111.  Handb.  Laubh  II. 
(1907)  193. 

Inflorescentia  corymboso-paniculata.  Flores  andro-polygami 
pentaraeri.  Stamina  8  bypogyna  in  flore  ^  perigyna.  Discus 
extrastaminalis.  Stylus  profunde  bilobatus.  Folia  membranacea 
pahiiatilobata,  lobis  integeriimis  gro.sse  sinuato-dentatis.  Species  2. 


C*iat'is  sitecieruin. 

"       Foliis    lobis    iobulatis    ol)tuse    acuminatis;     samarae    loculi 

tomentosi J .  Miyahei,  ^Pix. 

FoUis  lobis  integris  acuminatis  ;  samara  glabra 

A.  pictwii,  Thunb. 

27.     Acer  Miyabei,  Max.  (Tab.  XXXI.) 

C  J.  Maximowicz,  Mél,  Biol.  XII.  (1SS8)  72')  ; -Pax  in  Engl.  P.l  Keicli,  S  Heft  (190:)  53;  — 
Making,  Tokyo  Bot.  Mag.  (190i)  111  ;-C.  K.  Schv.  111.  Hanclb.  Liubh.  IL  (1907;  229;-Leveil. 
in  Bull.  Soc.  Bot.  Fr.  VI.  (190o)  591. 

A.  Haijatte,  Leveil.  et  Vst.  in  Bull.  Soj.  Bot.  Fr.  VI.  (190J)  590. 

Arbor  ramulis  novellis  glabris  vel  puberulis.  Gemmae 
perulae  interiores  oblongae  apice  rotundatae  vel  obtusae  extus 
pubescento-tomentosae,  exteriores  ovatae  coriaceae.  Folia  mem- 
branacea nascentia  utrinque  densius  adulta  subtus  ad  venas  parce 
pubescentia,  quinquenervia,  5-fida,  elongato-peliolata,  basi 
cordata;  lobis  obtusiter  acuminatis  utrinque  paucilobulolatis. 
Inflorescentia  paniculata;  floribus  foliis  coaetaneis  fulvo-sulp- 
hureis;   pedicellis  pilosis.     Sepala  5  oblonga  apice  rotundata  extus 


58  Art.  1  — G.  Koidzuini. 

pilosa.  Petala  5  pilosa  auguste  spath ulata  sepalis  aequilonga. 
Stamina  8  in  flore  -^  pamm  exerta,  antheris  rotunclatis  glabris. 
Discus  carnosus  8-crenulatus  extrastaminalis.  Ovaria  dense 
puberula  stylis  ultra  medium  bilobatis.  Samarae  loculi  semi- 
orbiculati  fulvescenti-velutini;  alae  oblongae  horizontaliter 
patentes. 

Nom,   Jap.      Kuruhi-ilaiia  (K.  Miyaf.e). 

Hab.  Yezo:  Prov.  Isbikari,  Iburi,  Hidaka;  Hondo  :  Kuro- 
begöri,  PIa_yacbine,  Imoda,  Shige,  Kadoma  (Rikuclnu). 

DrsTR.     endemica. 

28.     Acer  pictum,  Thunb.  (Tab.  XXXH.) 

Thunberg,  FI.  Jap.  (1734),  161,  et  Icon.  Fl.  Jap.  dec.  V.  t.  1  ;— Sieb,  et  Zucc.  Fl.  Jap.  Farn.  Xat. 
in  Abb.  Akad.  Münch.  IV.  2.  (1846)  156  ;— A.  Gray  in  Perrt,  Exped.  Jap.  310  ;— C.  Koch,  in. 
MiQ.  Ann.  Mus.  Lugd.  Batav.  I.  251  ;-Miq.  Prol.  Fl.  Jap.  (1866)  19  ;— Fr.  et  Sav.  En.  PI.  Jap.  I. 
(1875)  87,  et  II.  (lS79j  318:— Max.  Mél.  Biol.  X.  (18S0)  594;— Pax.  Engl.  Bot.  Jahr.  VII.  (1886) 
235,  et.  Pfl.  Eeicli.  8  Heft.  (1902)  47;— Leveil.  Bull.  See.  Bot.  Fr.  VI.  Engl.  (1906)  592 ;-C.  K. 
ScHN.  111.  Handb.  Laubh.  II.  (1907)  225  ;— C.  Koch,  Dendrol.  I.  p.  531  ;— Forb.  et  Hemsl.  Jour. 
Linn.  Soc.  XXIII.  1  tl  ;— Miyabs,  Fl.  Kuril.  223  ;— Xakai,  Fl.  Korea.  I.  p.  133  ;— P.^lib.  Consp. 
Fl.  Korea.  III.  p.  46  ;— Franch.  PI.  David.  77;— Diels  in  Engl.  Bot.  Jahrb.  XXIX.  499;— 
Saegent,  Fort  st  Fl.Jap.  28  ;— Koidz.  PI.  Sachal.  (1910)  89. 

A.  truncatiim,  Fb.  et  Sav.  En.  PI.  Jap.  I.  (1875)  87,  etl.  (1879;  320  (non  Bge.) 

A.  Mnno,  Max.  in  Bull,  Phys.  Math.  Akad.  Petrop.  XV.  126,  et  Prim.  Fl.  Amur.  GS  ;— Eegel, 
Fl.  Tss,  35  ;— Schmidt,  Fl.  Sachal.  119  ;— Komaro.  Fl.  Mansh.  IL  73. 

Ä.  picUim,  var.  inono,  Koesh.  in  Act.  Hort.  Petrop.  XII.  318. 

A.  laetinn,  var.  parvißorain,  Regel  in  Bull.  Phy.  Math.  Akad.  Petrop.  XII.  219. 

A.  amhifjiaivi,  Dippel  (vcc.  Heer)  Laubh.  IL  457,  fig.  218  ; — Pax  in  Engl.  Pfl.  lieich.  8  Heft. 
(1102)47.  (serculus  sterilis  !) 

A.  pictum,  var.  amhiguum,  Pas,  in  Engl.  Bot.  .Jahrb.  XVI,  401  ;—C.  K.  Schn.  111.  Handb. 
Laubh.  IL  (1907)  225. 

A.  jnctum,  var.  Paxi,  Gr.  v.  Schw.  in  Gart.  Fl.  (  )893)  458 

A.  BippeU,  Gr.  v.  Schw.  in  Gart.  FI.  XLII.  (1893)  460. 

A.  Mayri,  Gr.  v.  Schw.  in  Mittl.  D.  D.  G.  (1901)  58  ;— Matr,  Fremdl.  Wald  u.  Parkb.  fig.  150. 

Nom.  Jap.  Tohhra-l-aede,  Taiita-momijt]  Shirahi-l-aedc  (jCity'<  ¥. 
7}^^ m  >\-  Ä.  li^S^m^.  1809.) ;  Itagi,  Yurokho-lcaedc  (5:  -fc  -b  ^.  7K  ^  ^  X  Ä. 
:^^mmm.  1 8 1 0) ;  Itai/a-haede  (Ê,:^SfkM  m) ;  Oh -Imcdc  (M.  ^  3«  ^  Ä . 
:^^?is::f:);  O]i-tsuta-iuomiji\  Kilnim'-inomiji^  liagi-kaedc  (J.  Matsu- 
mura,  Shokubutsu-Meii.  no.  42);  Sandiciede,  Ao-hicde,  hranc-luicde 
(^:^^^^. ),    liaija  ;  Tuldwa . 

Arbor  magnus  ramis  ramulisque  ab  initio  glabris.  Gemmae 
perulae  interiores    oblanceolato-oblongae   apice  obtuse  rotundatae 


Tvevisio  Aceraccfirum  Japonicarum. 


59 


extus  ferrugineo-velutinae  aut  pubesceiites.  Folia  maxime 
variabilia  (Fig.  3 — G.)  membraiiacea  deinde  subchartacea  elongato- 
petiolata,  subtus  plus  minus  hirta,  ambitu  lotundata  semirotundata 
vel  orbiciilato-reniformia,  (3 — )  5  ( — 9)  lobata,  basi  subtrmicata 
cordata  aut  rotundata;  lobis  triangulari-ovatis  integris  subcaudato 
acuminatis  vel  acutis;  petiolis  gracile  elongatis.  Flores  andro- 
pol^-gami  sulphurei,  foliis  coaetanei,  in  paniculam  glabram 
plurifloram  dispositi;  pedicellis  ebracteolatis.  Calj^x  glaber  sepalis 
5  late  oblongis  apice  rotundatis.  Petala  5  anguste  oblonga  sepalis 
aequilonga.     Stamina  8   disci   medio  inserta;   antheris  rotundatis 


Via.  3 


60 


Art.  1. — G.  Koi'lzumi. 


Fig.  4. 


laevibus.  Discus  carnosus  extrastaminalis  margine  crenato. 
Ovaria  glabra  styles  profunde  bilobatis,  laciniis  revolutis  intus 
stigmatosis.  Samara  glabra,  alis  loculis  circiter  duplo  longioribus 
in  varus  angulis  divergentibus. 

Nom.  vulg.     Itaija-Kaede. 

Hab.     Saghalin,  Kuril,  Yezo,  Hondo,  Shikok  et  Kiushiu. 

DiSTE.     Korea,  China,  Manshuria,  et  Amur. 


Eevisio  Acorace".rum  Japonicaruui. 


61 


Fig.  G. 


a.       typicum,  Gr.    v.    ScHW.,  mGartn.  F1. 

XLir.     (1893)    45S;— Pax,   in    Engl.    Pfl.    Eeich    8    Heft. 
(1902)  47. 

Folia  glabra  vel  subtos  secus  nervös 
plus  minus  pubescentia  et  in  axillis 
costarum  tomentosa,  5  ( — 9)-lobata,  basi 
subtruncata,  vel  aperte  cordata  ;  lobis  ovatis 
vel  ovato-lanceolatis  longe  acuminatis. 
Planta  juvenilis,  serculi  steriles  et  inno- 
vatio  novus  foliis  valcle  abnormalibus 
(fig.  6.)  plerumque  basi  profunde  cordatis 
lobis  basalibus  parallel  is. 


Subvar.  1.     eupictum,  Pax,  ].  c.  47. 

A.  incium,  var.  eiij)iclinii,  Pax,  Bot.  Jahrb.  VIT.  236. 

Folia  plerumque  5-lojata  triloba  immixta;  alae  samarae  angulo 
acuto  divergentes. 


62 


Art.  1. — G.  Koidzumi. 


Nom.   Jai\     Itcujalcaede. 

Hab.  Yezo  :  Sapporo, 
Konuma  (Oshima).  Hondo: 
Aclzumasan,  Nikkosan, 
Togakushiyania,  Chichibu, 
Ki^o,  prov,  Hida.  Sliikok. 
ïosa. 

Subvar.  2.  Mono,  (Max.) 

Pax,    in   Engl.    Pfl.   Reich.   8  Hoft. 
(1902)  47;— Xakai,  I.e.  133. 

A.  Mono,  Max.  Prim.  Fl.  Amur.  68. 
A.  pictitm,  vir.  y.   Mono,  Max.  MJl. 
Biol,  X.  60O. 

A.  hietum,  v,.r.  parvifl.rum,  Regel,  in  Bull.  Akad.  St.  Petersbg.  XV.  (1857)  219 ;-C,  K.  Schn, 
1.  c.  225. 

Folia  5-lobata.     Samarae  akie  angulo  fere  recto  divergentes. 
Nom.   Jap.      Itcujakacàe. 

Hab.     Yezo:    Rishiri,   Shiribesi;    Hondo:    Iwatesan,    Nikko, 
Chichibu,  Hakusan. 


Fisr.  8 


Fiff.  9. 


Revisio  Aceraceanim  Japonicaru  ii.  63 

Subvar.   3.     Savatieri,  Pax,    m  engl.  Pot.  jahrb.  vir.  236,  et  pa.  Reich. 

I.e.  47;— Xakai,  I.e.  133  (fig.  7  ) 

Folia  plerumque  T-lobata  pentaloba  immixta,  basi  profunde 
cordata;  lobis  late  triangalaribus  subito  acuminatis.  Alae  samarae 
subhorizoDtaliter  vel  angulo  subobtu.so  divergente.5. 

Nom.  Jap.      Itomald-itaija. 

Hab.  Yezo:  Jozankei;  Hondo:  Chichibu,  Hakusan,  Nik- 
kosan,  Azumasan,  Iwatej^ama,  Togakusbiyama,  Hakkodasan, 
Seugantöge. 

forma     septenlobum,  m. 

forma     novemlobum,  m.  (fig.  5,  a.) 

Subvar.  4.     Mayri,  (Gr.  v.  Schw.)  m. 

Acer  2Iauri,  Or.  v.  Schw.  in  Mitt.  D.  D.  G.  1.  e  ;  Pax,  I.e. 

Folia  plerumque  5-lobata  7  loba  immixta.  Alae  Samarae 
rectae  paralleles  saepe  versus  apicem  conniventes. 

Nom.  Jap.     Itaijahacde. 

Hab.  Yezo:  Sapporo,  Konuma,  Jozankei;  Hondo:  Hakko- 
dasan; Sawanouchi,  Sengantöge  (Rikucbiu);  Azumasan;  Matsu- 
mine  (Shônai)  ;  Ikaho;  Nikko;  Togakusbiyama,  Ontake.  Sikok: 
Tsurugisan  ;  Nanogawa  (Tosa). 

Subvar.  5.  Futagoyama,  m.     (fig.  8.) 

Nom.  Jap.     Futago i/ ama  (m^  \ii). 
Hab.     in  bortis  culta. 

var.  ß.   Paxil,  Gr.  v.  Schw.  i.  c.  458  ;-Pax,  i.e.  47. 

Folia  5-lobata  subtus  pubescentia. 
Nom.  Jap.      Oni-itaya. 
Hab.     lidesan  (Uzen). 
DisTR.     Korea. 

var.  r,   disSeCtUm,  WeSMAEL,     in  BuU.  Soc.  Bot.  Belgique  (1S90)  56;— Pax, 

I.e.  47.  (Fig.  9,  a.) 

A.  jiicUim,  var.  angustilobum.  Making,  Bot.  Mi^g.  Tokyo  VI.  51.  (nomon  nudum) 


64  Art.  1. — Gr.  Koidzumi. 

Folia  glabra  su])tiis  in  axillis  costarum  tomenlella,  palmato 
5 — 7  partita;  laciniis  lanceolatis  vel  anguste  oblongis  acuminatis. 

Nom.  Jap.     Asahi-lxocdc. 

Hab.  Honclö:  Goyözan,  (Rikuchiu);  Awone  (Rikuzen); 
Fukushima  (Iwasliiro);  Nikküsan,  prov.  Közuke;  Hakone, 
Ohyanm,  Yugashima;  Chichibu;   Amagisan.   Sikok:  Tsurngisan. 

Subvar.       SUbtrifidum,  MakI>-0.    Bot.  Ma-  Tokyo  aOOl)  114. 

Folia  fere  semper  tripartita. 
Nom.  Jap.  Yü'jurunia-lMcde. 
Hab.     Takaoyama  (^lusashi). 

var.  ö.  glaucum,  m. 

Gemmae  perula  extus  puberula.  Folia  minora  cire.  7  cm. 
longa  ac  lata,  palmato  5-fida,  subtus  glauca  in  basin  costarum  vix 
pilosa;  lobis  lanceolato-ovatis  acuminatis.  Samara  glabra  minor 
16 — 20  m. m.  longa,  alls  subborizontaliter  patentil)ns  10 — 12  m. m. 
Ion  gis  7 — 8  mm.  latis. 

Nom.  Jap.      Vrajiro-itaija.  (nov.) 

Hab.     Mogariyama  (Uzen). 

subvar.     latilobum,  m. 

Folia  minor  -4 — 5  cm.  longa  ac  kita,  ambitu  quadrangulata  5- 
lobata  3-loba  (fig.  0.  b.)  immixta,  basi  truncata  rarius  rotundata, 
lobis  deltoideis  subito  acuminatis.     Samara  ut  in  praecedente. 

Nom.  Jap.      Urajiro-üomald-itaija.  (nov.) 

Hab.     Azumayama  (Uzen). 

Sect.  14.   Diüholica,    ni. 

Inflorescentia  cor\^mbosa.  Flores  dioeci.  Sepala  fl.  'S^  5 
inter  se  connata,  in  fl.  $  libera.  Petala  fl.  '^  1 — 5  cum  sepaJis 
connata,  in  il.  -Ç  libera.  Stamina  8 — 9  perigyna.  Stylus 
profunde  bilobatus.  Discus  extrastaminalis.  Folia  membranacea 
simplicia  palmatilobata.     Species  unica  in  Hondo  incola. 


lîovisio  Acf^raiCoainnii  Japonioarnui.  b-) 

29.     Acer  diabolicum,  Blume.  (Tab.  XXXTIT.) 

C.  Kocir,  in  Miq.  A71u.  Mus.  Lnj^d.  Bafcav.  T.  (ISGÎ-(M.)  2r>l;— MiQ.  Prol.  FI.  Jap.  (ISj.'j-CG) 
20;-Fk.  ot  S\v.  En.  Tl.  .Tax  I.  (187.'>)  87,  IL  (1879)  320  ;-Max.  Mel.  EioL  X.  (18S0)  593  ; -Pax, 
in  Enol.  Eot.  Jahrb.  VIL  (1S8G)  251  ;-Leveil.  BuU.  Soc.  Bot.  Fr.  VL  (1008)  590  ;-Pax  in 
Kngl.  Pfl.  Il(icli.  8  Heft  (1902)  71  :-C.  K.  Schn.  IlL  Handb.  Laubh.  IL  (1907)  241. 

A.  pitrpitrancena,  Fk.  et  Sav.  I.e.  IL  320  ;  — Max.  Mel.  Biol.  X.  (1850)  59^^  ;  ~Pax,  1.  c.  251,  et 
71  ;— Leveil.  1.  c.  592. 

A.  tliaboliciim,  Y.\T.  irarimrascota,  Rfhder  in  Sarg.  Tree.  Shr.  I.  (1905)  134.  ;—C.  K.  Sühn. 
I.e.  242. 

Nom.  Jap.  Om-momiji  (^  iEfe  A  ¥.  ^K  ^  Kl  A  it .  m  «"«  W.  ^,  fè  a.  1825)  ; 
KiriJia-haede  (mW^MM);  Kaji-hacdc  (J.  Matsumura,  Shokubutsii- 
Meii,  110.  43). 

Arbor  10 — 20  metralis,  ramuli  novelli  pubescentes.  Gemmae 
penilao  interiores  lanceolatae,  intimae  longe  spathulatae 
feiTLigineo-tomentosae.  Folia  membranacea  demum  subcbartacea, 
juvenilia  supra  pubescentia  subtus  villoso-tomentosa  ad  margines 
ciliata,  adulta  tantum  subtus  ad  venas  pubescentia,  circuinscri- 
ptione  suborbicularia,  basi  cordata,  quinque-lobata;  lobis  ellipticis 
grosse  obtuse  pauci-dentatis,  apice  obtuse  productis,  infimis  multo 
niinoril)iis;  petiolis  quam  lamina  bvevioribus  nunc  glabris  nunc 
versus  apicem  adpresse  pilosulis.  Umbella  5 — 10-flora,  floribus 
coaetaueis  mediocribus  purpurascentibus  ;  pedicellis  adpresse 
pilosis  inferne  bracteolatis  in  anthesi  nutantibus.  Calyx  cam- 
panulatus  sepalis  5  ellipticis  apice  rotundatis  in  flore  -^  connatis. 
Petala  5  albo-rosea  sepalis  aequilouga,  in  flore  -^  cum  sepalis 
adnata.  Discus  crenatus  extrastaminalis.  Stamina  8  filamentis 
antherisque  laevibus  in  anthesi  exerta.  Ovaria  tomentosa  stylis 
ad  basin  bipartitis.  Samara  cum  loculis  circ.  3 — 3,5  cm.  longa; 
loculis  in  faciebus  cristatis,  adpresse  ferrugineo-tomentosis,  extus 
distincte  carinatis;  alis  saepe  parallelis  vel  fere  aiigulo  recto 
divergentibus. 

Hab.  Hondo:  Nikkösan;  Ikaho;  Chicliilm;  Mitake  (Mu- 
sashi);  Usuitöge;  Hakone;  prov.  Owari.  Shikok:  Tsurugisaii  ; 
Nanogawamura  (Tosa).  Kiusiu:  Inutake  (Buzen). 

DisTPv.     endemica. 

ÇFinisy 


06 


Distributio  geographicalis  specierum  in  Arcipelago  Japonico. 


A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 
A. 


Loc. 

Spec. 

parvifloruiu 

clistylum 

crataegi folium  ... 

insulare  

Kawakaiuii 

ovatifoliuiii 

morrisonense.... 
carpinifolinin .... 

rnfinerve .. 

cnpillipos 

rubescens 

micrantbuiii 

TsclionosKii 

avf^ntuni 

cissifoliuîii 

rubrum 

trifidum  

Ginnala 

spicatum 

Olivei'iaiium...  . 
Sieboldianum.  . 
Sbirasawanum.. 

japonicura 

palmatum 

oblongum 

nikoense 

Miyabei 

pictura 

\aiabolicum. 


X 

X 
X 

X 
X 
X 
X 
X 

X 
X 

X 
X 
X 
X 


fM 


INDEX. 


sectionum,  specierum  et  synonymorum. 

Acer,  L 2 

Acer  albo-purpuraycens^  Hayata 55 

Acer  aiiihi(juuiii,  Dippel 58 

Acer  argutam,  Max 25 

Acer  Buerger iamuii,  Mjq 20 

Acer  Bih'njrrln tun,  xiw.  niiujpoensc,  Keuder.       ... 29 

Acer  Bu'dmhdla,  Buch-Ham 55 

Acer  capillipes,  Max 21 

Acer  capillipes,  ivu-.  fujisaneuso,  Koidz 21 

Acer  carpiuifolium,  S.  et  Z 17 

Acer  caudal  if oliuni,  Ha  yaï 15 

Acer  caudatifii,  Matsum. 15 

Acer  caudal  am,  Matsum  .  et  Ha  y  at ...  15 

Acer  caadatuiii,  Yi\Y.  uharundacnse,  Render ...  o2 

Acer  circamlobatuuij  Max 42 

Acer  circaiuluhalwii,  var.  Heijhachil,  Marino 41 

Acer  circaniîohaluiii,  var,  insulare,  Pax 42 

Acer  cissifolium,  Koch 20 

Acer  crataegifolium,  S.  et  Z 13 

Acer  crataegifolium,  rar.  Veitcliii,  NrcfioLS 14 

Acer  crataegifoliuiii,  var.  ti/piciiin,  Gr.  v.  Schw 13 

Acer  cucidlohraclealinn,  Levfal.  at  W'ST 13 

Acer  decoiiipositain,  Miq 53 

Acer  dedijle,  Max 32 

Acer  diabolicum.  Be 65 

Acer  diaholicam,  var.  purpurascens,  C.  K.  Sciin.         05 

Acer  diabolicwn,  ssp.  argiitain,  Wesml 25 

'  Acer  Dippeli,  Gr.  v.  Schw 58 

Acer  dissectum,  Tug 53 

Acer  distylum,  S.  et  Z 13 

Acer  duplicatoserratuin,  Hayat 50 


Üö  Index  Sectionuui,  b'pccieruui,  et  Synouyuioruui. 

Acer  Ginnala,  Max 30 

Acer  Ginnala,  var.  yezoense,  Koidz 31 

Acer  Haijatac,  .Ly^vl.  et  y:^T 57 

Acer  insulare,  Manixo 14 

Acer  Itoanum,  Hay  ata 55 

Acer  japonicum,  Tlig 39 

Acer  japoiiicLim,  car.  aurcuin,  Gii.  v.  Sciiw 41 

,,             ,,            i'är.  circumlobatum,  Koidz 42 

,,             ,,           car.  Heyhachii,  Matsüm 41 

,,              ,,            cdr.  Kasadü,  Koidz 42 

,,              ,,            t'cir.  Kükonoe,  Koidz 43 

,,              ,,            rar.  Matsuyoi,  Koidz 43 

,,              ,,            rar.  micro})]iy]liini,  Koirz 42 

,,              ,,            rar.  Parsunii,  Vkitcii 4] 

,,              ,,            i'cir.  Sayosigure,  Koidz 43 

,,              ,,            y i\v.  S ichoklianuin,  F II.  dt 'ë)AY 35 

,,              .,            car.  typicuin,  Gii.  v.  Sciiw 40 

,,              ,,            rar.  villosum,  Koidz 42 

Acer  Kawakaiuii,  Koidz 15 

Acer  laäiuii,  \i\y.  iHirci/lonuii,  'Regv.l, 02.58 

Acer  lasiocarpum,  Levl.  et  Vxt 32 

Acer  laurifolntiii,  D,  Dox 55 

Acer  Maximoiüicziaiiuin,  ]\Iiq  .         5G 

Acer  Mau rii,  Gr.  \.  Sciiw 03.58 

Acer  Meihots, '^lEJ} 46 

Acer  micranthnm,  S .  et  Z 23 

Acer  Miyabei,  Max 57 

Acer  Mono,  Max 02. 58 

Acer  morrisonense,  Hayata 16 

Acer  iiikoense,  Max 56 

Acer  oblongum,  Wall 55 

Acer  oblong  um,  var.  Itoaiuim,  Hayat 55 

Acer  Oliveriannn],  Pax 33 

Acer    Oliverianum,    rar.     Nakaharae,     scar,     formosanuni, 

Koidz 33 

,,              j,               ,,      Nakaharae,  Hayat 33 


Jutk'X  Scctiouum,  i^pcciorimi,  ct  Syuouyuiorum.  69 

Acer  Oliveriaiuiiii,  vdr.  Nakabarao,  scar,  trilubuni,  Koidz....  34 

Acer  ovatifoliun"«,  Koidz 16 

Acer  palmatuni,  Tito 44.43 

Acer  palmattmi,/.  decomposiluiii^  S.  et  Z 53 

,,          ,,          J\  dissecfwji^  ^,  at  Z.    ...     53 

,,          ,,           J .  linear ilohiuu ^  S.  et  Z 48 

Acer  palmatum,  siihsp.  geiiuinuni,  Koidz 44 

a.  amabile,  Koidz 45 

sulvar.  Kagiri,  Koidz 46 

b.  spectabile,  Koidz 44 

suhrar.  crispiim,  Koidz 45 

,,         Hikasayania,  Koidz 45 

Acer  palmatum,  .s?(6s;?.  Matsumurae,  Koidz 40 

b.     horteiisis,  Koidz 50 

•srrtr.  dissectum,  Koidz 53 

scar,  lieterolobiim,  Koidz 53 

scar.  ])ahiiatilobum,  Koidz 50 

scar,  palmatipartitum,  Koidz 50 

scar,  sessibfolium,  Koidz 54 

a.     spontaneum,  Koidz 4Ü 

scar,  elegaiis,  Koidz 50 

scar,  formosanum,  Koidz 50 

Acer  palmatum,  suhsp.  septenlobum,  Koidz 46 

car.  latilobatum,  Koidz 47 

car.  palmatipartitum,  Koidz 47 

scar,  linearilobum,  Koidz 48 

scar.  Tanabata,  Koidz 48 

car.  speciosum,  Koidz 47 

Acer  paliiial mil,  var.  dissectain,  Koch 53 

Acer  palmalwii,  var.  dissectwii,  1".  ruhri/oliuiii,  MiQ 53 

Acer  pahiiaf Ulli,  \'ài\  geniiinuiu^  S.  et  Z 44 

Acer  palinatuiii,  var,  Unearilohiuii,  Miq.,  S,  et  Z 48 

,,          ,,             ,,     multifidum,  Koch 53 

,,          ,,              ,,      palmatißdiun,  ^.  et  7j 46 

,,          ,,             ,,     palmatum,  Koch.         44 

,,          ,,             ,,     septenlobum,  Koch,  Miq 46 


70 


Index  Soctionum,  Spccioruui,  et  Synonyuioriiui. 


Acer  'palinatwii,  «.  Tliuithoyi,  Pax 

Acer  palniafwiij  «.  Thunhergi,  cupalmatuiii,  Gr.  v, 
Acer  paliuatiuiij  «.  Thimherrji,  b.  septenlohiim ,  Gr. 

Acer  palm atwu,  var.  trilobuiii,  Koch 

Acer  parviflorum,  Fr.  et  Say 

Acer  Paxi,  yhy.  ningpocnse,  Pax 

Acer  pcllucidohract.eatwu,  Levl.  et  V^'T 

Acer  pennsijlcaniciuii,  var.  capillipes,ViES^iL. 

,,  ,,  ,,     ru/incri'c,  Wesml 

,,  ,,  ,,     parvijlorwii,  Wes.ail.     .. 

Acer  pictuiii,  Tjig 

Acer  i)ictu]ii,  var.  ((;^/6/^H/y/,  Pax 

,,  ,,  ,,      cDigustilohwii,  .Makixo 

,,  ,,  ,,     dissectuiii,  AVesmael. 

,,  ,,  ,,     (lissectuii),  scar,  subtriluluin 

,,         ,,  ,,     glaucuin,  KoiDz 

,,  ,,  ,,     glaucuni,  scar,  latiluluiiii,  K 

,,  ,,  ,,     eupictuiii,  Pax 

,,  ,,  ,,     mono,  ^\\x.,  KoRsn 

,,  ,,  ,,     Paxi,  Gr.  v.  Schw.     

,,         ,,  ,,     typicLiiii,  Gr.  v.  Schw. 

Acer  pictuni,  a.  t3^picum,  eupictum,  Pax. 

,,         ,,         ,,  ,,         Futago^-arna,  Koidz 

,,  ,,         ,,  ,,  Alayrii,  Koidz. 

Mono,  Pax 

,,  ,,         ,,  ,,  Savatieri,  Pax. 

Acer  pühjmorphnm, '^.  Qt  Zi 

Acer  purpurascens,  Fr.  et  Sav 

Acer  pijcnanthum,  C.  Koch 

Acer  rubescens,  H ayat 

Acer  rubrum,  L 

Acer  rufinerve,  S .  et  Z 

Acer  rußnerve,  ExGL.,  Matsum.  et  Ito,  Ford,  et 
Acer  rufinerve,  f.  albo-]imbatum,  Hk.  I'il 

,,  ,,  f.  marginatum,  Pax 

,,  ,,  f.  marmoratum.  Pax 


Sciiw. 
V.  Schw 


OIDZ. 


>L\K 


Hemsl, 


44 
44 
46 
29 
11 
20 
24 
21 
19 
11 
58 
58 
03 
63 
64 
(34 
64 
61 
62.58 
63.58 
61 
61 
63 
63 
62 
63 
43 
65 
27 
21 
27 
19 
15 
20 
20 
20 


Tmlex  Pocfioniiin,  Spoeù^rnm,  ot  Synonynioruni. 


71 


Acer  ruf i nerve,  f.  normale^  (  Ju.  v.  St-nw. 

Acrr  seiniorhiciihdiiiii,  V.w 

Acer  scpknloJiniu^  Tin; 

Acer  serrulatwii,  Hayat 

Acer  sessilij'oUum^  S.  et  Z 

Acer  Shirasawanum,  Kotdz 

var.  tenuifolimn,  Koidz 

Acer  Sieboldianum,  MiQ.     

Acer  Sieboldianum,  var.  micro pliijlhim.^  Max. 

,,     tortuosum,  Max. 
,,     tsusimense,  Kotdz. 
,,      typicum,  Max. 
Acer  spicatum,  Lam.  rar.  ukuriinduense,  ]\1a 

Acer  spicaliiiii^  var.  ii.'isiiricnse,  Bud 

Acer  tftlaricuDi^  var.  acuiuivaiwu^  Franch 

,,  ,,  ,,      (liz-neiise^  Fr.  et  S  a  v.... 

,,  ,,  ,,      ewiiiinala,  Pax 

Acer  tdtaricum,  var.  (jinnala^  jMax 

,,  ,,  ,,      hiciniatwii.,  Regel. 

Acer  tri  fi  du  m,  Hook.  et.  Arn 

Acer  trifidum,  var.  formosanum,  Hayat. 

,,  ,,  var.  integrifoliuin,  Makiko. 

,,  ,,  var.  uingpoense,  Hange.... 

Acer  trinerve,  Dippel 

Acer  truncdiwu,  Fr.  et  Sav 

Acer  Ttrchonoski,  Max 

Acer  Tutclieri,  var.  Sliimadcie,  Hayat. 
Acer  iihir  undue  use,  Tr.  et  INIey.    ... 

Accra,  Juss 

Aceraceae,  DC 

Aceriueae,  DC.  

Adiscantha  V AX 

Arguta,  Rehder ... 

Campestria,  Pax.  

Carpinifolia,  Koidz 

Cissifolia,  Koidz 


10 
27 
40 

oo 

OO 

54 
38 
39 
35 
37 
37 
37 
30 
31 
32 
30 
30 
30 
30 
30 
29 
30 
30 
29 
29 
58 
24 
34 
32 

9 
9 

2 

4.20 

25 

4 

17 

20 


7Q 
^  Indox  Soctionnui,  Spocicruui,  it  Synonymornin. 

Dial)olica,  Kotdz 64 

Euaccr,  OvvA 2 

Extrastaminalia,  Pax 26 

Glabra,  Pax I] 

Iiidivisa,  Pax 12 

IntegrifoJia,  Pax 55 

Intrastaminalia,  Pax 11 

Lithocarpa,  Pax 3 

Macraiitha,  Pax 18 

Negundium,  Rafix 2 

Ncgumh,  Ludwig ...     2 

Negiindo,  Pax 4 

Ncguiido  cissifolia,  S.  ct  Z 26 

Negundo  nilcociw^  jMiq 26 

Ncgiindo  scssilifolia,  Miq ~A 

Palniata,  Pax 35 

Palnialoidoa,  Kotdz 22 

Parvillora,  Kotdz. 11 

Perigyna,  Pax 26 

Platanoiden,  Pax 27 

Rubra,  Pax 27 

Ridac,  Adaxs 2 

Saccharina,  Pax ...     4 

Sapindaccae^  ^Q\\  Acereat\  Bxii^j. 2 

Sapindaccac,  subord.  Aecrincae,  B.  et  H.,  Drut)e 2 

Spicata,  Pax 28 

Trifoliata,  Pax 56 


INDEX. 

Nominum  japonicarum. 


Agajinishiki    

Akabanauchiwakaede 

Akegarasu       

Akitsushima    

Akitsuta 

Aniakkokaede 

Amakogi 

Amakngi 

Aoba 

Aokaede 

Aokaede 

Aomomiji        

Aoshidare        

Aouri        

Arahaga    

Arahago   

Arahana 

Asahikaede      

Asaji 

Asanoha 

Asanohakaede 
Ashibosourinoki     . . . 

Aj^aigasa 

Chidorinoki     

Chidorinoki     

Chirimenrnomiji     . . . 

Chisato     

Chishio     

Chöjanoki        

Fuiri-kourikaede    . . . 
Fuiri-urikaede 

Futagoyama    

Hagoromokaede     . . . 


45 

40 

53 

51 

48 

26 

26 

26 

52 

19 

58 

19 

54 

14 

32 

17 

32 

64 

52 

50 

25 

21 

38 

13 

17 

54 

47 

44 

56 

14 

20 

63 

55 


Hakusanmomiji     ...     . 
Hanaizumiiiishiki ...     . 

Hanakaede      

Hanakaede      ...     ...     . 

Hanakaede      

Hanakazura    

Hananoki 

Hatsuyukikaede     ...    . 
Hauclnhakaede      ...    . 

Hibari       

Hikasayama    

Hina-itayameigetsu. 
Hina-ncbiwakaede.. .  . 
Himeogarabana  ...  . 
Himeuchihakaede  ...  . 
Hiroha-ashibosourinoki 
Hirobamomiji  ...  . 
Hitotsubakaede      ...     . 

Hon-uri    

Hosoekaede     

Hosoeurihada 

Hozakikaede 

Ichigyoin 

lizaku       

Irohamomiji    

Itagi 

Itagikaede        

Itagokaede       

Itaya 

Itayakaede     ... 

Jtayameigetsu 

Itomakikaede 

ItoDiaki-shimarnomJji  - 


24 
46 
14 
27 
30 
29 
27 
20 
40 
53 
45 
38 
39 
24 
37 
21 
.     ...         47 
.     ...         13 
.     ...         14 
.    ...         21 
.     ...         21 
.    ...         32 
.    ...         47 
.    ...         19 
.    ...    45.44 
.     ...         58 
.    ...         58 
.    ...         13 
.    ...         58 
63.62.60.58 
.    ...         36 
.    ...         63 
33 


74 


Index  Nouiiuuui  japonicarum. 


Iwanekaede     

58 

MegLiro     

5G 

Iwatobeni        

44 

Megusurinoki 

27 

JOnihitoye        

40 

Meigetsukaede 

40 

Kaede       

44 

Meurinoki        

14 

Kaedenoki       

44 

^linekaede       

24 

Kaerudenoki 

44 

Mitsudemomiji 

26 

Kageorinishiki 

51 

Mitsudekaede 

26 

Ivagiri       

40 

.Mitsudeshimamomiji 

34 

Kajikaede        

05 

Miyamainoiiiiji 

25 

Kakuniino       

'2U 

Mochinesso      

30 

Xanokogi 

:]() 

^[omiji      

44 

Karakogi  kaede 

m 

Moinijigasane 

37 

Ivarakaya 

:y> 

Monnisbiki      

51 

Kasado     ...     ...     ... 

42 

Musen       

44 

Kasatoriyama 

38 

Murasanie        

39 

Kibanauch  iwakaede 

30 

Murasakitaka 

52 

Kibunemomiji 

58 

Muieliibari       

53 

Kihachijö         

51 

Musasbino       

47 

Kirihakaede    

65 

Nanayeitaya    

30 

Koba-itayameigetsu 

37 

Nisbikigasane 

50 

Kobaonagakaede    ... 

10 

Nokibata 

51 

Kochönoki      

50 

Nomurakaede 

47 

Kokonoe 

43 

Ogarabana        

32 

Koniinekaede 

23 

Ogurayama      

40 

Komoninshiki 

45 

Ohbaurinoki    

19 

Kûmorikaede 

19 

Obgarabana     

21 

IvoDJinoki        

19 

Obitayaraeigetsu     . . . 

38 

Kourikaede      

14 

Obkaede 

58 

Kurabuyama 

52 

Obkarabana     

21 

Kurobi-itaya    

58 

Ohmeigetsu     

42 

IvQsunohakaede 

50 

Ohminekaede 

19 

Maikujaku       

41 

Ohmitsudekaede    . . . 

56 

Marubakaede 

13 

Obmomiji        

47 

Marubatôkaede       ...     . 

30 

Ohsakazuki      

48 

Matsukaze        

54 

Ohsiusbidare 

54 

jMatsnyoi ...     . 

43 

Obtsutamomiji        ...     . 

58 

Index  Xominum  japonicarum. 


V5 


Ohurikaede      

10 

Tanabata 

48 

Okushimo        

45 

ïaniasa     

17 

Onngakaede     

IG 

Tetsukaede      

11 

OuimoQiiji        

C5 

Tetsunoki        

11 

Oniitnj'a 

G3 

Tökaede    

..    29.30 

Oridononisliiki 

46 

Tokiwa      

58 

Sarukaede        

58 

Tokiwakaede 

58 

Sayoshigure     

43 

Tokonatsu        

51 

Seminoki 

56 

Tsubanoki        

17 

Senri 

47 

ïsucliigumo     

46 

Sensunagashi  

53 

Tsukubaue       

48 

Shigitatsu 

50 

Tsutamomiji    

58 

Shigarezome    

52 

Urajiroitaya     

64 

Shimamomiji  ...     ... 

33 

ürajiroitomakiitaya 

64 

Shimaurikaede 

15 

Uribadakaede 

19 

Sbimenouchi 

48 

Urii 

19 

Shirahashinoki 

14 

Urikaede 

14 

Shirakaede       

14 

Urikko      

19 

Shirakikaede    

58 

Urinoki     

19 

Sbirashide        

17 

Wabibito 

53 

Sbirobanaitayameigetsi 

j..         37 

Yacbiitaya 

30 

Sbirosbide        

17 

Yagurumakaede     . . . 

64 

Sodenoucbi      

37 

Yamaasagara 

32 

Takaomomiji 

47 

Yamakaede      

14 

Tak  asagomomij  i     ... 

50 

Yamamomiji 

49 

ïakasagotôkaede    ■ . . 

30 

Yamasbibakaede    . . . 

17 

Takasagourikaede  . . . 

17 

Yezokarakogi 

31 

Takasagourihada     . . . 

22 

Yezomeigetsu 

42 

Takinogawa 

52 

Yugure     

51 

Tainukeyama 

54 

Yorokkokaede 

58 

G.   KOIÜZCMl. 

BEVISIO  ACERACEARÜM  JAPOfilCARUM. 


PLATE     I. 


EXPLICATIO  TABULAE    I. 

Acer  inirdflorum^  Fr.  et  Sav, 

1.  Raiiuilus  fructifer  (mag.  nat.). 

2.  Flos  ^  (aug.). 

3.  Stamen  (valde  aug.). 

4.  FJos  hermaphroditus  (aug.)- 

5.  Verticalis  ejus  sectio  (aug.)- 


Jour.Sci.Coll.,Vol.KX){ll.,Art.l.PI.  I. 


G.Koidzumi    del. 


F.  Fujisawa  sculp. 


: 


G.   KOIDZUniI. 

REÏISIO  ACERJCEARUM  JAPONICABUM. 


PJL  ATE    II. 


EXPUCATIO  TABULAE    II. 


Acer  distyhim,  S.  et  Z. 

1.  Ramulus  fmctifer  (mag.  nat.). 

2.  Flos 'S^(ang.). 

3.  Sectio  ejus  verlicalis  (aiig.). 


Jour.  Sei.  OgIL,  Vol.  XXX/I.Jrt.J.  PI.  II. 


G.Knidzuiiii    del. 


F.  Fujisawa    <rulp. 


G.   ICOIDZUAII. 

REViSIO  ACERACEARUi  JAPOBICARUM. 


PLATE     m. 


EXPLICATIO  TABULAE    IM. 

Acer  craiacgifoliiuii,  S.  et  Z. 

1.  Eamuli  fructiferi  (mag.  nat.). 

2.  Flos  hermaphroditus,  sepalis  petalisque  decisis  (aug.). 

3.  Floris  ^  Sectio  verticalis  (aug.). 


Jour.Scl.Coll.,  Vol.  XXXII.,  Art.  I. PI. III. 


G.Koidzumi   del. 


F.  Fujisawa  .souli 


G.   KOIUZUMI. 

REÏISIO  ÄCERACEARÜM  JAPOBICARÜM.  ' 


PLATE     IV. 


EXPUCATIO  TABULAE    IV. 

Acer  insulare,  Makino. 

1.  Ramulus  fructifer  (mag.  nat.). 

2.  Flos  lierraaphroditus  (aug.)- 


Jour. Sei.  Coll.,  Vol.  XXXII ,  Art.  I. Fl. IV. 


(i- Koidzmiii    (1(1. 


F.  Fiiii.-.-,(«-,i   s,u\\'. 


G.   KOlDZUmi. 

REYISIO  ACERACEARUM  JAPONICARUM. 


PLATE    V. 


EXPUCATIO  TABULAE    V, 

Acer  Kaiüakamii,  Koidz. 

1.  RamiiluR  fructifer  (mag.  nat.). 

2.  Flos  heriimphroditus  (aug.)- 


Jour.Sci.  Coli,  Vol.  XXXII.,  ArtJ. PI.  V. 


•i.KniilzilIui     (Ici. 


F.  Kuji.s;uv;i   .sculp. 


a.  Koiuzunii. 

( 


REYISIO  ACERACEARIM  JAPOBICARUM. 


PLATE    VI. 


EXPLICATIO  TABULAE    VI. 

Acer  ovati folium.,  Koiclz. 

1.  Ramulus  fructifer  (mag.  nat.). 

2.  Folium  (mag.  nat.). 
3.-4.     Fmctus  (mag.  nat.). 


Jour. Sei.  Coll.,  Vol.  XXX IL ,  Art.  I.  PI.  VI.     \ 


F.  Fiijisawa  sculp. 


a,  KOIDZFMl. 

REYISIO  ÄCERACEARUM  JAPONICARUM. 


PLATE    VIL 


4 


EXPLICATIO  TABULAE    VII. 

Acer  inorrisoneiise,  Hayata. 

1.     Pars  plantae  tloriïerae  (mag.  nat.)- 
2.-3.     Flos-J^   (aug.)- 

4.     Folium  aclultum  (mag.  nat). 


Jour. Sei.  Coll.,  Vol.  XXXII. ,  Art.  I.  PI.  VII. 


G.  Koidzumi    del. 


F.  Fujisawa  sculp. 


C.  KOIDZUnil. 

REÏ1SI0  ACERACEARUM  JAPONICARDM, 


PLATE    VIII. 


EXPUCATIO  TABULAE    VIII. 


Acer  carpmifolnim ,  S.  et  Z. 

1.  Ramulus  fructifer  (mag.  nat.). 

2.-4.  Flos-S^Caug.). 

3.  Sectio  ejus  verticalis  (ang.). 

5.  Flos  hermapliroditus  (aug.). 


Jour.Sci.  Coll.,  Vol  XXXII.,  Art. I. PI.  Mill. 


ô- 


F.  Fujisawa  sculp. 


G.   KOIDZUMI. 

REVISIO  ACERACEARUM  JAPOUICARUM. 


PLATE    IX 


EXPLICATIO  TABULAE    IX. 

Acer  rußnervc,  S.  et  Z. 

1.  Kamulns  fructifer  (nat.  mag.). 

2.  Flos  hermapliroditus,  sepalis  petalisque  decisis  (aug.). 

3.  Verticalis  ejus  sectio  (aug.). 

4.  Sectio  verticalis  fl.  'S"  (aug.). 

5.  Sepal  urn  (aug.)- 
G.  Petaluin  (aug.). 


Jour. Sei.  Coll.,  Vol.  XXXII,  Art. I. PI.  IX. 


G.Koidziiiui    del. 


F.Fujisawa  sculp. 


G.   KOIDZUmi. 

REVISIO  ACERÄCEARÜM  JAPONICÄRUM. 


PLATE    X. 


EXPLICATIO  TABULAE    X. 

Acer  capillipcs,  Max. 

1.  Pars  plantae  (mag.  nat.). 

2.  Flos  hermaphroditiis  (aug.). 

3.  Sectio  ovarii  vcrticalis  (aug.). 

4.  Flos  -S^  (aug.). 


Jour.Scl.  Coll.,  Vol.  XXXII, Art.l.Pl.X. 


G.Koidzumi    del. 


F.  Fujisawa  sculp. 


G,   KOIDZÜMI. 

REYISIO  ACERACEÄRUM  JAPOHICARUM. 


PLATE    XL 


EXPLICATIO  TABULAE    XI. 

Acer  nihcscens,  Ha3^ata. 
Pars  planlae  fructiferae  (mag.  nat.). 


Jour.Sci.  Coll.,  Vol.  XXXII ,  Art.  l. PI.  XI 


G.Koidzumi    del. 


F.  Fujisawa   sculp.. 


G.  KOIDZUMI. 

REÏISIO  ACERACEAEUM  JAPONICARUM. 


PLATE    XII. 


EXPLICATIO  TABULAE    XII. 

Acer  micrantliiuu ,  S.  et  Z. 

1.  Pars  plantae  fr.  (mag.  nat.). 

2.  Sectio  verticalis  f1.  ^  (aug.). 

3.  Stamen  (valde  aug.)- 

4.  Flos  hermaphroditus  (ang.). 

5.  Fructus  alis  in  angulo  fere  recto  divergentibus    (leviter 

aug.). 
G.     Folium  (mag.  nat.). 


' 


Jour.Sci.  Coll..  Vol.  XXXII.,  Art.  I. PI.  XII. 


(î.Koidziiini    del. 


F.  Kuji.s;!»;!   sculp. 


G.   KOIDZUniI. 

REYISIO  ACERACEARUM  JAPOICARM, 


PLATE    XIII. 


EXPLICATIO  TABULAE    XIII. 


Acer  Tschonoshii,  Max. 


1.  Pars  plantae  fr.  (mag.  nat.). 

2.  Sectio  verticalis  fl.  "^  (aug.)- 

3.  Stamen  (aug.). 

4.  Flos  bermaphroditus  (ang.). 

5.  Sectio  ovarii  verticalis  (aug.). 


I 


Jour.  Sei.  Coll. ,  Vo  I.  XXXII.  ,Art.l.  PI.  XIII. 


n.Koidziiiiii     del. 


F.  Kiijisawa   snilp. 


J 


G.   KOIDZUMI. 

REÏISIO  ACERACEARUH  JAPOMICARÜl. 


PLATE    XIV. 


EXPLICATIO  TABULAE    XIY. 

Acer  argiitum,  Max. 

1.  Ramulns  fructifer  (mag.  nat.). 

2.  rios^(aug.). 

3.-4.  Verticalis  ejus  sectio  (aug.). 

5.  Flos  ^  (aug.). 

6.  Flos  $  cum  staminibus  abortivis  (aug.). 

7.  Sectio  ovarii  verticalis  (aug.  ). 


XIV. 


G.Koiflzumi    rlcl. 


F.  Fujisawa  sculp,  i 


f 


^1 

il. 


%• 


G.   KOIDZUMï. 

REVISIO  ACERÄCEARÜM  JAPOSiCARUlI. 


PLATE    XV. 


EXPLICATIO  TABULAE    XV. 

Acer  cissifoUum^  S.  et  Z. 

1.  Pars  plantae  fructiferae  (mag.  nat.). 

2.  Flos  ^  (valdc  aug.). 

3.  Flos  '^  (valde  aug.). 

4.  Fructus  mat.  (leviter  aug.). 


Jour.Scl.  Coll.,  Vol.  XXXII.,Art.LPI.  XV.  \ 


<i.Koi(]ziinii    del. 


F.  Fujisawa   sriilp. 


4.1 


G.    KOIDZUmi. 

REYISIO  ACERACEÄRUM  JAPONICARUM. 


» 


PLATE    XVI. 


EXPLICATIO  TABULAE    XVI. 


Acer  ruhniin^  L. 

1.  Pars  plantae  fiuctiferae  (mag.  nat.). 

2.  Folia  varia  (mag.  nat.). 


Jour.Scl. Coll.,  Vol  XXXII. ,  Art. LPI. XVI 


îi.Koidzumi    del 


F.  Fujisawa   setup. 


G.  KOIDZUMI. 

REYISIO  ACERACEARUM  JAPOMICARUM. 


PLATE    XVII. 


EXPUCATIO  TABULAE    XVII. 

Acer  tnfidnm,  Hook,  et  Arn. 

1.  Pars  plantae  (mag.  liât.). 

2.  Sectio  floris  "^  verticalis  (aug.). 

3.  Stamen  fl.  -^^Cang.). 

4.  Flos  hermaphrociitus  (aug.). 

5.  Verticalis  ejus  sectio  (aug.) 

6.  Sectio  ovarii  transversalis  (aug). 

7.  Fructus  (mag.  nat.). 

Acer  trificliun,  var.  formosanuni,  Hay. 
<S.     Folium  (mag.  nat.). 
9.     Fructus  (mag.  nat.). 


Jour.Sci.  Coll.,  Vol.  XXX/l.,Art.LPI.  XVII. 


F.  Fujisawa  srulp. 


il 


G.   KOIUZUMI. 

EEVISIO  ACERACEARUM  JAPONICARUI. 


PLATE    XVIII. 


EXPUCATIO  TABULAE    XVIII. 

Acer  Ginnala,  Max. 

1.  Pars  plantae  (mag.  iiat.)- 

2.  Flos  "^  (ang.). 

3.  Verticalis  ejus  sectio  (aug.). 

4.  Flos  hermaphroditus  (aug.). 

5.  Sectio  ovarii  verticalis  (aug.)- 
Acer  Ginnahi,  var.  yezoensc,  Koidz. 

6.  Fructus  (mag.  nat.). 


Jour.Sci.  Coll., Vol.  XXXIL,Art.LPI.  XVIII.  \ 


G.Koidziiini    del. 


v.  Fujis<iwa  sculp. 


fil 

Hi 


I 


§' 


Mi 


U.   iCOIDZUMI. 


REYISIO  ACERÄCEARUM  JAPONlCARUffl. 


PLATE    XIX. 


EXPLICATIO  TABULAE    XIX. 

Jeer  spicatuiii,  Lam.,  var.  ul'iirunih(c?ise,  Max. 

1.  Par?  plantae  (mag.  nat.). 

2.  Floy  ^  (ang.). 

3.  Verticalis  ejus  ^--ectio  (aug.). 

4.  Flos  liermaphroditus,  sepalis  petalis  staminibusque  dicisis 

(aug.). 

5.  Sectio  ovarii  verticalis  (aug.). 
G.     Fructus  (mag.  iiat.). 

7.     Stamen  (aug.). 


Jour. Sei.  Coll.,  Vol.  XXXII,  Art. I. PI.  XIX. 


G.Koidzumi    del. 


F.  Fujisawa  sculp. 


G.   KOIDZUMI. 

REVISIO  ACERACEARUM  JAPONICARÜM, 


PLATE    XX 


EXPLICATIO  TABULAE    XX. 

Acer  Oliver iaimm,  Pax,  var.  Nahaharae,  Hayata. 

1.  Pars  plantae  fr.  (nat.  mag.)- 

2.  Flos  "P  (aug.). 

3.  Flos  hermaphroditus(aug.). 

Svar.  forniosanum,  Koidz. 

4.  Folium  (mag.  nat.). 

Svar.  trilobatum,  Koidz. 

5.  Folium  (mag.  nat.). 


Jour.Sci.  Coll,,  Vol  XXXII,  Art.!. PI.  XX. 


G.Koidzumi    del. 


F.  Kujisawa   seul 


i 

■il  i 


H 


G.   KOIDZU3II. 

REVISIO  ACERACEARUM  JAPONICARUM, 


PLATE    XXL 


EXPLICATIO  TABULAE    XXI. 

Acer  SiehokUcmmii ,  j\Iiq. 

1.  Eamulus  fructifcr  (mag.  nat.). 

2.  Sectio  floris  "^  verticalis  (aug.)- 

3.  Stamen  (aug.). 

4.  Flos  hermaphroditus,  petalis  decisis  (aug.)- 

5.  Sepalum  (aug.). 

6.  Petalum  (aug.). 

7.  Fructus  (mag.  nat.). 

var.  ts2isimcnsc,  Koidz. 

8.  Folium  (mag.  nat.). 

subvar.  laxiflonim,  Koidz. 

9.  Folium  (mag.  nat.). 


Jour.Sci.  Coll.,  Vol.  XXX II.  Art. 1  PI.  XXI. 


•'■  l\<»i{|zuiiii    (Id. 


K.  Fu|isa«'.i   sc\i\\ 


I 


;     I 

J    I 


G.  KOIDZUIflI. 

REYISIO  ACERACEARUM  JAPOmCARÜM. 


PLATE    XXII. 


EXPUCATIO  TABULAE    XXII. 

Acer  Sliirasaicaniuii,  Koidz. 

1.  Pars  phintao  fr.  (mag.  nat.). 

2.  Foliis  lobiis  (mag.  nat.). 

3.  Flos  ^  (aug.). 

4.  Stamen  (ang.). 

5.  Flos  hermaphroditus,  (aug.)  sepalis  petalisque  decisis. 
G.  Fructus  (mag.  nat.). 


Jour.  Sei.  Go//. .  Vo/.  XXX //,  Art.  I.  Pi.  XX  ii. 


('"•Koi(|/-.iiiiii    (1(1. 


hi|is;i\\;i    sciilii 


Il 

1  I 


m 


G.   KOIDZUMÏ. 

EEÏISIO  ACERACEARUM  JAPOBICARUM. 


PLATE    XXIII. 


EXPUCATIO  TABULAE    XXIII. 

Acer  Shirasawanum,  Koidz.  var.  tcmdfoUum,  Koidz. 
1.     Ramulus  fructifer  (mag.  nat.). 


Jour.Scl.  Coll..  Vol  XXXII,  Art. I. PI.  XXIII. 


G.Koid'/.imii    ilcl 


F.'Fujisawa  sciUp. 


f' 


■F 

I 


G.    KOIDZUmi. 

REVISIO  ACERÄCEÄRUM  JÄP0N1CJRUM, 


PLATE    XXIV. 


EXPLICATIO  TABULAE    XXIV. 

Àccrj  ajionicun},  Tluiii]). 

1.  Pars  plantae  fr.  (mag.  iiat.). 

2.  Sectio  vertiealis  floris  -^  (aiig.)- 

3.  Flos  hermaphroditus  (aiig.). 

4.  Sectio  vertiealis  ovarii  (aug.). 

5.  Embryon  (aug.). 


Jour.Scl.Coll.,Vol.XXXII.,Art.l.PI.XXIV. 


1'.  I' iijixawa   sculp. 


G.    liOIDZUMl. 

REViSlO  ACERACEARUM  JAPONiCARUM. 


PLATE    XXV. 


EXPUCATIO  TABULAE    XXV. 

Acer  japonicuni,  Thanb. 
var.  typicum,  Gr.  v.  Schw. 
1.-2.-3.     Fructus  varii  (mag.  nat.)- 

var.  circnmlohaiwn ^  (Max).  Koidz. 

4.  Folium  (mag.  nat.). 

5.  Fructus  (mag.  uat.). 

var.  viicroph tjllwu ^  Koidz. 

6.  Folium  (mag.  nat.). 

var.  Hcyhachii,  ^latsum. 

7.  Folium  (mag.  nat.). 


Jour.  Sei.  Coll.,  Vol  XXX  II,  Art.  I.  PI.  XX  V. 


G.  Kojdzüiiii    fifl. 


F.  Fujisawa  xcu 


"Ip^      J 


^^1 


G.   ItOIDZUflll. 

REVISIO  ÄCERACEARUM  JäPOlCARUM. 


PLATE    XXVI. 


EXPUCATIO  TABULAE    XXVÏ. 

Acer  palmaüuu^  Tliuiib. 

subsp.  (jcnuinwn,  (S.  et  Z.)  Koidz. 

var.  (iniahile,  Koidz. 

1.  Pars  plantae  (mag.  iiat.)- 

2.  Sectio  vcrticalis  floris  ^  (aug.)- 

3.  Flos  Hermaphrodit  us  (aug.)- 

4.  Fructus  (mag.  uat.)- 

5.  Verticalis  ejus  sectio  (aug.)- 

6.  Embryon  (aug.). 

Acer  palmatwu,  Tliunb. 
subsp.    scptenlohum,  (Thg)  Koidz. 
var.  latilohatiim ,  Koidz. 

7.  Folium  (mag.  nat.). 

8.  Fructus  (mag.  uat.). 


Jour.Scl.  Coll.,  Vol.  XXXII.,  Art.  J. PI.  XX  VI. 


f  1.  Koidzuiiii    del. 


r  .I'm i'i sawn    scul|>. 


G.   KOIDZUfllI. 

REÏISIO  ACERACEARÜM  JAPOSlCAfiUM. 


PLATE    XXVII. 


EXPLICATIO  TABULAE    XXVII. 


Acer  palmatwn,  Thunb. 
subsp.  scptcnlohum ^  (Thunb.)  Koidz. 

1.     Pars  plantae  fructiferae  (mag.  nat.). 
2.-3.-4.-5.     Fructus  varii  (mag.  nat.). 


■Jnur.Soi.  Coll.,  Vol.  XXXII.,  Art.  I.  PI.  XXVII. 


G.Koidzunii    del 


1'".  I''iliis;i\v;i    .siiil|i. 


fi 

Si 


41 


I  ) 


l\ 


u 


G.   KOIUZUMI. 

REVISIO  ACERACEÄRUM  JAPONICARÜM, 


PLATE    XXVIII. 


EXPLICATIO  TABULAE    XXVIII. 

Acer  palmatum,  Thunb. 
subsp.  Matsumurae,  Koidz. 

1.  Eamulus  fructifer  (mag.  nat.). 

a.  spontaneum ,  subvar.  elegans,  Koidz. 

2.  Fractus  (mag.  nat.). 

Acer  palmatum,  Thg. 
subsp.  genuinum,  var.  spectabile.  Koidz. 

3.  Folium  (mag.  nat.). 


Jour.Sci.  Coll..  Vol.  XXKII.,  Art.  I.  Pi. XX  Vll/. 


fîKoidziiriii    del. 


I-'.  I'ujisiiwci    srulp. 


l 


1 


G.   KOIDZII31I. 

REÏISIO  ACERACEARUffl  JAPONICARUM. 


PLATE    XXIX. 


EXPLICATIO  TABULAE    XXIX. 

Acer  ohlonguvi,  Wall. 

1.  Pars  plantae  fr.  (mag.  nat.). 

2.  Flos  ^  (aug.). 

3.  Fructus  yarii  (mag.  nat.). 

4.  Folia  varia  (mag.  iiat). 


Jour. Sei.  Coll.,  Vol  XXX IL ,  Art.  I.  PI.  XXIX 


G.Koidzumi   del. 


F.  Fujisawa  sculp. 


Il 


1. 


G.   KOIDZUMI. 

KEVISIO  ACERACEÄRUM  JAPOHICARÜM, 


PLATE    XXX. 


EXPLICATIO  TABULAE    XXX. 

Acer  iiilcoense,  ^Tax. 


1.  Pars  plantae  fr.  (mag.  nat.). 

2.  Flos  ^  (aug.). 

3.  Flos  ^  (aug.). 


Jour.Scl.  Coll.,  Vol.  XXXII.,  Art.  I. PI.  XXX. 


''•  Ivnidziiini    del. 


V.  I''iiji.s;nv;i   kciiI|i, 


i 


I 


G.   KOIDZUniI. 

EEVISIO  ACERACEARUffl  JAP0N1CÄRÜM. 


PLATE    XXXI. 


EXPUCATIO  TABULAE    XXXI. 

Acer  Miijahei,  Max. 

1.  Pars  plantae  fr.  (mag.  nat.). 

2.  Sectio  vertical  is  fl.  "^  (aiig.)- 


Jour.Sci.  Coll.,  Vol.  XXXII.,  Art.  I  PI.  XXXI. 


(  i.  Koiil/.unii    ticl. 


F.  Fujisana    sriilp. 


I 


'-1 

5 


G.  Koiuzcrnii. 

REVISIO  ACERACEÄRIM  JAPOBICARUM. 


PLATE    XXXII. 


EXPLICATIO  TABULAE    XXXII. 

Acer  inctwu,  Thunb. 

a.     ti/picwii,  Gr.  V.  Schw. 
subvar.  eiipictiun,  Pax. 

1.  Pars  plaiitae  fructiferae  (mag.  nat.). 

2.  Fructns  (mag.  nat.). 

3.  Sectio  vcrticalis  floris  "^  (aug.  ). 

4.  Flos  hermaphroditus,  petalis  sepalisque  decisis  (aug.). 

5.  Verticalis  ovarii  scctio  (aug.). 

sub\'ar.  mono,  Pax. 
G.     Fructus  (mag.  nat.). 

subvar.  Savaticri,  Pax. 
7.     Fructus  (mag.  nat.). 

subvar.  Mafjrii\  Koidz. 
8.-9.     Fructus  (mag.  nat.). 

var.  fjJancum,  Koidz. 
10.     Fructus  (mag.  nat.). 


Jour.Scl.  Coll.,  Vol.  XXXIL,  Art.l.PI.  XXXII. 


(i.Koitlzuini    del. 


!•'.  Kujisiiwa    sculp 


G.    KOinZUMI. 

EEVISIO  ACERACEARÜM  JAPONICARUM, 


PLATE    XXXIII. 


EXPUCATIO  TABULAE    XXXIII. 

Acer  diaholictnn,  Bl. 


1.  Pars  plantae  fr.  (mag.  nat.). 

2.  Flos  ^  (ang.). 

3.  Flos  ^  (ang.). 

4.  Verticalis  ejus  sectio  (aug.). 


Jour.Sci.  Coll.,  Vol.  XXXII.,  Art. 1. PI.  XXXIII. 


G.Koidzumi    del. 


F.  Fujisawa   sculp. 


4 


4 


RXAL  OF  THE  COLLEGE   OF  SCIENCE,  IMPERIAL  UNIVERSITY,  TOICYO,  JAPAN. 

VOL.  XXXII.,  ARTICLE  2. 


Beobachtungen  an  einer  SüsswasserPeridinee. 

Yon 

N.  OhnO,  Rujakuhakuahi. 

Professor  der  Botanik  an  der  Landwirtschaftlichen  Fakultät 
der  Tnhoku  Kaiserlichen  Universität,  Sapporo. 


Mit  1    Tafeln. 


Im  Jahre  1901  und  in  den  darauf  folgenden  zwei  Jahreji 
wurde  eine  massenhafte  Entwickelung  einer  Peridineenart  in 
einem  kürzhcli  angelegten  kleinen  Teiche  in  dem  Botanischen 
Garten  der  Universität  zu  Tokyo  beobachtet,  einem  Teiche,  der 
für  die  Bepflanzuug  monokotjledoner  Wassergewächse  bestimmt 
war.  Die  Entwickelung  der  l)etreffenden  Organismen  schien  mit 
dem  Wandel  der  Jahreszeiten  eng  verknüpft  zu  sein.  Die  üppig- 
sten Schwärme  beobachtete  man  nämlich  im  kältesten  Moiiat  des 
Jahres  d.  i.  im  Februar;  schon  Mitte  März  trat  eine  l)eträchtliche 
Verminderung  der  Individuenzahl  ein,  und  je  weiter  die  Jahres- 
zeit vorschritt,  desto  geringer  wurde  die  Entwickelung,  um  all- 
mählich ganz  zu  verschwinden.  Diese  periodische  Erscheinung 
kehrte  in  den  erwähnten  drei  Jahren  mit  einer  ziemlichen  Regel- 
mässigkeit  wieder.  Die  absolute  Menge  der  Individuen  war  aber  in 
jedem  Jahre  eine  verschiedene.  Im  Frühjahr  1901,  war  die  Pro- 
duktion eine  ganz  ausserordentlich  grosse,  so  dass  sogar  das  Wasser 
cies  Grabens  auf  eine  ziemlich  grosse  Strecke  rostfarbig  wurde.' 
Die  phototaktischen  Eigenschaften  der  Organismen,  von  denen 
weiter  unten  die  Rede  sein  wird,  bedingten  den  Ort  der  Ansamm- 
lung. In  den  nächsten  zwei  Jahren  nahm  die  Menge  aus  unbe- 
kannten Gründen  beträchtlich  ab. 

1.  Das  massenhafte  Vorkommen  der  Meeresperidineen  ruft  bekanntlich  oft  die  Eotfür- 
Ijung  des  Meeres  hervor.  Dass  auch  die  Peridineen  im  süssen  Wasser  zviweilen  reichlich  genug 
auftreten,  um  eine  deuthcbe  Färbung  des  Wassers  zu  erzeugen,  -wurde  mehrfach  beobachtet. 
Vgl.  hieriibei  Butschli,  Protozoa  in  Bronns  Klassen  und  Ordnungen  des  Thier-Reichs.     S.  1025. 


78  Art.  2.-X.  Ohno  : 

Die  ^•o^liegencle  Abhandlung   bezweckt,    die   Eigenschaften 
dieser  Organismen  eingehend  zu  betrachten,  soweit  die  angestell- 
ten  Beobaclitungen  reichen.     Bevor  ich  die  systematische  Stel- 
lung dieser  Organismen  näher  betrachte,   möchte  ich  eine  kurze 
Beschreibung  derselben  vorauschicken.     Aus  dem  durch  das  dichte 
Anschwärmen    der   Peridinee   rotbraun    gefärbten    Grabenwasser 
wurde  ein  Quantum  mit  einer  Krystallisierschale  herausgeschöpft 
und   dieses  wurde   im   Laboratorium  mit  Leitungswasser   massig 
verdünnt.     Unter  diesen  Bedingungen  dem  diffusem  Lichte  aus- 
gesetzt,   gediehen  die  Organismen  gut   und  die   Kulturen  waren 
tagelang  lebensfähig.      Die    einzelnen    Individuen  konnte   man 
schon  mit  blossem  Auge  als  sich  bewegende  gelbbraune  Pünktchen 
in  der  Kulturflüssigkeit  wahrnehmen,  besonders  dann,  wenn  das 
Kulturgefäss  gegen  das  Licht  gehalten,    oder  auf  weisses  Papier 
gebracht  wurde.     Wenn  man  einen  Tropfen  der  Kultur  auf  den 
Objektträger  brachte  und  unter  dem  Mikroskop  beobachtete,   so 
bemerkte  man  eine  Anzahl  lebhaft  herumschwimmender  kleiner 
gelblich  brauner  Organismen.      Das  Lidividuum  ist  etwa  von  halb- 
kugeliger Gestalt.     Die  Bewegung  ist  eine  vorwärtsgehende  com- 
biniert  mit  einer  rotierenden.      Die  Rotationsrichtung  wechselt 
von  Zeit  zu  Zeit,    sie  geht  nämlich  bald  von  links  nach  rechts, 
bald  aber  von  rechts  nach  links.     Zuerst  ist  die  Bewegung  sehr 
lebhaft,  und  es  gelingt  äusserst  schwer,  die  Gestalt  und  die  anderen 
Beschaffenheiten  der  Organismen  näher  zu  ermitteln.     Nach  eini- 
gen Minuten  jedoch   wird  die  Bewegung  immer  schwächer  und 
schwächer  und  es  tritt  ein   Desorganisationsvorgang  ein.     Hierl^ei 
ist  es  beachtenswert,  dass  Individuen  aus  ganz  frischen  Kulturen 
d.   h.   solchen,   die  kurz  nach  dem  Einbringen  ins  Zimmer  dem 
Gefäss  entnommen  waren,  stets  etwas  früher  eine  Sistierung  der 
Bewegung  und  eine  darauffolgende  Desorganisation  erlitteji,    als 
diejenigen,  die  längere  Zeit  in  Zimmer  aufbewahrt  wurden.     Diese 
Tatsache  findet  aller  Wahrscheinlichkeit  nach  darin  ihre  Erklärung, 
dass  die  letzteren  sich  den  gegebenen  Bedingungen  in  höherem 
Grade  angepasst  haben.     Die  früher  oder  später  eintretende  Des- 
organisation der  Organismen  geht  zunächst  in  der  Weise  vor  sich. 


Beobachtungou  an  einer  Süsswasser-Peridinee. 


79 


dass  ihre  Bewegung  allmählicli  aufhört,  dann  beginnt  der  Vorgang 
des  Lostrennens,  oder  anderweitige  nekrobiotische  Erscheinungen 
treten  auf.  Von  den  beiden  Arten  der  Geissehi — der  Quer-  und 
Längsgeissel,  von  welchen  später  ausführhcher  gesprochen  werden 
soll — ist  die  Quergeissel  empfindhcher  gegen  schädhche  Einflüsse 
und  wird  bald  von  ihrer  Ausgangsstelle  abgeworfen  und  mit  einer 
sehr  eigentümlichen  Vorwärtsbewegung  von  dem  Körper  abge- 
stossen.  Die  auf  diese  Weise  losgetrennten  toten  Geissein  finden 
sich,  hie  und  da  als  lange  geschlängelte  Fäden.  Die  Längsgeissel 
hingegen  beginnt  ihre  Desorganisation  erst  später,  und  wird  nicht 
abgestossen. 

Dimcnsionsvcrhdîtnissc— Die  Grösse  der  schwärmenden  In- 
dividuen variirt  zwischen  engeren  Grenzen.  Im  folgenden  wer- 
den Messungen  die  für  25  Individuen  ausgeführt  wurden,  zusam- 
mengestellt, um  diese  Verhältnisse  einigermassen  anschaulich  zu 
machen.  Diese  Messungen  wurden  in  der  Weise  ausgeführt,  dass 
die  Bilder  der  zu  messenden  Individuen  mittelst  Abbeschem  Zei- 
chenajDparat  auf  Papier  genau  entworfen  wurden,  bei  einer  Ver- 
grösserung  von  460  Diametern;  und  die  Zeichnungen  wanxlen 
mittelst  Maasstab  der  Länge  und  Breite  nach  gemessen. 


L 

cinge 

Breite 

1 

21-7    [). 

15-2     [A 

2 

23-9    „ 

185     „ 

3 

20-6    „  ■ 

17-4     „ 

4 

250    „ 

210     „ 

5 

21-7    „ 

15  2     „ 

6 

22-6    „ 

18-5    „ 

7 

20-6    „ 

15-2     „ 

8 

25'0    „ 

20-6     „ 

9 

22-6    „ 

15-2     „ 

10 

20-6    „ 

17-4    „ 

11 

22-6    „ 

20-6     „ 

12 

23-9    „ 

15-2     „ 

13 

22-6    „ 

16-3     „ 

80 


Art.  2.-N.  OLno: 


L 

inge 

Breite 

14 

21-7    |J. 

17-4     [1. 

15 

206    „ 

152     ,. 

IG 

250    „ 

13-5     „ 

17 

21-7    „ 

152     „ 

IS 

20G    „ 

12  0     „ 

19 

20-G    „ 

15-2     „ 

20 

225    „ 

17-4     „ 

21 

20-6    „ 

15-2     ,. 

22 

23.9    „ 

18-5     „ 

23 

23  9    „ 

135    „ 

24 

206    „ 

15-2     „ 

25 

217    „ 

17-4    „ 

Durchschnitt 

22-2    (X 

169    iJ. 

So  Ijesitzt  unser  Organismus  im  Mittel  eine  Länge  von  22.2  /^ 
und  eine  Breite  von  IG '9  /-«. 

Begcissciungsvcrhältnissc. — Bekanntlich  sind  die  Peridineen 
mit  zweierlei  Arten  von  Geissein  versehen.  Die  eine  Art,  welche 
sich  in  der  Querfurche  herumbewegt,  ist  die  Quergeissel  und  die 
andere  Art,  welche  nacli  rückwärts  gerichtet  ist,  stellt  die  sog. 
Längsgeissel  dar.  Die  Quergeissel  wurde  lange  als  ein  an  der 
Querfurche  angesetzter  Cilienkranz  aufgefasst.  Erst  in  neuerer 
Zeit,  machte  Klebs^  eine  wiclitige  Entdeckung,  dass  es  keinen 
Cilienkranz  in  der  Querfurche  gibt,  sondern  eine  eigentümliciie 
Geissei,  welche  mit  einem  Ende  hm  Körper  haftet  und  in  der 
Querfurche  in  wellenförmiger  Bewegung  sich  befindet.  Dies  bildet 
zweifellos  eine  Errungenschaft  für  unsere  Kenntniëse  der  Organisa- 
tionsverhältnisse der  Peridineen.  Diese  ursprünglich  nur  an  Süss- 
wasserformen  gemachte  Beobachtung  konnte  Klebs  später  auch 


1.  G.  Klees,  Über  die  Organisation  einiger  Flagellatengruppen  und  ihre  Beziehungen 
zu  Algen  u.  Infusorieagruppen,  Unters,  aus  dem  botan.  Institut  zu  Tübingen,  Bi.  I.  S.  233-262, 
1S33. 


Beobachtungen  an  einer  Süsswasser-Peridinee.  81 

für  marine  bestätigen'.     Bütschli  konnte  diesen  Befund  ebenfalls 
bestätigen",  so  dass  er  den  Namen  Dinoflagellata  anstatt  Ciliofla- 
gellata  vorschlug.     Die  Längsgeissel  nun  wurde  schon  1833  und 
1834  von  Ehrenberg  bei  einem  Teil  der  marinen  sowie  der  Siiss- 
wasserformen  entdeckt";    und  dass  sie  in  Einzahl  vorhanden  ist, 
gilt  als  eines  der  Organisationsmerkmale  dieser  Organismengruppe. 
Während  ich  die  Bewegungsweise  der  im  Rede  stehenden  Orga- 
nismen mit  grossem  Interesse  beobachtete,   schien  es  mir  als  ob 
die    nach    hinten  meist  in    gerader  Richtung  befindliche  Geissei 
aus  einem    Paar  gleichlanger  und  gleichbeschaffner  Geissein  be- 
stände,  die  aus  der  Kreuzungsstelle   der  Quer-  und  Längsfurche 
hervorspringend   etwa   1^   mal  grösser  waren   als  die   eigentliche 
Körperlänge  des  Organismus.     Wiederholte  Beobachtungen  zeig- 
ten stets  dieselbe  Tatsache.     Anfangs  erwachte  in  mir  das  Beden- 
ken, ob  es  sich  hier  nicht  um  eine  optische  Täuschung  handelte, 
wie  in  dem  Falle,   wo  ein  rasch  hin  und  her  schwingender  Stab 
wie  zwei  Stäbe  aussieht.     Dieses  Bedenken  zu  beseitigen  wurden 
Fixierungs-  und  Färbungsmethoden  zu  Hilfe  gezogen.     Der  Trop- 
fen in  dem  die  Organismen  sich  in  aktiver  Bewegung  befanden, 
Avurde  nämlich  auf  einem  Deckglas  ausgebreitet  und  schnell  Os- 
miumdämpfen ausgesetzt  und  darauf  ausgetrocknet,   mit  Fuchsin 
gefärbt  und  dann  in  Kanadabalsam  eingeschlossen.     In  wohlge- 
lungen Fällen,   konnte  man  stets  zwei  unabhängig  von  einander 
stehenden   Längsgeisseln   nachweisen.     Da  auf  diese   Weise    die 
Tatsache,  dass  die  Längsgeisseln  bei  meinem  Objecte  in  Zweizahl 
vorhanden    sind,    endgiltig  festgestellt   wurde,    so  wurde   es  nun 
meine  Aufgabe,  nachzusehen,  ob  ein  ähnlicher  Fall  in  der  Littera- 
tur  zu  finden   sei.     Es  ergab  sich  alsbald,   dass  einige  Angaben 
hierüber  vorliegen.      So  schreibt  BiJTSCHLi*,  ,,Von   verschiedenen 
Beobachtern   wurde  berichtet,    dass  sowohl  bei  Cemtiam  tetraceros 

1.  G.  Klebs,  Ein  kleiner  Beitrag  zur  Kenntniss  der  Peridineen.    Bot.  Ztg.  Jahrg.  43,  1834 
S.  721-33  u.  737-45. 

2.  0.  BiJTSCHLi,  Einige  Bemerkungen  ü.  gewisse  Organisationsverhältnisso  der  Ciliofla- 
gellaten  u.  der  Noctiluca.     Morpholog.  Jahrb.     Bd.  X.  1885,  S.  529-77. 

3.  Ehrenberg,  Abhatjdl.  d.  Berliner  Akad.  1833  u.  1834. 

4.  O.  BüTscHLi,  Protozoa  in  Bronns  Xlassen  u.  Ordnungen  des  Thierreichs.  S.  958. 


82  Art.  2.-X.  Ohuo: 

(comiium)  wie  bei  Ceratium  Tripos  zuweilen  zwei  Längsgeisseln  statt 
der  einfachen  zu  finden  seien.  Für  die  erstgenannte  Form  be- 
merkten dies  sclion  Clapareds  und  Lachmanx,  welche  sicli  auf 
das  Zeugniss  von  Lieberkühn  beriefen.  Auf  den  prächtigen 
Originalabbildungen  des  letzteren  Forschers,  die  mir  durch  seine 
Güte  zugänglich  waren,  ist  die  zweite  Geissei  denn  auch  deutlich 
wiedergegeben.  Für  Ceratium  Tripos  machte  zuerst  Iîergh  auf 
eine  gelegentliche  Verdoppelung  der  Längsgeissel  aufmerksam. 
Man  könnte  nun  diese  Beobachtungen,  Avelche  aus  einer  Zeit 
stammen,  wo  die  Geissei  der  Querfurche  noch  unbekannt  war, 
wohl  mit  Klees  für  zweifelhaft  halten,  da  die  zweite  Geissei  event- 
uell die  Quergeissel  gewesen  sein  könnte,  welche  speciell  bei  den 
Ceratien  zuweilen  aus  der  Querfurche  hervorzutreten  scheint. 
Da  nun  aber  BÜtsciili^  bei  Ccr.  Tripos  gelegentlich  neben  der  Quer- 
geissel zwei  deutliche  Längsgeisseln  aufïand,  so  scheint  die  An- 
gabe der  früheren  Beol)achter  doch  gerechtfertigt."  Bütschli  ist 
also  der  Ansicht  dass  die  besprochene  Erschein ug  nur  ausnahms- 
weise bei  einigen  Individuen  der  Art  vorkommt,  also  etwa  als 
Abnormität  aufzufassen  ist.  In  einer  Arbeit  über  einige  Süsswas- 
serperidineen  macht  Folgner^  auf  eine  ähnliche  Tatsache  aufmerk- 
sam und  gil:)t  Näheres  darüber  an.  Die  Art  womit  er  sich  l)e- 
schäftigte,  war  Ceratium  Tetraceros  Schrank  (C.  cornuum  Claparede 
ET  Lachmann)  also  eine  der  oben  genannten  Arten.  Er  sagt  wört- 
lich: ,,Ehe  ich  die  Besprechung  des  Ceratium  tetraceros  schliesse, 
will  ich  doch  nicht  unterlassen,  noch  einer  Eigenthümlichkeit 
Erwähnung  zu  tun,  die  bei  dieser  Peridinee  zwar  schon  hin  und 
wieder  beobachtet  wurde,  niemals  aber  bisher  meines  Wissens  als 
eine  bei  ihr  anscheinend  allgemein  auftretende  Erscheinung  be- 
kannt geworden  ist,  nämlich  des  gleichzeitigen  Auftretens  von 
zwei   Längsgeisseln.       Bekanntlich  haben  schon  Claparede  und 


1.  O.  Bütschli,  Einige  Bemerkungen  über  gewisse  Organisationsverhältnisse  d.  Ciliofla- 
gellaten  u.  der  Noctiluca.     Morpholog.  Jahrbuch.     Bd.  X.  ]8S5.     S.  529-77. 

2.  V.  FoLGNER,  Beiträge  zur  Kenntniss  der  Entwicklungsgeschichte  einiger  Süsswasser- 
Peridineen.  Arbeiten  des  botanischen  Institutes  d.  k.  k.  deutächen  Universität  in  Prag.  Nr. 
XLII.  Oestsrreichische  bot.  Zeitschrift.     Bä.  XLIX  1S09. 


Boobauhtungen  an  einer  Süsswasser-Pcridinee. 


83 


Lachmann,  welche  sich  auf  die  Angaben  Liebekkühn's  stützen, 
und  nach  ihnen  Bütschli'  (Die  Angaben  von  ClaparÈde  und 
Lachmann  im  Original  habe  ich  nicht  gesehen.  Bergh  und  später 
BuTscHLi  beobachteten  auch  bei  Ceratium  tripos  das  ,, gelegent- 
liche" Vorkommen  von  zwei  Längsgeisseln.)  auf  dasselbe  auf- 
merksam gemacht,  es  jedoch  lediglich  als  eine  gelegentlich  vor- 
kommende Abnormität  betrachtet."'  Meines  Erachtens,  sind 
mit  diesen  bisher  erschienenen  Angaben  die  Beobachtungen  über 
die  Doppellängsgeisseln  erschöpft.  So  wäre  es  gewiss  von  eini- 
gem Interesse  ein  neues  Beispiel  und  zwar  bei  anderen  Form  an- 
zuführen; und  da  in  vorliegenden  Falle,  die  sämtlichen  Indivi- 
duen mit  z\Yei  gleichlangen  Längsgeisseln  versehen  sind,  so  ist  dies 
keineswegs  als  Abnormität  anzusehen.  Man  kann  wohl  mit  Sicher- 
heit sagen,  dass  in  gewissen  Süsswasserformen  von  Peridineen  die 
Längsgeissel  in  der  Zweizahl  vorhanden  ist.  Es  wäre  wünschens- 
wert in  dieser  Beziehung  auch  eine  Reihe  von  Meeresformen  der 
Prüfung  zu  unterwefen.  Die  xMöghchkeit  ist  wohl  nicht  ausge- 
schlossen, dass  die  Erscheinung  eine  zeimlich  verbreitete  unter  den 
Peridineen  sei.  Da  die  Zahl  und  Ausbildung  der  Geissein  wich- 
tige Organisationsmerkmale  bei  derartigen  Organismen  bilden,  so 
ist  die  Nachprüfung  umsomehr  von  Bedeutung. 

Bevor  ich  die  Beschreibung  der  Begeisselungsverhältnisse 
unserer  Peridinee  vollende,  möchte  iclr  noch  etwas  über  die  ab- 
normalen Verhältnisse  der  Geissein  angeben.  Folgner-'  macht  auf 
ein  eigentümliches  Gebilde  auf  der  Geissei  aufmerksam,  welches 
er  bei  Ccmtiwn  tetraceros  beobachtet  hat.  Nach  ihm  zeigte  das  In- 
dividuum eine  Eigentümlichkeit,  welche  der  Autor  ,, bisher  noch 
bei  keiner  anderen  Peridinee  in  irgend  welchem  Lebensstadium 
gesehen  und  auch  in  der  Litteratur  nirgends  erwähnt  gefunden 
habe,  nämlich  am  Ende  der  Längsgeisseln  Anhangsgebilde  resp. 
Anschwellungen  in  Form  von  ziemlich  grossen  Kugeln.  Ob  sie 
massive  Kugeln   oder  aber  Blasen,    also   Hohlkugeln  darstellen, 


1.  Siehe  oben. 

2.  FOLGNEE  1.  c.  S.     140-141. 

3.  FoLQNEE,  Ebenda. 


84  Art.  2.-X.  Ohno: 

gelang  mir  nicht  zu  entscheiden  ;  die  Gestalt  kreisförmiger  Schei- 
ben schienen  sie  jedoch  nicht  zu  besitzen.  ''  Er  beschreibt  dieses 
Gebilde  sehr  ausführlich.  Er  ist  aber  geneigt  dasselbe  mehr  als 
Missbildung  aufzufassen,  da  er  die  Entstehung  desselben  nicht 
beobachten  konnte,  das  Gebilde  eben  stets  schon  vorhanden  war 
beim  Sichtbarwerden  der  Geissein,  und  da  er  auch  während  der 
Beobachtung  für  längere  Zeit  keine  bemerkbare  Veränderung  an 
ihm  bemerkte.  Er  lässt  jedoch,  indem  er  ähnliche  Gebilde  bei 
Flagellaten  und  anderen  Organismen  in  Betracht  zieht,  die  Mög- 
lichkeit bestehen  dass  dies  als  eine  Desorganisationserscheinung 
aufzufassen  sei. 

Gelegentlich  meiner  Beobaclitung  habe  auch  ich  ein  Gebilde 
das  in  jeder  Beziehung  mit  dem  eben  beschriebenen  überein- 
stimmt, hin  und  wieder  gesehen.  Allein  diese  Gebilde  waren 
meist  erst  dann  sichtbar  als  das  Mattwerden  des  Individuums  schon 
ziemlich  weit  fortgeschritten  war.  Man  kann  die  Gebilde  von  den 
ersten  Anfängen  bis  zum  Grosswerden  verfolgen,  und  es  besteht 
kein  Zweifel,  dass  man  es  hier  mit  Desorganistationsvorgängen  zu 
tun  hat.  Fig  17  zeigt  ein  solche  Gebilde  auf  den  Enden  des  beiden 
Längsgeisseln.  Auf  den  Quergeissein  trifft  man  nie  solche  Ge- 
bilde; und  die  Verschiedenheit  in  der  Beschaffenheit  der  beiden 
Arten  von  Geissein  kommt  hierin  auch  zum  Ausdruck. 

lieber  die  Desorganisation  der  Geissein  in  allgemeinen 
machte  Bütschli  interessante  Beobachtungen.  Er  schildert  die 
Verhältnisse  wie  folgt:  ,,Die  Glenodinien  stellen  zunächst  allmäh- 
lich ihre  Bewegungen  ein  und  liegen  ruhig  da,  wobei  von  der 
hinteren  Geissein  nichts  mehr  zu  sehen  ist.  Dann  bemerkt  man 
plötzlich,  wie  sich  in  der  Gegend  der  Querfurche  eine  Geissei  zu 
einem  dichten  Korkzieherartigen  Gewinde  aufrollt.  Ganz  kurz 
darauf  löst  sich  diese  zu  einem  kleinen  Packet  aufgerollte  Geissei 
mit  einem  Ruck  von  dem  Körper  ab  und  bewegt  sich  ein  Stück 
weit  fort.  Dieses  kleine  Geisselpacket  kann  nun  zunächst  einige 
Sekunden  ruhig  liegen  bleiben  und  dann  plötzlich  in  heftig  um- 
herflatternde Bewegung  übergehen,  oder  es  schwimmt  gleich  nach 
der  Abstossung  in  dieser  Weise   weiter.      Diese   Bewegung  der 


Beobachtungen  can  einer  Süsswasser-Peridinee.  85 

abgelösten  Geissei  dauert  etwa  eine  Minute  oder  wenig  länger 
lebhaft  fort.  Dabei  bleibt  die  Geisse!  stets  eng  aufgerollt.  End- 
lich gelangt  sie  zur  Ruhe,  indem  sie  ohne  Zweifel  völlig  abstirbt."^ 
Diese  Beobachtung  weicht  von  meiner  insoweit  ab,  als  bei  un- 
serer Peridinee  die  beiden  Längsgeisseln  bis  zuletzt  sichtbar 
bleiben — also  lange  nach  dem  Wegwerfen  des  Quergeissels  noch 
dem  Körper  anhaften —  bis  die  oben  erwähnte  Desorganisations- 
erscheinung auftritt. 

Cystcnbildung.— Zwischen  schwärmenden  Individuen  und 
am  Boden  befindlichen  Schlamm  treten  uns  eine  Anzahl  encys- 
tierte  Individuen  entgegen.  Diese  stellen  den  Dauerzustand  un- 
serer Peridinee  dar.  Die  einzelnen  Cysten  erscheinen,  wie  Fig. 
18-21  und  23-26  zeigen  als  rundliche  Körper  mit  unregelmässig 
gewellter  Oberfläche.  Nicht  selten  findet  man  mit  Stacheln  ver- 
sehenen Formen.  Die  Cystenmembran  ist  ziemlich  dick.  Dass 
der  Stoff  der  Membran  hauptsächlich  aus  Cellulose  besteht,  lässt 
sich  zeigen,  indem  man  die  Membran  mit  Chlorzinkjod  behandelt, 
wobei  sie  rötlich  violett  tingiert  wird.  Der  Inhalt  der  Cysten 
besteht  aus  gelblich  brauner  körniger  Masse  mit  grösseren  oder 
kleineren  rötlich  braunen  Oeltropfen.  Ausserdem  ist  im  Innern 
der  Cysten  Stärke  in  reichlicher  Menge  aufgespeichert,  wie  es  mit 
Jod  leicht  nachzuweisen  ist,  und  in  diesem  Falle  scheint  deren 
Verteilung  eine  gleichmässige  zu  sein.  Ueberdies  werden  die 
vorderen  und  hinteren  Leibesenden  im  encystierten  Zustande 
unerkennbar.  Man  begegnet  zuweilen  den  in  Bildung  begriffen- 
en Cysten,  die  noch  mit  Geissein  versehen  sind  und  sich  noch 
bewegen,  wenn  auch  ihre  Bewegungsenergie  wesentlich  lierab- 
gesetzt  ist. 

Künstliche  Veranlassung  zur  Cystcnbildung. — Es  Avird  von 
einigem  Interesse  sein,  hier  einige  Versuche  zu  besprechen  über 
die  Tatsache,  dass  man  durch  Veränderungen  der  Lebensbedin- 
gungen die  schwimmenden  Individuen  zur  Cystcnbildung  veran- 

1.     O.  BÜTSCHLT,  Morphologisclic  Jahrbücher  1885,  Bei.  X.  S.  535. 


86  Art.  2— N.  Ohno: 

lassen  kann.  In  der  Krystallisierschale  kann  man  sie  wochen- 
lang halten,  ohne  class  sie  dabei  zur  Ruhe  gehen.  Aber  durch 
folgendes  Verfahren  kann  man  sie  unfehlber  zur  Ruhe  bringen. 
Einige  Tropfen  Wasser,  in  dem  die  schwimmenden  Individuen 
enthalten  sind,  waren  auf  der  in  Petrischer  Schale  vorher  erstarr- 
ten Agarplatte  ausgebreitet  (Die  Gallerte  bestand  aus  0.2%  Knop- 
scher  Nährlösung  und  1%  Agar).  Die  Individuen  schwimmen 
vor  ihrer  Encystierung  auf  den  dünnen  Wasserschichten  über  der 
Gallerte.  Die  Schale  wurde  dann  mit  dem  Deckel  bedeckt  und 
in  feuchter  Kammer  belassen.  Das  Verhalten  der  schwimmen- 
den Individuen  auf  der  Gallerte  wurde  von  Zeit  zu  Zeit  unter 
schwächerer  Vergrössezung  beobachtet.  Die  Bewegung  hört  bei 
dieser  Behandlung  sehr  frühzeitig  auf,  (d.  h.  nach  einigen  Stun- 
den). Schon  nach  etwa  einer  Woche  gingen  fast  alle  Individuen 
zur  Encystierung  über.  Diejenigen  Individuen,  welche  diese 
nicht  zeigten,  waren  desorganisierte.  Die  hierbei  beobachteten 
Cysten  sind  in  Fig.  27  bis  33  abgebildet.  Solche  Cysten  weichen 
etwas  von  den  natürlichen  ab,  und  sind  mehr  oder  minder  un- 
regelmässig gestaltet.  Die  Stacheln  welche  unter  natürlichen 
Bedingungen  entwickelt  werden,  wurden  hierbei  nicht 
ausgebildet,  wenn  auch  nicht  selten  Anfänge  dazu  sich  zeigten. 
Die  Oberfläche  der  Cysten  war  meist  mit  unregelmässigen  Undu- 
lationen  versehen.  Was  nun  die  dicke  der  Cystenmembran  selbst 
anbetrifft,  so  sieht  sie  bei  den  auf  Agar  gebildeten  der  normalen 
keineswegs  nach.  Der  jMembranstoff  wurde  hier  auch  als 
Cellulose  konstatiert  (nach  mehrtägiger  Behandlung  mit  Javelle' - 
scher  Lauge  und  darauf  folgender  Behandlung  mit  Chlorzinkjod). 
Der  Inhalt  der  Zellen  war,  der  Hauptsache  nach,  derselbe  wie 
derjenige  normal  gebildeter  Cysten,  d.  h.  rote  Oeltropfen, 
reichliche  Stärke). 

Anderweitige  Beobachtungen. — Dass  die  meisten  Peridineen 
ähnlich  wie  die  gefärbten  Flagellaten  und  die  Schwärmsporen  der 
Algen  phototaktisch  reagieren  ist  bekannt,  aber  eingehendere 
Untersuchungen  die  besonders  mit  dieser  Organismengruppe  an- 


Beobachtungen  an  einer  Süsswass-r-Peridinee.  87 

gestellt  wurden,   liegen  meines  Wissens  nicht  vor.'     Bei  meinen 
Versuchen  mit  unserer  Peridinee,  konnte  ich  vielfach  die  hervor- 
ragend phototaktische  Eigenschaft  derselben  beobachten.     Wenn 
man  das  Wasser  mit  schwimmenden  Individuen  in  eine  Krystal- 
lisierschale  von  etwa  10  cm  Durchmesser  bringt  und  in  f  m  Entfer- 
nung von  einem  hellen  Fenster  aufstellt,  so  sieht  man  schon  nach 
1 J  Minuten,  wie  sich  die  schwimmenden  Individuen  auf  dem  dem 
Lichte  zugewandten  Halbkreis  der  Schale  mehr  ansammeln,   um 
nach  3-6  Minuten  sich  vor  allem  auf  der  durch  das  reflektierte  Licht 
schärfer  erhellten  Partie   der  Schale  zu  konzentrieren.      Hierbei 
bildeten  sie  auch  noch,  von  oben  nach  unten  reichend,   vertikale 
Ansammlungsstreifen.      Einen  hübschen    Versuch    konnte    man 
fernerhin  in  der  W^eise  anstellen,  dass  man  einen  weissen  Porzel- 
lanteller nahm,  denselben  mit  Wasser  füllte,   das  durch  die  hohe 
darin  enthaltene  Individuenanzahl  scheinbar  braun  gefärbt  war, 
und  nun  im  Dunkelzimmer  mittelst  eines  Prismas  Sonnenspekt- 
rum darauf  fallen  liess.     Man  konnte  schon  nach  3-4   Minuten 
eine  beträchliche  Ansammlung  der  Organismen  unter  dem  farbi- 
gem Lichte  beobachten  und  nach  Entfernung  des  Prismas  behiel- 
ten die  angesammelten  Organismen  noch  für  einige  Zeit  deutlich 
das   Bild  des  Spektrums.      Die  stärkste  Ansammlung  zeigte  sich 
dabei  in  den  kurzwelligen  Teile  des  Spektrums,   nämlich  im  blau- 
violetten Teile. 

Ferner  sind  unsere  Kenntnisse  über  die  Fortpflanzungsver- 
hältnisse der  Peridineen  überhaupt  noch  recht  lückenhaft  und 
jede  genaue  Beobachtung  auf  diesem  Gebiete  ist  willkommen. 
Bei  meinen  Beobachtungen  konnte  ich  aber  keinen  neuen  Beitrag 
in    dieser   Richtung  liefern.      Betreffs   der   Art   und    Weise    der 

1.     Heber  die  phototakti sehen  Eigenscbaiten  der  verschiedenen  pflanzlichen  Organismen 
wird  auf  folgende  bekannte  Arbeiten  hingewiesen  :  — 

E.  Steasburger,  Wirkung  der  Wärme  und  des  Lichtes  auf  Schwärmsporen.  Jeu.  Zeitschr. 
Xaturw.     1S78.  12.  S.  551. 

F.  Oltmanns,  Heber  die  photometrischen  Bewegungen  der  Pflanzen.    Flora  1892.     75.  S. 
183. 

Man  wird  ferner  in  Pfeffer  :  Pflanzenphysiologie.    Bd.  II.  S.  753  u.   s.  w.  u.  Oltmanns: 

Morphologie  u.  Biologie  der  Algen.  2.  Bd.  S.  220.  Zusammenstellungen  finden. 


88  Art.  2.-N.  Ohno  ' 

Fortpflanzung,  ist  bis  jetzt  nur  die  vegetative  Zweiteilung  sicher 
gestellt,  und  diese  geschieht  entweder  im  beweglichen  oder 
Ruhezustände.  Es  gibt  Angaben  w^onach  Fortpflanzung  durch 
Konjugation  stattfindet,  diese  ist  jedoch  noch  nicht  sicher  gestellt 
und  bedürfen  weiterer  Bestätigung.  Bei  meinen  Beobachtungen 
traf  icli  mehrere  Male  auf  Individuen,  die  in  beweglichem 
Zustande  in  Teilung  begriffen  waren,  (Fig.  34-37).  Die  sich 
teilenden  Individuen  bieten  sein*  eigentümliche  Gestaltung  dar 
und  die  Teilungsebene  tritt  in  Schiefstellung  ein,  Konjugation 
konnte  ich  nicht  beobachten. 

Systematische  Stellung. — lieber  die  verwandtschaftliche  Be- 
ziehung der  Peridineai^  mit  anderen  Organismengruppen  sind  die 
Meinungen  noch  geteilt.  Die  Gruppe  ist  oft  mit  Diatomeen  in 
engste  Beziehung  gebracht  worden.  Näheres  darüber  kann  man 
bei  Schutt"  finden.  Aber  gleichzeitig  ist  ihre  verwandtschaftliche 
Beziehung  mit  den  Protococcacese  und  Volvocacese  von  ihm  auch 
anerkannt.^  Oltmanns^  kann  die  Auffassung,  dass  die  Gruppe  den 
Diatomeen  nahe  steht,  nicht  ganz  teilen.  Er  zieht  es  vor,  Bütschli 
folgend,  die  Beziehungen  zu  den  Cr^-ptomonadinen  zunächst  zu 
betonen,  damit  auch  die  Peridineen  als  Flagellaten  zu  betrachten 
und  den  Namen  der  Dinoflagellaten  beizubehalten.  Jedenfalls 
die  nahe  Verwandtschaft  mit  Flagellaten  und  verwandten  Formen 
ist  überall  ausgesprochen. 

Was  nun  den  systematischen  Wert  der  Zahl  und  Ausbildung 
der  Geissein  bei  der  Einteilung  der  Flagellaten  und  verwandten 
Formen  anbelangt,  so  bilden  sie  wichtige  Anhaltspunkte  für  die 
systematische  Einteilung  derartigen  Formen.  Bütschli  legte  z.  B. 
ein  Hauptgewicht  darauf.^     Klees''  aber  hat  seine  Einteilung  auf 

1.  Peridinese  (Klebs,  1883).  Peridiniales  (Schutt  in  En'Gleb  u.  Pbantl,  Die  nat.  Pfl.- 
familien.  L  Ti-il.  1896),  Cilioflagellata  (Claparkde  u.  Lächmann  1858-1861,  Arthroiele  Flagel- 
liten  (Stein  1833).  Dinoflagellata  (Butschli  1835). 

2.  Schutt,  Ppridiniales  in  Engler-Pranti,  :  Die  natürl.    Pflanz ?nfannlien  T.  Teil  1.  Abt.  b. 

3.  Schutt,  Ebenda  S.  2,  8  u.  lö. 

4.  Oltmanns.  Morphologie  und  Biologie  der  Algen,  1904.     Erster  Bd.  S.  35. 

5.  Hierüber  vgl.  man  Bütschli,  Protozoa,  in  Bronn's  Klassen  und  Ordnungen  des  Thier- 
Eeichs  1889  2te  Abt.    Mastigopbora. 

6.  Gr.  Klhbs  Flagellatenstudien.     Zeitschr.  f.  -wiss.  Zoologie.  Bd.  LV.  1892. 


Beobachtungen  an  einer  SüsswaSser-Peridinee.  ö«? 

die  gesammte  Organisation  des  Vorderendes  und  in  Verbindung 
damit  anf  die  Art  der  Naln-ungsaufnabme  gegründet.  Dieses 
Prinzip  ist  auch  von  Senn  in  seiner  Bearbeitung  der  Flagellaten^ 
in  erster  Linie  angewandt.  Senn  schätzt  ferner  den  systematis- 
clien  Wert  der  verschiedenen  morphologischen  Eigenschaften  wie 
folgt':  ,,in  zweiter  Linie  ist  für  die  Bildung  grösserer  Untergruppen 
die  Organisation  der  contractilen  Vaculolen  wichtig;  ob  sie  einfache 
in  Einzahl  oder  zu  mehreren  vorkommende  Bläseben  darstellen, 
oder  ob  sich  mehrere  zu  einem  mehr  oder  weniger  stark  differen- 
zierten Apparat  vereinigt  haben.  Ahnliclien  systematischen  Wert 
hat  wohl  auch  die  Kernstrnktur,  jedoch  ist  sie  vorläufig  noch  zu 
wenig  bekannt.  In  dritter  Linie  kommt  der  Grad  der  Ausbil- 
dung der  plasmatischen  Körperhülle  in  Betracht;  dabei  sind  aber 
die  Zellausscheidungen,  wie  Gallerthüllen,  Gehäuse  und  Stielbil- 
dungen auszuschliessen  ;  dieselben  können  zur  Begrenzung  von 
Gattungen  dienen,  von  Unterfamilien  nur  bei  grosser  Mannigfal- 
tigkeit der  Bildungen.  Erst  in  vierter  Linie  können  die  Geissein 
berücksichtigt  werden,  und  zwar  nicht  nur  ihre  Zahl  und 
Anordnung,  sondern  auch  ihre  Gestalt  und  Funktion.  Neben  der 
Geisseiausbildung,  muss  au(;h  der  Besitz  oder  Nichtbesitz  von 
Chromatophoren  und  die  Bildung  bestimmter  Stoff  wechselprodukte 
(Stärke,  Paramylon)  berücksichtigt  werden,  schliesslich  auch 
plasmatische  Zehanhänge  wie  Kragen  und  Peristombildungen." 

Er  gibt  ferner  an,  ,,Als  Gattungsmerkmale  kommen  in 
Betracht:  Zahl  und  Ausbildung  der  Geissein  bei  sonst  gleicher 
Organisation,  Metabolie  oder  Starrheit,  besondere  Mundapparate, 
unduherende  Membranen,  Gehäuse-,  Hüllen-,  Stiel-  und  Colo- 
nienbildung;  bei  grossen  Verschiedenheiten  der  Körpergestalt 
auch  diese;  sie  wird  sonst  nur  als  Artcharakter  benutzt." 

Für  die  Einteilungen  einiger  grösseren  Gruppen  jedoch  ist  die 
Zahl  der  Geissein  auch  gebraucht.  So  z.  B.  Avurde  die  Unterordnung 
Chrvsomonadinese  welche  von  Klebs^  nach  der  Ausbildung  der 


1.  G.  Senn,  Flagellata  in  Englee-Prantl,  Die  natürl.  Pfl.-fain.  I  Teil,  1  Abt.  S.  93. 

2.  Ebenda,  S.  109. 

3.  G.  Klebs,  Flagellatenst'idien  (Zeitschr.  f.  wiss.  Zool.  Bd.  LV.  1892). 


90  Alt.  2.— N.  Obno: 

Zellumhüllung  in  nackte  (Chrysomonaàina  nuàci)  in  einem  Gehäuse 
oder  Schale  sitzende  (loricata)  und  in  solche  mit  eng  anliegender, 
hautartiger  'HixWa {memhranata')  eingeteilt  worden  ist,  von  Senn^  nach 
der  Zahl  und  Ausbildung  der  Geissein  bestimmt.  Nach  seiner 
Ansicht  wäre  diese  mehr  als  genetisclies  Merkmal  aufzufassen, 
indem  die  Hüllenbildungen  der  Flagellaten  als  sekundär  erwor- 
bene Organe  aufzufassen  seien.  Er  teilt  nämlich  die  genannte 
Unterordnung  in  3  Gruppen:  Chromulinacea3  (mit  1  Geissei), 
Hymenomonadaceœ  (mit  2  ganz  oder  annähernd  gleichen  Geis- 
sein) und  Ochromonadacea)  (mit  2  ungleichen  Geissein). 

Bei  solcher  Sachlage  könnte  man  berechtigt  sein,  auch  eine 
neue  Gattung,  oder  selbst  höhere  systematische  Einheit  für  unsere 
Form  aufzustellen,  da  sie  Konstant  2  Längsgeisseln  trägt  im 
Gegensatz  zu  anderen  Peridineen  die  nur  1  Längsgeissel  besitzen. 
Dieses  um  so  mehr,  da  die  Gruppe  Peridinea?  sich  in  den  Begeis- 
selungsverhältnissen  sehr  einheitlich  verhält,  so  class  das  Vor- 
handensein von  1  Quergeissel  und  1  Längsgeissel  das  wichtigste 
Organisationsmerkmal  dieser  Gruppe  bildet. 

So  lange  wir  aber  nicht  im  Stande  sind  zu  beurteilen,  ob  das 
Vorhandensein  der  Längsgeisseln  in  der  2-Zahl  hier  als  primäres 
Merkmal  aufzufassen,  oder  vielmehr  als  ein  sekundäres  anzusehen 
sei,  wäre  es  besser,  wie  es  auch  in  dieser  Arbeit  geschieht,  vorläu- 
fig diese  Form  den  bekannten  Formen  mit  denen  sie  in  anderen 
Punkten  übereinstimmt,  anzureihen.  Es  ist  sehr  wünschenswert 
auch  bei  anderen  Süsswasser-sowie  Meeresforraen  die  Begeisse- 
lungs Verhältnisse  näher  zu  prüfen  und  reichlichere  Materialien  zu 
sammeln,  um  einiges  Licht  auf  den  systematischen  Wert  des  Vor- 
handenseins der  Doppellängseisseln  bei  den  Peridineen  zu  werfen. 
Dann  erst  kann  man  die  systematische  Stellung  der  in  Rede  ste- 
henden Form  richtig  beurteilen. 

Die  Gruppe  Peridiniales  teilt  man  in  3  Familien,^  Gymnodi- 
niaceie,   Prorocentraceaä  und  Peridiniacea?,   von  denen    die    erste 


1.  G.  Senn,  1.  c.  S.  152. 

2.  Schutt  in  Englee-Prantl,  Eie  nntürl.  Pfl.-familien.     I.  Teil. 


Beobaclitungen  an  einer  Süsswasser-Peridinee.  91 

Familie  ohne  Panzer  ist,  während  die  zwei  anderen  mit  Panzer 
versehen  sind.  Unsere  Form  kommt  also  unter  Gymnodiniacese, 
und  zwar  ist  sie  in  die  Gattung  Gymnodinium,  deren  Vertreter  so- 
wohl im  Meere  als  auch  im  Süsswasser  verbreitet  sind,  zu  bringen. 
Die  Artdiagnose  lautet  demnach  : 

Gymnodmium  bicflÊatum  nov.  sp. 

Bis  jetzt  der  kleinste  bekannte  Vertreter  der  Gattung.^  Länge 
22.2  <«  und  Breite  IG. 9  /^.  Die  seichte  Querfurche  den  Körper  in 
einer  kaum  ansteigenden  Schraubenlinie  umziehend.  Die  beiden 
Körperhälften  einander  nahezu  gleich.  Längsfurche  vertikal. 
Längsgeisseln  zwei,  gleich  lang  und  gleich  gebaut.  Ein  Augen- 
fleck nicht  vorhanden.  Durch  eine  bedeutende  Reprodnktions- 
kraft  ausgezeichnet  und  auch  zur  Cj^stenbildung  sehr  geneigt. 
Cysten  oft  stachelig,  derb,  durch  Chlorzinkjod  dunkelrotbraun; 
auf  Agar  künstlich  zu  Ency stierung  zu  veranlassen.  Süsswasser- 
bewohner,  in  Graben. 


1.  Gymnodinium  pusillum  Schilling  besitzt  die  Grosse  23'0  MX  i8'4  ß  und  ist  von  Schil- 
ling als  die  kleinste  Form  dieser  Gattung  angegeben  (Vgl.  Schilling  1.  c).  Die  kleinsten 
Formen  aus  anderen  Gattun^ea  sind  z.  B.  Amphidinium  lacustre  (23.0  AX  184  m)  und 
Glenodinium  pulvisculus  Stein  (23-Om  XlS-i^). 


92  Art.  2.— X.  Ohno: 


Figuren-Erklärung.  1 

Sämtliclie  Figuren  sind  mit  dem  Abbe'sclien  Zeiclieuapparat  und  zwar  wo  nichts  ; 

anders  angegeben  ist,  unter  Benutzung  von  Oc.  4  und  Obj.  F,  Zeiss,  entworfen. 

Fiö".  1.  Schwimmendes  Individuum.  , 

o  I 

,,     2  11-  3-     Fixierte  u.  gefärbte  Exemplare.  < 

,,     4  n.  5.     Korperform  im  Umriss.     (Mit  Geissehi). 

,,     6  11-  9,  ^1-  10  bis  16.     Variationen  in  Korperform.  (im  Umriss). 

,,     7  11-  8.     Kin  und  dasselbe  Individuum,  einmal  bei  oberer  Einstellung  (7)  ■ 

und  andersmal  tieferer  Einstellung  (8).  j 

,,     17-     Sich  desorganisierendes  Individuum,  dessen  Längsgeisselendeu  mit  i 

Bläschen  versehen  sind.  j 

,,     18  bis  21,  u-  23  liis  26.     Xatiirliclie  Cysten,  von  denen  19  mit  am  deut- 
lichsten ausgebildeten  Stacheln. 

,,     22.     Stärkekörner  aus  dem  Zellkörper. 

,,     27  Ijis  33.     Künstlich  auf  Agar  erzeugte  Ruhezellen. 

„     34  his  37.      Teilungsstadien.     36  (Oc.  2  +  ^  Horn.  Imm  ,  Zeiss.)  \ 


Jour.  Sei.  Coll.,  Vol,  XXXII.,  Art.  2.  PI.  I. 


N.   Ohno,  del. 


T 


JOURNAL  OF  THE  COLLEGE  OF  SCIENCE,  LMPEEIAL  UNIVERSITY. 
TOKYO.  JAPAN. 

VOL.  XXXII.,  ARTICLE  3. 


Observations  and  Experiments  on  the  Ctenophore  Egg: 

I.     The  Structure  of  the  Egg  and 
Experiments  on  Cell- division. 

By 
Naohide  Yalsu. 


Introduction. 

The  ctenophore  egg  has  become  classic  in  experimental  em- 
bryology through  the  invaluable  papers  of  Chun  ('92),  Driesch 
and  Morgan  ('95),  Ziegler  ('98,  '03)  and  Fischel  ('97,  '98  and 
'03).  Singularly  enough,  however,  since  the  appearance  of  these 
works  no  further  detailed  analytical  studies  on  the  interesting  egg 
have  been  undertaken.  At  the  suggestion  of  Professor  E.  B. 
Wilson,  the  writer  made  observations  and  experiments  upon  the 
egg  of  four  common  species  of  ctenophore  in  the  spring  of  190G, 
at  the  Naples  Zoological  Station.^ 

The  present  paper  deals  first  with  the  structure  of  the  egg 
with  a  note  on  the  polocy tes  and  on  fertilization  ;  secondly,  with 
observation  on  the  process  of  cell-division,  mainly  that  of  the  first 
cleavage;  and  thirdly  with  experiments  performed  upon  the  egg 
of  Beroë  ovata.  It  Avill  be  followed  by  two  other  papers;  one  on 
cytogeny  and  experiments  on  cleavage  physiology,  and  the  other 
on  germinal  localization. 

As  to  experimentation,  I  wish  to  lay  especial  stress  upon  the 
following  points.     Great  care  was  taken  to  secure  good  water  quite 

1  This  study  was  made  with  the  aid  of  a  grant  from  the  Carnegie  Institution  at 
Washington,  for  which  I  take  this  opportunity  to  express  my  gi-atitude.  I  also  have  pleasure 
L  n  acknowledging  my  indebtedness  to  the  staff  of  the  Naples  Zoological  Station. 


"2  Art.  ?.  -  Naohide  Yatau  : 

far  from  the  shore.  The  water  taken  near  the  city  of  Naples  was  so 
polluted  that  it  was  unfit  for  use  in  developing  egg-fragments  into 
embryos.  This  is  the  indispensable  condition  for  ctenophore 
experiments.  The  high  mortality  in  Driesgh  and  Morgan's  work 
seems  to  have  been  due  to  tlie  neglect  of  this  precaution  ('95  p. 
217).  To  obtain  eggs  two  or  three  animals  were  kept  in  a  rather 
small  cylindrical  jar,  so  that  they  stirred  the  water  more  or  less 
when  swimming  and  kept  the  eggs  they  laid  constantly  in  motion. 
If,  on  the  contray,  the  ctenophores  be  put  in  a  large  jar,  the  eggs 
are  liable  to  stay  near  the  surface  ;  there  they  l)ecome  weak  and 
give  rise  to  less  lively  larvae  or  fail  to  develop  at  all. 


I.      Structure  of  the  Egg. 

The  eggs  of  the  following  four  common  species  of  ctenophores 
were  studied;  namely  Beroë  ovatu,  B.fonkalii,  Callianira  hialata  and 
Eucharis  vmUicornis.     The  relative  sizes  of  the  eggs  of  these  forms 

are  shown  in  Fig.  I  {cf.  Chun 
'80  p.  100).  The  egg  of 
Beroë  ovala  was  the  one  most 
carefully  studied  and  exclu- 
sively used  in  experiments, 
being  peculiarly  suited  for 
the  purpose  on  account  of  its 
large  size  (1-1.2  mm)  and 
of  its  consistency. 

When  the  eggs  are  laid, 
they  are  found  entangled  in 
a  string-like  mass  of  jelly. 
Close  to  the  egg  is  a  thin 
gelatinous  covering  that 
turns  into  a  thick  layer  of 
jelly  after  fertilization.^  The 
egg  has   three    visible    con- 


PlG.    I. 

Diagram  showing  the  relative  sizes  of  ihe  eggs  of 
Callianira  hialata  (1),  Eucharis  viiilticoriu's  (2), 
Beroë  forskalii  ^3)  and  B.ovata  (4).     X6> 


1     The  eggs  of  Eucharis  multiconiis  can  be  more  easily  taken  out  of  the  jelly  than  those  of 
Beroë, 


Observations  and  Experiments  on  the  Ctenophore  Egg.  o 

«centric  differentiations;  namely  (a)  the  extremely  thin  homogene- 
ous outer  layer,  (b)  the  ectoplasm  and  (c)  the  entoplasm. 

The  outer  "membrane"   is  a  very  thin  semifluid  layer  free 

from  granules.     It  can  hardly  l)e  called  membrane.     It  is  difficult, 

if  not  impossible,   to  detect  it.     When  the  egg  is  compressed  or 

wherever  accumulation  of  the  ectoplasm  takes  place,   one  can  see  it 

fairly  well.     I   should  not  hesitate  to  homologize  this   with   the 

•ectosarc  described  by  Andrews  in  the  egg  of  Hydra  ('98)  and  with 

a  similar  layer  found  in  the  sea-urchin  egg,    to  Avhich  Zjeglee,  has 

.attached  so  much  importance  ('03,   '04).      It  can  not  be  looked 

upon    as    identical     with     the     hypothetical    cell-membrane     of 

Rhumbler  ('99).      What  rule  this  thin  layer  actually  plays  during 

•cleavage  is  problematical.     I  am  rather  inclined  to  think  that  it 

has  very  little  to  do  with  that  process. 

The  ectoplasm  is,  as  has  been  described  by  many  authors,  a 
rrather  thick  layer  of  finely  alveolar  plasm.  It  is  of  uniform  thick- 
ness until  fertilized.  This  layer  is,  contrary  to  Zieoler' s  view,  not 
.at  all  of  the  same  nature  as  the  "ectoplasm"  of  the  sea-urchin  egg. 
It  is  fluorescent  and  looks  green  under  reflected  light,  reminding 
•one  of  a  piece  of  uramium  glass  (cf.  Chun  '80  p.  100).  In  stimulating 
the  egg  with  a  weak  electric  current  this  layer  alone  seems  to  emit 
a  beautiful  greenish  light.  It  should  be  mentioned,  that,  as  the 
-development  advances,  the  ectoderm  comes  to  monopolize  this 
property.  In  a  Wood's  Holl  species,  Mncmiopsis,  the  egg  before 
-cleavage,  according  to  Peters  ('05),  was  not  phosphorescent. 

It  need  hardly  be  mentioned  that  the  entoplasm  is  a  coarse 
.alveolor  structure.  The  alveoles  of  the  Callianira  egg  are  much 
fewer  in  number  than  those  of  the  Beroë  egg,  the  individual  alveole 
■  of  the  former  being  much  larger  than  that  of  the  latter  (PI.  II, 
I^igs.  27  and  28).  On  crushing  the  Callianira  egg  I  could  count 
in  one  case  67,  in  another  64,  and  in  still  another  54  alveoles. 
Ziegler  calls  the  alveoler  substance  "  transparent  yolk  granules  " 
;('98  p.  36).  This  seems  to  have  given  Rhumbler  tlie  impression 
îthat  it  was  made  up  of  a  rather  highly  viscous  substance.  But 
.as::a  matter  of  fact,  this  is  not  so.  Chun  ('80)  has  called  it  "  Zell- 
rsaft."     Though  this  term  is  not  happily  chosen,  yet  it  is  far  better 


.4  •..  Art.  3.— Xaohide  Yatsu  : 

than  ''yolk,"  In  the  material  fixed  witli  Flemming's  fluid  the- 
alveolar  substance  is  completely  dissolved,  being  represented  bj^ 
holes.  Whatever  its  chemical  nature  may  be,  it  is  certainly  not 
similar  to  Avhat  commonly  goes  luider  the  term  yolk.  In  the 
entoplasm  no  axial  differentiation  can  be  detected;  tlie  size 
and  nature  of  the  alveoles  are  the  same  throughout  the  egg.' 


II.    Polocytes. 

I  have  nothing  to  add  about  the  formation  of  the  polocytes, 
excepting  that  in  one  case  a  maturation  spindle  with  no  pok'-ra3\s- 
was  met  with  in  an  ectoplasmic  accumulation  (Flemmixcj's  fluid 
material). 

In  tlie  living  egg  it  was  not  difficult  to  see  the  polocytes. 
Pseudopodia  are  seen  on  botli  the  polocytes.  The  first  polocyte 
loses  them  and  invariably  divides  in  two;  each  part  has  a  smootli 
surface.  The  pseudopodia,  however,  remain  on  the  second  polo- 
cyte (PL  I,  Fig.  4).  As  to  whether  or  no  the  polocytes  perform^ 
an  amoeboid  locomotion  b}-  means  of  the  pseudopodia  I  have  had 
no  means  of  determining.  Yet  I  am  rather  inclined  to  believe  that 
they  do  not  (cf.  Chun  '80  p  101).  As  is  shown  in  Figs.  1,  2  and  3 
(PL  I)  a  thickened  portion  of  the  ectoplasm  is  found  beneath  the- 
poloc^^tes.  The  egg  nucleus  undoul)tedl3'  lies  in  this.  The  ac- 
cumulation lasts  for  some  time  after  the  formation  of  the  second 
polocytes.     But  it  soon  disappears. 

III.     Entrance  of  the  Spermatozoon  into  the  Egg. 

As  already  mentioned,  when  the  eggs  are  discharged,  they 
have  a  thin  gelatinous  covering  about  them.  When  fertilized  tins 
membrane  changes  into  a  thick  layer  of  jelly.  I  could  not  ascer- 
tain how  the  process  took  place.  It  is  highly  j^i'obable  that  tlie 
change  is  of  the  same  nature  as  that  of  echinoderm  eggs.     In  any 

1    In  this  connection  it  may  be  of  some  interest  to  note  that  Fewkes  saw  in  the  egg  of 
Jgalma  a  mass  of  rosy  entoplasm  at  one  polo  ("85  p.  247). 


Observations  anel  Exper.ments  on  the  Otenophoro  Egg-. 

•event  tins  jelly  layer  gives  .a  splendid  criterion  for  distinguishing  at 
.a  glance  fertilized  egg  from  unfertilized. 

One  other  phenomenon  accompanying  fertilization  is  that  the 
•ooplasm  suddenly  acquires  greater  consistency.  This  is  especially 
frappant  after  the  formation  of  an  ectoplasmic  thickening  around 
the  spermatozoon.  One  can  hardly  fail  to  notice  this  change  when 
•experimenting  upon  various  stages  of  the  egg. 

The  entrance  of  the  spermatozoon  can  readily  he  seen  in  the 
■egg  of  Beroë.     Fig  5  (PL   I)  shows  a  surface  view   soon  after  its 
penetration  into  the  egg.     A  réfringent  hody  (acrosome?)  is  seen  a 
little   apart  from  the   head.     Behind   the  head  is  a  dark  body, 
sperm-centre,  provided  with  long  rays.     Soon,   however,   the  rays 
■disappear.     In  a  side  view  one  sees  an  entrance-cone  consisting  of 
a  thickened   external  homogeneous  layer  and   also   considerable 
.accumulated  ectoplasm.     The   entoplasmic  alveoles  are  arranged 
radially  (PI.  I,  Fig.  6).     In  them  no  rays  are  seen  in  the  living 
-egg.     In  section,    however,    distinct  long  rays  come  into   view, 
which  extend  from  the  straightened  alveolar  walls  of  the  ectoplasm 
into  those  of  the  entoplasm.      As  is  seen  in  Figs.   7  and  8  (PI.  I) 
the  ectoplasmic  accumulations  remain  for  some  time,    so  that  by 
them  one  can  tell  at  once  how  many  spermatozoa  have  entered  the 
•egg.       On  one  occasion  I  saw  an  egg  with  as  many  as  five  of  them 
in   it.       Polyspermy   in  this  form  is  not  at  all    a   physiological 
phenomenon.     It  usually    takes   place    when   eggs    are   kept  too 
-crowded  in  a  jar.     I  know  nothing  about  the  fate  of  those  sperm- 
nuclei  which  fail  to  unite  with  the  egg-nucleus.     Yet  judging  from 
the    fact    that    in    many   cases    polyspermy    does    not    lead  to 
.abnormal  cleavage  ;  those  sohtary  sperm-nuclei  seem  to  degenerate 
.■in  situ. 


IV.     Cell-division. 

a)     The  First  Cleavage. 
I    could    not    make    out    how    or    where    the   germ- nuclei 


6  Art.  3.  —  Xaohide  Yatsu  : 

meet.'  At  any  rate  prior  to  the  first  cleavage,  there  takes  place  a. 
change  in  the  distribution  of  the  ectoplasm:  it  thickens  consider- 
ably near  the  macromere  pole,'  while  at  tl)e  opposite  pole  it  thins 
out  a  great  deal  (PI.  I,  Figs.  9  and  10).  Cleavage  goes  on,  as  has 
been  observed  by  a  good  many  investigators.  Sometimes  the 
cleavage  fuiTOw  is  bent  sHglitly  to  one  side  near  its  completion 
(PL  I,  Fig.  11).  It  should  liere  l)e  especially  mentioned  tliat  the 
cleavage  is  not  strictly  unihiteral,  contrary  to  Ziegler's  observa- 
tion ('98.  p.  41;  '03  p.  159  and  his  diagram  Fig.  7),  and  also  to 
Rhumbler's  opinion  based  upon  Zieglers  results.  A  sliallow 
depression  is  always  present  at  the  micromere  pole,  as  is  seen  in. 
the  sketches  drawn  one  upon  another  at  different  periods  (PI.  I, 
Fig.  11).  And  it  will  be  also  noted  that  the  top  of  each  blas- 
tomere  becomes  more  rounded.  The  lateral  elongation  is  almost 
nil.  The  rate  of  tlie  cleavage  from  the  macromere  pole  is  <S-19/i 
per  minute,  17/<  on  an  average,  at  a  room  temperature  of  G5°-07°F. 
The  rate  of  cleavage  from  tlie  opposite  pole  is  a  little  slowei".  13/>- 
per  minute  on  an  average. 

To  supplement  Ziegleu's  observations,  the  j^rocess  uf  cleavage 
and  especially  the  "cleavage  head"  (Furchenkopf)  Avill  be  des- 
cribed in  some  detail.  In  the  beginning  a  slight  depression  ap- 
pears near  the  polocytes.  Its  optical  section  is  shown  in  Fig.  12. 
(PL  II).  Here  is  a  pair  of  prominences  in  the  outer  homogeneous 
layer  (cf.  Andrews  '98).  Sometimes  they  are  continuous,  forming, 
a  bridge  over  the  now  deepening  furrow.^  Sometimes  there  is  one 
process  on  one  side  and  two  on  the  other  (PL  IL  Fig.  12)  (rf. 
Kleinenbeeg  '72  p.  49.,  Taf,  4,  Fig  4;  Wager  '09  p.  23,  PL  III, 
Fig.  23a).  Around  the  process  a  fine  display  of  spinning  activity 
is  visible.  At  the  bottom  of  the  cleavage  depression  are  rays  in  tlie 
homogeneous  laver. 

1  As  is  seen  in  Figs.  10  and  11.  the  polocytes  ai-e  usually  situât  h1  at  some  distance  from- 
the  cleavage  furrow.  Whether  the  egg-nucleus  or  cleavage-nucleus  moves  a  little  from  the  spot 
■where  the  polocytes  liave  been  formed,  or  whether  the  polocytes  are  transported  by  some  means, 
is  not  certain.     But  the  latter  alternative  seems  to  be  the  more  probable  one. 

2  Macromere  pole  =  vegetative  pole  of  Hatschek  (Korschklt  and  Hkider  'OJ  p.  24). 
Throughout  the  i)lates  the  macromere  pole  is  above,  and  the  micromere  pole  below. 

3  The  bridge  is  not  so  distinct  as  was  obsei-ved  by  Tannreüther  in  the  Hi/dra  egg,  where- 
yolk  granules  were  seen  to  pass  from  one  blastomere  to  the  other  ('08  p.  2ö7). 


Observations  and  Experiments  on  the  Ctenophore- Egg.  T 

As  a  digession,  the  results  from  the  study  of  sections  of  this- 
stage  may  here  be  given.  Cleavage  begins  at  the  telophase  of  the 
first  mitosis.  Rays  (pole-rays)  of  a  considerable  length  extend 
into  the  entoplasm  of  both  blastomeres,  centering  about  the  newly 
formed  nuclei  (the  centres  are  in  all  probability  situated  very  close 
to  the  nuclei).  Besides,  something  like  sheath-rays  are  found 
between  two  asters.  They  are  evidentl}^  cut  apart  by  the  growing, 
"  cleavage  head."  A  similar  condition  has  been  observed  in  the 
Hydra  egg  by  Brauer  ('91  Taf.  IX,  Figs.  16  and  17). 

Now  coming  to  the  next  stage  (PL  II,  Fig.  13).  The  bottom 
of  the  depression  has  been  carried  farther  down.  Usually  the  clear 
protuberances  at  the  entrance  of  the  furrow  are  drawn  in.  Fine 
spinning  is  seen.  The  protoplasmic  threads  are  not  parallel  to  one 
another.  Often  they  decussate.  At  the  "head"  are  radiations- 
as  in  the  foregoing  stage. 

A  fully  formed  "  cleavage  head  "  is  shown  in  Fig.  14  (PI.  II). 
By  this  time  protoplasmic  spinning  is  restricted  to  the  entrance 
and  bottom  of  the  furrow.  The  "  head  ''  is  a  thickened  ectoplasm 
Here  one  notices  that  the  outer  homogeneous  layer  also  has  in- 
creased in  thickness.  Réfringent  alveoles  in  the  ectoplasm  are 
arranged  radially  as  extensions  of  fine  radiations  of  the  homoge- 
neous layer.  Towards  the  entoplasm  are  processes  (Zacken)  as- 
has  been  rightly  observed  by  Ziegler  ('98,  '03).  I  tried  hard  to 
detect  rays  extending  from  the  tips  of  the  processes  into  the 
entoplasm,  but  contrary  to  Rhumbler's  assumption  ('99  p.  203 
Fig.  12  and  p.  205  Fig.  13),  there  were  no  such  things;  here  the 
ectoplasm  simply  comes  in  contact  with  the  entoplasm.  The 
alveoles  of  entoplasm  liere  show  a  peculiar  arrangement  worth 
noting.  Those  along  tlie  walls  of  the  cleavage  furrow  seem 
to  have  been  carried  down  with  it  and  those  found  at  the  tip 
of  the  "  head  "  are  somewhat  flattened  (cf.  PI.  II,  Fig.  28).  This 
undoubtedly  shows  that  the  "cleavage  head"  pushes  downward 
instead  of  being  pulled  l)}^  the  contraction  of  rays,  stretching 
between  the  cleavage  head  and  the  micromere  pole.  At  the 
next  stage  the  alveoles  recede  from  the  median  plane  as  i& 
shown  in  Fig.   15  (PI.  II).     It  will  be  of  some  interest  here  to 


€  Art.  3.— Naohide  Yatsn  : 

•examine  two  cleavage  stages  of  the  egg  of  Berœ  forskhlii  (PI.  II, 
Figs.  10  and  20).  In  this  particular  egg  the  second  cleavage  has 
begun  before  the  first  has  come  down  nearly  two  thirds  of  this 
•entire  course.  By  the  precocious  second  division  tlie  typical 
alveolar  arrangement  has  been  considerably  disturbed,  a  flow-figure 
having  been  formed  in  each  blastomere.  For  all  that,  the  first 
•cleavage  cuts  through  the  egg  normally  as  through  nothing  had 
happened  near  the  macromere  pole. 

Pari  paH.m  with  the  coming-down  of  the  cleavage  furrow,  the 
•ectoplasm  thickens  near  the  micromere  pole  as  has  been  observed 
by  ZiEGLER.  And  the  "cleavage  head"  meets  the  ectoplasmic 
<iccumulation  there  (PI.  II,  Fig.  15,).  The  walls  (,f  the  cleavage 
furroAv  near  the  "head"  become  irregularly  wrinkled  and  the  spin- 
ning activity  increases  (cf.  Andrews  '98).  The  hole  now  assumes 
a  triangular  shape.  As  the  entoplasmic  alveoles  quickly  retreat, 
the  ectoplasmic  bi'idgc'  is  left;  between  two  fartmeres  (PI.  II, 
Figs.  17,  18,  21,  21).  It  is  interesting  to  recall  that  LoEB  observed 
•cytoplasm  flow  away  from  the  furrow  towards  the  end  of  unilateral 
•cleavage  ('06  p.  OG).  No  particular  movement  as  seen  by  Bunting 
in  Hijdraciinia  ('04  p.  216)  takes  place  {cf.  Ziegler's  experiments). 
Finally  the  ectoplasm  also  goes  into  the  blastomeres,  leaving 
behind  a  fine  thread  of  homogeneous  layer.  By  the  time  one 
finds  an  ectoplasmic  thickening  with  radially  arranged  entoplasmic 
alveoles  near  the  micromere  pole  of  each  blastomere  (PL  II,  Fig.  1 8) 
In  studying  this  stage  with  a  low  powder  one  soon  notices  that 
the  greater  part  of  the  ectoplasm  has  come  down  towards  the 
micromere  pole  (PI.  II,  Figs.  31,  22) 

Incidentally  I  might  mention  that  the  cell- wall  between  the 
two  blastomeres  of  Eucharis  and  CalUanira  has  a  sieve-like  appear- 
ance (PL  II,  Figs.  22,  24,  27).  In  the  latter  form  I  was  able  to 
see  this  peculiarity  between  two  entoderm  cells  as  late  as  the 
gastrula  stage.  The  fenestrated  appearance  is  due  to  the  lenticular 
accumulation  of  a  certain  fluid  as  correctly  observed  by  Chun  ('80 
p.  102)  (cf.  Fol  '73  Tf.  24,  Fig.  5). 


1     The  surface  of  the  bridge  has  r.idiatint'  wrinkles. 


Observations  and  Experiments  on  the  Ctenophoi'e  Egg.  «^ 

1))     Siihseiueiit  Cleavajes. 

Soon  alter  the  first  cleavage  is  completed,  the  ectoplasmic 
thickening  near  the  micromere  pole  disappears  and  at  the  same 
time  an  accumnlation  comes  in  view  over  the  macromere  pole 
(PI.  II,  Fig.  23).  The  second  cleavage  takes  place  in  exactly  the 
same  manner  as  the  first  (PI.  II,  Fig.  24).  In  the  beginning  of  the 
third  cleavage  an  accumnlation  of  the  ectoplasm  near  the  macro- 
mere  pole  is  also  seen  (PI.  II,  Figs.  25,  26,  27).  At  the  fourth 
division  the  micromeres  are  formed,  which  are  almost  entirely 
made  up  of  the  greenish  ectoplasm  (PL  11,  Figs.  29,  oO).  Sub- 
sequent divisions  of  the  micromere  are  carried  on  in  unilateral 
fashion  similar  to  the  division  of  the  entire  egg.  So  also  the 
divisions  of  the  macromeres  (Text  fig.  II). 


Fig.  II. 
Two  dividing  entoderm  cells  of  Beroë  ovata.  X60. 

V.     Experiments  on  Cell-division.^ 

Fifty  eggs  of  Beroc  ovata  were  operated  on  during  the  first 
«leavage  in  twelve  different  ways  with  the  intention  of  testing,  if 
possible,  hypotheses  hetherto  put  forth  regarding  the  division 
mechanism  of  the  ctenophore  egg.  On  account  of  the  large  size 
iind  favorable  consistency  of  the  egg,  the  operations  were  performed 
Avith  the  greatest  ease  by  means  of  a  small  knife.  Sometimes, 
however,    in  case  tlie  jelly  around  the  egg  was  unusually  hard, 

I   This  part  of  the  present  p  ip.^r  was  read    as    a    preliminary   note   before   the    Seventh 
International  Zoological  Congress  at  Boston  1907. 


10  Art.  3.— Naohide  Yatsu  : 

cutting  was  accompanied  with  some  difficulty.  Very  soon  after 
the  operation  the  cut  surfaces  close  ;  so  rapid  is  the  closure  that 
one  cannot,  as  a  matter  of  fact,  see  exposed  entoplasm(f/.  Maas 
'03  p.  45).  Each  of  the  eggs  operated  on  was  placed  in  a 
compressorium\  and  the  subsequent  progress  of  cleavage  was 
followed.  From  a  single  egg  several  sketches  of  successive 
stages  were  made.  In  the  plates  of  the  present  paper  in  most 
cases  only  the  first  and  the  last  stages  have  been  reproduced, 
since  the  intervening  ones  would  l)e  of  little  value  in  ihustrating 
the  following  experiments. 

a)     Experitiiciit  1     (four  cases). 

A  portion  uf  the  egg  was  cut  below  tlie  "  cleavage  head  "  at 
various  angles  and  along  various  levels,  and  the  enucleated  pieces 
were  watched  to  see  if  they  showed  any  sign  of  division  activitv-' 
Even  in  the  case  in  which  the  cutting  plane  passed  very  near  the 
"  cleavage  head,"  nothing  happened  in  the  enucleated  piece — it 
simply  rounded  up  and  ceased  to  develop  further  (PI.  Ill,  Fig.  ol). 

1))     Expciiment  II    (one  case). 

The  above  experiment  was  modified  in  tlie  following  waj'. 
An  incision  extending  two  thirds  of  the  diameter  of  the  egg  was 
made  below  the  "cleavage  head"  to  see  if  the  connection  with 
the  nucleated  part  of  the  egg  would  impart  some  division  activity 
(PL  III,  Fig.  32).  The  cleavage  went  on  normally,  cutting  the 
upper  part  in  two,  but  the  lower  past  remained  undivided  (PL  III, 
Fig.  33). 

The  above  two  experiments  (Exp.  I  and  Exp.  II)  clearly  show 
that  portions  devoid  of  the  "  cleavage  head  "  do  not  manifest  any 
division  activity  whatever. 


1  This  was  not  used  for  compressin.,' the  eggs  but  as  a  sort  of  live-box  for  keeping  theui 
alive. 

2  I  have  a  case  in  which  both  pieces  produced  by  a  horizontal  cut  cleaved.      It  can  hardly 
be  doubted  that  it  was  a  dispermic  egg. 


Observations  and  Experiments  on  the  Ctenophore  Egg.  l  * 

c)     Experiment  III    (thirteen  cases). 

Sections  were  made  on  the  egg  above  the  "cleavage  head"  at 
various  periods  and  along  various  planes,  and  the  behavior  of  the- 
enucleated  fragments  containing  the  "cleavage  head"  were  studied. 

In  eight  cases  out  of  thirteen  the  cleavage  went  on  normally. 
The    rate    of    downward    progress    of    the    furrow    was    normal, 
or  a  little  slower  than  the  normal  that   is  15/^   per    minutes  on 
an    average.      It    was    sometimes    10//  or    even  as   slow   as    8//- 
per   minutes.     In   Fig.    34  (PI.   Ill)  the   section  passed   through 
the   middle    of   the    cleavage   furrow  and   in    Fig.    39    (PL    III) 
the  cut  Avas  made  when  the  cleavage  had  just  begun.     In  both  the 
enucleated  pieces  cleavage  went  on  as  though  they  were  a  part  of 
the  Avhole  egg.     It  is  interesting  to  note  that  the  movement  of  the- 
ectoplasm  is  the  same  as  in  the  entire  egg,   that  is,   a  thick  ecto- 
plasmic  accumulation  is  formed  about  the  micromere  pole  towards 
the  end  of  the  division  (PI.  Ill,  Fig.  35),  and  after  that  thickenings 
in  the  neighborhood  of  the  macromere  pole  (PI.  Ill,  Figs.  36,  40). 
Here    one    notices    that  the   thickness  of   the    above    ectoplamic- 
accumulation  depends  upon  the  size  of  the   enucleated  pieces.     It 
also  may  be  remarked  that  similar  up-and-down  movement  of  the 
ectoplasm  takes  place  in  the  nucleated  pieces  (PI.  Ill,  Fig.  41). 

In  studying  carefully  the  relation  between  the  angle  of  the  cuts 
and  the  direction  of  the  cleavage  furrow,  the  following  results  were 
obtained.  If  the  section  be  made  when  the  cleavage  furrow  is 
shallow,  then  the  division  goes  on  normally  irrespective  of  the 
angle  of  the  cut  (PI.  Ill,  Figs.  39.  40).  If,  on  the  other  iiand,  the 
operation  is  performed  in  later  stages,  the  cleavage  is  usually  ac- 
complished normally  only  when  the  section  is  horizontal  or 
approximately  so  ;  if  otherwise,  the  furrow  is  bent  and  the  l)ending  is 
ahvays  towards  the  side  with  more  cytoplasm,  so  that  tJie  result- 
ing blastomeres  are  of  nearly  the  same  size  (PI.  Ill,  Figs.  37,  38). 

In  three  cases  out  of  thirteen  something  unexpected  hap- 
pened. One  of  these  cases  is  represented  in  Figs.  41  and  42 
(PL  III).  The  other  two  were  very  much  hke  this.  The  cleavage 
furrow  went  down  riear  the  micromere  pole  and  turned  upward  so 


12  i         Art.  3.— N.;ohide  Yatsu  : 

that  in  tlie  end  ii  bridge-  was  formed  between  the  two  blastomeres. 
The  bridge  was  not  cut  through.  One  case  of  this  peculiar 
mode  of  cleavage  was  the  result  of  a  horizontal  section.  In  the 
other  two,  the  curving  of  the  furrow  faced  the  side  of  more 
■cytoplasm  (in  Fig.  41,  PI  III  to  the  left).  Conklix  observed  in 
the  egg  of  Linenjes  iiiercwiu^  tlie  turning-up  of  the  cleavage  furrow 
v(Pl.  3,  fig.  13)  and  thought  it  probable  that  this  might  l)e  due  to 
the  flowing  of  cytoplasm  through  the  bridge  from  one  cell  to  the 
other  (p.  160).  In  the  Beroc,  egg  no  streaming  phenomenon  ac- 
■companies  this  curling-up  of  the  cleavage  furrow. 

In  two  cases  for  some  unknown  reason  tlie  division  stopped  at 
.a  certain  stage  and  <lid  not  cut  thi-ough.  In  one  case  at  tlie  begin- 
ning of  the  division  the  ''  cleavage  head  "  Avas  thick  but  as  it  pro- 
•€eeded  the  ectoplasm  thinned    out  somewhat  (PI.  Ill,  Figs  43,  44). 

From  this  experiment  it  may  be  concluded  that  after  the 
"  cleavage  head  "  is  established,  the  cell-division  is  accomplished 
without  the  instrumentalitv  of  either  the  uncleus  or  the  centro- 
somes.  Furthermore  it  should  be  noted  in  this  experiment  that 
there  is  no  perceptible  difference  between  the  cut  and  the  uncut 
-eggs  in  the  thickness  of  "cleavage  head,"  though  in  some  cases 
.a  little  retardation  of  the  progress  of  the  cleavage  is  seen  in 
the  cut  eggs.  The  up-and-lown  flow  of  the  ectoplasm  takes  place 
independently  of  the  presence  or  absence  of  either  the  nucleus  or 
the  centi'osome. 

d)     Expeiiment.  II'     (Six  cases). 

In  order  to  test  whether  either  the  nucleus  or  the  centrosome 
•exerts  any  influence  on  the  deepening  cleavage  furrow,  the 
nucleated  portion  of  one  side  was  cut  off  at  various  angles  and 
periods,  and  the  behavior  of  the  cleavage  furrow  was  studied. 

In  all  cases  division  took  place  undisturbed  by  the  operation 
'(PI.  Ill,  Figs. 45, 46).  In  one  case,  however,  the  cleavage  furrow  was 
bent  near  its  end  towards  the  nucleated  side,  but  such  a  bending 
occurs  so  often  in  eggs  not  operated  upon  (Fig  11  PI.  1)  that  the 
•cause  of  the  bending  in  this  particular  case  cannot  be  ascribed  to 
the  operation. 


Observations  and  Experiments  on  the,  Ctenophoro  Eg^.  Xq 

This  experiment  streiigtheiis  the  results  obtained  from  tho 
preceding  one  (Exp.  Ill);  that  is,  the  nucleus,  centrosome,  and 
the  amount  of  cytoplasni  above  the  cutting  plane  have  little  to  da 
with  the  growth  and  direction  of  the  cleavage  fuiTOw\ 

;  e)     Experiment  V     (One  case). 

To  slightly  modify  the  above  experiment,  the  cut  w^as  extend- 
ed to  the  middle  of  the  nucleated  part  as  is  shown  in  Fig.  47  (PL 
IV).  The  "cleavage  head"  came  down  unintemipted  by  the 
operation.  But  the  cleavage  furrow  stopped  at  a  certain  point 
(PI.  IV,  Fig.  48)  OAving  probably  to  the  fact  that  the  nuclei  had 
already  enterd  the  phases  of  the  second  cleavage. 

f)     Experunent  VI     (Seven  cases). 

Id  tlris  experiment  the  "cleavage  head  "  were  split  in  two 
at  various  stages  by  a  vertical  cut  to  see  if  the  furrow  proceeds 
from  the  end  of  the  cut  or  from  some  other  place. 

In  no  cases  were  two  "cleavage  heads"  formed.  In  one  case 
out  of  seven  the  cleavage  furrow  made  its  appearance  at  the  bottom 
of  the  incision,  dividing  the  egg  in  two  equal  blastomeres. 

In  the  six  cases  a  remarkable  phenomenon  was  met  with.  A 
new  "  cleavage  head  "  emerged  on  one  side  of  the  incision  below 
the  original  bottom  of  the  furrow  and  in  most  cases  a  little  above 
the  end  of  the  cut.  It  proceeded  almost  at  right  angles  to  the  old 
cleavage  plane.  The  portion  of  the  cut  below  the  new  cleavage 
furrow  dwindled  and  was  either  obliterated  entirely  or  remained  as 
a  hole  for  a  fairly  long  time.  In  Fig.  49  (PI.  IV)  the  operation  was 
made  when  the  cleavage  furrow  was  very  shallow,'  and  the  result 
was  Fig.  50  (PI.  IV).  In  passing,  it  may  be  remarked  that  in  the 
egg  of    Pennaria    Hargitt  ('00)    observed    a   similar    figure    (his 

PL  II,  Fig.  2). 

It  is  interesting  to  note  that  in  this  case  an  ectoj)lasmic  thiclvcn- 
ing  was  foomed  at  the  side  of  the  egg  (PL  IV,  Fig.  50).  ^^Vhen  the 

1    'i'he   original  bottom  of    tbe   cleavage  furrow    is    marked   with    X  X    ia  this  and  the- 
following  figures. 


14  Art.  3,-Xaoliide  Yatsu  : 

operation  was  performed  at  a  later  stage  (PI.  IV,  Figs.  51,  52)  a 
new  accumulation  of  the  ectoplasm  appeared  at  the  end  of  the 
incision  which  reinforced  that  of  the  old  "cleavage  head"  to  form 
the  new  one.  The  resulting  cleavage  furrow  was  exactly  the  same 
.as  that  in  the  foregoing  case. 

In  five  cases  the  new  '^  cleavage  head  "  was  formed  on  the 
side  with  more  cytoplasm,  while  in  one  case  it  was  just  reverse. 
This  may  be  interpreted  in  two  difïerent  ways,  viz.,  (a)  that  a  new 
•cleavage  furrow  is  formed  on  the  side  of  more  cytoplasm,  or 
(b)  on  the  side  of  more  ectoplasm  or  "cleavage  head ''-substance 
•due  to  oblique  section.  The  above  experiment  fails  to  decide 
-which  of  the  alternatives  is  the  correct  view.  The  following 
«experiments  were  especially  directed  to  this  point. 

g)     Experiment  I'll     (One  cass). 

The  "cleavage  head"  was  split  vertically  and  a  nucleated 
portion  was  cut  off  as  is  shown  in  Fig.  52  (PI  IV).  A  new  cleavage 
furrow  was  formed  at  an  angle  to  the  old  and  on  the  side  of  more 
•cytoplasm. 

h)     Experiment  VIII    (Three  cases). 

For  the  same  purpose,  the  '  '  cleavage  head  '  '  was  split  and  a 
portion  of  the  cytoplasm  was  cut  off.  In  all  cases  as  in  Exp.  VI  a 
new  "cleavage  head"  ajopeared  at  some  angle  to  the  old  one. 
In  one  case  a  new  division  plane  was  directed  towards  the  cut  sur- 
face (PI.  IV,  Figs.  55,  56)  while  in  the  other  two  the  \\q\y  furrow 
was  formed  on  the  side  of  the  larger  cytoplasmic  mass  and  turned 
upwards  as  in  some  cases  in  Exp.  Ill  (PI.  IV,  Figs.  58,  59). 

This  experiment  clearly  shows  that  the  new  "  cleavage  head" 
develops  on  the  side  of  larger  cytoplasmic  portion. 

i)     Experiment  IX    (One  case). 

In  one  egg  the  "cleavage  head"  was  split  and  another 
incision  was  made  on  one  side  of  it  (PL  IV,  Fig.  GO).  A  new 
■"  cleavage  head  "  was  formed  on  the  left  hand  side  which  turned 


Observations  and  Experiments  on  the  Ctenophore  Egg.  J  5 

to  the  right  (PL  IV,  Fig.  Gl).     Tliis  cleavage  furrows  finally  cut 
off  an  enucl  ;ated  portion  (PI,  IV,  Fig.  62). 

j)     Kxpcrimeitt  X     (One  case). 

This  is  a  niodification  of  Exp.  VIII.  The  "  cleavage  head  " 
was  cut  in  two  by  a  vertical  incision,  a  small  nucleated  portion 
was  cut  off  (from  the  left  side  of  the  figure  Fig.  63,  PL  V),  and  the 
micromere  portion  was  removed.  A  new  cleavage  plane  was 
established  at  the  end  of  the  vertical  incision  that  cut  through 
almost  straight.  This  failed  to  yield  anything  of  interest,  being 
exactl}^  the  same  as  Figs.  55,  56  (PL  IV). 

k)     Experiment  XI     (Eleven  cases). 

In  this  experiment  an  incision  was  made  in  the  egg  at  the 
micromere  pole  to  see  if  that  would  affect  the  course  of  cleavage.  In 
nine  cases  out  of  eleven  the  cleavage  furrow  passed  by  the  incision 
as  though  nothing  had  happened  to  the  egg  (PL  V,Figs.  65,  67,  68). 
One  notices  in  Fig.  66  an  ectoplasmic  accumulation  at  the  left  hand 
corner.  In  one  case  the  cleavage  plane  stretched  towards  the 
incision  and  became  continuous  with  it  (PL  V,  Figs.  61.  70  ).  In 
another  case  the  cleavage  furrow,  which  had  attained  a  consider- 
able length,  dwindled  owing  to  the  operation,  and  two  new  furrows 
were  formed  giving  rise  to  a  three  lobed  egg  (PL  V,  Figs.  71,  72, 
73).  As  the  original  cleavage  furrow  shriveled  up,  a  peculiar  ray- 
like arrangement  of  entoplasm  was  seen.  Whether  it  was  due  to 
the  effect  of  the  incision  or  to  preparation  for  the  formation  of  two 
new  "heads"  I  could  not  determine.  At  an3^  rate  this  double 
"  headed  "  cleavage  seemed  to  be  an  exceptional  case,  and  should 
not  be  taken  as  of  constant  occurrence. 

1)     Experiment  XII    (One  case). 

Two  cuts  were  made  on  one  egg  (Fig.  75  PL  V)  and  the 
result  was  as  in  Fig.  76  (PL  V).  The  cleavage  furrow  stretched 
tOAvards  one  of  the  incisions  and  cut  through  the  egg. 


IG  Art.  3.— Naohicle  Yatsu  : 

VI.     Summary. 

Observational  part  : 

1.  The  ctenophore  egg  is  composed  of  (a)  the  outer  homo- 
geneous laj^er,  (b)  ectoplasm  and  (c)  entoplasm. 

2.  The  outer  homogeneous  layer  is  homologous  with 
Ziegler's  "  hj^aline  Aussenschicht  "  of  the  echinoderm  egg. 

o.  The  ectoplasm  is  an  alveolar  plasm  and  rays  may  be 
formed  in  it. 

4.  The  ectoplasm  is  phosphorescent. 

5.  The  sperm-rays  and  pole-rays  of  the  first  division  enter 
the  entoplasm,  the  alveolar  walls  of  the  latter  taking  a  radial 
arrangement. 

6.  Polyspermie  eggs  may  clenve  Jiomially. 

7.  Cleavage  is  not  strictly  unilateral,  the  furrow  being  formed 
in  the  micromere  region. 

8.  Fine  spinning  of  the  homogeneous  layer  can  be  seen  at 
entrance  and  at  the  bottom  of  the  cleavage  furrow. 

9.  In  the  "cleavage  head"  radiations  are  seen  in  the 
homogeneous  outer  layer  and  ectoplasm,  but  they  do  not  extend 
into  the  entoplasm. 

10.  Beneath  the  "cleavage  head"  tlie  ontoplasmic  alveoles 
are  considerablj^  compressed. 

11.  The  micromeres  consisting  almost  entirely  of  the 
ectoplasm  cleave  very  similarl}^  to  the  Avhole  egg. 

12.  In  the  beginning  of  each  cleavage,  ectoplasmic  accumula- 
tion is  seen  at  the  macromere  pole.  Towards  the  end  of  cleavage, 
an  accumulation  appears  in  the  micromere  region. 

Experimental  Part  : 

13.  Enucleated  fragments  destitute  of  the  "  cleavage  head  " 
not  manifest  any  division  activity. 

14.  The  cleavage  plane  is  not  predetermined  in  the  egg. 

15.  An  enucleated  piece  provided  with  the  "cleavage  head" 
divides  by  itself  without  the  aid  of  either  nucleus  or  centrosome. 
Nor  is  the  ray  system  necessary  for  the  cleavage  of  enucleated 
pieces. 


Observations  and  Experimonts  on  the  Ctenopborc  Ei^<^.  17 

10.  Tlie  cytuplasin  iibove  the  level  of  the  cleavage  bead 
lias  little  iniluenee  upon  the  accomplishment  of  the  division. 

17.  The  accumulation  of  ectoplasm  over  the  micro-^  and 
macromere  poles  is  formed  in  enucleated  fragments  in  the  same 
way  as  in  the  whole  egg. 

18.  If  the  removal  of  the  nucleated  portions  is  done  at  the 
beginning  of  the  division,  the  cleavage  furrow  goes  on  normally 
irrespective  of  the  angle  of  the  section.  If,  however,  the  same 
operation  is  ^jerformed  upon  an  egg  in  which  the  cleavage  has 
further  advanced,  the  division  plane  is  in  most  cases  turned  towards 
the  side  of  larger  amount  of  cytoplasm,  the  enucleated  fragment 
being  divided  into  nearly  equal  parts." 

19.  Sometimes  in  the  egg  operated  on  the  cleavage  furrow 
curls  up  towards  the  macromere  pole. 

20.  If  the  "cleavage  head"  be  split  lengthwise',  a  new 
head  "  forms  nearly  at  right  angles  and  towards  the  portion  with 
larger  amount  of  cytoplasm. 

1^1.  If  an  incision  is  made  in  the  egg  in  the  micromero 
region  the  cleavage  is  not  affected. 

VII.     General  Discussion. 

Three  views  have  been  put  forth  regarding  the  mechanism 
of  the  cytodieresis  of  the  ctenophore  egg.  Zieglee  maintains  that 
the  cleavage  is  accomplished  by  the  constriction  of  a  meridional 
ectoplasmic  thickening, Mvhicli  is  in  turn  caused  by  the  "  action  at 
a  distance  '"  of  the  centres  and  no  rays  are  necessary  for  cleavage 
('03  p.  162).' 


1  strictly  speaking-,  at  the  en  I  of  the  cleavage  furrow,  since  the  a-îcumuLition  takes  pLicj 
at  the  side  of  the  e<;g  in  case  a  new  cleavage  furrow  is  formed  at  right  angles  t j  the  old,  c.jj., 
PI.  IV,  Fig.  50. 

2  This  result  Mas  obtained  when  a  large  portion  was  out  off  from  the  og^.  Xo  experi- 
ment was  carried  out,  to  my  regret,  to  test  whether  or  no  the  removal  of  a  small  amount  of 
cytoplasm  from  an  egg  with  an  already  far  advanced  cleavage  furrow  affects  the  remaining 
course  of  the  cleavage  plane. 

3  It  is  interesting  to  recall  liow  Kowalkvskt  was  impresso  I  when  he  observed  cleaving 
ctenoi^hore  eggs:  "wie  sonderbar  es  auch  klingen  mag,  so  sc'ieint  mich  doch  diese,  S3  zug 
sagen,  todte  unbewegliche  centrale  Masse  ganz  der  mechinischen  Pressung  von  Aussen  ;  u 
und  keine  innfn  active  Kraft  zu  besitzen''  ('67  p.  3). 


18  Art.  3.-Naolii(le  Vatsn 


Based  upon  Zii;(;LKn,'s  ohservatioiis  Kiumülimi  tried  to  explain 
the  e^^todieresis  of  tlie  cteiiopliore  egg  ])V  adding  a  few  sul)8idiavy 
assumptions  to  liis  own  theory  of  eell-division  in  general  ])reviously 
put  forth  ('99),  viz.,  that  (a)  the  nuclear  fluid  is  present  along  the 
axis  of  the  egg;  (h)  at  the  expense  of  the  nuclear  fluid  the  mem- 
brane gi'ows  rapidly;  (c)  the  "  cleavage  head  "  is  a  structure  com- 
parable to  the  centrosome,  and  (d)  tlie  rays  radiating  from  tlie 
"  cleavage  head  ' '  contract  and  pull  down  the  cleavage  furrow  to 
the  micromen^  pole. 

FiSGiiEL  expresses  his  view  of  the  probable  existence  of  the 
pole-rays,  which  function  as  in  ordinary  cases  of  cytodieresis  ('08 
p.  G20  et  àry.). 

My  experimental  study  on  the  ctenophore  egg  makes  it  im- 
possible for  me  to  accept  anv  of  the  above  three  hypotheses  for  the 
following  reasons.  If,  as  Zieulkr  maintains,  the  cleavage  is  due 
to  the  contraction  of  an  elastic  ring  around  the  egg,  the  curling-up 
of  the  cleavage  furrow  towards  the  maciomere  pole  after  the 
removal  of  the  nucleated  portion  is  a  thing  not  easily  accounted 
for.  Still  more  difficult  is  it  to  apply  his  view  to  the  case  in  which 
a  new  cleavage  furrow  is  formed  at  right  angles  to  the  old.  The 
above  two  facts  are  also  against  Hitumuler's  assumption.  And 
the  fact  that  there  are  no  rays  radiating  from  the  "cleavage  head" 
into  the  entoplasm  makes  his  view  untenable.  It  is  certain  that 
the  cleavage  is  not  accomplished  bv  the  contraction  of  i)ole  rays, 
as  FiscHEL  incidentally  states,  as  is  seen  in  the  cases  in  which 
the  nucleated  part  is  removed. 

In  his  paper  on  the  development  of  Liuenjes  mrirtinus,  Conklix 
puts  forth  the  view  that  the  unilateral  cleavage  of  the  cœlenterate 
egg  in  general  is  at  least  in  part  due  to  the  structure  of  the  oöplasin 
itself,  that  is,  thin  central  entoplasm  with  a  firmer  peripheral  layer 
('08  p.  167).  This  we  have  no  reason  to  deny,  yet  how  such 
a  structure  is  favorable  to  one-sided  constriction  is  hard  to 
understand.  When  we  come  to  study  the  unilateral  cleavage  of 
the  njicromeres  of  the  ctenophore  egg,  which  are  almost  entirely 
made  up  of  ectoplasm,  it  becomes  doubtful  how  much  influence 
the  origir.al  structure  of  the  ooplasm  exerts  on  the  ]H'i-forniance  of 


Observations  and  Experiments  on  the  Ctenophoro  Eu'< 


1Ü 


^ucli  cell  division.  It  is  also  interesting  to  note,  as  I  have  done 
elsewhere,  that  unilateral  cleavage  is  seen  in  some  parthenogeneti- 
cally  developing  sea-urchin  eggs  and  also  in  lamprey  eggs,  whose 
ooplasm  is  uniform ely  laden  with  yolk  granules. 

At  present  I  am  not  in  a  position  to  construct  any  hypothesis 
to  account  for  the  cleavage   mechanism   of  the  ctenophore   egg. 
Further   detailed    biophysical   experimentation    on  the    egg    will 
undoubtedly  shed  a  new  light  on  the  problem.      As  a  working- 
hypothesis  this  much  can  be  said.      Through    the   action   of  the 
centres  (centrosomes)  sin-face  tension  is  increased  along  the  cleavage 
plane  first  at  the  animal  region  and  then  towards  the  micromere 
pole^  and  thus  the  ectoplasm  is  graduall}^  collected.     The  optical 
section  of  the  bottom  of  the  cleavage  furrow  is  the  ''  cleavage  head, 
that  is  a  passive    structure.     The  entoplasm  now  tends  to  round 
up  around  two  centres  (geometrical)  and  the  two  blastomeres  are 
formed.     My  experimental  study  seems  to    have    furnished  two 
important     data    regarding    the     above     rather    vague     general 
interpretation  of  the  cleavage  phenomenon.     Firstly,  the  cleavage 
furrow  tends  to  divide  the  egg  equally,   as  for  instance  in  the 
cases  where    a   portion   of   ooplasm   is   removed    and    thus    the 
sjanmetr^^    is   disturbed,    the    new   cleavage    furrow   being  bent 
toward  the  larger  mass  of  cytoplasm.     Secondly,   the   ectoplasm 
flows  up  and  down  just  as  Avell  without  the  nuclei  and   centres, 
as    with    them.      This  change  may  l)e   caused    b}^  the    unequal 
increase  of  surface  tension  due  to  the  internal  division  phases. 
At  any  rate   my  results  do  not  indicate  that  the  ectoplasm  alone 
is  an  active  cleaving  agent  as  Ziegler  and  Rlumbler  seem   to 
believe. 

Misald  Marine  Biological  Station 
Aucr.  n.  1910. 


1  As  has  been  pointed  out  by  Zieglïb  unilxt->ral  cleavaje  is  doubtless  in  some  way  con- 
nected with  the  eccentric  position  of  the  nuclei  and  centres.  But  it  should  be  noted  that  their 
being  in  the  ectoplaaui  is  not  ai  esi  ^ntiil  condition  of  ou»-siied  cleavag».  In  the  egg  of  a 
good  many  cœlenterates  the  nuclei  are  in  the  'n'oplas  ii  and  the  clewage  is  unilateral,  c.  r., 
L/H;'r^('s  (Conklin),  (',i')-ji(>iii<(  (Fp[.),  Hijilru  (Bkaui'^r,  T  vuneedther). 


20  Art.  3— Naohide  Yatsu 


LITERATURE. 

Andrews,  E.  A.,  ''J8. — Some  ectosai-cal  plienoimiüi   in  tlic  egg  of  Ui/ilr»  :  Johns 

Hopkins  University  Circulars  18. 
Brauer    A.,  '91. — Über  die  Entwicklung  von  H;/(b<i  :   Zeit.  wiss.  Zool.   52. 
Bunting    M.,    ö'i- — The  origin  of  the  sex-cells  in ////<^//örd/»/'f  and  ZWocv»;//»«-,  and 

the  development  of  H>jdractinia  :  Journ.   Morph,  i). 
Chun     C.      80. — Die  Ctenophoren  des  Golfes  von  Neapel  :  Fauna  uud  Flora  1. 
Chun,  C,  '92.  — Die  Dissogony  der  Rippenquallen  :  Festsch.  f.  Leuckart  1. 
Conklin,  E.   G.,  '08, — ^The  habits  and  early  dovclopmont  of   Lincniex   uit'rcuiiii.s  : 

Carnegie  Institution  Publication  103. 
Driesch,  H.  and  Morgan,  T.   T.,    "95.     Zur  Analysis  der  ersten  Enlwicklungs 

Stadien  des  Ctenophoreneies  :  Arch.  f.  Entwm.  2. 
FewkeS     J.  W.,     85. — On  the  development  of  Aijahiut  :    lUill.  Mns.  Conjp.  Zool. 

Harvard  Coll.  11. 
Fischel,  A.,  '97.— Experimentelle    I'liteisncl.ungen    an    Ctenophoronei    I     Teil: 

Arch-  f.  Entm.  ü. 
Fischel,  A.,   '98.^ — ^Experimciitelle  Untersuchungen  an  Ctcnophorcnei  Fortsetzung  : 

Arch.  f.  Entm.  7. 
Fol,   H.,   '73. — Die  erstL'  Entwicklung  der  Gerionideu-Eies  :  Jen.  Zeit.  7. 
Hargitt.,   C.  W,,    OO. — A  contribution  to  the  natural  history  and  development  of 

I'ennarin  :  Am.  Nat.  34. 
Hertwig,   0.,   '78. — Beitrage  zur  Kenntnis  der  Pilding,  Befruchtung  ui:d  Thoilung 

des  tierischen  Eies  :  Morph.  Jahrb.  4. 
Kleinenberg,  N.,  '12.  —  ]hiilia-tme  anatomisch-(ntAvi(k]uiigsgeschiclitliche  Unter. 

suchung 
Korschelt,  E.  and  Heider,    K.,    09. — Lehrbuch    der  vergleich.    Ent\Yicklungs- 

geschichte  der  wirbellosen  Tiere.  AUg.  Teil,  3  Lief. 
Kowalevsky,     A.,    'CG.— Entwicklungsgeschichte  der   Eippenquallen  :   Mein,    de 

l'acad.  des  sciences  de  St.  Petersburg,  7  série  10. 
Loeb,  J.,   '06. — Dynamics  of  living  matter. 

Maas,   0.,   '03. — Einführung  in  die  experimentelle  Entwicklungsgeschichte. 
Peters,  A.  M.,  '05. — Phosphorescence  in  ctenophores -.  Journ.  Exp.  Zcol.  2. 
Ehumbler,  L.,   '99. — Die  Furthung  des  Ctenophoreneies  nach  Ziegler  und  deren 

Mechanik  :  Arch.  f.  Entm.  8. 
Tannreuther,   J.  W.,    OS.— Development  of  iiy^/r«  :  £ioh  Bull.   14. 
Wager,  R.  E.,  '09.— lie  ccgene.^is  and  early  development  of  H»/(/;v/  :  Biol.  Bull.   18. 
Yatsu,   N.,   '10. — An  experimental  ^tudy  on  the  cleavage  of  the  clerophoie  egg  : 

Proceed.   Seventh  Intern.  Zool.  Congress. 
Zifgler,  E.   H.,    "18,— E5itrin(nte]le   Studien  über  die  Zelltheilung  :    III   Die 

FiircbungszelJen  von  Eeroe  ornta  :   Arch  f.   Entm.  G. 


Observations  and  Experiments  on  the  Ctenophore  E^g.  21 

Ziegler,  E.  H.,  '08.— Expei-imentelle  Studien  über  die  Zelltheihmg:  IV  Die 
Zelltheilnug  der  Fnrchungszellen  bei  Jieroc  und  Echiims  :   Arch.  f.  Entm.  IG. 

Ziegler,  E.  H.,  '04. — Die  erste  Eutwicklangsvorgänge  der  Echinoderraeneies 
insbesondere  die  Vorgänge  am  Zellkörper  :  Festschrift  zu  E.  Haeckel. 


N.  YATSU. 
OBSERVATIONS  AND  EXPERIMENIS  ON  THE  CTENOP.IORE  EGG. 


PLATE    I. 


PLATE    I. 

Figs.  1-8,  10  and  11  Uerue  oiahi. 
Fig.  9   l'"iicli(iris  ))iuUicor)iin. 

Figs.  1,  2  and  3.     Three  successive  stages  of  tbe  division  of  tlie  first  [lolocyto.     x270. 
Fig.    4.     Surface  view  of  three  polocytes.     Clear  spot  indicates  tlie  egg-nnclens. 

x270. 
Fig.    5-     Sperm-head  in  thi-  egg  (surface  view).     Notice  an  aster  around  the  centre 

and  a  réfringent  body  situated  a  little  apart  from  the  sperm-head.      x270. 
Fig.    6.     Side  view  of  an  ectoplasmic  acciunnlation  caused  hy  the  spermatozoon. 

X27(). 
Fi-g    7-     Tbc  same  drawn  from  a  Polyspermie  egg.      x  270. 
Fig.    8.     Trispermic  egg  (surface  view).     xGO. 
Fig.    9.     Dividing  egg  of  Kucltaiis  (optical  section),  a  stage  preceding  Figs  23  and 

24  (PL  II),  3.23  P.M.     xl40. 
Fig.  10-     Dividing  egg,  the  cleavage  having  proceeded  nearly  one  third  the  diameter. 

xCO. 
Fio-  11-     Dividing  egg,  showing  outlines  of  four  stages  ;  a-a'  10  A.M.,  h  10.8  A.M., 

e-c'  10.2.5  A.M.,  and  d-d'  10  3.5  A.M.     xGO. 


N.  Yatsu 


r-^ 


Jour.  Soi.  Coll.,  Vol.  XXXII.,  Art.  3.  PI.  I. 


N.  YATSÜ. 
OBSERVATIONS  AND  EXPERIMENTS  ON  THE  CTEN0P210RE  EQQ. 


PLATE     II 


PLATE    II. 

Figs.  12-18,  21,  25  ^nd  26  Beivë  oiaio 
FigS-  19  aiid  20  Bcroé  J'orskiilii 
Figs-  22,  27-30  CalUonira  hialaUt. 
FigS-  23  aiid  24  J'.xchaii'i  iiiiilticoruis. 

Fig-  12-     Very  young  cleavage-head,     x  270. 

Figs-  13  and  14-     Two  stages  of  the  elongation  of  the  cleavage  furrow.     x270. 

Fig-  15-     Cleavage-head    having    fused    with    the    ectoplasm    in    the     micromere 

region.     x27Ü. 
Figs-  16-18-     Last  three  stages  of  the  first  division,     x  27Ü. 
Figs-  19  ai^d  20-     T^^'o  «tages  of  the  first  division  of  the  egg  of  7>.     forskalii,  drawn 

respectively  at  10.  t  A.M.  and  10.15  A  M.     x71. 
Fig-  21-     Early    two-cell    stage    showing    the    ectoplasmic    accumulation    in    the 

luicroiuere  region.      xOO. 
Fig-  22-     Early  two-cell  stage  of  CalUan'oa.     Notice  fenestrated  cell- wall  and  thicker 

ectoplasm  at  the  micromere  region.     x390. 
Fig-  23-     Two-cell  stage  of  Knchnis,  in  which  the  cleavage  is  about  to  begin.     3.55 

P.M.     X140. 
Fig- 24-     The  same;  in  one  of  the  blastomeres  the  second  cleavage  has  been  taking 

place.     XI 40. 
Fig- 25-     Blastomere  of  the  four-cell  stage.    x71. 
F^ig-  26-     Beginning  of  the  third  division.   x60. 

Fig-  27-     Beginning  of  the  third  division  (Callianira)  ;  cf.  Fig.  22.   x390. 
Fig-  28-     Blastomere  of  the  four-cell  stage,  in  which  the  third  cleavage  lias  more 

advanced  {CdllUtnlra).   x200. 
Fig-  29-     t'pper  cell  of  the  eight-cell  stage,  giving  off  a  micromere  towards  the 

micromere  pole  {CaUianira)  x390. 
Fig  3D-     Lower  cell  of  the  eight-cell  stage  giving  off  a   micromere   horizontally 

{Callianira).    X390. 


Jour.  Sei.  Coll.,  Vol.     XXKII.,  Art.  3.  PI.  II. 


N.  YATSU. 
OBSERVATIONS  AND  EXPERIMENTS  ON  THE  CTENOPHORE  EGG. 


PLATE     III. 


PLATE    m. 


Fig. 

31. 

Fig. 

32. 

Fig. 

33. 

Fig. 

34. 

Fig- 

35- 

Fig. 

36- 

Fig- 

37. 

Fig. 

38- 

Fig. 

39. 

Fig. 

40. 

Fig.  41. 


Fig.  42- 


Fig.  43- 


Fig.  44. 
Fig.  45. 


Fig.  46. 


Beroë  ovafa.  xGO  (with  the  exception  of  Fig.  44  xl02). 

Egg  cut  horizontally  below  the  "  head."  The  enucleated  cytoplasm  cut  off 
did  not  show  any  division  activity. 

Egg  cut  below  the  "  head  "  liorizontally  nearly  two  third  of  its  diameter. 
10.15  AM. 

The  same.     10.38  AM. 

Egg  cut  in  two  horizontally  at  the  stage  when  the  "  head  "  liadcome  down 
nearly  two-tbirds  of  its  course,     10.10  AM. 

Enucleated  fragment  of  the  same  showing  ectoplasmic  thickening  in  the 
micromere  region  as  in  the  normal  case.  11;25  AM. 

The  same  showing  a  very  thin  ectoplasmic  layer  over  the  cut  surface. 
11.4-)  AM. 

Egg  cut  obliquely  just  above  tlie  "  head."     11.22  AM. 
Enucleated  fragment  of  the  same  divided  by  a  slightly  curved  cleavage 
plane.     0.10  PIM. 

Egg  cut  a  little  above  the  "  head."'     9.85  AM.  " 

Enucleated  fragment  of  the  same  showing  a  thick  ectoplasmic  accumulation 
along  the  cut  surface.  Tlie  nucleated  part  was  at  the  four-cell  stage. 
10,55  AM. 

Egg  cut  horizontally  a  little  above  the  "  head."  Notice  the  thickening  of 
the  ectoplasm  along  the  cut  surface  of  the  nucleated  pieces.  2.40  Pm. 
The  "  head  "  at  a,  2.50  PM.,  and  at  b,  3.25  PM. 

Enucleated  piece  of  the  same.  4.40  PM.  At  G  PM.  tlie  upper  bridge 
became  broader. 

Enucleated  fragment  obtained  by  a  horizontal  cut  (outline  in  dotted  line) 
10.2  AM.  ;  the  same  (in  full  line)  11.15  AM.  In  the  beginning  the  "  head  " 
came  down  at  the  rate  of  18  ;«  per  minute,  and  stopped  at  the  spot  repre- 
sented in  the  latter. 

Cleavage  farrow  and  "  head  "  of  the  same  magnified  XI 02.     11.15  AM. 
Egg   from   which  the   right-hand  nucleated  portion  has    been    cut    off 
horizontally.     Irregular  mass  attached  to  the  nucleated  fragment  is  the 
portion  that  flowed  out.     10.45  AM. 
The  same.     11.27  AM. 


N.  Yatsu 


Jour.  Sei.  Coll.,  Vol.  XXXII..  Art.  3.  PI.  III. 


Il 


f 


I 


à 


N.  YATSU. 
OBSERVATIONS  AND  EXPERIMENTS  ON  THE  CTENOPHORE  EÖQ. 


PLATE     IV 


PLATE    IV. 

lieio'é  ovata.     XGO, 
X  X  ill  sürui!  of  tlio  tigures  in  this  plate  indicate  the  bottom  of  cleavage  furrow  at 
the  time  of  operation. 

Fig-  47-     Egg  partially  cut  in  two  liorizontally,  a  little  above  the  "  head  "  Ü.45  PM. 

Fig.  48-     The  same.     2  PM. 

Fig-  49-     Ego  an   which  young  "  head  "  has  been  split  lengthwise  by  a   vertical 

incision.  9.55  AM. 
Fig.  50-     The  same  11.35  AM. 
Fig.  51-     Operation  similar  to  Fig.  49.     The  ''  head  "  had  got  halfway  when  cut. 

11.15  AM. 
Fig.  52-     The  same.  1 1.35  AM.     The  cleavage  furrow  cut  through  the  egg  at  0.5  PM. 
Fiö".  53.     Egg  with  two  cuts,  one  longitudinal  and  the  other  horizontal,  separating 

nucleated  portion  from  right-hand  prominence.     10.29  AM. 
Fig.  54.     The  same.     11.24  AM. 
Fig-  55-     Egg  from  which  a  portion  of  cytoplasm  has  been  cut  off  and  the  "  head  " 

split  lengthwise.     11.15  AM. 
Fig-  56-     The  same.  11.55  AM. 
Fig-  57-     Egg  operated  on  similarly  to  Fig.  55. 
Fig.  58-     Cleavage  furrow  has  been  formed  from  the  bottom  of  the  cut  towards  the 

left. 
Fig.  59-     Cleavage  furrow  has  cut  through  the  egg.     Nucleated  portions  of  both  the 

blastomeres  have  divided. 
Fig-  60-     Egg  in  which  the  "  head  "  has  been  split  lengthwise  and  an  incision  made 

at  the  micromere  pole.     9.42  AM. 
Fig.  61.     The  same.     10.40  AM. 
Fig.  62-     The  same.     Cleavage  furrow  has  cut  through  the  egg,  forming  an  enucleated 

mass  on  the  right. 


Jour.  Sei.  Coll.,  Vol.  XXXII.,  Art.  3.  PI.  IV. 


N.  YATSU. 
OBSERVATIONS  AND  EXPERIMENTS  ON  THE  CTENOPHORE  E(iQ. 


PLATE    V. 


PLATE    V. 

Ik'fo'é  ucata.      X  60. 

Fig.  63.  ligg  with  three  cuts,  namely,  tlic  nucleated  part  was  cut  off  from  the  left 
half,  which  is  found  attached  to  the  right  portion,  the  ''  head  "  was  spht 
lengthswise  (it  was  at  X  X  when  operated  upon)  and  a  portion  of  cytoplasm 
was  cut  off  obliquely  from  the  micromere  region.     11.10  AM. 

Fig.  64.     The  same.     0.35  PM. 

Fig.  65-     £^gg  "\^'itii  '^11  incision  at  the  micromere  region.     9  AM. 

Fig.  66.  The  same.  Ectoplasmic  thickening  is  very  conspicuous  over  a  prominence 
to  the  left  of  the  incision.     11.45  AM. 

Fig.  67-     i^gg  ^"»'ith  an  incision  on  the  right  side.     0.5  PM. 

Fig.  68.     Til«  same-     O-SO  PM. 

Fig.  69.     Egg  with  an  incision  at  the  micromere  region.     9.45  AM. 

Fig.  70-  The  same.  Cleavage  furrow  has  become  continuous  with  the  cut.  10.18 
AM. 

Fig.  71-     Egg  with  a  vertical  incision  at  the  micromere  region.     2.15  PM. 

Fig.  72-     The  same.     "  Head  "  has  become  irregular  and  a  flow  figure  is  seen. 

Fig.  73.  The  same.  Cleavage  furrow  has  fused  and  two  new  furrows  have  been 
formed.  3.11  PM.  (at  2.55  PM.  two  slight  indentations  were  formed). 

Fig.  74.     The  same.     4.18  PM. 

Fig-  75.     Egg  with  two  incisions  on  both  sides.     10.53  A^I. 

Fig.  76.     The  same.     11.28  AM. 


N.  Yatsu 


Jour.  Sei.  Coll.,  Vol.  XXXII.,  Art.  3.  PI.  V. 


i 

é 


4 


JOURNAL  OF  THE  COLLEGE  OF  SCIENCE,  TOKYO     DIPERIAL  UNIVERSITY. 

VOL.  XXXIT.,  AKTICLE,  4. 


Études  Anthropologiques. 
Les  Aborigènes  de  Formose. 


(2®  Fascicule.) 

par 

R.  Torii. 

Chargé  de  cours  d'Anthropolorjie  à  l'Université  IvipCriale  de  Tôhyô,  memlre  du  bureau  des  afain 
concernant  les  Aborigènes,  département  de  V Administration  civile.  Gouvernement  de  Formose. 


Arec  7  planches. 

I.    Caractères  physiques. 
A.     Tribu  Yami 

Cette  étude  sur  les  "Aborigènes  de  Formose"  comprend  2" 
parties  : 

1".     Caractères  physiques. 
2°.     Mensurations. 

Je  commencerai  par  décrire  les  "Caractères  physiques"  des 
Aborigènes,  en  passant  en  revue  les  9  tribus  qui  peuplent  l'île. 

Ce  premier   chapitre    est   consacré  à   l'étude    des   caractères 
physiques  de  l'une  de  ces  tribus:  les  Yami  qui  habitent  Kö-tö-shö. 

J'ai  déjà  publié  un  album  de  photographies  des  Indigènes  de 
Kö-tö-shö  ^'^  et  une  note  sur  les  coutumes  locales  de  cette  île. — ^^^' 
M.  Otto  Scheerer^^^  a  fait  paraître  une  traduction  en  allemand  de 
ces   2   travaux   dans   la   revue:       "Mitteilungen    der   deutschen 
Gesellschaft  für  Natur  und  Völkerkunde  Ostasiens." 

(1)  R.  Törii,  "  Kö-tö-shö  shashin-shü  "  Tokyo  1839. 

(2)  E.  Törii,  "  Kö-tö-shö  clözoku  hökoku  "  Tokyo  1902. 

(3)  Otto  Scheerer,  Ein  Ethnographischer  bericht  über  die  Tnsel-Botel  Tob  ige  in  "  Mitteil- 
ungen der  deutschen  Gesellschaft  für  Natur  und  Völkerkunde  Ostasiens  "—Band  XT.  1906. 


•^  Art.  4.— Torii: 

Déjà,  auparavant,  le  même  auteur  avait  publié  dans  la  même 
revue,  une  étude  sur  les  relations  d'origine  entre  les  aborigènes  de 
Lu'jon  (Philippines)  et  ceux  de  Kö-tö-shö,^'^  dans  laquelle  il  cite 
les  2  ouvrages  dont  je  viens  de  parler,  et  celui  de  M.  Davidson/^^ 
M.  Davidson  et  ]M.  Fischer^^^  ont  visité  Kö-tö-shö,  et  ont  publié  le 
Tésultat  de  leurs  investigations  sur  les  aborigènes  de  cette  île. 
M.Davidson  s'exprime  ainsi  au  sujet  de  leurs  caractères  physiques: 
'*  The  natives  are  small,  averaging  only  five  feet  two  inches  in 
height.  They  are  yellowish  brown  in  color;  and,  with  one  indi- 
vidual exception,  possess  straight  hair,  black  with  a  brownish  tint. 
While  thus  conforming  with  the  Malayan  type  in  their  straight 
hair  and  complexion,  they  appear,  so  far  as  face  character  is 
concerned,  to  be  two  types.  We  find  the  rather  small  nose  and 
non-protruding  lips  of  the  Malay;  and  again  a  type  with 
projecting  eyebrows,  deeply  sunk  orbits,  short  noses  rather 
depressed  at  the  root  and  with  large  nostrils, — in  fact  almost  a 
Negrito  nose, — together  with  the  comparatively  large  mouth  and 
thick  lips  of  the  pure  papuan  type.  One  of  the  chief  characteristics 
of  the  Papuan  is  his  frizzled  hair,  of  which  no  trace  is  found  among 
the  Botel-Tobago  natives,  save  in  the  case  of  one  male  adult  who 
possesses  hair  distinctly  curly.  So  much  for  their  physical  character- 
istics. In  their  usages  and  rites,  their  canoes  and  dwellings,  they 
possess  much  in  common  with  the  Papuan,  and  in  some  instances 
opposed   to   the  Malay " 

A  la  suite  des  travaux  ci-dessus  mentionnés,  que  j'ai  publiés 
au  retour  de  mon  voyage  à  Kö-tö-shö,  je  vais  maintenant  parler 
des  caractères  physiques  des  aborigènes  de  cette  île. 

"Kö-tö-shö"  est  une  petite  île  située  au  sud-est  de  Taï- 
Wan/*^  Les  Européens  l'appellent  "Botel-Tobago."  Elle  est 
habitée  par  des  aborigènes  qui  l'appellent  "  Yami,"  et  se  nomment 
eux-mêmes  "  Gourougourousera." — 

(1)  0.    Scheerer,  Zur  Ethnologie  der  Inselkette  zwischen  Luzon  und  Formosa  in  "  Mitteil. 
•der  deutsch.  Gesell,  für  Nat.  und  Volk.  Ost  "-Band  XI.  1906. 

(2)  Z.  W.  Davidson,    The  Island  of  Formosa..  London  and  New  York  1903.  (p.  585-6). 

(3)  A.  Fischer,  Streif züge  durch  Formosa.    Berlin  1900.  (p.  359). 

(4)  Formose. 


Études  Anthropologiques.  o 

Il  y  a  encore,  au  sud  de  Ko-tô-shô,  une  autre  petite  île  qui 
s'appelle  "  Ri -Taiwan."  Les  Européens  l'appellent  "  Little -Botel- 
To-bago."  Elle  n'est  pas  habitée. 

L'île  de  Kö-tö-shö  mesure  22  à  23  ri^^-*  environ  de  pourtour. 
L'île  étant  entièrement  formée  de  montagnes  rocheuses,  les  villages 
-des  aborigènes  sont  situés  sur  le  bord  de  la  mer. 

Les  aborigènes  de  "  Kô-tô-sho  "  nomment  leurs  villages 
^  '  nahmen.  '  '  On  n'  en  compte  que  8  à  savoir  :  Ya3^ou,  Ibatashi, 
Irarai,  Iwao,  Ibariminouk,  Ibarinou,  Imorod,  et  Iratai. 
Le  village  de  Yayou  compte  un  nombre  de  maisons  assez  élevé,  50 
environ.  Le  village  d' Imorod  en  compte  30,  et  chacun  des  villages 
d' Irarai,  d' Ibariminouk,  d'Ibarinou,  25.  Il  n'y  en  a  que  3  au 
village  d'Iwao,  et  2  au  village  d'Iwatashi. 

D'après  les  légendes  des  aborigènes  de  Kö-tö-shö,  à  une  époque 
reculée,  il  n'y  avait  dans  l'île  qu'un  homme  et  une  femme,  au 
village  d' Irarai.  Ils  eurent  une  nombreuse  descendance.  L'homme 
mourut  de  bonne  heure.  La  femme,  restée  seule,  éleva  les  enfants. 

Un  jour,  des  Indigènes  le  l'île  d'Ibatan  étant  arrivés  dans  l'île, 
s'emparèrent  de  la  femme  et  l' emmenèrent  avec  eux.  Dans  la 
suite,  les  enfants,  étant  devenus  grands,  se  marièrent  entre  eux. 
Ce  fut  l'origine  du  village  d' Irarai. 

Les  légendes  du  village  d'Ibarinou  racontent  que  des  indi- 
gènes de  l'île  d'Ibatan,  venant  du  village  d' Irarai,  où  ils  avaient 
enlevé  cette  femme,  ayant  eu  toutes  sortes  de  tribulations  sur  mer, 
ne  purent  retourner  dans  leur  pays,  et  s'arrêtèrent  à  Ibarinou. 
De  là  date  la  fondation  de  ce  village. 

Au  village  d' Imorod,  existe  la  légende  suivante:  Les  ancêtres 
des  habitants  de  ce  village  étaient  des  indigènes  de  l'île  d'Ikoubarat. 
Un  jour,  alors  qu'ils  étaient  sortis  pour  pécher,  un  grand  vent 
s'éleva  tout-à-coup.  Emportés  par  le  courant,  ils  atterrirent  à  l'en- 
droit où  se  trouve  aujourd'hui  le  village  d' Imorod. 

Ils  prirent  des  femmes  dans  le  village  voisin  d' Iratai.  De 
ces  unions  naquirent  des  enfants,  qui,  s' étant  peu  à  peu  multipliés, 
formèrent  le  village  d' Imorod. 

(1)     Le  ri  équivaut  à  3  kiloin.  927. 


Art.  4.-Torii; 


Les  légendes  que  je  viens  de  rapporter,  parlent  d'îles  nommées- 
Ibatan  et  Ikoubarat. 

Où  étaient  situées  ces  îles?  J'ai  fait  des  recherches  à  ce  sujet. 
Je  suppose  que,  comme  ces  îles  étaient  situées  au  sud  du  détroit  de- 
Bashi,  elles  doivent  avoir  quelque  rapport  avec  les  îles  Batan. 

Ibatan  serait  alors  Batan,  et  Ikoubarat  serait  l'île  Ibayat 
qui  fait  partie  du  groupe  des  îles  Batan. 


Fig.  1.     Vue  de  l'île  Botel-ïobagû. 

Il  nie  parait  donc  évident  que  les  aborigènes  de  Kö-tö-shö  ont 
eu  quelque  relation  avec  les  aborigènes  du  nord  des  Philippines. 

La  population  totale  des  S  villages  de  Kô-lo-shô,   s'élève  au 
chiffre  de  1300  habitants  environ. 

J'estimais  que  des  recherches  anthropologiques  sur  les  abori- 
gènes de  Kô-tô-shô  offriraient  un  grand  intérêt. 

L'Université  Impériale  de  Tokyo  m' 3^  envoya  dans  ce  but. 

Parti  de  Ke-Lung  le  21  octobre  189G,  j'arrivai  à  Kö-tö-shö  le 
26  du  même  mois.  Je  séjournai  d'abord  quelque  temps  sous  la  tente, 
au  village  d'Imorod;  x^^^is,  je  visitai,  ultérieurement,  tous  les  autres 
villages. 


Etudes  Anthropologiques. 


J'ai  poursuivi  ces  recherches  anthropologiques  jusqu'au  29 
•décembre,  c'est-à-dire,  pendant  70  jours. 

Il  n'y  a  à  Kô-tû-sho,  ni  Japonais,  ni  Chinois.  Aucun  navire 
ne  faisant  le  service  entre  cette  île  et  Taiwan  (Forrnose),  le  Gouver- 
neur de  Formose,  M.  le  Baron  Nogi  eut  la  bonté  d'en  mettre  im 
spécialement  à  ma  disposition  pourque  je  pusse  me  rendre  à  Kô-tô- 
shô. 

Je  lui  adresse  ici  l'expression  de  ma  respectueuse  gratitude. 


Fig.  2.     Un  village  de  l'ile  Botel-Tobago. 

Je  suis  également  respectueusement  reconnaissant  à  M.  K. 
Ouchida,  gouverneur  civil  de  Formose,  ainsi  qu'à  M.  R.  Otsu,  qui 
ont  bien  voulu  me  faciliter  l'accomplissement  de  ma  mission. 

C'est  à  l'appui  de  M.  le  Frofes.  S.  Tsuboï,  que  je  dois  d'avoir 
été  envoyé  à  Kö-tö-shö.  M.  le  Profes.  J.  Ijima  m'a  aimablement 
prêté  son  concours  pour  la  publication  de  cet  ouvrage.  M.  S.  Hashi- 
moto a  bien  voulu  également  m' aider  dans  les  calculs  des  men- 
•surations.  J'ai  enfin  reçu  l'aide  ahiiable  et  les  conseils  de  Mrs.  U. 
Mori  membre  du  bureau  des  affaires  concernant  les  aborigènes, 


b  Art.  4.— Torii: 

T.  Goto,  Y.  Noro  et  K.  Mi3'ajinui.  Enfin,  Mr.  le  docteur  Chemin  a 
bien  voulu  se  charger  de  la  traduction  en  franf^ais  de  cet  ouvrage. 
Je  leur  adresse  à  tous  mes  bien  sincères  remerciements. 

Je  dois  aussi  remercier  mon  assistant  Mr.  T.  Nakajima,   dont 
l'aide  m'a  été  précieuse  dans  l'accomplissement  de  ce  travail. 


CHAPITRE    I 


Caractères  Descriplifs. 


Couleur  de  la  Peau 

La  peau  est  fine  et  lisse.  Je  n'ai  trouvé,  au  village  cl'Imorod, 
qu'un  seul  individu  présentant  des  boutons  sur  la  figure  ;  beaucoup, 
cependant,  seraient  atteints  de  cette  particularité  pathologique. 

Les  riches  s'enduisent  la  peau  de  graisse  de  porc;  la  peau 
devient  luisante,  et  dégage  une  odeur  désagréable;  mais  ils  en 
retirent  une  certaine  considération. 

Pour  mes  recherches  sur  la  couleur  de  la  peau  chez  les  abori- 
gènes de  Kô-to-shô,  je  me  suis  servi  de  la  table  qui  se  trouve  dans 
l'ouvrage  publié  par  la  Société  Anglaise   d'Anthropologie.^'^ 

Broca  donne  également,  dans  son  ouvrage,^"-*  une  table  des 
différentes  couleurs  de  la  peau:  (couleurs  de  la  peau  et  du  sj^stème 
pileux.)  Mais  le  nombre  des  couleurs  donné  dans  le  manuel  de  la 
Société  Anglaise  d'Anthropologie  est  moins  élevé  que  celui  que 
donne  Broca.  Il  n'y  en  a  que  10.  Le  n°  1,  la  couleur  la  plus 
fon(^ée  est  le  "  coal-black  (noir  de  charbon)  "  ;  la  plus  claire,  le  n° 
10  est    "florid,    or  rosy  (couleur  fleurie  ou  rosée)."  Au  milieu  se 

trouvent  le  "  yellow  (jaune)  "  ;  le  "  brown   (brun)  "   etc Cette 

table  est  très  commode  pour  l'explorateur. 

(1)  Notes  aad  queries  on.  Anthropology.    London  1892.     (p.  16;  pL  III.) 

(2)  P.  Broca,  Instructions  anthropologiques  générales.     Paris  1879. 


Études  Anthropologiques.  7 

J'ai  observé  la  couleur  de  la  peau  sur  les  parties  suivantes: 
front,  dos  et  paume  de  la  main. 

La  couleur  de  la  peau  est  la  môme  chez  les  hommes  et  chez 
les  femmes. 

Couleur  du  front. 
La  couleur  du  front  est  le  '' brown  (brun)  "  ;    elle  n'est  pas 
tout-à-fait  aussi  foncée   que  la  couleur  n°  5  de  la  table   '  '  copper 
coloured  (cuivrée)."  Elle  me  parait  être  un  mélange  des  couleurs 
n°  9  et  n°  7  (n°  30  de  Broca). 

Couleur  de  la  Paume  de  la  Main. 

Cette  couleur  diffère  de  celles  du  front  et  du  dos  de  la  main. 
C'est  le  n°  10  de  la  table,  c.  à.  d.  "  florid  (fleurie) ''  ou  "rosy 
(rosée)." 

J'ai  fait  porter  mes  recherches  à  ce  sujet  sur  des  individus  de 
20  à  50  ans,  hommes,  femmes,  vieillards,  enfants,  indifféremment. 

Ils  ont  tous  la  même  couleur  de  peau. 

Sur  les  reins,  elle  ne  diffère  pas  lorscj^u'ils  sont  nus,  môrae  à 
l'endroit  où  ils  portent  d'habitude  le  "  gigat."*^^^  V.  la  fig.  A  de 
la  pi.  XIV  de  l'Introduction. 

J'ai,  cependant,  observé,  au  village  d'Ibarinou,  un  sujet, 
nommé  Shaman-Barrou  dont  la  peau  présentait  la  couleur  n°  9 
"pale-white  (blance  pâle),"  au  niveau  du  grand-trochanter,  à 
l'endroit  où  la  peau  était  cachée  par  son  "  gigat."  J'ai  aussi 
observé,  ultérieurement,  le  même  fait  sur  2  autres  sujets. 

Chez  les  enfants,  la  couleur  du  front  et  du  dos  de  la  main  est 
un  peu  plus  claire  que  celle  des  adultes  qui  est  le  "  yellow-brown 
(jaune  brun)."     Elle  un  peu  plus  foncée  que  le  n°  10  de  la  table. 

Couleur  des  Yeux. 

En  prenant  comme  base,  pour  observer  la  couleur  des  yeux, 
"l'Échelle  chromatique  des  yeux  "    de  Broca, -'"^  j'ai  constaté  que 

(1)  Sorte  de  pagne  de  3  mètres  de  longueur  et  de  15  centimètres  de  largeur,  en  toile  de 
chanvre,  que  les  indigènes  portent  enroulé  autour  de  la  ceinture  à  partir  de  l'âge 'de  7  à  8  ans. 

(2)  Loc.  cit. 


s 


Art.  4— Torii 


la    couleur  des  yeux  des  aborigènes  de  Kö-tö-shö,  répond   aux 
numéros  1,2  et  3  de  la  série  "  brun-noir,"  de  cette  échelle. 

Cheveux. 

Les  cheveux  des  indigènes  de  Kö-tö-shö  sont  absolument 
droits.  Leur  couleur  est  noire,  un  peu  mélangée  cependant  de 
brun.     Ils  sont  fins  et  souples,  et  sont  très  fournis. 

Les  enfants  ont  les  cheveux  bruns  au  début;  mais,  à  mesure 
qu'ils  grandissent,  la  couleur  de  leurs  cheveux  se  rapproche  peu  à 
peu  de  celle  des  cheveux  des  adultes. 

Voici  une  section  des  cheveux,  examinée  au  microscope: 
<fig.  3). 


/^/^ 


7.S/Ù 


(C) 


B 


D.  D  I  2.  zeiss 


//Jl 


Fis.  3 


A.     Section  à  la  racine.        B.     Section  à  la  partie  moyenne. 

C.     Section  à  l'extrémité. 

Comme  on  le  voit  par  ces  figures,  la  section  à  la  racine  est 
légèrement  triangulaire;  elle  est  ronde  à  la  partie  moyenne  et 
absolument  circulaire  à  l'extrémité. 

Dans  le  nombre,  il  y  a  des  cheveux  de  forme  très  irrégulière  ; 
mais  ils  sont  rares. 

Au  milieu  de  la  section,  il  y  a  un  canal. 

J'ai  trouvé  au  village  d'Imorod,  un  sujet  du  nom  de  Shikashi, 
•dont  les  cheveux  étaient  frisés.     Il  est  représenté  en  B  dans  la  fig.  4. 


Études  Anthropologiques.  9 

Les  indigènes  actuels  de  Ko-tö-shö  ont  le  type  que  Mr. 
Deniker  appelle  "  indonésien  "^'^  Je  n'ai  jamais  trouvé  le  type 
''Negrito."  Tous  les  autres  indigènes  de  Kö-tö-shö  que  j'ai  exa- 
minés, ayant  les  cheveux  droits,  je  pense  que  ce  cas  de  cheveux 


frisés  chez  le  nommé  Shikashi  est  unique  (B),  et  je  suis  porté  à  le 
considérer  comme  un  phénomène  de  retour,  les  ancêtres  des  indi- 
gènes actuels  s'étant  métissés  autrefois,  ailleurs,  avec  le  type  negrito, 
avant  leur  arrivée  à  Kö-tö-shö. 

Le  type  A  de  la  fig.  4,  qui  est  extrait  de  l'album  du  Dr. 
]\jyei.(2)  ^p2^  XII)  est  un  indigène  venu  des  Philippines,  habitant 
Calayan,  dans  le  nord  de  Lu(;on,  et  issu  d'un  métissage  entre  malais 
et  negrito.  En  les  comparant  ensemble,  on  peut  constater  qu'ils 
se  ressemblent. 

Barbe. 

Il  ressort  des  recherches  que  j'ai  pratiquées  sur  79  indigènes 
de  Kö-tö-shö,  que  leur  barbe  présente  3  variétés. 

(1)  J.  Deniker,  Les  races  et  les  peuples  de  la  terre.    Paris  1900. 

(2)  A.  B.  Meyer,  Album  von  Philippinen  typen.  Dresden  1835  (PI.  XII). 


10 


Art.  4.— Torii 


Type  a)    Sujets  ayant  de  la  barbe  sur  quatre  parties  du  visage  : 
des  moustaches,  la  mouche,  de  la  barbe  au  menton  et  sur  les  joues. 

Cette  barbe  est  extrêmement  rude,  et  peu  abondante. 

Sur  79  sujets  observés,  2  à  peine,  aj^pelés  l'un  Manigad,  l'autre 
Shimagon,  tous  deux  du  village  d'Ibarinou  appartenaient  à  ce  type. 


Type  />).  Sujets  n'ayant  de  barbe  que  sur  3  parties  du  visage: 
moustaches,  mouche  et  barbe  au  menton.  C'est  le  type  le  plus 
fréquent:  73  sujets  sur  71)  le  présentaient. 

Type  c)  Sujets  n'aj^ant  pas  du  tout  de  barbe  sur  les  quatre 
parties  du  visage.  Je  n'ai  trouvé  que  4  sujets  de  ce  type  sur  les  79 
que  j'ai  observés:  2  étaient  originaires  du  village  d'Ibarinou;  ils 
s'appellent  Shaman-Shagerrou  et  Kourongo;  le  troisième  Shaman- 
Shabougai,  est  du  A^illage  d'Imorod;  je  n'ai  pu  savoir  le  nom  du 
quatrième,  originaire  du  village  d'Iratai. 

Poils  du  Corps. 

On  peut  établir  dans  chaque  race  4  divisions  relatives  au 
degré  d'abondance  des  poils:  absents,  rares,  moyennement  abon- 
dants, abondants. 

Chez  les  indigènes  de  Kö-tö-shö,  la  variété  ''rares"  est  fré- 
quente ;  mais  parmi  ceux  qui  composent  cette  variété,  il  y  en  a 
chez  qui  les  poils  sont  presque  "  absents,''  de  même  que  d'autres 
sont  à  la  limite  de  la  variété  "  moyennement  abondants."  Voici 
le  résultat  de  mes  recherches  relatives  à  l'abondance  du  système 
pileux  sur  120  sujets. 

«)     Sujets  aj^ant  des  poils  fins  aux  jambes.      Les  sujets  de 


Études  Anthropologiques.  it 

cette  variété  sont  nombreux  à  Kô-tô-shô;  j'ai,  cependant,  observé 
que  les  poils  du  membre  inférieur  poussaient  très  rudes  chez  un 
individu  du  nom  de  Shaman-Manigad,  au  village  d'Ibarinou. 

h)  Sujets  ayant  des  poils  fins  aux  avant-bras  et  aux  jambes. 
Ils  sont  également  nombreux. 

c)  Sujets  ayant  des  poils  fins  à  la  fois  aux  épaules,  aux  avant- 
bras  et  aux  jambes.  Dans  cette  variété,  les  poils  sont  plus  rares 
que  dans  les  variétés  précédentes.  Je  n'ai  trouvé  qu'un  seul  sujet, 
du  nom  de  Shaman-Barrou,  du  village  d'Ibarinou,  appartenant  à 
ce  groupe. 

d)  Sujets  qui  n'ont  pas  de  poils  du  tout  sur  le  corps.  2  in- 
dividus seulement,  Shaman-Shagai,  et  Shaman-Kwakou,  tous  deux 
du  village  d'Imorod,  appartenaient  à  cette  variété. 

t')  Sujets  dont  presque  tous  les  poils  du  corps  sont  rudes. 
Je  n'ai  trouvé  qu'un  seul  sujet  de  cette  variété.  Cet  individu, 
nommé  Shap-Makarrou,  du  village  d'Iratai,  était  âgé  de  50  ans  en- 
viron. Les  poils  étaient  rudes  partout  où  ils  poussaient,  sur  les 
épaules,  sur  le  dos,  les  bras,  le  ventre,  les  jambes.  Ces  poils,  assez 
abondants,  avaient  la  longueur  suivante:  sur  les  bras,  26"™';  sur 
les  épaules,  24'^"'-;  aux  jambes,  26.™"'.  On  peut  le  ranger  dans, 
la  variété  '*  abondants." 

Sourcils. 

La  couleur  des  sourcils  est  noire  ;  ils  sont  généralement  épais, 
la  queue  étant  plus  fournie.  Chez  4  sujets  sur  130  examinés,  des 
poils  fins  poussaient  entre  les  sourcils.  L'un  d'entre  eux  était  du 
village  d'Ibarinou;  je  n'ai  pu  avoir  son  nom;  un  autre,  Shap- 
Makarrou,  était  du  village  d'Iratai;   enfin, 


les  deux  autres  étaient  de  jeunes  garçons 
'"■    ■  de  15  à    16    ans,    l'un    Sekoa,    du    village 

d'Imorod,  l'autre  Sheramaza,  du  village  d'Iratai. 

Si  l'on  regarde  sans  trop  d'attention  ces  4  sujets,  leurs  sour- 
cils semblent  se  rejoindre,  en  raison  des  poils  fins  qui  poussent  au 
milieu  (fig.  6). 


12 


Art.  4— Torii  : 


Insertiois  des  Cheveux  sur  le  Front. 

Ayant  examiné  le  mode  d'implantation  des  cheveux  sur  le 
front  sur  80  indigènes,  j'ai  trouvé  les  3  formes  suivantes  (fig.  7). 

a)     Insertion     à 
-courbe  très  prononcée. 

h)     Insertion     à 
courbe  moyenne.  CO  ^  0 

c)    Insertion      à  ^.    ^ 

•courbe  peu  prononcée. 

Sur  les  80  sujets  examinés,  48  présentaient  la  forme  a),  14  la 
forme  ?>),  10  la  forme  ^).  Le  forme  ci)  semble  donc  être  la  plus  fré- 
quente. 

Face. 

La  face  est  plate.  Elle  est  déprimée  à  sa  partie  médiane.  Sa 
forme  est,  le  plus  souvent  ronde. 

Le  front  est  un  peu  incliné  en  arrière  et  bombé. 

L'os  malaire  est,  le  plus  souvent  proéminent;  l'angle  du 
maxillaire  inférieur  fait  saillie.  Il  en  résulte  que  la  forme  de  la 
face  semble  carrée.  ^=>':^^ 


Fis.  8 


Voici  quelques  spécimens  de  profils  de  la  face  chez  les  in- 
digènes de  Kö-tö-shö  (fig.  8). 


Études  Anthropologiques. 


13" 


On    peut  les   comparer  ici,   avec   les   profils  de  Japonais  et 
d'Européens  qui  sont   reproduits  dans  la  fig.  8/'^^-^ 

1,  2,  3.    Sont  des  profils   de 
Japonais. 

4-5.     d'Européens. 
\  Les  lèvres  des  aborigènes  de- 

J     Kô-tô-shô  sont  épaisses. 
-'  La  bouche  est  large. 


Yeux. 

Les  yeux  ont  des  dimensions 
moyennes.  Leur  direction  est 
toujours  horizontale. 

Ayant  examiné  attentivement 
la  forme  des  yeux  sur  98  indivi- 
dus, j'ai  constaté  que  des  2  formes 
d'œil  que   donne  M.  Topinard/-^ 
presque   tous   les    sujets   avaient 
l'œil  '  '  européen. '  '    Je  n' ai  trouvé 
que  chez  un  seul  individu  nommé 
Sheriton,   du  village  d'Ibarinou, 
l'œil  '  'oblique,  '  '  caractéristique  de 
la  race  mongole, 
le  globe  de  l'œil  s'y  enfonce  profondé- 


Fisr.  9. 


L' orbite  est  très  excavé 


ment. 


La  couleur  des  yeux  est,  comme  je  l'ai  déjà  dit  plus  haut, 
brun-noir." 

La  paupière  supérieure  présente  deux  rephs. 


Nez. 

^Dans  mes  recherches  sur  les  profils  des  nez  chez  les  aborigènes 
de  Kô-tô-shô,  j'ai  trouvé  3  types.     Dans  ces  3  types,   le  sillon  qui 

(1)  Tiré  de  J.  Eanke,  Der  Mensch  II.  Leipzig  189.  (p.  278). 

(2)  P.  Topinard,  Eléments  d'anthropologie  générale.     Paris  1885  (-p.  909). 


14 


Art.  4.— Torii: 


Fier.  10. 


sépare  les  ailes  du  nez  des  joues,  est  profondément  marqué,  l'ouver- 
ture des  narines  est  large. 

Voici  un  exposé  détaillé  des  caractéristiques  de  chacun  de  ces 
types  (fig.  10). 

a)  La  forme  de  ce  nez  est  un 
mélange  du  n°  1  et  du  n°  7  de  la 
table  des  "Types  de  nez;  profils" 
de  Topin ard. 

18  individus  sur  85  présentaient 

ce  type.     J'ai  dessiné  la  forme  de 

ce  nez  d'après  un  indigène  du  village 

d'Ibarinou,  nommé  Shaman- Ayou. 

h)     19  sujets  sur  85.    La  forme 

■de  ce  nez  a  été  prise  sur  un  nommé  Shapoun-Magaro,  du  village 

d'Imorod. 

c)  Au  milieu  du  dos  du  nez,  il  existe  une  dépression;  les 
ailes  sont  larges;  l'ouverture  des  narines  est  grande.  48  sujets 
présentaient  ce  type.  Chez  l'un  deux,  nommé  Shap-Magarrou,  du 
village  d'L-atai,  existe  une  profonde  dépression;  à  proprement 
j)arler,  il  semble  qu'il  n'y  ait  pas  de  dos  du  nez;  les  ailes  seules 
ressortent.  Ce  type  extrême  de  nez  me  fait  penser  au  type  negrito, 
dont  il  se  rapproche  beaucoup.  J'ai  dessiné  ce  type  de  nez 
d'après  un  nommé  Kateshana-Yappanko,  du  village  d'Imorod. 

De  ces  3  types  de  nez,  le  type  c  est  donc  le  plus  fréquent. 
Les  profils  ont  été  pris  dans  les  villages  d'Imorod,  d'Ibarinou  et 
-d'Iratai.  Le  t^^pe  a  est  fréquent  au  village  d'Ibarinou;  j'ai  vu 
souvent   le  type  c  au  village  d'Imorod. 

La  forme  du  nez  est  caractéristique  d' une  race.  C'est  vrai  pour 
les  aborigènes  d'Australie,  les  Papous,  les  Boshimen,  les  Hotten — 
tots,  demême  que  pour  les  Européens  et  les  Juifs. 

Les  formes  extremes  des  types  a  et  c  devaient  exister  avant 
l'arrivée  des  aborigènes  dans  l'île;  la  forme  />  est  probablement 
le  résultat  du  mélange  des  types  a  et  c  après  l'arrivée  dans  l'île. 

J'ai   trouvé    dans  l'album  des  photographies   des  types   des 


Études  Anthropologiques.  15 

Philippines  du  Dr.  Myer/'^  des  formes  de  nez  semblables  au  type  c 
<îes  aborigènes-de  Kö-tö-shö  chez  les  Igorrotes  et  les  Tinguianen. 
Il  y  a  donc  probablement  une  relation  d'origine  entre  les  deux. 

Dents. 

Les  aborigènes  de  Kö-tö-shö  ont  généralement  de  bonnes 
dents  ;  elles  sont  réguHèrement  plantées.  Chez  certains  d'entre  eux, 
le  bord  de  l'arcade  dentaire  inférieure  est  horizontal.  Sur  73 
sujets  de  20  à  50  ans  que  j'ai  examinés,  je  n'en  ai  vu  qu'un,  nommé 
Shaman-Magaro,  à  qui  manquaient  les  dents  molaires. 

Oreille. 

Tantôt   le   bord   du   pavillon   de  l'oreille  est  droit,    tantôt  il 
présente   une  échancrure;  de   même,   chez  les  uns,   le  lobule   est 
nettement  séparé  de  la  tête,  tandis  que  chez  d'autres  il  y  est  rattaché. 
j  Dans  le   dessin    I   de  la  fîg.    11,  ci- 

dessus,  en  a  le  bord  du  pavillon  est  très 
droit,  le  lobule  nettement  séparé  de  la 
paroi  du  crâne. 

En  h  le  lobule  est  également  séparé 
de  la  paroi  du  crâne;    mais    le  bord  du 
d  -/       pavillon  présente  une  échancrure. 

jy  En  II,    le  bord  du  pavillon  de  c  est 

droit;  mais  le  lobule  est  rattaché  à  la  paroi 
du  crâne;  en  cl  le  pavillon  est  droit,  le 
lobule  adhérent. 

Sur  73  sujets  examinés  46  présentaient 
le  type  a,  17  le  type  6,  14  le  type  c;  enfin 
3  seulement  répondaient  au  type  d. 

Les  femmes  percent   dans  le  lobule 

Fig.  11.  . 

de  l'oreille  un  petit  trou,  dans  lequel  elles 
introduisent  un  fil  auquel  est  attaché  un  pendant  d'oreilles  que 
l'on   appelle  dans  le  pays  "oubai."^-^ 

(1)  Meyer,  Album  von  Philippinen-typen  (PI.  XIV  et  XXIV). 

(2)  Coquillage  taillé  en  forme  de  8  provenaut  du  Xautilus  PompUlua  L. 


16  Art.  4.— Toni: 

Les  hommes  ne  portent  pas  de  pendants  d'oreille,  mais  ils 
présentent  encore  une  petite  cicatrice  en  trou  d'aiguille  dans  le 
lobule. 

Voici  le  résultat  de  mes  recherches  sur  115  sujets: 

a)  Sujets  chez  lesquels  le  petit  trou  du  lobule  est  encore 
ouvert. 

h)     Sujets  chez  lesquels  persiste  une  cicatrice  des  deux  côtés. 

c)  Sujets  chez  lesquels  cette  cicatrice  ne  persiste  que  d' un  coté. 

d)  Sujets  ne  portant  plus  aucune  trace  du  trou. 

Un  seul  sujet  nommé  Sliaman-Shabougai,  du  village  d'Imorod 
appartenait  à  la  catégorie  a.  Parmi  99  sujets  venant  se  ranger  dans 
la  catégorie  h,  un  d'entre  eux,  nommé  Kourougo,  du  village 
d'Imorod,  avait  le  lobule  fendu.  Enfin,  il  y  avait  un  seul  sujet 
dans  la  catégorie  c,  et  5  dans  la  catégorie  d.  Les  sujets  présentant 
encore  des  vestiges  du  trou  dans  le  lobule,  sont  donc  les  plus 
nombreux.  (99  sur  115). 

D'après  les  indigènes  de  Kô-tô-shô,  ceux  des  îles  Ibatan 
(Batan)  et  Ikoubarat  (Ibayat)  se  perdent  un  grand  trou  dans  le 
lobule  de  l'oreille,  et  y  introduisent  des  morceaux  de  bambou  ou 
de  bois.  En  outre,  les  aborigènes  de  ces  îles  s'appelleraient  du 
nom  de  "  gourougourousera '''^  de  même  que  ceux  de  Kô-tû-shû. 
On  peut  donc  en  déduire  que  les  aborigènes  de  Kô-tô-shô  se  per- 
çaient également  autrefois  de  grands  tious  dans  le  lobule  de  l' oreille. 

Dimensions  de  l'index  et  de  l'annulaire. 

Sur  95  indigènes  chez  lesquels  j'ai  recherché  les  dimensions  de 
l'index  et  de  Tannulaire,  84  avaient  l'index  plus  long;  chez  8 
autres  ces  deux  doigts  étaient  d'égale  longueur.  Chez  3  seule- 
ment, l'annulah-e  était  plus  long.  L'index  est  donc  généralement 
plus  long  que  l'annulaire. 

Forme  de  l'Ongle. 
Les  3  formes  suivantes  existent: 

(1)     Hommes  qui  ont  les  cheveux  taillés  en  rond,  (de  "  gourou,"  cercle  et  "  sera,"  homme.) 


Etudes  Anthropologiques.  1  < 

Sur  70    sujets,    13  présentaient  la  forme  a,  48  la  forme  h,   9 

I 1      I      I    seulement  la  forme  c  ;   la  forme  h,  est  done 

J       celle  que  l'on  trouve  le  plus  fréquemment. 


a  4 

Fie:.  12. 


Courbe  du  bras. 


Voici  le  résultat  de  mes  recherches  sur  le  degré  de  courbure- 

du  bras  chez       i 

97  sujets: 

^  Tig.  13.  -^ 

En  faisant  tendre  le  bras  autant  que  la  force  le  permet,  94 
sujets  sur  les  97  présentaient  la  forme  en  a),  3  seulement  la  forme 
en  V).  Le  bras  ne  présente  donc  pas  de  courbure;  il  est  le  plus- 
souvent  horizontal. 

Degré  n' écartement  des  membres  inférieurs. 

Ayant  fait  placer  79  sujets  dans  la  position  debout  pour  recher- 
cher le  degré  d'écartement  des  membres  inférieurs,  j'ai  constaté 
que  74  les  avaient  écartés,  alors  que  5  seulement  les  avaient  réunis. 
L'écartement  est  donc  la  généralité. 

Dimensions  des  1™  et  2^^^^^  orteils. 

Le  deuxième  orteil  est  généralement  plus  long  que  le  premier, 
ainsi  qu'il  résulte  de  ce  tableau  établi  d'après  des  recherches  prati- 
quées sur  82  sujets. 

a)     I>II     8  sujets. 

h)     I  =  II     23      ,, 

c).   I  <:II     51      ,, 

Chez  un  sujet,  en  outre,  la  longueur  des  T''  et  2"'°  orteils  du 
pied  droit  et  du  pied  gauche  différaient:  d'un  côté  le  V  était  plus 
long  que  le  2*^,  tandis  que  c'était  l'inverse  de  l'autre  côté. 

Mes  recherches  ont  surtout  porté  sur  les  hommes;  je  n'ai  pas 
étudié  les  femmes. 

Les  hommes  que  j'ai  examinés  sont  surtout  des  adultes;  j'ai 
laissé  de  côté  les  enfants  et  les  vieillards. 


18 


Art.  4. — Torii  : 


Voici  la  liste  des  sujets  sur  lesquels  j'ai  pratique'  des  mensura- 
tions ■•  j 

(*  Ce  siçfne  indique  les  enfants  au  dessous  de  12  :vns). 


Nuint'ros 

Xoms  des  Villages 

Xoms  des  Individus. 

1 

Imorod 

Kateshana  Yappanko 

2. 

>> 

Shapou  Magaro 

3 

>> 

Sbennikotan 

4 

Ibarinou 

Jaraboi 

5 

Ibariminoiik 

Sheriton 

6 

Ibarinou 

Shenemararop 

7 

? 

Sbipanigaman 

8 

Ibariminouk 

Sbaman  Garap 

9 

>> 

Sbamau  Jagaraou 

10 

>» 

Banaibinko 

11 

Ibarinou 

Sliaman  Manigad 

12 

>> 

Sbigatok 

13 

Imorod 

Sbaman  Norain 

14* 

Ibarinou 

Sbijariga 

15 

Iratai 

Sbap  Makarou 

16* 

Ibarinou 

Sbitokourin 

17 

Imorod 

Sbagerrou 

18 

Ibarinou 

Sbaman  Noyau 

19 

9 

Sbaman  Ranrounkô 

20 

Iratai 

Sbegawosbi 

21 

Ibarinou 

Panaman 

22* 

)> 

Sbijempou 

23 

Imorod 

GaraiD 

24 

Ibarinou 

Sbaman  Barrau 

25 

Imorod 

Sbaman  Sbabougai 

26 

jj 

•? 

f'> 


Études  Anthropologiques. 


19 


Numéros 

Noms  des  Villages 

Noms  des  individus. 

27 

Iratai 

Shaman  Kouyamako 

28 

Imorod 

Shaman  Kwako 

29* 

Ibariminouk 

Shiramaya 

30 

Ibarinou 

Shaman  Panama 

31 

j> 

Shimagou 

32 

jj 

Shaman  Jokourin 

33 

jj 

Shaman  Jonanko 

34 

Imorod 

Amanikotan 

35 

5> 

Shaman  Maroukanko 

36 

9 

Shaman  Harongan 

37 

Ibarinou 

Shaman  Jüaji 

38 

5> 

? 

39 

Imorod 

? 

40 

5) 

Shaman  Eoumai 

41 

Ibarinou 

Pountan  Shagai 

42 

Imorod 

Shaman  Kiprin 

43 

j> 

Shap  kara  Watouko 

44 

9 

? 

45 

Imorod 

9 

46 

)> 

Chikojiratai 

47 

Iratai 

Shaman  Joumanoud 

48 

Imorod 

Shaman  Eoup 

49 

Ibarinou 

9 

50 

Ibariminouk 

Shaman  Mararou 

51 

Iratai 

Shaman  Kwako 

52* 

Imorod 

Shekoa 

53* 

JJ 

Chichiagod 

54 

>> 

Shap  Morainbinko 

55 

JJ 

Shaman  Panigoun 

56 

Ibarinou 

Shaman  Chinapi 

57 

JJ 

Shaman  Kotouko 

20 

Art.  4.— Toni: 

Xuméros 

Noms  des  Villages 

Noms  des  individus. 

58 

Imorod 

Shaman  Karawako 

59 

5> 

Shaman  Shadaji 

60 

Ibarinou 

Kakoutoükü 

61 

Iratai 

Shaman  Kwakù 

62 

Imorod 

Shaman  Garouwash 

63 

Iratai 

Shaman  Chinapi 

64 

Ibarinou 

Shap  Megatok  Shoun. 

65 

Imorod 

Shaman  Matapo 

66 

Iratai 

Kagon 

67 

Imorod 

Set'machiang 

68 

Ibarinou 

Shaman  Masori 

69 

)> 

Sham  Norain 

70 

j> 

Shaman  Joubout 

71 

Yayou 

Shépoposou 

72 

Ibarinou 

Shiraton 

73 

9 

Shaman  Shagerroui 

74 

Ibarinou 

Kaurougo. 

75 

>» 

Shaman  Patö 

76 

j» 

Shaman  Magato 

77 

Yayou 

Matounara 

78 

5> 

Shaman   Jagagi 

79 

Iratai 

Shaman  Gararop- 

80 

Imorod 

Shenegapouri 

81* 

? 

Shenaoijei 

82* 

Imorod 

Shijabat 

83* 

Iratai 

Sherakounko 

84 

Ibarinou 

Shaman  Jombou 

85 

Iratai 

Shaman  Japourrou 

86 

>> 

? 

87 

Iratai 

Maroshi 

88 

j> 

? 

Études  Anthropologiques. 


21 


CHAPITRE    II. 
Mensurations. 

Mensurations  de  la  tête  et  de  la  face. 


I.     Diamètre  antéro-postérieur  maximum. 

Mensurations  pratiquées  sur  48  sujets. 

Le  chiffre  moyen  est  de  178  millimètres  ;  les  dimensions  ex- 
trêmes sont  de  169  millim.  et  de  189  millim. 
Voici  le  tableau  de  ces  mensurations. 


Xuméros  d'ordre 

Diam.  ant.  jjost.  maxim. 

Xuméros  d'ordre 

Diam.  ant.  i)ost.  maxim. 

1 

176  '"'"• 

34 

178  '"'^• 

2 

184 

35 

184 

3 

180 

36 

174 

4 

175 

39 

177 

6 

170 

40 

180 

6 

172 

41 

182 

7 

182 

42 

182 

9 

169 

43 

183 

11 

172 

47 

188 

12 

180 

48 

174 

13 

174 

50 

177 

15 

176 

51 

180 

17 

174 

55 

185 

18 

175 

57 

172 

21 

177 

58 

171 

23 

179 

59 

184 

24 

184 

60 

184 

25 

172 

61 

179 

27 

182 

62 

179 

28 

176 

63 

170 

30 

175 

64 

176 

31 

178 

65 

172 

32 

176 

66 

188 

33 

175 

74 

189 

22 


Art.  4— Toiii  : 


Diamètre  transversal  maximum. 

Mensurations  pratiquées  sur  44  sujets. 

La   moyenne  est   de   làl'^'^^o;    les  dimensions  extrêmes  de 
130™"-,0  et  de  153™"-,0. 

Voici  le  tableau  de  ces  mensurations. 


Numéros  cVordre 

Diarii.  transv.  maxim. 

Numéros  cVordre 

Diam.  transv.  maxim.. 

1 

130  ™™- 

35 

153  "^"^■ 

2 

137 

36 

140 

3 

137 

39 

146 

4 

139 

40 

140 

5 

144 

41 

145 

G 

137 

42 

138 

7 

139 

! 

43 

140 

9 

1149 

47 

138 

11 

jl40 

\         48 

138 

12 

'l53 

55 

138 

13 

150 

1 

57 

138 

15 

1130 

58 

132 

17 

il45 

1 

59 

143 

18 

|l41 

60 

145 

21 

!l39 

1 

61 

139 

25 

!l51 

62 

189 

27 

161 

63 

130 

28 

il40 

64 

141 

30 

145 

65 

139 

31 

145 

66 

144 

32 

139 

74 

143 

33 

141 

— 

— 

34 

140 

— 

— 

Etudes  Anthropologiques. 


23 


Longueur  totale  du  visage. 

(Du  point  mentonnier  à  la  racine  des  cheveux). 

Sur  une  série  de  45  individus,   la  longueur  moyenne  a  été  de 
178"™-,  4;  les  variations  extremes  de  16r"%0  et  de  198"™-, 0. 


Numéroa  d'ordre 

Long.  tot.  du  visage 

Numéros  d'ordre 

Long.  tot.  du  visage 

1 

183  "^°^- 

36 

172  "^• 

2 

182 

39 

176 

3 

185 

40 

198 

4 

164 

41 

184 

5 

161 

42 

191 

6 

171 

43 

182 

8 

191 

47 

177 

11 

171 

48 

180 

12 

173 

50 

183 

15 

182 

51 

176 

17 

176 

55 

181 

18 

172 

58 

174 

21 

184 

59 

186 

23 

170 

60 

189 

24 

181 

61 

171 

25 

164 

62 

174 

27 

189 

63 

175 

28 

179 

64 

176 

31 

165 

65 

186 

32 

172 

66 

191 

33 

177 

67 

177 

34 

167 

74 

187 

35 

181 

— 

— 

^4 


Art.  4.— Torii  : 


Longueur  totale  de  la  face. 

(du  point  sns-nag;il  au  point  mentonnior.) 

Sur  une  série  de  47  individus,  le  chiffre  moyen  a  été  de  10G'"'"',0 
les  variations  extremes,  de  94"""-, 0  et  de  116'""\,0. 


Numéros  cVordre 

Long.  tot.  de  la  face 

Numéros  dordre 

Long.  tot.  de  la  face 

1 

108  '"™- 

36 

103  '"'" 

2 

107 

39 

107 

3 

103 

40 

ni 

4 

101 

41 

112 

5 

101 

42 

111 

6 

111 

43 

113 

8 

100 

47 

112 

11 

104 

48 

105 

12 

104 

50 

107 

15 

107 

51 

104 

17 

104 

55 

106 

18 

101 

57 

107 

21 

103 

58 

101 

23 

105 

59 

108 

24 

106 

60 

111 

25 

95 

61 

107 

27 

109 

62 

103 

28 

106 

63 

105 

30 

110 

64 

94 

31 

103 

65 

110 

32 

107 

66 

116 

33 

112 

67 

100 

34 

104 

74 

113 

35 

106 

— 

— 

Études  Anthropologiques. 


25 


Largeur  totale  de  la  face  (distance  bi-zygomatique  maxiaaaa). 

Examen  pratiqué  sur  49  individus. 

Largeur  moyenne   128™", 6;   variations  extrêmes  lOO'^^^jO   et 


139"^-,  0. 


Numéros  d'ordre 

Larg.  tot.  de  la  face 

Numéros  d'ordre 

Larg.  tot.  de  la  face 

1 

137  '""• 

36 

122  '"™- 

2 

136 

39 

130 

3 

131 

40 

128 

4 

127 

41 

138 

5 

128 

42 

133 

6 

125 

43 

139 

8 

125 

47 

136 

9 

127 

48 

125 

11 

127 

50 

125 

12 

135 

51 

137 

15 

121 

54 

134 

17 

130 

55 

135 

18 

124 

57 

122 

21 

116 

58 

124 

23 

121 

59 

127 

24 

109 

60 

138 

25 

125 

61 

130 

27 

125 

62 

128 

28 

129 

63 

128 

30 

132 

64 

124 

31 

127 

65 

136 

32 

122 

66 

128 

33 

114 

67 

136 

34 

135 

74 

129 

35 

139 

— 

— 

26 


Art.  4.— Torii  ; 


Hauteur  du  front. 


Le  chiffre  moyen  de  la  hauteur  du  front  mesurée  chez  4G 
sujets  est  de  72^"-,4  ;les  variations  extrêmes  sont  de  60™™-,0et  9r™%0. 


Numéros  d'ordre 

Haut,  du  front 

Numéros  d"ordre 

Haut.  du.  front 

1 

75  "'"^• 

36 

69  "'^■ 

2 

75 

39 

69 

3 

82 

40 

87 

4 

63 

41 

72 

5 

60 

42 

80 

6 

;  65 

43 

65 

8 

'  91 

47 

65 

11 

67 

48 

75 

12 

69 

50 

76 

15 

■  75 

51 

72 

17 

;  72 

53 

71 

18 

;    71 

55 

75 

21 

81 

58 

73 

23 

!  65 

59 

78 

24 

;     75 

60 

78 

25 

69 

61 

64 

27 

80 

62 

71 

28 

73 

63 

70 

31 

62 

64 

82 

32 

.   65 

65 

76 

33 

65 

66 

75 

34 

63 

67 

77 

35 

75 

74 

74 

J 


Études  Anthropologiques. 


27- 


Longueur  du  nez. 

La  longueur  mo^^enne,  prise  sur  48  individus  est  de  42"""-,  9, 
les  variations  extremes  de  34°™-,  0  et  53^°^-, 0. 


Xuméros  d'ordre       Lo] 

ag.  du  nez 

Numéros  d'ordre 

Long,  du  nez 

1 

36  "'"• 

36 

43  °""- 

2 

45 

39 

45 

3 

46 

40 

43 

4 

46 

41 

47 

5 

41 

42 

45 

6 

43 

43 

48 

8 

42 

47 

44 

11 

42 

48 

40 

12 

42 

50 

46 

15 

34 

51 

45 

17 

45 

54 

51 

18 

50 

55 

39 

21 

41 

57 

44 

23 

39 

58 

38 

24 

41 

59 

41 

25 

39 

60 

45 

27 

44 

61 

37 

28 

42 

62 

42 

30 

49 

63 

43 

31 

45 

84 

37 

32 

41 

65 

39 

33 

53 

66 

,47 

34 

41 

67 

40 

35 

42 

74 

42 

128 


Art.  4.— Torii  : 


Largeur  du  nez. 

Largeur  moyenne  prise  sur  une  série  de  48  individus:  40 
;  variations  extrêmes  :  32'""'\0  et  50'^"'-,0. 


lUam.   o 


Numéros  d'ordre 

Larg.  du  nez 

Numéros  d'ordre 

Larg.  du  nez 

1 

^]^   mm. 

36 

^]^   mm. 

2 

46 

39 

46 

3 

38 

40 

40 

4 

35 

41 

42 

5 

34 

42 

39 

6 

45 

43 

40 

8 

40 

47 

34 

11 

41 

48 

42 

12 

35 

50 

40 

15 

33 

51 

43 

17 

41 

54 

41 

18 

39 

55 

42 

^1 

41 

57 

43 

23 

l35 

58 

42 

24 

;37 

59 

38 

125 

32 

60 

41 

27 

42 

61 

38 

28 

41 

62 

44 

30 

50 

63 

87 

31 

37 

64 

40 

32 

38 

65 

43 

33 

42 

66 

44 

34 

42 

67 

41 

35 

50 

74 

36 

Études  Anthropologiques. 


29 


Largeur  palpébrale   (Longueur  de  l'œil). 

Largeur  moyenne  sur  49  individus:  31"™', 8.    Variations  extrê- 
mes :  24'™"-,  Oet  39™"-,0. 


Xuméros  cVordre 

Larg.  palj)ébralo 

Numéros  d'ordre 

Larg.  imlpébrale 

2 

28  ™'»- 

40 

32  °"^- 

3 

31 

41 

31 

4 

24 

42 

31 

5 

35 

43 

34 

6 

28 

47 

39 

9 

29 

48 

29 

11 

29 

50 

34 

12 

32 

51 

33 

15 

30 

52 

29 

17 

28 

53 

30 

18 

31 

54 

31 

21 

33 

55 

34 

22 

28 

57 

31 

23 

32 

58 

31 

24 

25 

59 

31 

25 

33 

60 

33 

27 

33 

61 

28 

28 

36 

62 

33 

31 

30 

63 

34 

32 

30 

64 

34 

33 

36 

65 

32 

34 

37 

66 

35 

35 

37 

67 

36 

36 

33 

74 

31 

39 

33 

— 

— 

30 


Art.  4— Toni 


Largeur  bicaronculaire  (distance  minima  des  yeux). 

La  largeur  bicaronculaire   moyenne   sur  49   individus  est  de 
.34'^"^9.     Les  variations  extremes  sont  de  2S'^™-,0  et  44"™\0. 


Numéi'oa  d'oi'dre 

Larg.  iDicaroncnl. 

Numéros  d'ordre 

Larg.  ))icaroncu]. 

2 

33  mm- 

40 

34  mm. 

3 

37 

41 

38 

4 

36 

42 

33 

5 

39 

43 

38 

6 

36 

47 

35 

9 

40 

48 

31 

11 

30 

50 

34 

12 

34 

51 

34 

15 

37 

52 

32 

17 

37 

53 

29 

18 

36 

54 

42 

21 

S3 

55 

33 

22 

30 

57 

35 

23 

35 

58 

29 

24 

30 

59 

34 

25 

28 

60 

35 

27 

32 

61 

32 

28 

35 

62 

35 

31 

39 

63 

32 

32 

35 

64 

35 

33 

44 

65 

35 

34 

39 

66 

38 

35 

40 

67 

32 

36 

34 

74   . 

38 

39 

36 

— 

— 

Études  Anthropologiques. 


31 


Largeur  buccale. 

48   sujets  observés.     Largeur  mo_yenne:    53"™, 3  Chiffres  ex- 
trêmes: 40"""-,0  et  G4"™-,0. 


Numéros  cVordre 

Larg.  ):)uccale 

Numéros  d"ordre 

Larg.  buccale 

1 

60  '""■ 

35 

59  '""■ 

2 

60 

36 

55 

3 

40 

40 

41 

4 

51 

41 

63 

5 

49 

42 

54 

6 

54 

43 

54 

8 

47 

47 

51 

9 

50 

48 

53 

11 

55 

50 

52 

12 

58 

51 

55 

15 

59 

54 

54 

17 

57 

55 

57 

18 

54 

57 

54 

21 

52 

58 

54 

23 

52 

59 

47 

24 

52 

60 

49 

25 

51 

61 

56 

27 

64 

62 

56 

28 

56 

63 

52 

30 

57 

64 

46 

31 

51 

65 

49 

32 

46 

66 

61 

33 

55 

67 

52 

34 

53 

74 

51 

32 


Art.  4;.— Torii  : 


Longueur  des  oreilles. 

Chiffre  moyen  sur  51  individus:  59""", 7.     Chiffres  extrêmes 
46'"'"-,0  et  69"^-,  0. 


Numéros  clordre 

Long,  des  oreilles 

Numéros  d'ordre 

Long,  des  oreilles 

1 

62  "™- 

36 

63   "='" 

2 

65 

39 

64 

3 

52 

40 

59 

4 

59 

41 

58 

5 

61 

42 

63 

6 

61           :■: 

43 

66 

8 

60 

47 

56 

9 

60 

48 

56 

11 

62 

50 

62 

12 

57 

51 

58 

15 

69 

53 

59 

17 

61 

54 

61 

18 

60 

55 

58 

21 

57 

57 

67 

23 

59 

58 

63 

24 

63 

59 

61, 

25 

46 

60 

51 

27 

46 

61 

54 

28 

56 

62 

65 

29 

55 

63 

62 

30 

57 

64 

59 

31 

57 

65 

59 

32 

65 

66 

56 

33 

67 

67 

59 

34 

59 

74 

61 

35 

68 

— 

— 

Études  Anthropologiques. 


3S 


Indice  céphalique. 

Le  chiffre  moyen  de  l'indice  céphalique,  sur  une  série  de  44 
individus,  est  de  79°™, 4.  Les  chiffres  extrêmes  sont  de  TO^'^^ô  et 
88™"-,  T). 

Cette  moyenne  correspond  aux  Sous-dohchocéphales  de 
"  l'Indice  céphalique  sur  le  vivant  "   de  Deniker/'^ 

Voici  le  tableau  de  ces  indices. 


Xuruéros  d'ordre 

Indice  céphal. 

Numéros  d'ordre 

Indice  céphal. 

1 

73,9 

35 

83,2 

2 

74,5 

36 

80,5 

3 

76,1 

39 

82,5 

4 

79,4 

40 

77,8 

5 

84,7 

41 

79,7 

6 

79,7 

42 

75,8 

7 

76,4 

43 

76,5 

9 

87,6 

47 

73,4 

11 

81,4 

48 

79,3 

12 

85,0 

55 

74,6 

13 

86,2 

57 

80,6 

15 

73,9 

58 

77,2 

17 

83,3 

59 

77,7 

18 

80,6 

60 

78,8 

21 

78,5 

61 

77,6 

25 

87,8 

62 

77,6 

27 

88,5 

63 

70,6 

28 

79,6 

64 

80,1 

30 

82,9 

65 

80,8 

31 

81,5 

66 

76,6 

32 

79,0 

74 

75,7 

33 

80,6 

— 

— 

34 

78,6 

— 

— 

(1).     J,  Deniker,  Races  et  peui^lf  3  de  la  terre,  p.  89.     Paris  1900. 


54 


Art.  4. — Torii  : 


En  établissant  un  rapport  entre  ces  indices  céphaliques  et  le 
nombre  des  individus  observés,  on  obtient  ce  qui  suit  : 


70,1     — 

71,0 

71,1     - 

72,0 

72,1     ~ 

73,0 

73,1     — 

74,0 

74,1     — 

75,0 

75,1     — 

76,0 

76,1    — 

77,0 

77,1     - 

78,0 

78,1     — 

79,0 

79.1     — 

80,0 

80,1     — 

81,0 

81,1     — 

82,0 

82,1     — 

83,0 

83,1     — 

84,0 

84,1     — 

85,0 

85,1     — 

86,0 

86,1    — 

87,0 

87,1     — 

88,0 

88,1    — 

89,0 

89,1    — 

90,0 

90,1    — 

91,0 

44 


Dolichocépliales . 


Sous-dolicliocéphales. 


Mésocéphales. 


Sous-brachycépliales. 


Brachycéphales. 


Hyperbrachycéphales. 


•Ce  rapport  est  figuré  dans  la  courbe  ci-dessous  : 


71     nz     T^    7f    TS    76    77    73    7?    80    BT    &Z    63    8"^     S5   66    87    83    8?    90 


Études  Anthropologiques.  35 

Indice  du  visage. 

(obtenu  en  comparant  la  longueur  totale  du  visage,  de  la  racine 
des  cheveux  au  point  mentonnier,  à  la  largeur  totale  de  la  face  ou 
distance  bizygomatique  maxima.) 

Chiffres  obtenus  sur  une  série  de  32  individus  : 

Indice  moyen:  71,9.     Variations  extrêmes:  60,2  et  80,8. 


Numéros  d'ordre 

Indice  du  visage 

Numéros  d'ordre 

Indice  du  visage 

1 

74,9 

27 

66,1 

2 

74,7 

28 

72,1 

3 

69,3 

31 

77,0 

•1 

77,4 

32 

70,9 

5 

79,5 

33 

64,4 

6 

73,1 

34 

80,8 

8 

65,4 

35 

76,8 

11 

74,3 

36 

70,9 

12 

78,0 

39 

73,9 

15 

66,5 

40 

64,6 

17 

73,9 

41 

75,0 

18 

72,1 

42 

69,6 

21 

63,0 

43 

76,5 

23 

71,2 

47 

76,8 

24 

60,2 

48 

69,4 

2.5 

76,2 

50 

68,3 

Indice  facial. 

(obtenu  en  comparant  la  longueur  totale  de  la  face,  (du  point 
sus-nasal  au  point  mentonnier),  à  la  distance  bi-zygomatique 
maxima.) 


36 


Art.  4. -Toni 


Chiffres  obtenus  sur  une  série  de  47  individus  :  indice  moyen 
121,1.     Variations  extremes:  101,8  et  130,0 


Numéros  d'ordre 

Indice  facial 

Nnméro&  d'ordre 

Indice  facial 

1 

126,9 

36 

118,4 

2 

127,1 

39 

121,5 

3 

127,2 

40 

115.3 

4 

125,7 

41 

123,2 

5 

126,7 

42 

119,8 

6 

112,6 

43 

123,0 

8 

125,0 

47 

121,4 

11 

122,1 

48 

119,0 

12 

129,8 

50 

116,8 

15 

113,1 

51 

131,7 

17 

125,0 

55 

127,4 

18 

122,8 

57 

114,0 

21 

112,6 

58 

122,9 

23 

115,2 

59 

117,6 

24 

102,8 

60 

124,3 

25 

131,5 

61 

121,5 

27 

114,7 

62 

124,3 

28 

121,7 

63 

121,9 

30 

120,0 

64 

131,9 

31 

123,3 

65 

123,6 

32 

114,0 

66 

110,3 

33 

101.8 

67 

136,0 

34 

129,8 

74 

114,2 

35 

131,1 

— 

^— 

Études  Anthropologiques. 


37 


Indice  nasal. 


Indice  moyen  sur  une  série  de  48  individus:  94,3.     Chiffres 
extrêmes:  7G,1  et  119,0. 


Xuméros  d'ordre 

Indice  nasal 

Xuméros  d'ordre 

Indice  naäal 

1 

113,9 

36 

95,3 

2 

102,2 

39 

102,2 

3 

82,8 

40 

93,0 

4 

76,1 

41 

89,4 

5 

82,9 

42 

86,7 

6 

104,7 

43 

83,3 

8 

95,2 

47 

77,3 

11 

97,6 

48 

105,0 

12 

83,3 

50 

87,0 

15 

97,1 

51 

95,6 

17 

91,1 

54 

80,4 

18 

78,0 

55 

107,7 

21 

100,0 

57 

97,7 

23 

89,7 

58 

110,5 

24 

90,2 

59 

92,7 

25 

82.1 

60 

91,1 

27 

95,5 

61 

102,7 

28 

97,6 

62 

104,8 

30 

102,0 

63 

86,0 

31 

82.2 

64 

108.1 

32 

92,7 

65 

110,3 

33 

79,2 

66 

93,6 

34 

102,4 

67 

102,5 

35 

119.0 

74 

85,7 

38 


Art.  4.— Torii  : 


B.    Mensurations  du  tronc  et  des  membres. 


Taille. 


La  moyenne  de  la  taille,  sur  une  série  de  46  sujets,  est  de 
160°'"-5.     Les  chiffres  extrêmes  sont  de  lör'^^O  et  de  172''"% 0. 

Cette  moyenne  correspond  à  la  "  Taille  au  dessus  de  la 
moyenne  "  de  M.  Topinard/'-* 


Numéros  d'ordre 

Taille 

Numéros  d'ordre 

Taille 

cm. 

cm. 

1 

161,0 

33 

165,5 

2 

159,0 

34 

157,5 

3 

165,0 

35 

169,0 

4 

154,5 

36 

158,5 

5 

154,5 

39 

159,5 

6 

163,0 

40 

159,5 

7 

155,0 

42 

165,5 

8 

154,0 

47 

161,0 

11 

160,0 

48 

158,5 

12 

160,0 

50 

162,5 

13 

168,0 

51 

160,5 

15 

163,0 

54 

162,0 

17 

163,5 

55 

155,5 

18 

156,0 

57 

154,0 

20 

172,0 

58 

162,0 

21 

155,0 

59 

157,0 

23 

162,5 

60 

162,0 

24 

156,0 

61 

154,0 

25 

159,0 

62 

161,0 

27 

163,5 

63 

151,0 

28 

171,5 

65 

170,0 

29 

163,5 

66 

164,5 

31 

156,5 

67 

158,0 

(1)    p.  Topinard.    Elements  d'anthropologie  générale.     Paris  1885.    (p.  402.) 


Études  Anthropologiques. 


3c^ 


Si  l'on  établit  un  rapport  entre  ces  chiffres,  en  partant  des  j^lus 
faibles  pour  arriver  aux  plus  forts,  et  le  nombre  des  individus  ob- 
servés, on  obtient  ce  qui  suit  : 


Taille 

KoDibre  des  sujets 

cm.            cm. 

151,1-152,0 

1 

152,1-159,0 

— 

153,1-154,0 

— 

154,1-155,0 

3 

155,1-156,0 

4 

156,1-157,0 

3 

157,1-158,0 

2 

158,1-159,0 

2 

159,1-160,0 

4 

160,1-161,0 

4 

161,1-162,0 

4 

162,1-163,0 

3 

163, 

164,1 

165,1 

166,1- 

167,1 

168,1 

169,1 

170,1 

171,1 

172,1 


-164,0 
-165,0 
-166,0 
-167,0 
-168,0 
-169,0 
-170,0 
-171,0 
-172,0 
-173,0 


46 


C'est  ce  rapport  que  représente  également  la  courbe  ci-jointe: 


1^0    isr    752    7Î3    rÇ4    755    756    75r  7^8   759    760    767    t62   163    764  765    Vo6    76r   768    769    770    777    112   773. 


40 


Art.  4— Torii  : 


Grande  envergure. 
La  longueur  moyenne  sur  une   série  de  43  individus,   est  de 
IGo'^^^e.    Les  variations  extrêmes  sont  de  156'"-, 5  et  178'"'-,0. 


Numéros  cVordre 

Grande  envergure 

Numéro.?  d'ordre 

Grand  envergure 

cm. 

cm. 

1 

171,0 

34 

163,5 

2 

165,5 

35 

177,0 

3 

173,0 

36 

161.0 

4 

156,5 

39 

162,5 

5 

156,5 

40 

163,0 

6 

165,5 

42 

173,5 

7 

165,0 

47 

170,0 

11 

165,0 

48 

160,5 

12 

167,0 

50 

165,5 

13 

176,5 

51 

167,5 

15 

165,5 

54 

168,0 

17 

173,0 

55 

157,5 

18 

157,0 

57 

165,0 

21 

157,5 

58 

170,0 

23 

173,5 

59 

165,0 

24 

160,0 

60 

169,0 

25 

169,0 

61 

160,0 

27 

167,5 

62 

171,0 

28 

176,5 

63 

161,0 

29 

162,5 

65 

178,0 

31 

169,5 

66 

169,5 

33 

174,5 

— 

— 

Si  l'on  établit  un  rapport  entre  ces  chiffres,  en  partant  des 
plus  faibles  pour  arriver  aux  plus  élevés,  et  le  nombre  des  individus 
observés,  on  obtient  ce  qui  suit  : 


Grande  envergure 

Nombre  des  sujets 

cm.            cm. 

155,1-156,0 
156,1-157,0 
157,1-158,0 

3 

Etudes  Anthropologlqiies. 


41 


Grande  envergure 

Nombre  des  sujets 

cm.                cm. 

158,1-159,0 

2 

159,1-160,0 

— 

160,1-161,0 

2 

161,1-162,0 

3 

162,1-163,0 

— 

163,1-164,0 

3 

164,1-165,0 

1 

165,1-166,0 

4 

166,1-167,0 

4 

167,1-168,0 

1 

168,1-169,0 

8 

169,1-170,0 

2 

170,1-171,0 

4 

171,1--172,0 

2 

172,1-173,0 

— 

173,1-174,0 

2 

174,1-175,0 

2 

175,1-176,0 

1 

176,1-177,0 

— 

177,1-178,0 

8 

178,1-179,0 

1 

43 
Rapport  do  la  grande  envergure  à  la  taille. 
Ayant  recherché  ce  rapport  sur  une  série  de  43  individus,  j'ai 
constaté  qu'  à  l'exception  d'un  seul  individu  (le  n°  29  de  la  série, 
dont  la  taille  est  de  163'=^'%5  et  la  grande  envergure  de  162°^^-,5),   la 
grande  envergure  était  toujours  supérieure  à  la  taille.      Le  rapport 
moyen  de  la  grande  envergure  à  la  taille  est  de  103,7. 
Voici  le  tableau  de  ce  rapport  : 


Numéros  d'ordre 

Grande  envergure 

Taille 

Différence 

Eapport  entre  la  gr. 
enverg.  et  la  taille 

1 

2 
3 

cm. 

171,0 
165,5 
173,0 

cm. 

161,0 
159,0 
165,0 

cm. 

10,0 
6,5 

8,0 

106,2 
104,1 
104,9 

42 

Art.  4.— Torii: 

Numéros  d'ordre 

Grande  envergure 

Taille 

Différence 

Rapport  entre  la  gr. 
enverg.  et  la  taille 

cm. 

cm- 

cm. 

4 

156,5 

154,5 

2,0 

101,3 

5 

156,5 

154,5 

2,0 

101,3 

6 

165,5 

163,0 

2,5 

101,5 

7 

165,0 

155,0 

10,0 

106,5 

11 

165,0 

160,0 

5,0 

103.1 

12 

167,0 

160,0 

7,0 

104,4 

13 

176,5         ! 

168,0 

8,5 

105,1 

15 

165,5 

163,0 

2,5 

101,5 

17 

173,5 

163,5 

10,0 

106,1 

18 

157,0 

156,0 

1,0 

100,6 

21 

157,5 

155,0 

2,5 

101,6 

23 

173,5 

162,5 

11,0 

106,8 

24 

160,0 

156,0 

4,0 

102,5 

25 

169,0 

159,0 

10,0 

106,3 

27 

167,5 

163,5 

4,0 

102,4 

28 

176,5 

171,5 

5,0 

102,9 

29 

162,5 

163,5 

-1,0 

93,9 

31 

169,5 

156,5 

13,0 

108,3 

33 

174,5 

165,5 

9,0 

105,4 

34 

163,5 

157,5 

6,0 

103,8 

35 

177,0 

169,0 

8,0 

104,7 

36 

161,0 

158,5 

2,5 

101,6 

39 

162,5 

159,5 

3,0 

101,9 

40 

163,0 

159,5 

4,5 

102,2 

42 

173,5 

165,5 

8,0 

104,8 

47 

170,0 

161,0 

9,0 

105,6 

48 

160,5 

158,5 

2,0 

101,3 

50 

165,5 

162,5 

3,0 

101,8 

51 

167,5 

160,5 

7,0 

104,4 

54 

168,0 

162,0 

6,0 

103,7 

55 

157,5 

155,5 

2,0 

101,3 

57 

165,0 

154,0 

11,0 

107,1 

58 

170,0 

162,0 

8,0 

104,9 

59 

165,0 

157,0 

8,0 

105,1 

60 

169,0 

162,0 

7,0 

104,3 

61 

160,0 

154,0 

6,0 

103,9 

62 

171,0 

161,0 

10,0 

106,2 

63 

161,0 

151,0 

10,0 

106,6 

65 

178,0 

170,0 

8,0 

104,7 

66 

169,5 

164,5 

5,0 

103,0 

Études  Atithropologiqnes. 


43 


Hauteur  du  conduit  auditif  au-dessus  du  sol. 

La  hauteur  moyenne  est  de  147™', 0  d'après  les  mensurations 
prises  sur  une  série  de  25  sujets. 

Les  variations  extremes  sont  de  140™, 0  et  IST'^^jo. 


Numéros  dordre 

Haut,  du  cond.  audit. 

Eapport  de  la  haut,  du 
cond.  audit,  à  la  taille 

cm. 

4 

141,0 

91,3 

5 

144,0 

93,2 

11 

147,0 

91,9 

12 

149,5 

93,4 

13 

157,5 

93,8 

17 

150,0 

91,7 

18 

143,0 

91,7 

21 

140,0 

90,3 

23 

145,0 

89,2 

24 

140,5 

90,1 

25 

144,5 

90,9 

27 

143,5 

87,8 

28 

153,0 

89,2 

29 

147,0 

89,9 

31 

143,0 

91,4 

32 

151,0 

93,1 

33 

154,0 

92,1 

34 

145,0 

92,3 

35 

156,0 

94,3 

36 

149,5 

90,7 

43 

143,5 

89,0 

47 

146,0 

92,3 

48 

141,0 

89,9 

50 

150,0 

— 

67 

142,0 

— 

44 


Art.  4.— Torii 


Hauteur  clu  bord  inférieur  du  menton  au-dessus  du  sol. 

La  hauteur  moyenne,  sur  25  sujets,  est  de  ISO*""-,!. 

Les  variations  extrêmes  sont  de  13P™',5  et  152™', 5. 

Le  rapport  de  la   hauteur  du  menton  au-dessus  du  sol  à  la 

taille  est  de  8G,7. 


Xuméros  d'ordre 

Haut,  du  menton 

Rapport  de  la  haut,  du 
ment,  à  la  taille 

cm. 

4 

132,5 

85,8 

5 

135,0 

87,4 

11 

137,5 

85,9 

12 

149,5 

93,4 

13 

149,0 

88,7 

15 

142,0 

87,1 

17 

141,5 

86,5 

18 

135,5 

86,9 

21 

132,5 

85,5 

23 

139,5 

85,7 

24 

131,5 

81,3 

25 

137,0 

86,2 

27 

137,5 

84,1 

28 

145,0 

84,6 

31 

135,5 

86,6 

32 

141,5 

92,2 

33 

152,5 

87,3 

34 

137,5 

87,3 

35 

147,5 

86,1 

36 

136,5 

85,1 

43 

135,0 

85,5 

47 

137,0 

87,1 

48 

135,5 

84,2 

50 

141,5 

— 

67 

133,0 

— 

Études  Anthropologiques. 


Hauteur  de  racromion  au-dessus  du  sol. 

Hauteur  moyenne:  133''"-,  1  (sur  une  série  de  21). 
Variations  :    123^-^-, 5  et  U5'^\0.     Rapport  de  la  hauteur  de 
r  acromion  à  la  taille  :  82,9. 


Numéros  d'ordre 

Haut,  de  racrom. 

Rapport  de  la  haut,  de 
l'acrom.  à  la  taille 

4 

cm. 

130,5 

84,5 

11 

130,5 

81,6 

13 

143,5 

85,6 

17 

137,5 

84,1 

18 

127,5 

81,7 

21 

131,5 

84,8 

28 

135,0 

83,1 

24 

127,5 

81,7 

25 

130,5 

82,1 

27 

134,5 

82,3 

28 

145,0 

84,6 

31 

123,5 

78,9 

32 

133,5 

85,7 

33 

135,0 

82,0 

34 

138,5 

83,6 

36 

132,5 

81,1 

43 

130,5 

83,6 

47 

130,5 

83,1 

48 

132,5 

82,0 

50 

135,0 

— 

67 

129,5 

— 

46 


Art.  4. — Torii 


Hautour  de  l'extrémité  inférieure  du  médius  au-dessus  du  sol. 

Hauteur  moyenne  :  57°"'-,  3  (sur  une  série  de  24).  Variations 
extrêmes:  49""-, 5  et  65°'"-,0. 

Le  Rapport  de  la  hauteur  de  l'extrémité  inférieure  du  médius 
à  la  taille,  sur  une  série  de  22,  est  de  35, 8. 


Xuméros  d'ordre 

Haut,  de  l'extrém.  du 
médius 

Eapport  de  lu  haut,  du 
med.  à  la  taille 

cm. 

4 

56,5 

36,6 

5 

55,5 

35,9 

11 

58,0 

36,3 

12 

62,0 

38,8 

13 

61,5 

36,6 

15 

58,5 

35,9 

17 

61,5 

37,6 

18 

62,0 

39,7 

21 

55,0 

35,5 

24 

49,5 

31,7 

25 

62,5 

39,3 

27 

60,0 

36,7 

28 

57,0 

33,2 

31 

56,0 

35,8 

32 

60,5 

32,6 

33 

54,0 

36,5 

34 

57,5 

38,5 

35 

65,0 

35,6 

36 

55,5 

31,1 

43 

50,0 

33,1 

47 

50,0 

36,0 

48^ 

52,5 

34,8 

50 

58,5 

— 

67 

55,0 

— 

Études  Anthropologiques. 


47 


Hauteur  de  l'ombilic  au-dessus  du  sol. 

Hauteur  moyenne  :  95°""-,  3  (sur  une  série  de  23). 
Variations  extrêmes:  88°"^, 0  et  103°™-,5.    Rapport  de  la  hauteur 
de  l'ombilic  à  la  taille,  (sur  une  série  de  21):  59,3. 


Numéros  d'ordre 

Haut,  de  rombilic 

Rapport  de  la  haut,  de 
î'ombil.  à  la  taille 

cm. 

4 

88,0 

57,0 

5 

92,5 

59,9 

11 

94,5 

59,1 

12 

99,5 

62,2 

13 

103,5 

61,6 

15 

98,5 

60,4 

17 

93,5 

57,0 

18 

89,5 

57,4 

21 

92,5 

59,7 

23 

96,0 

59,1 

24 

92,5 

59,3 

25 

94,5 

59,4 

27 

97,5 

59,6 

28 

99,5 

58,0 

32 

94,5 

61,9 

33 

102,5 

58,4 

34 

92,0 

60,1 

35 

101,5 

57,7 

36 

91,5 

57,5 

43 

92,5 

60,3 

47 

92,5 

59,4 

48 

95,5 

— 

50 

96,5 

— 

48 


Art.  4.-Torii 


Hauteur  du  grand  trochanter  au-dessus  du  sol. 

Hauteur  moyenne:  92"'', 0  (sur  une  série  de  20.) 
Variations  extrêmes  :  SS'^^'^O  et  lOO^^'^O. 

Rapport  de  la  hauteur  du  grand  trochanter  à  la  taille  :  57, 2 
(sur  une  série  de  19). 


Numéros  d'ordre 

Haut,  du  grand  troch. 

Eapijort  de  la  haut,  du 
gd.  troch.  à  la  taille 

cm. 

2 

84,5 

53,1 

4 

87,5 

56,6 

5 

91,5 

59,2 

11 

83,0 

5L9 

12 

96,5 

60,0 

13 

100,0 

59,5 

15 

93,5 

57,4 

18 

87,5 

56,1 

21 

87,5 

56,5 

23 

95,0 

58,5 

24 

89,5 

57,4 

25 

91,5 

57,6 

27 

93,5 

57,2 

28 

98,5 

57,4 

31 

90,0 

57,5 

32 

93,0 

60,4 

33 

100,0 

55,9 

34 

88,0 

58,0 

35 

98,0 

56,8 

47 

91,5 

— 

Études  Anthropologiques. 


49 


Hauteur  du  vertex  dans  la  position  assise. 

Hauteur  moyenne:  Sô^'^jQ  (sur  une  série  de  25). 
Variations  extrêmes:  SS'^^^O  et  93'='°-,5. 

Rapport  de  la  hauteur  du  vertex   dans  la  position  assise  à  la 
taille  :  54,0  (sui-  une  série  de  24). 


Numéros  d'ordre 

Haut,  du  vertex  dans  la 
posit,  assise 

Eapport  de  la  haut,  du 
vertex  pos.  ass.  à  la  taille 

2 

cm. 

84,0 

52,8 

3 

85,0 

51,5 

4 

86,5 

56,0 

5 

88,5 

57,3 

6 

85,0 

52,2 

11 

85,0 

53,1 

l'2 

83,0 

51,9 

15 

86,5 

53,1 

17 

88,0 

53,8 

18 

86,5 

55,5 

21 

85,5 

55,2 

23 

89,5 

55,1 

24 

84,0 

53,9 

25 

85,5 

53,8 

27 

85,5 

52,3 

28 

88,5 

51,6 

31 

88,5 

56,6 

32 

90,5 

55,6 

33 

92,0 

55,2 

34 

87,0 

55,3 

35 

93,5 

55,2 

36 

87,5 

53,4 

47 

86,0 

54,3 

48 

86,0 

51,7 

50 

84,0 

— 

50 


Art.  4.-Torii: 

Distance  bi-acromiale. 


La  moyenne  est  de  40°°"-,  5  (sur  une  série  de  39,)  les  variations 
extrêmes  étant  de  Sl'^^^O  et48'=™-,0 

Le  rapport  de  la  distance  bi-acromiale  à  la  taille  est  de  25,3 
(sur  une  série  de  33). 


Numéros  d'ordre 

Dist.  bi-acromiale 

Eapport  de  la  dist.  bi- 
acrom.  à  la,  taille 

cm. 

2 

43,0 

27,1 

4 

42,5 

27,5 

5 

38,5 

24,9 

9 

39,0 

26,9 

11 

43,0 

28,6 

13 

48,0 

25,6 

18 

40,0 

25,2 

21 

39,0 

24,6 

23 

40,0 

25,0 

24 

39,0 

25,8 

25 

41,0 

25,7 

27 

42,0 

23,9 

28 

41,0 

25,5 

30 

39,0 

27,6 

31 

40,0 

25,7 

32 

44,0 

23,3 

34 

43,5 

23,8 

35 

43,5 

26,0 

36 

37,0 

24,5 

39 

38,0 

24,3 

42 

43,0 

26,2 

44 

43,0 

24,9 

47 

39,5 

23,5 

48 

38,5 

28,8 

50 

42,5 

25,0 

51 

40,0 

26,2 

54 

38,0 

25,5 

55 

37,0 

24,7 

57 

38,5 

23,9 

Études  Anthropologiques. 


51 


Numéros  d'ordre 

Dist.  bi-acromiale 

f  apport  de  la  dist   bi- 
aorom.  à  la  taille 

cm. 

58 

42,5 

26,2 

59 

40,0 

23,5 

61 

38,0 

24,3 

62 

38,5 

26,9 

63 

39,5 

— 

64 

37,0 

— 

65 

40,0 

— 

66 

40,0 

67 

42,5 

74 

39,0 

— 

Circonférence  de  la  poitrine. 

Moyenne  prise  sur  une  série  de  37  .'individus:  Bô^'^^^O.    Varia- 
tions extrêmes:  78"", 0  et  94"", 0 


Numéros  d'ordre 

Circonf .  de  la  ijoitr. 

Numéros  d'ordre 

Circonf.  de  la  poitr. 

cm. 

cm. 

2 

85,5 

42 

9],0 

4 

84,0 

43 

90,0 

9 

94,0 

47 

89,5 

n 

85,0 

48 

83,5 

13 

86,0 

50 

84,0 

17 

88,5 

51 

82,0 

18 

81,0 

54 

86.0 

21 

92,0 

55 

87,0 

23 

89,0 

57 

87,0 

24 

85,0 

58 

87,5 

25 

85,0 

59 

88,5 

27 

92,0 

61 

78,0 

28 

89,0 

62 

84,5 

30 

93,5 

63 

86,0 

32 

83,5 

64 

80,5 

33 

89,0 

65 

910 

34 

86,0 

66 

84,0 

35 

91,0 

77 

88,0 

39 

86,0 

— 

— ■ 

52 


Art.  4.-Torii: 


Longueur  du  membre  supérieur. 

(de  l'acromion  à  l'extrémité  du  raédius\ 

Longueur  moyenne:  To™", 2  (sur  une  série  de  19).     Variations 
extrêmes  :  ßö'™- 0  et  82'='^-,5. 


Numéros  d'ordre 

Long,  du  memb.  sup. 

Rapport  de  la  long,  du 
m.  sup.  à  la  taille 

cm. 

4 

74,0 

47,2 

11 

72,5 

45,3 

13 

82,0 

48,8 

17 

75,5 

46,2 

18 

65,0 

41,7 

21 

76,0 

49,0 

24 

78,0 

50,0 

25 

68,0 

42,8 

27 

77,5 

47,4 

31 

67,5 

43,1 

33 

72,5 

43,8 

34 

77,5 

49,2 

35 

73,0 

43,2 

36 

77,5 

48,9 

40 

80,0 

50,2 

47 

82,5 

51,2 

48 

79,5 

50,2 

50 

76,5 

47,1 

67 

74,5 

47,1 

Longueur  du  membre  inférieur. 

(hauteur  du  grand  trochanter  au-dessus  du  sol.) 

Longueur  moyenne:  9P",9  (sur  une  série  de  20).     Variations 
extrêmes  :  SS'^'^O  et  lOQ-^^^^O. 


Numéros  d'ordre 

4 
5 


Longueur  du  m.  infér. 
cm. 

87,5 
91,5 


Études  Anthropolo^ques. 


53 


Numéros  d'ordre 

Longueur  du  m.  infer. 

cm. 

11 

83,0 

12 

96,5 

13 

100,0 

15 

93,5 

18 

87,5 

21 

87,5 

23 

95,5 

24 

89,5 

25 

91,5 

27 

93,5 

28 

98,5 

31 

90,0 

32 

93,5 

33 

100,0 

34 

88,0 

35 

91,5 

47 

90,0 

48 

90,0 

Rapport  entre  la  longueur  du  membre  supérieur  et  du 
membre  inférieur. 

La  moyenne  de  ce  rapport  est  de  82,5  sur  une  série  de  14. 
Les  chiffres  extrêmes  sont  72,5  et  91,7. 


Numéros  d'ordre 

Longueur  du  m.  super. 

Longueur  du  m.  infér. 

Eapport  entre  ces 
2  long. 

cm. 

om. 

4 

74,0 

87,5 

84,6 

11 

72,5 

83,0 

87,4 

13 

82,0 

100,0 

82,0 

17 

75,5 

? 

? 

18 

65,0 

87,5 

74,4 

21 

76,0 

87,5 

86,9 

24 

78,0 

89,5 

87,2 

25 

68,0 

91,5 

74,3 

54 


Art.  4.— Torii  : 


Numéros  d'ordre 

Longueur,  du  in.  super 

Longueur  du  m.  infér. 

Eapport  entre  ces 
2  long. 

cm. 

cm. 

27 

77,5 

93,5 

82,9 

31 

67,5 

90,0 

75,0 

32 

'? 

93,5 

? 

33 

72,5 

100,0 

72,5 

34 

77,5 

88,0 

88,1 

35 

73,0 

91,5 

79,8 

36 

77,5 

? 

? 

43 

8,00 

? 

? 

47 

82,5 

90,0 

91,7 

48 

79,5 

90,0 

88,3 

50 

76,5 

? 

? 

67 

74,5 

? 

? 

Circonférence  du  bras. 

Moyenne  :  26''"',0  (sur  une  série  de  43).   Dimensions  extrêmes; 

20'='"-,5  et  29""-,0. 


Numéros  d'ordre 

Circonf.  du  bras 

Numéros  d'ordre 

Circonf.  du  bras 

cm. 

cm. 

4 

26,0 

39 

27,5 

5 

24,5 

41 

25,5 

6 

25,0 

42 

27,5 

8 

28,0 

43 

27,5 

9 

27,0 

47 

26,5 

11 

26,0 

48 

23,5 

12 

26,5 

50 

24,0 

13 

26,5 

51 

26,5 

17 

28,0 

54 

26,0 

18 

25,0 

55 

27,0 

21 

25,5 

57 

25,0 

23 

26,0 

58 

25,0 

24 

25,0 

59 

26,5 

25 

26,0 

60 

27,0 

27 

27,0 

61 

23,5 

28 

29,0 

62 

24,0 

Etudes  Anthropolo^quGs. 


55 


Numéros  d'ordre 

Circonf.  du  bras 

Numéros  d'ordre 

Circonf.  du  bras 

cm. 

cm. 

30 

28,5 

63 

26,0 

31 

27,0 

64 

22,0 

33 

20,5 

65 

26,5 

34 

27,5 

66 

26,0 

35 

26,0 

67 

28,0 

36 

25,0 

— 

— 

Circonf éreD ce  de  la  cuisse. 

Moyenne  :  47°", 3  sur  une  série  de  43.    Dimensions  extrêmes: 
39"°^-,  5  et  5r™-,0. 


Numéros  d'ordre 

Cire,  de  la  cuisse 

Numéros  d'ordre 

Cire,  de  la  cuisse 

4 

cm. 

46,5 

41 

cm. 

46,0 

5 

46,5 

42 

47,5 

9 

47,0 

43 

50,5 

11 

48,5 

47 

48,5 

12 

49,5 

48 

45,0 

15 

45,5 

50 

47,0 

18 

49,5 

51 

49,5 

21 

51,0 

54 

45,0 

23 

50,0 

55 

50,0 

24 

47,0 

57 

47,0 

25 

47,0 

58 

44,0 

27 

50,0 

59 

46,0 

28 

48,5 

60 

43,0 

30 

49,0 

61 

44,5 

31 

49,0 

62 

49,0 

32 

43,0 

63 

48,0 

33 

47,5 

64 

39,5 

34 

48,0 

65 

49,0 

35 

49,0 

66 

46,0 

36 

44,5 

67 

48,5 

39 

45,5 

74 

48,5 

40 

50,0 

— 

— 

56 


Art.  4  — Torii  ; 


Circonférence  du  mollet. 

Moyenne:  32°'"-,3  sur  une  série  de  44.     Dimensions  extrêmes; 

28'"-,5  et  36°"-,  0. 


Numéros  d'ordre 

Cire,  du  mollet 

Numéros  d'ordre 

Cire,  du  mollet 

cm. 

cm. 

2 

31,5 

39 

32,0 

4 

31,0 

40 

31,0 

5 

32,5 

42 

33,0 

9 

33,0 

43 

36,0 

11 

32,5 

47 

33,0 

12 

32,5 

48 

31,0 

13 

32,0 

50 

31,0 

17 

35,5 

51 

30,0 

18 

33,0 

54 

31,0 

21 

83,0 

55 

32,0 

23 

33,0 

57 

33,5 

24 

29,0 

58 

31,0 

25 

32,0 

59 

31,0 

27 

33,0 

60 

31,0 

28 

32,0 

61 

30,0 

30 

35,5 

62 

35,0 

31 

33,0 

63 

33,0 

32 

32,0 

64 

28,5 

33 

35,0 

65 

32,0 

34 

33,0 

66 

33,0 

35 

33,5 

67 

32,0 

36 

30,5 

74 

33,5 

Études  Anthropologiques. 


57 


Longueur  totale  du  pied. 

Les  indigènes  de  Kô-tô-shô  marchant  toujours  pieds-nus,  la 
plante  de  leur  pied  est  très  large. 

La  longueur  moyenne  est  de  25°°'',  1  (sur  une  série  de  43),  les 
variations  extrêmes  étant  de  'IV'^.O  et  de  29°"^-, 5. 


Numéros  d'ordre 

Long.  tot.  du  pied 

Numéros  d'ordre 

Long.  tot.  du  pied 

om. 

cm. 

4 

23,0 

40 

25,0 

5 

21,0 

41 

25,5 

9 

24,5 

42 

25,5 

11 

27,0 

43 

25,5 

12 

26,0 

47 

25,0 

13 

26,5 

48 

24,0 

15 

24,5 

50 

24,5 

16 

25,5 

51 

24,5 

18 

23,5 

54 

•25,5 

21 

29,5 

55 

25,5 

23 

25,5 

57 

24,5 

24 

24,5 

58 

25,0 

25 

24,5 

59 

24,5 

27 

25,0 

60 

25,5 

28 

29,0 

61 

2.5,5 

30 

25,5 

62 

26,5 

31 

24,0 

63 

24,5 

32 

26,5 

64 

23,5 

34 

24,5 

66 

25,0 

35 

25,0 

67 

25,0 

36 

25,5 

74 

24,5 

39 

25,0 

— 

— 

58 


Art.  4.-Torii 


Largeur  maxima  du  pied 

Largeur  moyemie:  Q"""',  36  (sur  une  série  de  43).  Variations  ex- 
trêmes: 8°'"-,0  et  10°'"-,5. 


Numéros  cVordre 

Larg.  uiax.  du  pied 

Numéros  d'ordre 

Larg.  max.  du  pied 

cm. 

cm. 

4 

8,5 

40 

9,5 

5 

10,0 

41 

8,5 

9 

10,0 

42 

10,0 

11 

9,0 

48 

10,0 

12 

9,0 

47 

9,0 

13 

9,5 

48 

9,5 

15 

9,5 

50 

9,0 

16 

8,5 

51 

9,5 

18 

9,0 

54 

9,0 

21 

9,5 

55 

9,5 

23 

10,0 

57 

9,0 

24 

9,0 

58 

8,0 

25 

9,5 

59 

8,5 

27 

10,0 

60 

10,0 

28 

10,0 

61 

10,0 

30 

10,5 

62 

9,0 

31 

9,5 

63 

9,5 

32 

9,0 

64 

9,5 

34 

10,0 

66 

9,0 

35 

10,0 

67 

8,0. 

36 

8,5 

74 

10,0 

39 

10,0 

— 

— 

Etudes  Anthropologiques. 


59 


TABLE  DES  MENSURATIONS  ET  INDICES  D'ABORIGENES  DE  KÜTÖSHÖ 


Numéros. 

1 

2 

3 

4 

5 

6 

mm 

172 

137 

171 

111 

125 

65 

43 

45 

28 

36 

54 

61 

7 

3 
'S 

d 

3  S 

■3  ^ 

ci  '^ 
u  ^ 

fl       1 

œ 

•Il 
'^ 

o 

PI 
2   . 

.2  3 

to 
Ö 

S 

i  Diam.  ant.  post.  max.    ... 

Diam.  transv.  max 

Jjong.  tot.  du  visage 

Long.  tot.  de  la  face 

Larg.  tot.  de  la  face 

Haut,  du  front    

Long,  du  nez 

Larg.  du  nez 

Larg.  palpébrale 

Larg.  bicaroncul 

Larg.  buccale       

^  Long,  des  oreilles     

mm 

176 

130 

183 

108 

137 

75 

36 

41 

60 
62 

mm. 
184 
137 
182 
107 
136 
75 
45 
46 
28 
3H 
60 
65 

mm 

180 

J37 

185 

10! 

131 

82 

46 

38 

31 

37 

40 

52 

mm. 

175 

139 

164 

101 

127 

63 

46 

35 

24 

36 

51 

59 

mm 

170 

144 

161 

101 

128 

60 

41 

34 

35 

39 

49 

61 

mm. 
182 
139 

^Taille ... 

Gr.  enverg 

Haut,  du  cond.  audit.    ... 

Haut,  du  menton     

Haut,  de  Tacrom 

Haut,    de   l'extrémi.    du 
médius 

Haut,  de  l'ombil 

Haut,  du  gr.  troch    

Haut,  du  vertex,  dans  la 
posit,  as&ise     

Dist.  bi-aciomiale      

Circonf.  de  la  poit 

Long,  du  memb.  sup.     . . . 

Long,  du  memb.  inter.  ... 

Circonf.  du  bras 

Cii'c.  de  la  cuisse       

Cire,  du  mollet    

Long.  tot.  du  pied   

^Lars:.  max.  du  pied 

cm. 
161,0 
171,0 

cm. 
159,0 
165,5 

84,5 

84,0 
43,0 

85,5 

31,5 

cm. 
165,0 
173,0 

85,0 

cm 
154  5 
156,5 
141.(1 
132,5 
130,5 

56,5 
88.0 
87,5 

86.5 
42,5 
84.0 
74,0 
87,5 
26.0 
46  5 
31,0 
23.(1 
8.5 

CDl 

154,5 
156.5 
144,0 
135,0 

55,5 
92,5 
91,5 

88,5 
38.5 

91,5 
24,5 
465 
32.5 
210 
10,0 

cm 
163,0 
165,5 

85,0 
25,0 

cm. 
155.0 
165,0 

Indice  céphal 

Indice  du  visage 

Indice  facial 

Indice  nasal 

73,9 

74,9 

126,9 

113,9 

74,5 

74,7 

127,1 

102.2 

76,1 

69,3 

127.2 

82,8 

79,4 

77.4 

125,7 

76,1 

84,7 

79  5 

126,7 

82,9 

79,7 

73,1 

112,6 

104.7 

76,4 

p 

^Rap.  entre  la  gr.  enverg. 

et  la  taille 

Rap.  de  la  haut,  du  cond. 

audit,  à  la  taille    

Rap.  de  la  haut,  du  ment. 

à  la  taille    

Rap.     de    la      haut,     de 

Tacrom  à  la  taille 

Raj).  de  la  haut,  du  méd. 

à  la  taille  

Rap.  de  la  haut,  de  l'omb. 

a  la  taille    

Rap.  de  la  haut,  du   gd. 

troch.  à  la  taille    

Eap.  de  la  haut,  du  ver- 
tex pos.  ass.  à  la  taille. 
Rap.  de  la  dist.  bi-acrom.  à 

la  taille 

Rap.    de   la  long,   du  m. 

sup.  à  la  taille 

''Raj).  entre  ces  2  long.  ... 

106.2 

104,1 

53,1 

52,8 
27,1 

104,9 
51,5 

101,3 
91,3 

85,8 

84,5 

36,6 

57,0 

56,6 

56,0 

27,5 

47,2 
84,6 

101,3 
93,2 

87,4 

35,9 
59,9 
59,2 
57,3 
24,9 

101,5 
52,'/ 

106,5 

60 


Art.  4. — Torii 


Numéros. 

8 

9 

10 

Il 

12 

J3 

14 

mm. 

lum. 

nmi 

mnj. 

mm. 

lum 

mm. 

'  Diam.  ant.  post.  max.    . . . 



169 



172 

18) 

174 



o 

Diam.  transv.  max 



149 



140 

153 

150 



-JJ 

Long.  tot.  du  visage 

191 



171 

173 



Long.  tot.  de  la  face 

100 





104 

104 





ni   c8 

Larg.  tût.  de  la  face 

125 

127 



127 

135 

_ 



Haut,  du  front    

91 





67 

69 



2  m 

Long,  du  nez 

42 

— 



42 

42 

_ 

__ 

ts-S 

Larg.  du  nez 

40 





41 

35 

— 

. 

Larg.  palpébrale       



29 

_ 

29 

32 

_ 

, 

Larg,  bicaroncul 

— 

40 

— 

30 

34 

— 

— 

0) 

Larg.  buccale      

47 

50 



55 

;8 

_ 



\  Long,  des  oreilles     

60 

60 

— 

62 

57 

- 

— 

cm 

cm. 

cm 

CUi. 

cm. 

cm 

cm. 

/Taille 

154,0 

— 

— 

160,0 

160,0 

168,0 

Grr.  enverg 

— 

— 

— 

165,0 

167,U 

176,5 



Haut,  du  cond.  audit.   ... 

— 

— 

— 

147,0 

149,5 

157,n 



■OQ 

Haut,  du  menton     

— 

— 

— 

137,5 

149,5 

149,0 



13 

Haut,  de  Tacrom 

— 

— 

_ 

130,5 

— 

143,5 



-1-3 

0) 

Haut,    de    Textrémi.    du 

o 
C 
g 

mécUus 

— 

— 

— 

58,0 

62,0 

61,5 



Haut,  de  l'ombil 

— 

— 

— 

94,5 

99,5 

103,5 



•*^  «J 

Haut,  du  gr.  troch 

— 

— 

— 

83,0 

S6,5 

100,0 

. — 

S  a 

Haut,  du  vertex,  dans  la 

posit,  assise      

— 

— 

— 

85,0 

83,0 

— 

— 

II 

"S 

Dist.  bi-acromiale     

— 

39,0 

— 

43,0 

- 

48,0 



Circonf .  de  la  poit 

— 

94,0 

— 

85,0 

— 

b6,0 

— 

^ 
ö 

Long,  du  memb.  sup.     ... 

— 

— 

- 

72,5 

— 

82,0 

— 

Long,  du  memb.  infér.  ... 

— 

— 

— 

83,0 

96,5 

100,0 

— 

« 
g 

Circonf.  du.  bras 

28,0 

27,0 

— 

26,0 

26,5 

26,5 

— 

Cire,  de  la  cuisse     

— 

47,0 

— 

48,5 

49,5 

— 



Cire,  du  mollet   

— 

33,0 

- 

32,5 

32,5 

32,0 

— 

Long.  tot.  du  pied 

^  Larg.  max.  du   pied     ... 

— 

24,5 

— 

27,0 

26,0 

26,5 

— 

~ 

10,0 

— 

9.0 

9,0 

95 

— 

CQ 

Indice  céphal 

_ 

87,6 

81,4 

85,0 

86,2 



S  J  lu  dice  du  visage 

65,4 

— 



74,3 

78,0 

- 

— 

^    Indice  facial 

125,0 





l:i2,l 

129,8 

— 

— 

l-H 

Indice  nasal 

95,2 

— 

— 

97.6 

83,3 

— 

/Eap.  entre  la  gr.  enverg. 
et  la  taille 

— 



_ 

103,1 

104,4 

105,1 

— 

Raj).  de  la  haut,  du  cond. 

audit,  à  la  taille 

— 

— 



91,9 

93,4 

93,8 

— 

Rap.  de  la  haut,  du  ment. 

à  la  taille    

_ 





85,9 

93,4 

88,7 

— 

Rap.     de     la    haut,     de 

l'acrom  à  la  taille 

— 

— 

— 

81,6 

— 

85,6 

— 

Raj).  de  la  haut,  du  méd. 

to 

à  la  taille  

— 

— 

— 

36,3 

38,8 

36,6 

— 

ft^ 

Rap.  de  la  haut,  de  l'omb. 

§: 

à  la  taille    ,  . 

— 



— 

59,1 

62,2 

61,6 

— 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille    

— 

— 

— 

51,9 

60,0 

59,5 

— 

Rap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 

— 

— 

— 

53,1 

51,9 

— 

— 

Rap.  de  la  dist  bi-acram  à 

la  taille 

— 

26,9 

— 

28,6 

— 

25,6 

— 

Kap.  de  la  long,   du  m. 

sup.  à  la  taille 

— 

— 

— 

45,3 

— 

48,8 

— 

^Raj).  entre  ces  2  long,   ... 

— 

■ — 

— 

87,4 

8i',0               — 

Études  Anthropologiques. 


61 


Numéros. 

15 

16 

17 

18 

19            20 

21 

lUIQ 

mm 

mm 

mm 

mm 

mm 

mm. 

„       /Diam.  an  t.  post.  max.    ... 

176 

— 

174 

175 

__ 



177 

Diam  transv.  max 

130 

— 

145 

141 

_ 

139 

-»^ 

LoDg.  tot.  du  visage 

182 

— 

176 

172 



184 

c3     , 

Long.  tot.  de  ]a  face 

107 

— 

104 

101 



103 

Larg.  tot.  de  la  face 

121 

— 

130 

124 

_ 



116 

2  cä 

/  Haut,  du  front    

75 

— 

72 

71 

._ 

81 

§'! 

\  Long,  du  nez 

34 

— 

45 

50 

_ 

41 

1^ 

Larg.  du  nez 

33 

— 

41 

39 





41 

Larg.  palpébrale        

30 

— 

28 

31 





33 

œ 

01 

Larg.  bicaroncul 

37 

— 

37 

36 



33 

Larg.  buccale       

VLong.  des  oreilles    

59 



57 

54 



52 

S 

69 

— 

61 

60 

— 

— 

57 

cm 

cm 

cm 

cm 

cm 

cm 

cm. 

/Tnille 

163,0 

— 

163,5 

156.0 

172,0 

155,0 

Gr.  enverg 

165,5 

— 

173,0 

157,0 



— 

157,5 

Haut,  du  cond.  audit.    ... 

— 



150,0 

143,0 





140,0 

O! 

Haut,  du  menton     

142,0 

__ 

141,5 

135,5 





132,5 

'S 

Haut,  de  l'acrom 

— 



137,5 

127,5 





131.5 

<» 

Haut,    de    l'extrémï.   du 

2   . 

médius 

58,5 

— 

61,5 

62,0 





55,0 

Haut.de  l'ombil 

98,5 

— 

93,5 

89,5 





92.5 

^J    00 

Haut,  du  gr.  troch   

93,5 





87,5 



_ 

87,5 

=1  ä 

Haut,  du  vertex,  dans  la 

c»   s 

posit,  assise      

86,5 

— 

88,0 

86,5 





85,5 

2  a 

Dist.  bi-acromiale     ..     ... 

— 

— 

- 

40,0 





39,0 

^ 

Circonf.  de  la  poit    

— 

— 

88,5 

81,0 





92,0 

s 

Long,  du  memb.  gup.     ... 

— 

— 

75,5 

65,0 

— 

. — 

76,0 

m 

p 

Long,  du  memb.  infér.  ... 

93.5 

— 

— 

87.5 





87,5 

03 

Circonf.  du  bras 

— 

— 

28,0 

25,0 





25,5 

Cire,  de  la  cuisse      

45,5 

— 

— 

49,5 



— 

51,0 

Cire,  du  mollet    

— 

— 

35,5 

33,0 

— 

— 

33,0 

Long.  tot.  du  pied    

24,5 

25.5 

— 

23,5 

— 

— 

29,5 

^Larg.  max.  du  pied 

9,5 

8,5 

— 

9.0 

— 

,5 

XQ 

Indice  céphal 

73,9 



83.3 

80,6 

_ 

78,5 

S     Indice  du  visage 

66,5 

— 

73,9 

72,1 

— 

— 

63,0 

'g  '  Indice  facial 

113,1 

— 

125,0 

122,8 

— 

— 

112.6 

h- 1 

Indicenasal 

97,1 

— 

91.1 

78,0 

— 

— 

100,0 

^Eap.  entre  la  gr.  enverg. 

et  la  tiille 

101,5 

— 

106,1 

100,6 

— 

— 

101,6 

Eap.  de  la  haut,  du  cond. 

audit,  à  la  taille    



— 

91,7 

91,7 

— 

90,3 

Rap.  de  la  haut,  du  ment. 

à  la  taille    

87,1 

— 

86,5 

8a.9 

— 

— 

85,5 

Eap.     de     la     haut,    de 

l'acrom  à  la  taille 



— 

84,1 

81,7 

— 

— 

84,8 

. 

Eap.  de  1  a  haut,  du  méd. 

CO 

à  la  taille    

35,9 

— 

37,6 

39,7 

— 

— 

35,5 

Rap.  de  la,  haut,  de  l'omb. 

à  la  taille    

60,4 

— 

57,0 

57,4 

— 

— 

59,7 

p^ 

Eap.  de  la  haut,  du  gd. 

troch.  à  la  taille    

57,4 

— 

— 

56,1 

— 

— 

56,5 

Eap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 

53,1 

— 

53,8 

55,5 

— 

— 

55,2 

Eap.  de  la  dist.  bi-acrom. 

à  la  taille    

— 

— 

— 

25,2 

— 

— 

24,6 

Rap.  de  la  long,  du  m. 

sup.  à  la  taille 

— 

— 

46,2 

41,7 

— 

— 

49,0 

^Eap.  entre  ces  2  long.   ... 

— 

~ 

— 

74,4 

— 

86,9 

62 


Art.  4— Torii  ; 


Numéros. 

22 

23 

24 

25 

26 

27 

28 

mm. 

mm. 

mm. 

miu. 

mm. 

mm 

mm. 

Diam.  ant.  p^st.  max.    ... 

— 

179 

1Ö4 

172 

— 

182 

176 

S 

1  iam.  transv.  max 



— 

— 

151 

— 

161 

140 

■-S 

Long.  tot.  du  visage 



170 

181 

164 

— 

189 

179 

Lonii.  tot.  de  la  face     ... 

— 

105 

106 

95 

— 

109 

106 

Larg.  tot.  de  la  face 

— 

121 

109 

125 

— 

125 

129 

Hiut.  dufront    

— 

65 

75 

69 

— 

80 

73 

Long,  du  nez 

— 

89 

41 

39 

— 

44 

42 

•H      O 

Larg.  du  nez 

— 

35 

37 

32 

— 

42 

41 

Larg.  palpébrale       

28 

32 

25 

33 

— 

33 

36 

CO 

Ijarg.  bicaroncul 

30 

35 

30 

28 

— 

32 

35 

0 

Larg.  buccale       ..     

— 

52 

52 

51 

— 

64 

56 

^  Long,  des  oreilles     

— 

59 

63 

46 

— 

4« 

56 

CQl. 

cm 

cm 

cm               cm. 

cm 

cm. 

/Taillft 

162,5 

156.0 

159,0 

— 

163,5 

171,5 

Grr.  enverg 

_ 

173,5 

160  0 

169,0 

— 

167.5 

176,5 

Haut,  du  cond.  audit.    ... 



145,0 

140,5 

144,5 

— 

143,5 

153,0 

CO 

Haut,  du  menton      



139,5 

131,5 

137,0 

— 

137,5 

145,0 

-ö 

Haut,  de  l'acrom 



135,0 

127,5 

130,5 

— 

134,5 

145,0 

0. 

Haut,    de    Textrémi.    du 

CJ 

médius 



— 

49,5 

62,5 

— 

60,0 

57.0 

§ 

Haut,  de  l'ombil 



96,0 

92,5 

94,5 

— 

97,5 

99,5 

^  œ 

Haut,  du  gr.  troch 

— 

95,0 

89,5 

91,5 

— 

93,5 

98,5 

Hiut.  du  vertex,  dans  la 

es 
m  a 

posit,  assise      



89,5 

84,0 

85,5 

— 

85,5 

88,5 

o  g 

Dist.  bi-acromiale     



40,0 

39,0 

41,0 

— 

42,0 

41,0 

^ 

Circonf.  de  la  poit 



89,0 

85,0 

85,0 

— 

9i,0 

89,0 

g 

Long,  du  memb.  äup.     ... 



— 

78,0 

68,0 

— 

77,5 

— 

00 

Long,  du  memb.  infér.  ... 



95,5 

89,5 

91.5 

— 

93,5 

98,5 

Circonf.  du  bras 



2K,0 

'25.0 

26,0 

— 

27,0 

29,0 

S 

Cire,  de  la  cuisse 



50,0 

47,0 

47,0 

— 

50,0 

48,5 

Cire,  du  mollet    



33.0 

29,0 

32,0 

— 

33,0 

32,0 

Long.  tot.  du  pied   

— 

25,5 

24,5 

24,5 

— 

25,0 

29,0 

1  Larg.  max.  du  pied 

— 

10. <> 

9.0 

9  5 

— 

10.0 

10.0 

m  j 

Indice  céuhal 



— 



87,8 

— 

88,5 

79,6 

.1]  Indice  du  visage 



71,2 

60,2 

76,2 

— 

66,1 

72,1 

'S  /  Indice  facial 



115,2 

102,8 

131,5 

— 

114,7 

121,7 

^ 

Indice  nasal 

— 

89,7 

90.2 

82.1 

— 

95.5 

97,6 

/Rai3.  entre  la  gr.  enverg. 

et  la  taille 



106,8 

102,5 

106,3 

— 

102,4 

102,9 

Eap.  de  la  haut,  du  cond. 

audit,  à  la  taille    

— 

89,2 

90,1 

90,9 

— 

87,8 

89,2 

Rap.  de  la  haut,  du  ment. 

à  la  taille    

— 

85,7 

84,3 

86,2 

— 

84,1 

84,6 

Rap.     de     la     haut,    de 

Tacrom.  à  la  taille 

— 

83,1 

81,7 

82,1 

— 

82,3 

84,6 

Rap.  de  la  haut,  du  méd. 

oo 

à  la  taille    

— 

— 

31,7 

39,3 

— 

36,7 

33,2 

o 

Rap.  de  la  haut,  de  l'omb. 

è 

à  la  taille    

— 

59,1 

59,3 

59,4 

— 

59,6 

58,0 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille    

— 

58,5 

57,4 

57,6 

— 

57,2 

57,4 

Rap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 

— 

55,1 

53,9 

53,8 

— 

52,3 

51,6 

Rap.  de  la  dist  bi-acr  jm. 

à  la  taille    

— 

25,0 

25,8 

25,7 

— 

23,9 

25,5 

Eap.  de  la  long,    du   m. 
sup.  à  la  taille 

_ 



500 

42,8 



47,4 

— 

^Rap.  entre  ces  2  long.    ... 



~ 

87,2 

74,3 

' 

82,9 

Études  Anthropologiques. 


63 


Numéros. 

29 

30 

31 

32 

33 

34 

1      35 

1 

mm 

mm. 

mm. 

mm. 

mm. 

mm 

mm. 

S 

^ 

'  Diam.  ant.  post.  max.  . . . 

— 

175 

178 

176 

175 

178 

184 

Diam.  transv.  max 

— 

145 

145 

139 

141 

140 

153 

Long.  tot.  du  visage     ... 

— 

— 

165 

172 

177 

167 

181 

c3 

L-^ng.  tot.  de  la  face 

— 

110 

103 

107 

112 

104 

106 

0)   o 

Larg.  tot.  de  la  face 

— 

132 

127 

122 

114 

135 

139 

05      - 

Uaut.  du  front    

— 

— 

62 

65 

65 

63 

75 

2    CD 

Long,  du  nez 

— 

49 

45 

41 

53 

41 

42 

l-ë 

LarJ^.  du  nez 

— 

50 

37 

38 

42 

42 

50 

2^ 

Larg.  palpébrale       

— 

— 

30 

30 

36 

37 

37 

Larg.  bicaroncul 

— 

— 

39 

35 

44 

39 

40 

a) 

Larg.  buccale      

— 

57 

51 

46 

55 

53 

59 

,  Long,  des  oreilles     

55 

57 

57 

65 

67 

59 

68 

cm. 

cm 

cm 

cm 

cm. 

cm 

cm. 

/Taille 

163,5 

— 

156,5 

_ 

165,5 

157,5 

169,0 

Gr.  enverg 

162,5 

— 

169,5 

— 

174,5 

163,5 

177,0 

Haut,  du  coad.  audit.    ... 

147,0 

— 

143,0 

151,0 

154,0 

145,0 

156,0 

03 

Haut,  du  menton      

— 

— 

135,5 

141,5 

152,5 

137,5 

147.5 

■^ 

Haut,  de  l'acrom 

— 

— 

123,5 

133,5 

135,0 

13S,5 

— 

+2 

Haut,    de    l'extrémi.    du 

o 

d 

o 

médius 

— 

— 

56,0 

60,5 

54,0 

57,5 

65,0 

Haut,  de  Tombil 

— 

— 

— 

94,5 

102,5 

92,0 

101,5 

Haut,  du  gr.  troch    

— 

— 

90,0 

93,0 

100,0 

88,0 

98,0 

-1' 

2  § 

Haut,   du  vertex,  dans  la 

posit,  assise 

— 

— 

8S,5 

90,5 

92,0 

87,0 

93,5 

d  S 
•2  '^ 

Dist.  bi-acromiale     

— 

39,0 

40,0 

44,0 

— 

43,5 

43,5 

r^ 

Circonf.  de  lapoit 

— 

93,5 

— 

83,5 

89,0 

86,0 

91,0 

2 

3 

Long,  du  memb.  sup.     ... 

— 

— 

67,5 

— 

72,5 

77,5 

73,0 

Long,  du  memb.  infer.  ... 

— 

— 

90,0 

93,5 

1000 

8S,0 

91,5 

11 

Circonf.  du  bras 

— 

28,5 

27,0 

— 

20,5 

27,5 

26,0 

Cire,  de  la  cuisse       

— 

49,0 

49,0 

43,0 

47,5 

48,0 

49,0 

Cire,  du  mollet    

— 

35,5 

3(,0 

32,0 

35,0 

33,0 

33,5 

[jong.  tot.  du  pied    

— 

25,5 

24,0 

26,5 

— 

24,5 

25,0 

^Larg.  max.  du  pied 

— 

10,5 

9,5 

9.0 

— 

10,0 

10.0 

EQ 

Indice  céphal 



82,9 

81,5 

79,0 

80,6 

78,6 

83,2 

O     Indice  du  visage 

— 

— 

77,0 

70,9 

64,4 

80,8 

76,8 

'^1  Indice  facial 

— 

120,0 

123,3 

114,0 

101,8 

129,8 

131,1 

d 

h- 1 

Indice  nasal  

— 

102.0 

82.2 

92.7 

79.2 

102.4 

119,0 

y  Rap.  entre  la  gr.  enverg. 

et  la  taille 

93,9 

— 

103,3 

— 

105,4 

103,8 

104,7 

Rap.  de  la  haut,  du  cond. 

audit,  h  la  taille    

89,9 

— 

91,4 

93,1 

92,1 

92,3 

94,3 

Rap.  de  la  haut,  du  ment. 

à  la  taille    

— 

— 

86,6 

92,2 

87,3 

87,3 

86,1 

Rap.     de     la     haut,    de 

racrom  à  la  taille 

— 

— 

78,9 

85,7 

82.0 

83,6 



03 

Rap.  de  la  haut,  du  méd. 

O 

à  la  taille    

— 

— 

35,8 

32.6 

36,5 

38  5 

35,6 

|< 

Rap.  de  la  haut,  de  l'omb. 

à  la  taille    

— 

— 

— 

61,9 

58,4 

60,1 

57,7 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille    

— 

— 

57,5 

60,4 

55,9 

58,0 

56,8 

Rap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 

— 

— 

56,6 

55,6 

55,2 

55,3 

55,2 

Rap.  de  la  dist.  bi-acrom. 

à  la  taille    

— 

27,6 

25,7 

23,3 

— 

23,8 

26,0 

Rap.  de  la  long,  du   m. 

sup.  à  la  taille 

— 

— 

43,1 

— 

43,8 

49,2 

43,2 

^Rap.  entre  ces  2  long.   ... 

■ — 

— 

75,0 

— 

72,5 

88,1 

79,8 

64 


Art.  4.— Torii  : 


Numéros. 

36 

37 

38 

39 

40 

41 

42 

jym 

mm. 

mm 

mm 

mm 

mm 

mm. 

,  Diam.  ant.  post.  max.    ... 

174 

— 

— 

177 

180 

182 

182 

-2 

'S 

Diam.  transv.  max 

140 

— 

— 

146 

140 

145 

138 

Long.  tot.  du  visage 

172 

— 

— 

176 

198 

184 

191 

es    , 

Long.  tit.  de  la  face 

103 

— 

— 

107 

m 

112 

111 

01    u 

Larg.  tot.  de  la  face 

122 

- 

— 

130 

128 

138 

133 

"5( 

13     QJ 

Haut,  du  front    

69 

— 

— 

69 

87 

72 

80 

Long,  du  nez 

43 

— 

— 

45 

43 

47 

45 

Larg.  du  nez 

41 

— 

— 

46 

40 

42 

39 

Larg.  palptîbrale       

33 

— 

— 

33 

32 

31 

31 

to 

Ö 
0) 

Larg.  bicaron  cul 

34 

— 

— 

36 

34 

38 

33 

Larg.  buccale      

55 

— 

— 

— 

41 

63 

54 

s 

Long,  des  oreilles     

63 

— 

— 

64 

59 

58 

63 

cm- 

cm. 

cm 

cm 

ciu 

cm. 

cm. 

/Taille 

158,5 

— 

— 

159,5 

159,5 

— 

165,5 

Gr.  enverg 

161,0 

— 

- 

162,5 

163,0 

— 

173,5 

Haut,  du  cond.  audit.    ... 

149,5 

— 

— 

— 

— 

— 

— 

Haut,  du  menton     

136,ri 

— 

— 

— 

— 

— 

— 

-0 

Haut,  de  l'a  crom 

132,5 

— 

— 

— 

— 

— 

— 

Haut,    de    l'extrémi.    du 

o 

médius 

55,5 

— 

— 

— 

— 

— 

— 

2«> 

Haut,  de  l'ombil 

91,5 

— 

— 

— 

— 

— 

— 

Haut,  du  gr.  troch    

— 

— 

— 

— 

— 

— 

— 

TS  g  i 

Haut,  du  vertex,  dans  la 

posit,  assise      

87,5 

— 

— 

— 

— 

— 

— 

§a 

Dist.  bi-acromiale     

37,0 

- 

— 

38,0 

— 

— 

43,0 

•i-i 

Circonf .  de  la  poit    

— 

— 

— 

86,0 

— 

— 

91,0 

1 

Long,  du  memb.  sup.     ... 

77,5 

— 

— 

— 

80,0 

— 

— 

Long,  du  memb.  infér.  ... 

— 

— 

— 

— 

— 

— 

— 

ai 

Cii"co:f.  du  bras 

25,0 

— 

— 

27,5 

— 

25,5 

27,5 

g 

Cire,  de  la  cuisse       

44,5 

— 

— 

45,5 

50,0 

46,0 

47,5 

Cire,  du  mollet   

30,5 

— 

— 

32,0 

31,0 

— 

33,0 

Long.  tot.  du  pied    

25,5 

— 

— 

25,0 

25,0 

25,5 

25,5 

^  Larg.  max.  du  pied 

8,5 

-- 

— 

10,0 

9.5 

8.5 

10,0 

rn 

Indice  céphal 

80,5 





82,5 

77,8 

79,7 

75,8 

.^     Indice  du  visage 

70,9 

— 

— 

73,9 

64,6 

75,0 

69,6 

'^  '  Indice  facial 

118.4 

— 

— 

1215 

115,3 

123,2 

119,8 

>-i     Indice  nasal 

95.3 

— 

— 

102,2 

93,0 

89.4 

86.7 

^Rap.  entre  la  gr.  enverg. 

et  la  taille 

101,6 

— 

— 

101,9 

102,2 

— 

104,8 

Rap.  de  la  haut,  du  cond. 

audit,  à  la  taille    

90,7 

— 

— 

— 

— 

— 

— 

Rap.  de  la  haut,  du  ment.- 

-     .  -, 

à  la  taille    

85,1 

— 

— 

— 

— 

— 

— 

Rap.     de    la     haut,     de 

l'acrom.  à  la  taille 

81,1 

— 

— 

— 

— 

— 

— 

œ 

Rap.  de  la  haut,  du  méd. 

f-< 

à  la  taille    

31,1 

— 

— 

— 

— 

— 

— 

l< 

Rap.  de  la  haut,  de  l'omb. 

i' 

à  la  taille    

57,5 

— 

— 

— 

— 

— 

— 

P^ 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille    

— 

— 

— 

— 

— 

— 

— 

Rap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 

53,4 

— 

— 

-- 

— 

— 

— 

Rap.  de  la  dist.  bi-acrom. 

à  la  taille    

24,5 

— 

— 

24,3 

— 

— 

26,2 

Rap.  de  la  long,  du  m. 

sup.  à  la  taille 

48,9 

— 

— 

— 

50,2 

— 

— 

^Rap.  entre  ces  2  long.  ... 

Études  Antbropologiques. 


65 


Numéros. 


0,5 


/  Diaui 
l 'iau!. 
Long. 
Long-. 
Larg. 
Haut. 
Long-. 
Larg. 
Larg-, 
Larg. 
T.arg". 
Lonof. 


.  ant.  post.  max. 

transv.  max.  ... 

tot,  du  visage 

tot.  de  la  face 
tot.  de  la  face 
du  fi-ont    

du  nez 

du  nez 

p^ilpébrale 

bioaroncul 

buccale 

des  oreille-!     ... 


^Taille 

Gr.  envèrg 

Ilaut.  du  cond.  audit.  ... 

Haut,  du  menton     

Haiit.  de  Tacrom 

Haut,   de    l'extrémi.    du 

médius  ... 

Haut,  de  l'ombil 

Haut,  du  gl',  troch    

Haut,  du  vertex,  dans  la 

posit,  assise      

Dist.  bi-acromiale     

Circonf .  de  la  poit 

Long,  dvi  meml).  sup.  ... 
Long,  du  memb.  infér.  ... 

Circonf.  du  bras 

Cire,  de  la  cuisse       

Cire,  du  mollet    

Long.  tot.  du  pied    

.Larg.  max.  du  pied 


43 


mm 

183 

140 

182 

113 

139 

65 

48 

40 

34 

38 

54 

66 


143,5 
135,0 
130,5 

50,0 
92,5 
92,5 


90,0 


50,5 
36,0 
25,5 
10,0 


44 


45 


46 


43,0 


47 


iiiin 

188 

138 

177 

112 

136 

65 

44 

34 

39 

35 

51 

56 


cm 
161,0 
170,0 
146,0 
137,0 
130,5 

50,0 
92.5 
91,5 

86,0 
39,5 
89,5 
82,5 
90,0 
26,5 
48.5 
33,0 
25,0 
9,0 


48 


mm 

174 

138 

180 

105 

125 

75 

40 

42 

29 

31 

53 

56 


cm 
158,5 
160,5 
141,0 
135,5 
132,5 

52,5 
95,5 


83,0 
3S,5 
83,5 
79,5 
90,0 
23,5 
45,0 
31,0 
24,0 
9,5 


49 


Indice  céphal 

S  J  Indice  du  visage... 
'Ö  j  Indice  facial 

Indice  n^sal 


/Rap.  entre  l;i  gj*.  enverg. 

et  la  taille 

Rap.  de  la  haut,  du  cond. 

audit,  à  la  taille    

Eap.  de  )a  haut,  du  ment. 

à  la  taille    

Eap.     de     la     haut,    de 

l'acrom  à  la  tai  lie 

Rap.  de  la  haut,  du  méd. 

à  la  taille    

Raj:).  de  la  haut,  de lomb. 

à  la  taille 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille  

Rap.  de  \%  haut,  du  ver- 
tex 130S.  ass.  à  la  taille. 
Eap.  de  la  dist.  bi-acrom. 

à  la  taille    

Eap.  de   la  long,  du  m. 

sup.  à  la  taille 

Rap.  entre  ces  2  long.   ... 


76  5 

76,5 

123,0 

83.3 


1^ 


89,0 
85,5 
83,6 
33,1 
60,3 


24,9 


73,4 

76,8 

121,4 

77,3 


79,3 

69,4 

119,0 

105.0 


105,6 
92,3 
87,1 
83,1 
36,0 
59,4 

54,3 

23,5 

51,2 
91,7 


101,3 
89,9 
84,2 
82,0 
34.8 


51,7 

23,8 

50,2 
88,3 


66 


Art.  4.— Torii  : 


Numéros. 

50 

51 

52 

53 

54 

55 

56 

inin. 

mai 

miu. 

lllUl. 

nui). 

iiiiii 

Ullll. 

o 

^ Uiaui.  ant.  post.  max.    ... 

177 

180 





1V5 

-l-i 

Diam.  trausv.  max 

— 



_ 



138 

+3 

Long.  tot.  du  visage 

183 

176 

__ 

_ 

_ 

181 

■—1    O 

Long.  tot.  de  la  face 

107 

104 



_ 

_ 

106 



Larg.  tot.  de  la  f^ice     ... 

J25 

137 





134 

135 



s  e  i 

Haut,  du  front    

76 

72 



71 

Long,  du  nez 

4ti 

45 



51 

39 



"^  "* 

Larg.  du  nez 

40 

43 





41 

42 



Ë  o 

Larg.  palpébrale       

34 

33 

29 

30 

31 

34 



!  arg.  bicai-oncul 

34 

34 

32 

29 

42 

3! 



CD 

Larg.  buccale       

\  Long,  des  oreilles 

!■)? 

55 



54 

r»7 



«2 

58 

— 

59 

61 

r^H 

— 

CUl. 

tin. 

cm. 

cm. 

OUI 

cm. 

cm. 

/'l'aille •■■   

162,5 

160.5 



_ 

162,0 

155.5 



(jr.  enverg 

16.5.5 

167,5 



168  0 

157,5 



Havit.  du  cond.  audit.    ... 

15'\0 

Haut,  du  menton     

141,5 

_ 

ns 

Haut,  de  Tacr  im      

1:55.0 





■+3 

Haut,    de    l'extrémi.    du 

o 

sa 

médius 

58,.=^ 

_ 



o 

Haut,  de  rombil 

96,5 

Haut,  du  gr.  troch    

'S -2 

Haut,  du  vertex,  dans  la 

•S  ^ 

Ijosit  assise       

81,0 

_ 

_ 

Dist.  bi-acromiale     

42,5 

40  0 





3S,0 

37,0 



1^  ■ 

Circonf .  de  la  poit 

84.0 

82,0 



86.0 

87,0 



Tiong.  du  memb.  sup.     ... 

76,5 







Long,  du  memb.  iofér.... 



O 

Ci  rconf .  du  bras 

24,0 

26,5 



26,0 

27,0 



f^ 

<'irc.  de  la  cuisse       

47,0 

49.5 



. 

45,0 

50,0 



Cire,  du  mollet    

31,') 

30  0 





31,0 

32.0 



1 

Long.  tot.  du  pied  

24,5 

24,5 





25,5 

i'5.5 



l 

,  Lars".  max.  du  pied 

90 

9.5 

— 

— 

9.0 

9.5 

— 

»j 

Indice  céphal 



74,6 

2     Ind'ce  du  visage 

68,3 











'g  '   Indice  facial 

116,8 

131,7 

_ 

_ 



127,4 

_ 

>— ( 

Indice  nasal  

87.0 

95.6 

— 

— 

8^.4 

107.7 

— 

fRap.  entre  la  gr.  enverg. 

!'      et  la  taille 

101.8 

104,4 

._ 

_ 

103,7 

101,3 



Kap.  de  1+  haut,  du  cond. 

audit,  à  la  taille    

Rap.  de  la  haut,  du  ment. 

à  la  taille  

_ 

Rap.     de    la     haut,     de 

l'acrom.  à  la  taille 



. 

Rap.  de  la  haut,  du  méd. 

o 

à  la  taille 





_ 

a.' 

Rap.  de  la  haut,  de  l'omb. 

^ 

à  la  taille    





_ 

f-H 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille    



._ 







Rap.  de  la  haut,  du  ver- 

tex pes.  ass.  à  la  taille. 







_ 



Rap.  de  li  dist.  bi-acrom. 

à  la  taille    .. 

25,0 

26,2 

_ 



25,5 

24,7 

— 

Rap.  de  la  long,   du  m. 

sup.  .à  la  taille 

47,1 







_ 





\Eap.  entre  ces  2  long.    ... 

— 

— 

— 

— 

— 

— 

— 

Etudes  Authropologinvep, 


67 


Nuinevop. 

57 

58 

59 

60 

61 

ez 

<ôà 

luiu 

mill 

11. m 

i,iiii 

IMIIJ 

mm 

iiini. 

(  Diam.  an  t.  post.  max.     ... 

\li 

171 

184 

18i 

179 

179 

170 

-S 

1  Diam,  transv.  max 

13S 

132 

143 

145 

13  > 

139 

130 

-i3 

Long-,  tot.  du  visage 

— 

174 

ISG 

189 

171 

174 

175 

^^    O 

Long.  tot.  de  la  fa.ce 

107 

101 

108 

111 

107 

103 

lü5 

1-    o 

Larg.  tot.  de  la  face 

122 

124 

127 

138 

130 

128 

128 

Haut,  du  front    ..-   

— 

73 

78 

78 

64 

71 

70 

Long,  du  nez 

44 

38 

41 

45 

37 

42 

43 

Larg.  du  nez 

43 

42 

3S 

41 

38 

44 

37 

ë-ë 

Larg.  palpébralo        

31 

31 

31 

33 

i8 

33 

34 

Larg.  bicaroncul 

35 

29 

34 

35 

32 

35 

32 

Larg.  buccale      

54 

54 

47 

49 

56 

56 

52 

S 

\  Long,  des  oreille  =■       

67 

63 

61 

•=•1 

54 

65 

62 

(JIU 

cal 

cm. 

Ciu 

cm 

cm 

cm. 

/Taille 

l.ï4.,0 

162.0 

157,0 

162.0 

154,0 

161,0 

151,0 

Gr.  enverg 

IG>,0 

170,0 

165,0 

169,0 

160,0 

171,0 

161,0 

Haut,  du  cond.  audit.    ... 

— 







— 

— 

— 

50 

Haut,  du  men  toi     

— 





_ 

— 

— 

— 

t) 

Haut,  de  Tacrom      

— 



— 

— 

— 

— 

— 

-P 

Haut,    de   l'extrîmi.    du. 

O 

médius 

— 

— 

— 

— 

— 

— 

— 

g 

Haut,  de  lombil 

— 

_ 



— 

— 

— 

^ 

Haut,  du  gr.  troch    ...    ... 

— 

— 

— 

— 

— 

— 

— 

—  ^ 

Haut,  du  vertex,  dans  la 

'^  3  ' 

posit,  assise     ...  

— 

— 

— . 

— 

— 

— 

•2  ^ 

Dist.  bi-acromiale     

38,5 

43.5 

40,0 

— 

38,0 

33,5 

3J,5 

Circonf .  de  la  poit    

87,0 

87,5 

88,5 

— 

78,0 

84,5 

86,0 

a 

Long,  du  me mb.  sup.     ... 

— 

— 

— 

— 

— 

— 

— 

m 

Long,  du  memb.  infJr.  ... 

— 

— 



— 

— ' 

— 

— 

a 

Circonf.  du  bras 

25.0 

25,0 

2(5,5 

27,0 

23,5 

24,0 

26,0 

s 

Cire,  de  la  cuisse       

47,0 

44,0 

46,0 

43,0 

44,5 

49.0 

48,0 

Cire,  du  mollet    

33,5 

31,0 

31,0 

31,0 

30,0 

35,0 

33.0 

LoDg.  tot.  du  pied    

24,5 

25,0 

24,5 

25,5 

25,5 

26  5 

24,5 

^Laro-,  mix.  du  pied  ... 

n.o 

8.0 

85 

10.0 

10.0 

90 

9,5 

=«•( 

'  Indice  céphal 

80,6 

77,2 

77,7 

78,3 

77,6 

77,6 

70,6 

5  l  Indice  du  visage 

— 





— 

— 

— 

— 

^  J  Indice  facial 

114.0 

122,9 

1 17,6 

124,3 

121,5 

124,3 

121,9 

^( 

Indice  nasal  

97.7 

110.5 

92.7 

911 

102.7 

104  8 

86.0 

/Rap.  entre  la  gr.  enverg. 

et  la  taille 

107,1 

101,9 

105,1 

104,3 

103,9 

103,2 

106,6 

Rap.  de  la  haut  du  cond. 

audit,  à  la  taille    







— 

—  - 

— • 

Rap.  de  la  haut,  du  ment. 

à  la  taille   

— 





— 

— 

— 

— 

Rap.     de     la     haut,    de 

Tajroui.  à  la  tai lie 



_ 





— 

— 

— 

M 

Eap.  de  la  haut,  du  méd. 

-U 

à  la  taille  

— 





^- 

— 

— 

^ 

Rap.  de  la  haut,  de  Tomb. 

Cu\ 

à  la  taille    

— 





— 

— 

— 

— 

M    ; 

Rap.  de  la  haut,  du  gd. 

troch.  à  la  taille 

— 





— 

— 

— 

— 

Rap.  de  la  haut,  du  ver- 

tex, pos.  ass.  à  la  taille. 

— 





— 

— 

— 

— 

Rap.  de  la  dist.  bi-acrom. 

à  la  taille    

•-'3,9 

26,2 

23,5 

— 

24,3 

26,9 

— 

Rap.  de  la  long,   du  m. 

sup.  à  la  taille 

—  1 



— 

— 

— 

— 

— 

> 

Rap.  entre  ces  2  long.  ... 

" 

— 

— 

i 

~ 

68 


Art,  4.— Toi  ii: 


Numéros. 

64 

65 

66 

67 

68 

69 

70 

niiu 

mm 

mm 

mm 

ujm 

lUlIl 

n.tû. 

a> 

,''  Uiam,  ant.  post.  max.    ... 

176 

172 

]8S 

•o 

Diam.  transv.  max 

111 

139 

144 

-4-i 

Lont!'.  tot.  du  visage     ... 

176 

186 

191 

177 

'~'  6 

Long.  tot.  de  la  face     ... 

94 

110 

116 

100 

_ 

"^M 

Larg.  tot.  de  Iq,  face 

12i 

136 

123 

136 

)  Haut,  du  front    

Long,  du  nez 

82 
37 

76 
39 

7^ 
47 

77 
40 

- 

- 

— 

1'^ 

Larg.  du  nez 

40 

43 

44 

41 



D    0) 

Larg.  loalpébrale       

34 

32 

35 

36 

§ 

Lai'g.  bicaroncul 

35 

35 

38 

32 

<D 

'S 

Larg.  buccale       

46 

49 

61 

52 



r=H 

,  Lono-.  des  oreilles     

.59 

59 

56 

59 

— 

— 

— 

/Taille 

Gr.  enverg 

cm. 

cm. 
170,0 
178,0 

cm. 
164,5 
169,5 

OUI. 

158,0 

cm 

cm. 

cm. 

CO 

a; 
-(J 

Haut,  du  cond.  audit.    ... 

Haut,  du  menton     

Haut,  de  lacrom 

Haut,    de    1  extrémi.    du 

142,0 
133,0 
129,5 



— 



o 
d 

2   . 

médius 

Haut,  de  lombil 

- 

- 

- 

55,0 

- 

— 

- 

-&S 

Haut,  du  gr.  trocli    

3    t' 
CO    ?   ■ 

.2    ^    : 

Haut,  du  vertex,  dans  la 

posit,  assise      



List,  bi-acromiale     

3,70 

40,0 

40,0 

42,5 

_ 

"S 

Circonf.   delà  poit 

8,05 

91,0 

84,0 



p 

Long,  du  memb.  sup.     ... 



. 

74  5 

S 

Long,  du  memb.  infer.  ... 



® 

^ 

Circonf.  du  bras 

22.0 

26,5 

26,0 

28,0 

z 



Circ.  de  la  cuisse      

Cire,  du  mollet 

39,5 

28  5 

49,0 
32,0 

46,0 
33,0 

48,5 
32,0 

— 

— 

— 

1  Jjong.  tot.  du  pied    

\  Lars',  max.  du  pied 

^  f  Indice  céphal 

23.5 
9..Ö 



25,0 
9.0 

25,0 
8,0 

— 

— 



80,1 

80,8 

76,6 

.^  _  Indice  du  visage 





'^1  Indice  facial 

131,9 

123,6 

110,3 

136,0 



^  i^    ^• 

indice  nasal  

108.1 

110.3 

93,6 

102.5 

— 

— 

— 

^Bap.  entre  la  gr.  enverg. 

et  la  taille 



104,7 

103,0 

_ 

Eap.  de  la  haut,  du  cond. 

audit,  à  la  taille 

Kap.  de  la  haut,  du  ment. 

à  la  taille 

Eap.     de     la     haut,     de 

l'acrom.  àlataille 



'f- 

Eap.  de  la  haut,  du  méd. 

1 

il  la  taille    



o 

Bap.  de  la,  haut,  de  l'omb. 

a 

à  la  taille    

_ 



_ 

Ph 

Eap.  de  la  haut,  du  gd. 

troch.  à  !a  taille   





Eap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 





_ 



Eap.  de  la  dist.  bi-acrom. 

à  la  taille    





Eap.  de  la  long,   du   m. 

sup.  à  la  taille 







47,1 



— . 

^ Eap.  entre  ces  2  long.  ... 

— 

— 

— 

— 

— 

— 

I 


Etudes  Anthropoloü'iqnes. 


69 


Numéros. 


3    Ol 


f  Diain 

Diani, 

Long-. 

Lono-. 

Larg. 

Haut. 

Long. 
\  Larg. 

Larg. 

Larg. 

Larg. 
\  Loner. 


.  ant.  jxist.  max. 

,  transT.  max.    . 
tot.  (lu  visage 
tot.  de  la  face 

tot.  de  la  face 

du  front    

du  nez 

du  nez. 

palpébrale 

bicaroncul...     .. 

buccale      

des  oreille^     ... 


o  a 


Taille 

Gr.  enverg 

Haut,  du  cond.  audit.    ... 

Haut,  du  menton     

Haut,  de  Tacrom 

Haut,    de    l'extrémi.    du 

médius 

f  Haut,  de  Tombil 

Haut,  du  gr.  troch    

Haut,  du  vertex,  dans  la 
posit,  assise      

Dist.  bi-acromiale     

Circonf .  de  la  poit 

Long,  du  memb.  sup.     ... 

Long,  du  memb.  inf«r.  ... 

Circonf.  du  bras 

Cir(3.  de  la  cuisse       

Cire,  du  mollet    

,  Long.  tot.  du  i-)ied    

\Larj-.  max.  du  pied 

f  Indice  céphal 

Indice  du  visage 

Indice  facial 

Indice  nasal 


71 


/RajD.  entre  la  gr.  enverg. 

et  la  taille 

Rap.  de  la  haut,  du  cond. 

audit,  à  la  taille    

Rap.  de  U  haut,  du  ment. 

à  la  taille    

Ri.p.     de     la     haut,    de 

racrom.  à  la  taille 
.     Rap.  de  la  haut,  du  méd. 

à  la  taille    

Rap.  de  la  haut,  de  l'omb. 

à  la  taille   

^     Rap.  de  la  haut,  du   gd. 

troch.  à  la  taille    

Rap.  de  la  haut,  du  ver- 
tex pos.  ass.  à  la  taille. 
Rap.  de  la  dist.  bi-acx-om. 

à  la  taille  

Rap.   de  la  long,  du  m. 

sup.  à  la  taille 

VRap.  entre  ces  2  long.  ... 


72 


73 


74 


muj 

189 

143 

187 

113 

129 

74 

42 

36 

31 

38 

51 

«1 


39,0 


48  5 
33,5 

24,5 
10.0 


75.7 

114,2 
85.7 


75 


76 


77 


83,0 


70 


Art.  4.— Torii 


Numéro-. 

78 

79 

eo 

81 

82 

£3 

84 

llIUi 

mill 

mai 

11111,. 

uiai.;            mm. 

mm. 

/Diam.  ant.  post.  max.    ... 

— 

— 

— 

— 



— 



-M 

Dism.  trans V.  max 















■"■  a. 

Long.  tot.  du  visage 















Long.  tot.  de  la  face 









_ 





0)    Ü 

'0^ 

l.arg.  tot.  de  la  face 

— 













Ö  ^' 

Haut,  du  front    

, 

— 









.o '^  )  long,  du  Lez 



- 





_ 





"ë'^  ä  Lai-o-.  du  nez 

—  1             

— 









3     * 

Larg.  palpéljrale       

— 

— 

— 

— 



— 

_ 

Larg.  bicaroncul 

— 

— 

— 









Ol 

13; 

Larg.  buccale       















f^ 

..Long-,  des  oreilles     

—  !        — 

— 

— 

— 

— 

— 

/Taille 

Gr.  euverg 

cm.              cui. 

cm. 

cm 

cm.              cm. 

cm. 

Haut,  du  cond.  audit.    ... 

_ 



m 

Haut,  du  menton     

-è 

Haut,  de  Tacrom 

_ 



^ 

H^ut.    de    lextrémi.    du 

0 
g 

mrdius 

_ 



_ 



Hiut.de  l'oml  il  ..    





^A 

Haut,  du  gr.  troch    

_ 

_ 

^■%i 

Haut,  du  vertex,  dans  la 

.!§ 

posit,  assise      















Dist.  bi-acrou:iale     







"ce 

Circonf.  de  la  puit 

._ 

_^ 



_ 

_ 



K 

p 

Long,  du  meml).  sup.     ... 



_ 











œ 
PI 

Long,  du  memb.  inéir,  ... 



. 









_ 

Ä 

Circonf.  du  bras 

1 





_ 



ë 

Cire,  de  la  cuisse      













— 

Cire,  du  mollet    















Long.  tot.  du  pied  

^Laro-.  max.  du  pied 









_ 





— 

— 

— 

— 

— 

— 

— 

Indice  céphal       

.2  F  Indice  du  vi.sage 

'^  '  Indice  facial ...  











__ 

_ 



_ 

._ 



•-^ 

Indice  na.sal         

— 

— 

— 

— 

— 

— 

- 

/•Eap.  entre  la  gr.  enverg. 

et  la  taille 











_ 



Rajj.  de  la  haut,  du  cond. 

audit,  à  la  taille    



_ 

_ 







Ryp.  de  la  haut,  du  ment. 

à  la  taille    







_ 

__ 



. 

Eap.    de     la     haut,     de 

l'acrom.  à  la  taille 

— 



— 





— 

— 

^ 

Eap.  de  la  haut,  du  méd. 

■i 

à  la  taille    

— 













Eap.  de  la  haut,  de  Tomb. 

ce 

à  la  taille 





— 







_-. 

W 

Eap.  de  la  haut,  du  gd. 

troch.  H  la  taille  



.. 

— 









Eap.  de  la  haut,  du  ver- 

tex pos.  ass.  à  la  taille. 

— 



— 





— 



Eap.  de  la  dist.  bi-acrom. 

à  la  taille  

— 

• — 

— 

— 

— 

_ 

— 

Eap.  de  la  long,  du   m, 

sup.  à  la  taille — 

— 

— 

-- 

— 

— 

— 

\  K'ap.  entre  ces  2  long.   ...               — 

— 

— 

— 

— 

-- 

— 

Etudes  Anthropolojii  |u(  s. 


71 


Numéros. 

85 

86 

87 

88      1 

-p 

Cä   "^ 

C    Ü 

CO 

Ö 

o; 

o 
o 

Ö 

p 
P   '-' 

m  a 
o  S 
.2  a 
'-fi 
ce 
u 

ta 

a 

o 

f' Diam.  ant.  jjost.  max.    ... 

Diam.  transv.  max 

Long.  tot.  du  visage 

Long  tot.  de  la  face 

Larg.  tot  de  la  face 

Haut   du  front    

Long,  du  nez 

Larg.  du  nez 

Larg   palpébrale       

Larg.  Iiicaroncul 

Larg.  buccale       

^Lono".  des  ore  lie«     ...    ... 

iiiiii 

mm. 

iiiiii 

IIIIU. 

- 

_  ! 

-i 

-S 

/Taille 

Gr.  enverg 

Haut,  du  cond.  audit.    ... 

Haut,  du  menton     

Haut,  de  Tacrom 

Haut,   de    Textrémi.    du 
médius  ...   

Hant.de  l'ombil 

HauT.  du  gr.  troch    

Haut,  du  vertex,  dans  la 
posit,  assise      

Dist.  bi-acromiale     

Circonf.  de  li  poit 

Long,  du  memb.  sup.    ... 

Long,  du  memb.  infér.  ... 

Circonf.  du  bras 

Cire,  de  la  cuisse     

Cire,  du  mollet  

Long.  tot.  à\\  pied    

\LaT£r.  ma.y.  du  pied 

cm. 

cm. 

cm 

z 

cm.J 

g  (■  Indice  céiîhal 

^     Indice  du  visage 

'g  1  Indice  facial 

^      Indice  nasil 

— 

— 

— 

Kapports. 

K  1  p  entre  la  gr.  enverg. 

et  la  taille 

Eaj).  de  la  Vaut,  .du  cond. 

audit,  à  la  taille    

Rap.  de  la  haut,  du  ment. 

à  la  taille    

Knp.     de     la     haut,     de 

1  acrom.  à  la  1  aille 

Ka-iJ.  de  la  haut,  du  méd. 

à  )a  taille    

Eip.  de  la  haut,  de  Tomb. 

à  la  taille    

Kap.  de  la  haut,  du  gd. 

troch.  à  la  taille    

Kap.  de  la  haut,  du  ver- 
tex pos.  ass.  à  la  taille. 
Ri) p.  (le  la  dist.  bi- acrom. 

à  la  taille    

Kap.    de  la    long,  du  m. 

snp.  M  II.  taille 

Rip.  entre  ces  2  long.    ... 

— 

— 

— 

—  Il 

J 

J 

.^) 


73 


Table  des  Matières. 

Page. 

Introduction    1 

Chapitre  I.     Caractères  descriptifs  6 

Couleur  de  la  peau  6 

Couleur  du  front  7 

Couleur  de  la  paume  de  la  main     :  —  7 

Couleur  des  yeux 7 

Cheveux    8 

Barbe 9 

Poils  du  corps  10 

Sourcils 11 

Insertion  des  cheveux  sur  le  front 12 

Face   12 

Yeux 13 

Nez 13 

Dents 15 

Oreille    15 

Dimensions  de  l' index  et  de  l' annulaire  16 

Forme  de  l' ongle      16 

Courbe  du  bras 17 

Degré  d' écartement  des  membres  inférieurs 17 

Dimensions  des  P""' et  2me.  orteils     17 

Noms  des  villages  et  des  Aborigènes     •••  17 

Chapitre  IL     Mensurations 21 

A.  Mensurations  de  la  tête  et  de  la  face   21 

Diamètre  antéro-postérieur  maximum   21 

Diamètre  transversal-maximum 22 

Longueur  totale  du  visage  23 


74  Tal. le  des  Matières. 

Longueur  totale  de  la  face  24 

Largeur    totale    de    la    face     (distance     bi-zygomatique 

maxima)    25 

Hauteur  du  front 26 

Longueur  du  nez 27 

Largeur  du  nez 28 

Largeur  palpébrale    29 

Largeur  bicaronculaire 30 

Largeur  buccale    31 

Longueur  des  oreilles  .- 32 

Indice  céphalique 33 

Indice  du  visage   35 

Indice  facial 35 

Indice  nasal  37 

B.  Mensurations  du  tronc  et  des  membres    38 

Taille 38 

Grande  envergure     40 

Rapport  de  la  grande  envergure  à  la  taille    41 

Hauteur  du  conduit  auditif  au-dessus  du  sol     43 

Hauteur  du  bord  inférieur  du  menton  au  dessus  du  sol  ...  44 

Hauteur  de  V  acromion  au-dessus  du  sol    45 

Hauteur  de  l'extrémité  inférieure  du  médius  au-dessus  du 

sol  40 

Hauteur  de  l'ombilic  au  dessus  du  sol   47 

Hauteur  du  grand  trochanter  au-dessus  du  sol 48 

Hauteur  du  vertex  dans  la  position  assise    49 

Distance  bi-acromiale 50 

Circonférence  de  la  poitrine    ol 

Longueur  du  membre  supérieur  (de  1' acromion  à  l'extré- 
mité du  médius)    52 


Tal)le  dos  Matières.  75 

Longueur  du  membre  inférieur  (hauteur  du  grand  tioch- 

anter  au-dessus  au  sol)     52 

Rapport    entre  la  longueur  du  membre  supérieur  et  du 

membre  inférieur 53 

Circonférence  du  bras  54 

Circonférence  de  la  cuisse    55 

Circonférence  du  mollet  56 

Longueur  totale  du  pied 57 

Longueur  maxima  du  pied 58 

Table  des  mensurations  et  indices  d'aborigènes  de  Kôtô- 

shô 59 


I 


K.  TORII. 

LES  ABOBICENES  DE  FORMOSE. 


PLANCHE   I. 


Explication  de  la  PI.  I. 

A.  Carte  montrant  les  montagnes,  les  rivières  et  les  villages  des 
Aborigènes  de  Kö-tö-sbö  (Botel-Tobago).  La  petite  Carte  qui 
est  annexée  à  la  carte  de  Kô-tô-sliô,  montre  la  situation  relative 
de  Kû-to-slif)  et  de  Formose. 


B.         Carte  montrant  la  situation  respective  de  Kô-to-sbo  (Botel- 
Tobago)  et  des  îles  qui  sont  situées  au  Nord  des  Pbilipines. 


ft- 


>< 


Co 


o 


4. 


Ä    —     -3       m 


a: 


5  7D 


O 

Q 


< 

a 

o 

H 


pa 


•3 


H.   TORI  I. 

LES  ABORICÈBES  DE  FORMOS? 


PLANCHE    IL 


Explication  de  la  PI.  II. 

A.  Aborigène  de  Kô-t5-sli5  (Botel-Tobago). 

B.  Le  même,  de  profil. 

C.  Aborigène  de  Kô-to-sliô  (Botel-Tobago). 

D.  Le  même,  de  profil. 


/?.  Torii. 


Jour  Sei.  Coll.,  Vol.  XXXII.,  Art.  4.  PI.  II. 


\ 


3D 


I 

i 


R.  Tomi. 

LES  ABOBIGEBES  DE  FORMOSE. 


PLANCHE    III. 


Explication  de  la  Pi.  HI. 

A.  Aborigène  de  Ko-to-shô  (Bote! -Tobago). 

B.  Le  même,  de  profil. 

C.  Aborigène  de  Kö-tö-shö  (Botel-Tobago). 

D.  Aborigène  de  Kö-tö-shö  (Botel-Tobago). 


/?.  Torii. 


Jour.  Sei.  Coll.,  Vol.  XXXII.,  Art.  4.  PI.  III. 


kTimmmmim 


ä 


R.   TORII. 

LES  ABORIGÈNES  DE  FORMOSE, 


PLANCHE    IV. 


Explication  de  la  PI.  IV. 

A.  Aborigène  de  Kô-to-slio  (Botel-Tobago). 

B.  Le  même,  de  profil. 

C.  Aborigène  de  Kô-tô-sliô  (Botel  Tobago). 

D.  Le  même,  de  profil. 

E.  Femme  de  Kô-tô-shô  (Botel  Tobago). 

F.  La  même,  de  profil. 


ft. 


O 


R.   TORII. 

LES  ABORIGÈKES  DE  FORMOSE. 


PLANCHE    V. 


Explication  de  la  PI.  V. 

A.  Groupe  d'Aborigènes  (hommes)  de  Kö-to-shö-(BoteJ-Toba- 
go.) — On  voit,  en  bas  de  la  photographie,  un  modèle  des 
"  maisons  de  travail  " — La  plupart  des  aborigènes  possèdent 
deux  maisons:  une  où  ils  habitent  et  une  autre  où  ils  travail- 
lent. Celles-ci  sont  plus  élevées,  et  ont:généralement  un  étage. 
Ceux  qui  sont  pauvres,  n'ayant  pas  de  "  luaison  de  travail," 
travaillent  en  plein  air,  devant  leur  maison  d'habitation. 

B.  Groupe  d'aborigènes.  Les  2  individus  qui  sont  accroupis, 
en  bas,  au  premier  rang,  sont  des  hommes.  Devant  eux,  se 
trouvent  2  spécimens  d'une  sorte  de  coiffure  en  argent,  de 
forme  conique,  qu'ils  mettent  à  l'occasion  des  cérémonies 
religieuses  .Le  2*"  rang  est  entièrement  composé  de  femmes. 
A  droite,  sont  de  jeunes  garçons. 


/?,  Torii. 


Jour.  Sei.  Coll.,  Vol.  XXX 1 1.,  Art.  4.  PI.  V. 


.=i^L^^ 


R.  TORIl. 

LES  ÂBORICÈBES  DE  FORMOSE, 


PLANCHE    VI. 


Explication  de  la  PI.  VI. 

A.  "  Maison  de  travail."  2  indigènes  sont  occupés  à  fabriquer 
un  modèle  de  pirogue.  Celui  du  bas  qui,  se  nomme  Shika- 
shi,  a  les  cheveux  frisés,  particularité  dont  il  est  parlé  au 
cours  de  cette  étude. 

B.  Aborigènes  occupés  à  produire  du  feu  par  le  frottage. 


/?.  Torii. 


Jour.  Sei.  Coll.,  Vol.  XXXII.,  Art.  4.  PI.  VI. 


Explication  de  la  PI.  VII.  i 

A.  Un  Village  de  Ko-tô-shô  (Botel-Tobago).  En  bas,  à  droite 
de  la  photographie  est  une  maison  d'habitation.  Celles  qui, 
dans  le  village,  sont  plus  élevées  que  les  autres,  sont  des 
"  maisons  de  travail."  Au  premier  plan,  on  voit  des  femmes 
et  des  petites  filles. 

B.  Groupes  d'aborigènes  (hommes  et  femmes),  devant  une 
'*  maison  de  travail." 


/?.   Torii. 


Jour.  Sol.  Coll.,  Vol.  XXXII.,  Art  4.  PI.  VII. 


B 


JOURNAL  OF  THE  COLLEGE  OF  SCIENCE,  TOKYO  IMPERIAL  UNIVERSITT. 

VOL.  XXXII.,  ARTICLE  5. 


Climatic  Changes  in  Japan  since  the  Pliocene  Epoch. 

By 
Malajiro  Yokoyama,  B'Kjakuhakuski. 

Professor  of  Pahrontolony,  Imjjci'ial  Univertiit)j  of  Tohjo. 


With  1  Plate. 


As  is  well  known,  one  of  the  striking  features  of  the  climate 
of  the  primEeval  world  was  the  occasional  interruption  of  a  com- 
paratively warm  and  uniform  climate  by  one  of  intense  cold.  The 
time  during  whicli  such  a  cold  climate  prevailed  is  called  an  ice-age, 
because  of  the  enormous  quantities  of  ice  which  in  the  form  of 
glaciers  covered  the  land,  smoothing,  polishing  and  scratching  the 
rocks  over  which  they  moved,  and  carrying  with  them  erratics  and 
moraines,  just  as  they  do  to-day  in  the  Alps  and  in  Arctic 
countries.  Geologists  have  ascertained  that  such  ice-ages  have 
visited  the  earth  at  least  tliree  times  during  the  past.  The  first 
visit  was  during  the  Eozoic  or  Precambrian,  the  period  in  which 
the  first  dawn  of  life  appeared  in  the  world.  Evidences  of  this  age 
have  been  discovered  in  China,  Canada  and  Northern  Norway. 
The  second  was  toward  the  close  of  the  Palœozoic  era,  in  a  period 
called  Permian.  This  time  the  ice  chiefly  invaded  the  countries 
around  the  Indian  Ocean — India,  Australia  and  Southern  Africa. 
The  so-called  Glussopteris  flora,  which  resembles  that  of  the  succeed- 
ing Mesozoic  more  than  that  of  the  Palaeozoic  and  which  flourish- 
ed during  the  Permian  period  in  the  above  named  countries,  is 
often  brought  into  connection  with  this  second  glaciation,  on  the 
assumption  that  it  was  the  result  of  the  transforming  power  of  the 
cold  acting  on  j^lants  of  the  Palaeozoic  which  remained  unchanged 
until  its  close  in  places  where  there  was  no  glaciation. 


'^  Art.  5.— Matajiro  Tokoyama  : 

The  third  and  the  last  ice-age  was  of  a  comparatively  late 
date.  It  was  m  the  Diluvial.  During  the  Tertiary,  a  period 
which  immediately  preceded  this  age,  the  climate  of  Europe  and 
America  was  very  warm,  so  warm  in  the  beginning  that  tropical 
plants  grew  in  Southern  England  and  chelonians  and  crocodiles 
inhabited  its  waters.  This  great  heat,  however,  gradually 
diminished  as  time  went  on,  becoming  subtropical  in  the 
Miocene  and  temperate  in  the  Pliocene,  the  last  subdivision 
of  the  Tertiary.  Within  this  Pliocene,  too,  the  lowering  of 
temperature  still  went  on  from  tlie  beginning  to  the  end,  a  sure 
indication  of  the  approacl)  of  an  ice-age.  And  this  is  nowhere 
more  clearly  mirrored  than  on  the  molluscous  animals  of  the 
so-called  Crag  Formation  of  England  which  belongs  to  the  above- 
said  Pliocene  Epoch.  The  Mollusca  in  the  lowest  division  of  this 
Crag,  called  the  Coralline  Crag,  in  spite  of  an  admixture  of  a  few 
northern  or  boreal  forms,  still  bears  in  general  the  stamp  of  a  very 
genial  climate.  But  in  the  Bed  Crag,  the  Crag  next  above  it,  the 
number  of  boreal  forms  increases  to  10%  and  in  the  still  higher 
Norwicli  Crag  to  still  more,  until  at  last  in  the  uppermost  Crags, — 
the  Chillesfonl  and  Weghowiie  Crags — their  number  is  so  great  that 
the  fauna  may  be  called  really  Arctic,  and  there  is  even  a  doubt 
whether  these  Crags  might  not  be  better  classed  among  the 
deposits  of  the  ice-age  itself. 

With  the  dawn  of  the  Diluvial  Epoch,  the  whole  aspect  of 
Europe  and  America  was  changed.  Enormous  glaciers  were 
moving  everywhere.  They  formed  a  continuous  sheet  of  ice 
several  thousands  of  feet  in  thickness  and  covering  the  greater  part 
of  the  two  continents.  They  looked  very  much  like  those  now 
found  in  the  interior  of  Greenland  or  on  the  Antarctic  continent. 
This  ice-age,  however,  was  not  one  continuous  age  of  ice.  There 
were  also  times  in  which  the  ice  partly  melted  and  shrunk  and 
the  climate  became  comparatively  mild.  Such  times  are  called 
Inter  glacials,  their  number  varying  in  different  regions  but  amount- 
ing to  as  many  as  six,  as  has  been  ascertained  in  America.  Thus 
the  ice-age  after  several  fluctuations  in  the  extent  of  the  ice  at  last 
disappeared,    and  in  the  Alluvi'àTlSF' ]\r6dern  EpÔcTr"wT'see*tKe 


Climatic  Changes  in  Japan  since  Pliocene  Epoch.  3 

once    ice-covered    Europe    and    America    again     covered     with 
meadows  and  woods,  and  quite  as  inviting  as  in  by-gone  ages. 

Hereupon  a  thinking  mind  is  naturally  led  to  ask  whether 
this  state  of  things  was  limited  to  the  above  two  continents,  or 
was  more  world-wide  in  nature,  in  which  case  the  old  remains  of 
erratics  and  moraines  and  of  polished  and  scratched  rocks  should 
also  be  found  in  other  parts  of  the  world.  And  so  when  Japan 
Avas  opened  to  international  traffic  and  geologists,  both  foreign 
and  native,  began  to  scour  the  country,  they  naturally  looked  for 
■evidences  of  glaciers.  But  strange  to  say,  they  were  nowdiere  to 
be  found.  They  were  not  found  in  Honshu,  nor  in  the  Hokkai- 
do, nor  even  in  the  cold  island  of  Sakhalin  where  even  in  the 
southernmost  part  the  mean  January  temperature  falls  far  below 
the  freezing  point,  to  — ^13°C,  a  temperature  w^hich  w^e  find  in 
Labrador  and  Southern  Greenland.  From  this  negative  evidence 
they  were  obliged  to  infer  that  glaciers  had  never  existed  in  Japan, 
probably  because  the  climate  had  never  been  cold  enough  to 
generate  them.  But  why  had  it  not  been  cold?  There  was  no 
one  who  could  answer  this  question. 

Since  about  a  year  ago,  I  have  been  studying  our  Pliocene 
fossils  found  at  a  place  called  Koshiba,  some  eleven  kilometres 
south  of  Yokohama  and  beautifully  situated  on  the  shore  of  the 
Tokyo  Bay.  Tlie  rock  in  which  the  fossils  are  entombed  is  a 
coarse  tufaceous  sandstone,  sometimes  so  coarse  as  to  look  like  a 
conglomerate,  thus  betraying  the  shallowness  of  the  sea  in  which 
it  was  deposited.  The  fossils  are  chiefly  Molluscs  and  Molluscoids 
with  some  Ecliinodcrms^  Tiibicolous  Annelids,  Balanids,  Fish-teeth,  etc. 
The  Molluscs  seem  to  be  very  rich  in  species,  while  the  case  seems 
to  be  quite  the  contrary  with  the  Molluscoids,  though  they  are  rich 
in  individuals.  The  number  of  the  species  of  these  two  groups  of 
animals  which  I  have  been  able  to  distinguish  up  to  this  time, 
amounts  to  seventy-one,^-'  of  which  the  following  thirty-nine  are 
living  ones:  — 

1.     Cylichna  cylindracea  Pennant. 

1)     Detailed  descriptions  of  these  fossils  will  appear  in  a  future  numbar  of  this  journal. 


Art.  5.— Matajiro  Yokoyama  : 

2.  Conus  siehokli  Reeve. 

3.  Pleurotoma  hamahirana  Pilsbry. 

4.  Mangilia  rohiisticostata  Smith. 

5.  Admete  viridula  Fabricius. 

6.  Voluta  megaspim  Sowerby. 

7.  Mitra  ehenvs  Lamarck. 

8.  Chrysodoimis  pliœniceiis  Dali. 

9.  Chrysodomus  pericochlion  Schrenck. 

10.  Troplion  dathratus  Linné. 

11.  Priene  orcgonensis  Redfield. 

12.  Natica  clausa  Broderip  et  Sowerby. 

13.  Leptotliyra  amussitata  Gould. 

14.  Puncturella  conica  Orbigny. 

15.  Acmœa  heroldi  Dunker  var.  pygmcca  Dunker 

16.  Patella  pallida  Gould. 

17.  Dcntalium  lücinkaiiß  Dunker. 

18.  Corbula  veniista  Gould. 

19.  Lucina  horcalis  Linné. 

20.  Cardium  modcstum  Adams  et  Reeve. 

21.  Aiiomia  cijtœum  Gray. 

22.  Lima  goliatli  Smith. 

23.  Lima  smitlvi  Sowei'by. 

24.  Lima  japonica  A.  Adams. 

25.  Pectcn  swiftii  Bernardi. 

26.  Pectcn  yessoensis  Jay. 

27.  Pectcn  vesiculosits  Dunker. 

28.  Pectcn  similis  Lasky. 

29.  Pectcn  iircgidaris  Sowerby. 

30.  Pectcn  tujerrimis  Müller. 

31.  Ostrea  gigas  Thunberg. 

32.  Area  hohcltiana  Pilsbry. 

33.  Area  decussata  Sowerby. 

34.  Area  stcarnsii  Pilsbry. 

35.  Limopsis  crenata  A.  Adams 

36.  Lcda  ramsayi  Smith. 

37.  Nucida  insignis  Gould. 


Climatic  Changes  in  Japan  siac9  Pliocene  Epoch.  5 

38.  Terehmtidina  crosm  Davidson. 

39.  Terehratidina  cailleti  Crosse. 

I  also  found  three  forms  which,  if  not  quite  identical  with, 
are  at  least  close  to,  living  species.     They  are 

40.  Sipho  cf.  gracilis  Da  Costa. 

41.  Natica  cf.  pallida  Broderip  et  Sowerby. 

42.  Fissuriaea  cf.  tanncri  Verrill. 

The  remaining  twenty-nine  species  are  those  which  are  not 
yet  known  to  be  living,  and  belong  to  the  genera  Conus,  Pleurotoma, 
Drillia,  Mangilia,  Mitra,  Troplion,  Bittium,  Trichotropis,  TurriteUa^ 
Solarium,  Acrilla,  Scala,  Dentaliwn,  Diplodonta,  Liicina,  Astarte^ 
Cardita,  CrassateUa,  Mijodora,  Pcctcii,  Östren,  Neniodon,  Terehratidina^ 
etc.  They  amount  to  about  40%  of  the  whole,  and  even  when 
reduced  by  future  discoveries,  can  hardly  be  imagined  as  falling 
below  20%.  From  this  we  may  safely  infer  that  the  fauna  can 
not  be  younger  than  the  Mildle  Pliocene. 

But  it  is  not  this  high  percentage  of  extinct  forms  which  has 
struck  me  most.  It  is  the  decidedly  boreal  character  of  the 
entire  fauna.     The  following  species  are  those  which  point  to  it:  — 

1.  Admete  viridida. 

2.  Chrijsodomus  phœniceus. 

3.  Chrysodomus  pericocldion. 

4.  Sijjho  gracilis. 

5.  Trophon  clathratus. 

6.  Priene  oregoncnsis. 

7.  Natica  clausa. 

8.  Natica  pallida. 

9.  Leptothyra  amussitata. 

10.  Puiicturella  conica. 

1 1 .  Corhula  veniista. 

12.  Pectcn  yessocnsis. 

13.  Pecten  siüiftii. 

14.  Leda  ramsayi. 

15.  Nucula  insignis. 

Troplion  clathratus  and  Admete  mridula  are  well  known  circum- 
boreal  species.      Chrysoloivis  phoe:ilceus  is  now  found  on  the  coast 


Ö  Art.  5. — Mata  jiro  Yokoyama  : 

of  British  Columbia  near  51°  N.  Lat.  and  at  a  depth  of  240  fathoms 
where  the  temperature  of  the  water  is  7°C.  Puncturella  conica  is 
now  living  only  near  the  Falkland  Islands.  Leda  ramsayi  is  found 
in  New  South  Wales,  but  at  a  depth  of  950  fathoms.  Also  all  the 
others  are  now  living  north  of  Tokyo  Bay,  and  the  three  species 
of  Lima  before  mentioned,  though  existing  near  Central  Japan, 
have  never  yet  been  met  with  in  the  shallow  waters  of  the 
coast.  Although  the  boreal  forms  together  with  these  deep-water 
ones  make  up  about  one-half  of  the  living  species,  there  is  not  a 
single  one  which  is  limited  to  the  warmer  seas.  Moreover,  the 
occurrence  of  such  genera  as  Tricliotropis  and  Astarte  among  the  new 
species  clearly  indicates  that  the  tempei'ature  of  the  waters  in  which 
the  Koshiba  shells  had  lived  must  have  been  a  pretty  low  one. 

Now  what  makes  this  boreal  character  of  the  Koshiba  fauna 
the  more  important  is  the  less  boreal  nature  of  the  MollusccC-^  of 
the  upper  Pliocene  found  in  the  immediate  neighbourhood  of 
Tokyo,  at  Oji,  Shinagawa,  Tabata,  etc.,  which,  when  compared 
with  the  recent,  are  still  boreal  enough.  From  these  facts,  I  am 
forced  to  conclude  that  the  climate  of  Central  Japan  during  the 
Pliocene  Epoch  was  on  the  whole  colder  than  now,  and  indeed, 
colder  in  the  earlier  than  in  the  later  part  of  it.  This  is  quite  in  accord- 
ance with  the  conclusion  already  arrived  at  by  Prof.  Nathorst"-^ 
from  studying  our  Pliocene  plants.  This  eminent  palœobotanist 
recognized  plants  of  the  said  epoch  occurring  in  a  rock  exposed  at 
the  sea-coast  near  Yokohama  and  also  at  Mogi*-*  as  corresponding 
to  those  now  growing  on  our  higher  mountains  and  not  on  the 
lowlands,  as  the  situation  of  the  fossils  would  naturally  suggest; 
and  although  he  does  not  touch  the  question  of  the  rise  of  tem- 
perature in  the  course  of  the  Pliocene,  he  advances  the  view  that 
the  Yokohama  plants  are  probably  upper  Pliocene  and  are  younger 

1)  These  fossils  have  been  studied  by  Dr.  Tokunaga  and  the  results  given  in  bis  "  Fossils 
from  the  Environs  of  Tokyo,"  article  2,  vol.  XXI  of  this  journal.  TJd fortunately  he  took  them 
for  Diluvial,  probably  led  by  the  boreal  nature  of  the  fossils  of  the  same  age  in  Europe,  which 
can  not  be,  for  they  contain  at  least  about  10%  of  extinct  form=.  Tokunaga  himself  describe» 
more  than  20%  of  extinct  species. 

2)  Contribution  à  la  flore  fossile  du  Japon,  18S3.     Zur  fossilen  Flora  Japans.  1888. 

3)  Xear  Nagasaki. 


Climatic  Changes  in  Japan  since  Pliocene  Epoch.  7 

than  those   of  IMogi,    a  view  wliich   can   partly  be  proved  both 
geologically  and  palseozoologically. 

As  soon  as  I  became  aware  of  the  above  stated  increase  of 
temperatm-e,  it  occurred  to  me  that  the  so-called  coral-bed  of  Noma^^ 
in  the  southern  part  of  the  Sôbô  Peninsula  might  belong  to  the 
succeeding  Diluvial  age.  The  bed  is  a  muddy  sand  filling  the 
valley-bottoms  between  the  hills  of  the  Pliocene  rocks  and  not  far 
from  the  sea.  The  fossils  consist  of  large  masses  of  corals  mixed 
with  shells  and  possess  a  very  young  looking  aspect,  on  which 
account  they  were  hitherto  supposed — quite  vaguely  of  course— to 


be  Prehistoric.     But,  as  we  do  not  find  now  such  large  corals  in  the 
neighbouring  seas,  no  one  could  tell  how  they  happened  to  be 


1)     Near  the  town  of  Tateyama  in  Awa  ;  latitude  35°  N.  and  Longitude  139°  50'  E. 


ö  Art.  5.  — Matajiro  Yokoyaina  : 

found  there.  Therefore  I  immediately  took  up  the  examination  of 
the  shells  of  the  bed  and  was  agreeably  surprised  to  find  them  to  l^e 
such  as  can  only  be  interpreted  as  Diluvial,  not  younger,  not  older; 
for,  although  the  thirty-five  species'^  which  I  have  been  able  to  dis- 
tinguish are  all  living,  yet  fourteen  of  them  are  now  living  only 
south  of  Kyushu — in  the  China  Sea,  in  the  Philippines  and  the 
tropical  portions  of  tlie  Pacific  and  Indian  Oceans,  etc.  Such  are 

1.  Purpura  alveolata  Reeve. 

2.  Cuma  rugosa  Born. 

3.  Triton  ohscurus  Reeve. 

4.  Cyprœa  carneola  Linné. 

5.  Trochus  atropurpureus  Gould. 

6.  Fissuridea  rupellini  Sowerby 

7.  Cytherca  tigrina  Lamarck. 

8.  Venus  lacerai  a  Han  ley. 

9.  Venus  cf.  juhesi  Deshayes. 
10.      Chama  multisquamosa  Gmelin. 
IL      Cardita  cf.  guhernaculnm  Reeve. 

12.  Area  kraussi  Philippi. 

13.  Arcafusca  Brugiere. 

14.  Perna  marsupium  Lamarck. 

Of  the  remaining  twenty-one  species,  fourteen  are  those  living 
in  tropical  as  well  as  in  Japanese  seas,  while  only  six  are  purely 
Japanese. 

As  to  the  corals  which  are  found  together  with  these  shells,  I 
have  not  yet  been  able  to  determine  their  species;  but  this  much 
is  certain  that  they  are  true  reef-building  corals  belonging  to  such 
genera  as  Ileliastraea,  Cijpliastraea,  Prionastraea,  3Iussa,  Goniophora, 
Stylophora,  Alveopora,  Domoseri^,  Madrepora,  several  genera  of 
Fungidœ,  etc.,  all  which  we  do  not  find  now  living  north  of  the 
Ryukyus  (the  northernmost  is  28"  20'  N.  Lat.)  or  of  tlie  Bonins 
(about  27°  N.  Lat). 

From  this  we  see  that  we  have  here  a  layer  which  corres- 
ponds to  the  Diluvium  of  Europe.  Just  as  the  latter  contains 
many  forms  which  have  since  retreated  to  the  north,  the  Noma 

1)     These  will  be  described  in  a  future  number  of  this  journal. 


Climatic  Changes  in  Japxn  since  Pliocene  Epoch.  9 

bed  contains  those  which  now  inhabit  tropical  seas  only.  This  is 
an  unmistakable  sign  of  the  very  warm  climate  which  then 
prevailed  in  that  part  of  Japan,  much  warmer  than  that  of  to  daj^, 
for  the  sea  near  Noma  now  cools  down  to  about  10°C,  while 
the  minimum  temperature  of  the  water  in  which  the  reef-corals 
live  is  19^G.  Indeed  I  am  quite  sure  that  while  the  Occident  was 
buried  under  the  heavy  burden  of  ice  millions  of  tons  in  weight, 
Centra]  Japan  was  exposed  to  the  lieat  of  tlie  tropical  sun. 
Beyond  in  the  West,  one  speaks  of  the  ice-age;  here  in  the  Far 
East  we  can  talk  only  of  tlie  coralline.  Such  being  the  case  it  is 
quite  natural  that  geologists  should  have  been  unable  to  find  any 
glacial  remains  in  this  part  of  the  world. 

It  may  perhaps  be  asked  whether  in  Japan  there  are  no  beds 
which  are  the  equivalents  of  the  Interglacials  and  Postglacials  of 
the  West.  I  think  there  are.  Several  years  ago  I  was  passing  by 
the  town  of  Mobara  in  Kazusa  situated  on  the  Pacific  side  of  the 
Söbö  Peninsula  and  much  to  the  north-east  of  Noma,  when  I 
discovered  a  sand-layer  exposed  along  the  two  sides  of  a  stream 
flowing  through  a  coastal  plain  at  the  foot  of  the  Tertiary  hills. 
In  this  layer  I  found  about  sixty  species  of  marine  Mollusca  all 
living.  Some  of  these  are  either  tropical  or  are  at  least  not  yet 
known  to  exist  in  the  neighbouring  seas.  Such  are  Area  symmetrica 
Reeve,  Tapes  d.  quadriradiata  Deshayes,  Venus  cî.  listeri  Gray,  Eulima 
solida  Sowerby  and  Suhemarginiila  carinata  Reeve.  I  also  examined 
shells  brought  from  a  sand-layer  at  Cape  Daito,  somewhat  to  the 
south-east  of  the  above  mentioned  place,  and  found  them  to 
contain,  besides  one  or  two  tropical  forms,  a  species  of  Turritella 
which  hitherto  has  l)een  known  to  occur  only  in  the  Koshiba 
Pliocene.  And  as  these  layers  contain  no  snch  large  corals  as 
those  found  at  Noma,  they  must  be  considered  to  be  deposits 
of  a  time  or  tim^es  in  which  the  climate  was  much  more 
temperate.  This  leads  us  to  ask,  if  they  are  not  IntercoraUine  or 
PostcoraUine.  '  ^ 

From  all  that  I  have  stated  above,  it  follows  quite  naturally 
that  the  temperature  in   Central  Japan  has  gradually    increased 

1)     Similar  shell-layers  seem  to  occur  also  on  the  coast  between  Xoma  and  Kachiyama. 


10  Art.  5. — Matajiro  Yokoyama  : 

silice  the  earlier  Pliocene,  attaining  its  maximum  in  the  Diluvial 
and  then  again  decreasing  down  to  the  present  time.  This  is,  as 
every  one  can  see,  a  state  of  things  just  the  reverse  of  what  we  find 
in  Europe  and  America,  a  singular  contrast  for  which  there  must 
surely  be  a  cause. 

But  before  entering  into  the  discussion  of  this  cause,  let  us  go 
back  a  little  to  the  Miocene  Epoch  when  Switzerland  is  said  to 
have  enjoyed  a  climate  such  as  we  find  now  in  the  southernmost 
cape  of  Kyushu  (Sata-no-misaki  in  Osumi  with  a  mean  annual 
temperature  of  18°C)  and  Amami-Oshima  (the  northenmost  of  the 
Ryukyu  Islands  with  20,8°C).  The  Japanese  fauna  of  this  epoch 
has  not  yet  been  fuU}^  studied,  but  we  know  something  of  its 
plants  from  the  investigations  of  Nathorst.^'  This  palaeobotanist 
found  them  to  consist  of  a  njixture  of  the  European  Miocene  and 
of  the  so-called  Arctic-Tertiary  flora,  indicating  that  the  climate 
of  Japan,  at  least  between  35°  and  40°  N.  Lat.  in  which  the  plants 
were  collected,  was  not  in  the  least  warmer  than  now.  This  \vould 
naturally  lead  us  to  assume  that  a  difference  in  climate  had 
already  at  that  time  existed  between  Europe  and  Japan,  but  that 
this  difference  was  not  so  marked  as  in  later  epochs.  And  I  think 
this  is  quite  in  accordance  with  the  already  known  fact  that  the 
further  back  we  go  into  the  past,  the  more  uniform  the  climate 
becomes  throughout  the  world. 

About  the  Pre-Miocene  Tertiary  fossils  of  Japan  we  do  not 
yet  know  much.  And  the  few  that  I  myself  have  lately  described^-* 
are  not  enough  to  enable  us  to  dra^v  any  conclusion  as  to  the 
climate  of  those  times. 

The  phenomena  of  Nature  which  have  been  already  set  forth 
as  the  probable  causes  of  the  climatic  changes  of  the  past,  and 
especially  of  the  ice-age,  are  partly  astronomical  and  partly  physical. 
Among  the  former  we  may  mention  the  change  in  the  eccentricity 
of  the  earth's  orbit  or  in  the  obliquity  of  the  ecliptic,  the  preces- 
sion of  the  equinoxes,  the  displacement  of  the  poles  and  the 
formation  of  the  smaller  planets.     Among  the  latter  we  may  count 

1)     Zur  fossilen  Flora  Japans,  1888. 

•J)     Some  Tertiary  Fossils  from  the  Miike  Coal-fielJ;  1911. 


Climatic  Changes  in  Japan  since  Pliocene  Epocb.  iJ 

the  variation  of  the  quantity  of  carbonic  acid  gas  in  the  air  due  to 
the  greater  or  less  frequency  of  volcanic  eruptions  and  the  different 
distribution  of  land  and  water  in  past  times. 

Among  these  supposed  causes,  those  which  can  be  brought 
into  connection  with  oar  Coralline  Age  are  only  three,  viz.,  the 
displacement  of  the  poles,  the  quantity  of  carbonic  acid  gas  and 
tlie  distribution  of  land  and  water;  as  for  the  others,  if  they  were 
ever  real  causes,  they  must  have  been  of  a  more  general  character, 
either  affecting  the  whole  earth  at  once,  or  at  least  one-half  of  it, 
the  Northern  or  the  Southern  Hemisphere,  and  not  one-lialf  of 
the  same  hemisphere  as  in  our  case. 

That  carbonic  acid  gas  is  very  effective  in  keeping  the  air 
Avarm  b}^  preventing  the  too  rapid  radiation  of  heat  from  the 
ground,  is  quite  true.  Therefore,  if  it  is  used  in  explaining  the 
occurrence  of  our  Coralline  Age  only,  it  seems  to  work  very  well; 
for  we  may  assume  that  the  volcanic  eruptions  were  quite  violent 
at  that  time,  so  many  active  volcanoes  still  exist  in  our  country. 
But  then,  how  can  we  explain  the  temperate  climate  of  our 
Miocene  and  the  cool  climate  of  our  Pliocene,  when  volcanic 
eruptions  were  at  least  equally  as  violent  as  in  the  later  times? 
That  such  was  the  case,  we  know  by  the  profuse  occurrence  of 
liparites  and  andésites  and  of  their  respective  tuffs  containing 
either  jMiocene  or  Pliocene  fossils,  tlave  we  any  evidence  that 
the  European  Miocene  and  Pliocene  were  richer  in  volcanic 
eruptions  than  ours?  Can  any  one  prove  that  the  quantity  of 
carbonic  acid  gas  has  been  increasing  in  Europe  and  decreasing  in 
Japan  since  the  Diluvial  Epoch?  I  think  this  gas,  if  it  ever  be 
used  to  explain  the  change  of  climate  in  the  past,  can  be  more 
advantageously  applied  in  the  case  of  the  Pre-Tertiary  or  at  least 
the  Pre-Miocene  period,  when  the  climate  of  the  world  was  more 
uniform  throughout. 

The  distribution  of  land  and  water  also  can  hardly  be  said  to 
give  a  more  satisfactory  explanation  ;  for  the  Diluvial  was  a  time 
in  which  the  configuration  of  the  land  was  not  much  different 
from  the  present.  Looking  at  our  Pacific  side,  the  ocean  itself 
must   then    have    been    long  in    existence,    since    it   is    a    great 


12  Art.  5. — Matajiro  Yokoyama  : 

géosynclinal,  as  the  geologists  call  it,  whose  formation  is  said  to  go 
back  at  least  as  far  as  the  Mesozoic.  Then  the  American  and  the 
Asiatic  coast-regions,  Japan,  Formosa,  the  Philippines  and  even 
the  greater  part  of  the  Eyukyu  Islands  were  already  in  a  form 
very  nearly  as  we  see  them  to-day.  And  if  any  one  expects  to 
find  out  any  great  change  in  the  form  of  the  land,  it  would  be  just 
at  the  head  of  the  Tokyo  Ba}^.  This  place  is  coloured  in  our 
geological  maps  as  Diluvial  which,  if  marine,  would  transform  the 
Söbö  Peninsula  at  that  time  into  an  island.  But  what  is  here 
called  Diluvial  is  a  thick  layer  of  subaerial  loam  evidently  derived 
from  a  volcanic  ash  which  had  fallen  on  the  preexisting  land  made 
up  of  Pliocene  strata.  From  this  we  know  that,  in  the  Diluvial 
Epoch,  the  topography  of  the  region  surrounding  the  Tokyo  Bay 
was  not  much  different  from  wdiat  we  see  at  present. 

It  is  a  well  known  fact  that  ocean-currents  exercise  a  great 
modifying  influence  on  the  climate  of  a  country  near  which  they 
flow.  And  as  the  Kuroshio'^  flows  just  south  of  the  Söbö  Penin- 
sula, one  might  think  that  by  a  little  change  of  its  course,  it  might 
come  to  wash  the  shores  of  the  Peninsula,  thus  raising  the  tempera- 
ture of  the  water  to  a  degree  sufficient  to  fit  it  for  the  growth  of  the 
reef-corals.  But  such  a  change  in  the  course  of  the  current  can 
never  have  been  brought  about  without  a  marked  change  in  the 
configuration  of  the  surrounding  countries.  Those  who  look  only 
on  maps  and  are  not  well  acquainted  with  actual  meteorological 
conditions  are  easily  misled  by  the  proximity  of  the  Kuroshio 
and  over-estimate  its  warming  power  on  our  country.  One  must 
always  bear  in  mind  that  in  winter  in  Japan  a  cold  wind  sets  in 
from  the  Asiatic  continent  either  as  a  north  wind  or  a  north-west 
wind,  which  not  only  cools  down  the  land  to  a  temperature  quite 
low  in  comparison  with  its  latitude,  but  also  prevents  the  warm  air 
floating  ov^er  the  Kuroshio  from  ever  approaching  it  and  even 
causes  the  current  itself  to  deviate  a  little  to  the  south.  Therefore 
the  warming  power  of  this  current  at  this  season  is  reduced  to  a 

1)  Kuroshio  means  black  current  or  black  salt-w.iter,  and  not  black  salt  as  often  trans- 
lati'd  in  European  books.  This  mistake  is  undoubtedly  due  to  our  shio  |g  salt  and  shio  JU  salt- 
water being  phonetically  the  same,  though  different  in  written  characters.  Another  name  for 
Kuroshio  is  Ivurosegiwa  which  signifies  dark  rapid  rinr. 


Climatic  Chaiigf  s  in  Japan  since  Pliocejio  Epoch.  13 

minimum,  and  if  there  is  an}^  influence  from  it,  as  Prof.  Schott 
rightly  remarks,  ^^  it  would  be  on  the  Japan  Sea  side,  where  flows 
a  branch  stream  along  the  coast  Avhose  influence,  however,  is  of 
course  much  weaker  and  only  limited  to  that  side.  As  the  result 
of  this  cooling  power  of  the  wind  on  the  land,  the  waters  of  the 
immediate  neighbourhood  of  the  coast  are  also  cooled  down, 
often  to  about  10°C  as  has  been  already  mentioned.  Also  we 
must  not  forget  to  mention  that,  except  in  midsummer,  there 
is  always  a  counter-current  intervening  between  the  coast  and  the 
Kuroshio  which  is  taken  b}^  Schott  for  its  backset,  but  which  from 
its  comparatively  low  temperature  was  formerly  even  thought  to  be 
a  southern  continuation  of  the  cold  Kurik  Current  which  comes 
down  at  least  to  the  north-eastern  shores  of  Honshu.  Under  such 
circumstances  it  would  be  impossible  to  attribute  a  spcciallij  Kcirmincj 
power  to  the  current  just  for  the  Diluvial  age.  But  supposing  that 
we  can  do  it  for  some  reason  or  other,  can  we  then  attribute  a 
great  cooling  power  to  the  Gulf  Stream  which  now  exercises  such  a 
great  influence  on  the  climate  of  Western  Europe,  simply  because 
there  was  an  ice-age  on  that  side  of  our  Northern  Hemisphere?  It 
goes  without  saying  that  such  an  arbitrary  way  of  explaining 
things  is  of  no  value  at  all  in  science. 

Accordingly  only  one  cause  remains,  viz.,  the  displacement 
of  the  poles  or,  as  it  may  perhaps  better  be  expressed,  a  change 
in  the  position  of  the  earth's  axis.  This  is,  I  believe,  the  most 
plausible  explanation  in  a  case  like  ours. 

Neumayr^  in  1887  and  Nathorst^^  in  1888  already  attempted 
to  explain  the  curious  geographical  distribution  of  the  Arctic- 
Tertiary  flora  of  the  Miocene  period  by  this  change  of  the  earth's 
axis.  The  plants  of  this  flora  which  betray  a  rather  temperate 
climate  of  8° — 10°C  in  the  yearly  average,  a  climate  roughly 
corresponding  to  that  of  our  Southern  Hokkaido,  were  found  in 
Grinell  Land  (81°  45'  N.L.),  Greenland  (70'  N.L.),  Spitzbergen 
(77 J  and  77§  N.L.),  the  Lower  Lena  (65°  N.L.),  the  Lower  Amoor 

1)  Oberflächen-Temperaturen  und  Strömungen  der  ostasiatischen  Gewässer,  p.  45. 

2)  Erdgeschichte  IL  p.  511. 

3)  Zur  Fossilen  Flora  Japans,  p.  53. 


14  Art.  5. — Matajiro  Yokoyama: 

(Bureja),  Sakhalin  (51°  N.L.),  Kamtchatka.  Alaska,  etc.,  forming 
so  to  say  a  wreath  around  the  present  North  Pole,  but  far  nearer 
to  it  on  the  Atlantic  side  than  on  the  Pacific.  Neumayr  wislied 
to  bring  the  pole  ten  degrees  nearer  to  the  Asiatic  side  on  the 
meridian  of  Ferro,  while  Nathorst  increased  the  displacement  to 
twenty  degrees  on  the  meridian  of  120°  E.  long.,  which  would 
bring  the  position  of  the  supposed  North  Pole  in  the  lower  region 
of  the  Olenek  just  west  of  the  Lower  Lena.  Nathorst  also  sought 
to  account  for  the  smallness  of  the  leaves  of  the  fossil  plants  found 
at  Lena  by  the  proximity  of  the  pole  and  the  temperate  character 
of  our  Miocene  plants,  whicli  according  to  him  contain  not  a  single 
element  whicli  points  to  a  climate  warmer  than  the  present.  But 
Neumayr  in  the  second  edition  of  his  Erdgeschichte'^  seems  to  be 
inclined  to  renounce  his  former  assumption,  because  of  the  dis- 
covery of  ordinary  Arctic-Tertiary  plants  in  one  of  the  New 
Siberian  Islands  which  are  not  far  off  from  the  supposed  North 
Pole  of  Nathorst. 

Now,  if  I  may  be  allowed  to  express  my  own  opinion  on  the 
above  subject,  I  would  say  that  it  is  not  at  all  necessary  to  fix  the 
position  of  the  pole  so  as  to  make  it  as  equidistant  as  possible  from 
the  various  fossil  localities.  It  may  as  Avell  be  taken  as  lying 
more  to  the  east,  nearer  to  a  meridian  passing  through  the  Bering 
Strait.  In  saying  this,  however,  I  am  by  no  means  trying  to 
establish  the  position  of  the  Miocene  pole.  On  the  contrary,  I 
think  it  is  very  difficult  to  locate  this,  because  the  distribution  of 
land  and  water  was  then  very  different  from  what  it  is  to-day. 
Furthermore,  other  factors  which  determined  the  geographical 
distribution  of  plants  at  that  time  are  utterly  unknown  to  us. 
Therefore  I  simply  say  that  the  North  Pole  may  have  been  then 
in  a  different  position  from  now,  but  that  the  data,  now  at  our 
disposal,  are  too  scanty  to  justfy  us  even  in  guessing  at  it. 

The  case  becomes  different  in  the  succeeding  ages.  Here  the 
climatic  contrasts  in  the  East  and  the  West  are  very  strong,  and 
always  in  such  a  w^ay  that  when  it  is  cold  on  one  side  it  is  warm 


1)     Vol.  II,  p.  385,  1895. 


Cliuialic  Changes  in  Japan  since  Pliocene  Epoch.  15 

Oil  the  other,  and  vice  versa.  This,  I  think,  can  only  be  explained 
by  the  movement  of  the  poles  to  and  fro.  We  may  assume  that 
during  the  earlier  Pliocene,  the  North  Pole  was  more  to  the 
Asiatic  side.  Then  it  began  to  move  to  the  Atlantic  side  until 
the  Diluvial,  when  that  side  fell  into  ice-age  and  the  Asiatic  side 
into  coralline  age  ;  after  that  it  may  be  taken  as  having  made  a 
backward  movement,  that  is  to  say,  back  again  to  the  Asiatic  side. 
So  far  as  I  know,  this  mode  of  explanation  meets  with  no  serious 
objection.  Therefore,  if  theories  are  made  to  explain  a  phenome- 
non whose  cause  is  unknown,  and  if,  among  these  theories,  the 
one  which  explains  it  in  the  easiest  and  most  unconstrained  way  is 
the  most  plausible,  then  the  movement  of  the  poles  to  and  fro  must  be 
regarded  as  the  most  plausible  explanation  of  the  climatic  changes 
of  the  Northern  Hemisphere  since  the  Pliocene  Epoch. 

It  is  a  singular  coincidence  that  Prof.  Simroth  of  Leipsic,  led 
by  a  peculiar  geographical  distribution  of  the  organisms  in  the 
present  creation,  had  already  tried  to  explain  it  also  by  the  so- 
called  "  Theory  of  the  Oscillation  of  the  Poles,"  first  propounded 
by  Reibisch  to  account  for  the  displacement  of  the  beach-line.  In 
tliis  theory, '-^  the  North  and  South  Poles  are  made  to  swing  to  and 
fro  on  the  meridian  of  10°  E.L.  which  corresponds  to  170°  W.L. 
on  the  Pacific  side,  a  line  just  passing  through  the  Bering  Strait. 
This  meridian  has  been  calletl  by  Reibisch  the  circle  of  oscillation. 
Now  Simroth  had  recognized  a  more  or  less  symmetrical  distribu- 
tion of  similar  or  vicarious  foi'ms  either  east  or  west  of  this  circle  or, 
if  under  the  same  circle,  north  and  south  of  the  equator.  The  first 
is  called  by  Simroth  the  transversal  sijininetrtf,  and  is  said  to  be 
caused  by  the  organisms  diverting  to  the  east  or  the  west  as  the 
quickest  means  of  evasion  of  the  approach  of  a  pole  or  of  the 
equator.  The  second  is  called  by  him  meridial  s)jmmetrij,  and  is 
thought  to  be  caused  by  organisms  on  the  approach  of  the  equator, 
climbing  up  high  mountains  on  which  they  can  wander  south  and 
even  cross  the  equator  beyond  which  they  can  again  come  dow^n  to 
the  low-land,  where  the  climate  is  suited  for  their  existence.     As 

1  )    Simroth.     Die  Peadnlationstheorie,   1907. 


16 


Art.  5.— Matajiro  Yokoyama  ; 


examples  of  the  transversa]  symmetry  he  mentions,  among  others, 
the  occurrence  of  the  giant  salamander  in  Japan  and  of  Menopoma  in 
the  United  States,  of  the  genus  Alligator  in  China  and  America,  of 
Ceratodus  in  Queensland  and  of  Lepidosiren  in  South  America,  etc. 
Of  course  it  is  not  my  object  here  to  reproduce  all  the  details 
given  by  Simroth  in  his  book.  I  only  mention  them  to  show 
how  this  eminent  naturalist  was  led  to  assume  the  movement  of 
the  poles  from  the  distribution  of  recent  organisms,  just  as  I  have 
been  led  by  that  of  the  fossils. 

Lastly,  it  may  perhaps  be  asked:  If  the  poles  ever  moved, 
in  what  position  were  they  during  the  Japanese  Coralline  Age? 
To  such  a  question,  I  can  only  say  that  I  have  found  only  enough 
evidence  to  suggest  the  movement  of  tlie  poles  and  no  more.  I 
even  doubt  if  palaeontologists  will  ever  be  able  to  establish  the 
position  of  the  poles  in  the  past  b}^  simply  studying  the  fossils. 
Tlierefore,  let  it  suffice  for  me  to  say  that  during  our  Coralline 
Age,  the  poles  were  in  such  a  position  as  to  cause  the  sun  to  shine 
on  the  Sübö  Peninsula  with  about  the  same  intensity  as  it  now 
shines  at  least  on  the  Ryukyus  or  the  Bonin  Islands. 


m.    TiOKOVASI/V. 

CLIMATIC  CHANCES  IN  JAPAN  SINCE  THE  PLIOCENE  EPOCH, 


PLATE     I. 


Explanation  of  the  Plate  il). 
Some  of  the  Noma  Fossils. 

Fig.  1-  Heliastnea  sp.  ^/^  nat.  size. 

Fig.  2-  Do.     A  fragment  in  natural  size. 

Fig.  3.  Madrepora  sp.  A  fragment. 

Fig.  4-  Perna  marsupium  Lam.     (Tropical  species). 

Fig.  5-  Cyprsea  carneola  Linné.     (Tropical  species). 

Fig.  6.  Cytherea  tigrina  Lam.     (Tropical  species). 

Fig.  7.  Triton  obscurus  Rve.     (Tropical  species). 

Fig.  8-  Area  fusca  Brug.     (Tropical  species). 

Fig.  9.  Triton  costatus  Born.     (Tropical  and  Japanese  species). 


i 


Jour.  Sei.  Coll.,   Vol.  XXXII.,  Art  5,  PI.  I. 


M.    Yokovawa  :  Xowa  Fossils. 


h 


JOU  r.NAL  OP  THE  COLLEGE  OP  ECIP.KCP,  IMPEHTAL  TTNIVEIISTTT, 
TOKYO,  JAPAN. 

VOL.  XXXII.,  ARTICLE  6. 


On  Nepheline-basalt  from  Yingé-mên,  Manchuria 

By 

B.    Koto,   Ph.  D.,  Rhjakithakushi 
rrofesaor  of  Gi'ology,  Science  College,  Imperial  University,  Tôki/ô 


With  2  Plates 


The  present  short  paper  deals  with  the  first  genuine  occurrence, 
so  far  as  I  am  aware,  of  nepheline-basalt  in  the  Koreo-Japanese 
and  Chinese  regions.  For  this  reason  it  may  be  of  some  interest 
to  petrologists,  who  seem  at  present  to  attach  special  importance  to 
any  new  find  of  feldspathoid  rocks  in  the  subalkaline  circum- 
Pacific  region. 

It  is  the  current  opinion  that  the  "Pacific  region"  is  charact- 
erized by  the  predominance  of  subalkaline  igneous  rocks  in  contrast 
to  the  alkaline  rocks  of  the  "  Atlantic  region  "^'*;  but  as  there  are 
many  exceptions  in  tlie  latter,  so  we  find  alkaline  rocks  also  in 
the  former  ;  and  these  apparently  aberrant  forms  seem  to  be  in- 
creasing in  number,  as  our  petrological  knowledge  of  their  dis- 
tribution in  the  "  Pacific  region  "  by  degrees  accumulates. 


]  )  W.  Cross  in.  his  recent  paper  seems  to  discredit  broad  generalizations  concerning  the 
genetic  relations  and  regional  distribution  of  igneous  rocks,  termed  the  Atlantic  and  Pacific 
kindred.  After  closely  examining  the  existing  analyses  of  the  Hawaiian  lavas,  he  was  forced  to 
the  conclusion  that  "  the  Hawaiian  magmas  tend  to  show  that  the  generalizations  as  to 
geographic  distribution  or  the  genetic  relations  of  the  alkali  and  mhalkali  grouj^s  included  in 
current  definitions  of  the  Atlantic  and  Pacific  branches  or  Si'ppen  are  far  from  correct.  Hence  in 
their  present  form  they  can  have  no  place  in  a  pétrographie  system."  Whitman  Cross,  "  The 
Lavas  of  Hawaii  and  Their  Relations."  Jour.  IVasliington  Acad.  Sei.,  vol.  1,  Xo,  3,  August,  1911. 
How  far  Becke's  view  on  the  Sippe,  endorsed  by  Harker,  can  withstand  the  criticism  of  age,  I 
cannot  toll  now.     {Added  while  in  press.) 


2  lî.  Eotô  : 

Referring  mainly  to  the  western  Pacific,  WiCHMANN'Mong  ago 
made  known  a  melilite-nepheline  basalt  from  Oalui  in  the  Hawaiian 
Islands,  and  Ceoss^^  has  acquainted  us  with  the  occurrences  of  a 
trachyte  (acid  phonolite)  and  a  nepheline-basalt  from  one  of  the 
same  island  group.  Alkaline  rocks  are  said  to  occur  in  Tahiti,  in 
the  Viti  Archipelago,  and  also  in  Timor^-*;  while  leucite-basalts  are 
found  in  Java^\  in  the  southern  Celebes^\  and  also  in  Masbate, 
one  of  the  Philippine  islands*'^  A  limburgite  is  known  in  the 
Samoa  group,  and  a  nepheline-basalt  in  the  Caroline  islands'^-*. 


As  to  the  alkaline  rocks  of  Japan,  I  noticed  some  fifteen  years 
ago^^  an  allied  rock  from  the  Nemuro  headland  at  the  east  end  of 
Hokkaido,  where  it  is  said  to  occur  in  the  terrane  of  the  Cretaceous. 
In  1907,  I  found  the  same  kind  of  rock  occurring  in  the  Shirétoko 
promontory  in  Japanese  Sakhalin,  making  a  sheet  or  dyke  in  the 
coal-bearing  Tertiary.  Lately  I  have  received  dioritic-looking 
specimens,  sent  by  Déguchi,  from  Tendai-san,  in  the  islet  of 
Hattaku-to,  in  the  Pescadores,  and  from  Reisuiko  near  Taihoku, 
Formosa.  They  all  proved  to  be  the  same  kind  of  rock  as  that 
above    mentioned.      Tbese   are   doleritic   both    in    mineralogical 


1)  Neues  Jaltrhich  f.  Min.  etc.,  1S75,  \i.  172. 

2)  Jour.  Geol.,  vol.  xii.,  1901,  p.  510. 

3)  E.  Daly,  "  Origin  of  the  Alkaline  Eocks."     Bull.  Geol.  Soc.  Amer.,  vol.  xxi.,  1910,  p.  105. 

4)  Verbeek  et  Fennema,  "  Description  géologique  de  Java  et  Madoura,"  1896. 

5)  Harker,  "  Natural  History  of  Igneous  Eocks,''  p.  98.    Details  are  given  in  W.  Bucking, 

"  Leucitbasalt  aus  der  Gegend  von  Pangkadjene  in  Süd-Celebes."    Berichte  <1.  Nuturforsch.  Gessel- 
z.  Freiburg  i.  Br.  Bd.  XI.  Heft  2,  1899. 

C)  Iddings,.  "  The  Petrography  of  Some  Igneous  Eocks  of  the  Philippines."  Philipinne  Jour. 
Set,  section  A.  (1910)  5,  p.  164. 

7)  Harker,  loc.  cit.,  p.  98.  M.  Weber  recently  acquainted  us  with  the  occurrences  in  the 
Samoa  group,  of  an  alkaJi-trachyte,  phonolite,  trachydolerite,  nepheline-basalt  and  nephelino- 
basanite  besides  the  visual  plagioclase-basalt  and  andésite.  Ahh.  h.  Bayr.  Akad.  d.  TF/ss.  II.  Kl. 
24.  II.  Abt.  1909,  S.  290-310. 

8)  "  Notes  on  the  Geology  of  the  Dependent  Islands  of  Taiwan."  Jour.  Coll.  Sei.  Imp.  Univ. 
Tokyo,  vol.  xxii.,  p.  44. 


On  Neplicliuc-basalt  from  Yingc-mcn,  Manchuria.  3 

composition  and  texture,  the  cuneiform  spaces  left  by  fresli, 
polysynthetic  tabular  plagioclase  being  filled  up  with  analcime. 
They  may  be  analcime-diabase  (basalt)  or  teschenite,  and  seem  to 
be  akin  to  those  of  California,  described  by  Fairbanks^-*,  and 
many  others.  Since  I  could  not  examine  the  mode  of  their 
occurrence,  and  also  as  I  was  unable  to  find  a  sure  trace  of  either 
nepheline  or  leucite,  I  have  simply  left  them  undescribed.  A 
short  description  has,  however,  already  been  given  of  the  analcimc- 
hasalt  of  the  Pescadore  group  (HOko-tô)  iu  Taiwan^'. 

Three  years  ago,  I  found  in  a  Geological  Survey  specimen  from 
the  islet  of  Matsushima,  Kyushu,  a  rock  ]-esembling  an  aegirine- 
trachyte  on  which  Kozu^^  has  very  recently  given  a  preliminary 
note.  It  is  a  grayish,  trachytic-lookiug  laurvikose  soda-trachyte 
with  calcium-bearing  anorthoclase.  The  alkali-feldspar-bearing 
basalts  from  northern  parts  of  Kyû-shû  are  also  brought  to  our 
notice  by  the  same  writer'^  What  seem  to  be  barkevikite-bearing 
rhyolites  or  andésites,  I  have  several  times  observed  from  Kodzn- 
shima,  one  of  the  Idzu  islands,  and  also  from  the  islet  of  Koto-sho 
(Botel-Tobago),  Taiwan.  From  the  above  brief  account,  which 
might  be  multiplied  if  careful  searcli  were  made,  we  see  that  even 
alkaline  effusives  of  basic  and  acid  natures  are  by  no  means  rare  in 
Japanese  islands. 


1)  "  On  Analcite-diabase  from  San  Luis  Obispo  County,  California."  Bull.  Geol.  Depart.  Univ. 
Cal.,  vol.  T.,  p.  273.  I  am  always  watching  with  keen  interest  the  progress  on  the  knowledge  of 
the  Miocene  analcite-diabase  (augite-teschenito  or  basalt)  of  California  by  American  writers.  If 
there  is  any  thing  which  may  be  called  a  petrographical  province,  it  is  this  very  rock -group 
which  unites  both  sides  of  the  Xorth  Pacific.  There  are,  as  it  is  already  stated,  many  localities 
m  Japan  where  the  so-called  analcime-diabase  occurs  in  dykes  or  sheets,  and  one  of  the  allied 
rocks  is  the  "  don  "  which  produced  natural  cokes  by  its  contact  action  in  many  collieries  in 
northern  Kyûshû.  The  most  interesting  point  in  the  studies  of  these  rocks  centers  in  the 
presence  of  analcime  which  was  at  one  time  supposed  to  be  dirived  from  nepheline,  and  at  other 
times  from  decomposition  of  labradoiite.  The  latter  view  is,  I  think,  still  entertained  by  L, 
Haehl  and  E.  Arnold.     (Proc.  Philos.  Soc.  vol.  XLTII.  Xo.  175.) 

2)  Koto,  loc.  cit.,  p.  42. 

3)  Preliminary  Xotes  on  Some  Igneous  Eocks  of  Japan."  J.  Jour.  Geol.,  vol.  xix.  1911,  p.  555. 

4)  "  Preliminary  Notes  etc."    III.  Loc.  cit.,  p.  566. 


4  B.  Koto  : 

We  have  as  yet  scarcely  any  information  respecting  the 
alkaline  rocks  in  China  and  the  lands  adjoining  that  part  of  the 
continent.  But  so  far  as  the  writer's  knowledge  of  them  goes,  the 
granitic  rocks  on  the  southern  border  of  the  Mongolian  plateau, 
extending  from  In-shan  to  Manchuria  (Jwidii-slian),  are  mainly  of  a 
reddish,  coarse-aplitic,  microdine-YlGh  variety^-.  The  same  group 
which  is  often  mylonitized,  forms  the  foundation  of  the  Koreo- 
Manchurian  highland  in  contrast  to  the  granodiorites  which  are 
l^revalent  in  Japan.  A  nepheline-syenite  is  said  to  occur  in 
southern  China' \  I  have  a  specimen  of  riebeckite-granite  from  a 
quarry  near  the  city  of  Foo-cliou,  in  the  province  of  Fokien. 

As  to  effusives,  lack  of  knowledge  is  also  deeply  felt  here.  It 
is  well  known  since  the  explorations  of  R.  Pumpelly"-*  and  the  ]ate 
VON  RicHTiioFEN^^  that  basalt  is  widely  distributed  over  the  southern 
Mongolian  plateau  as  the  counterpoise  of  the  vast  basaltic  mesa  of 
the  east  Koreo-Manchurian  landmass.  P.  Vénukofï*^  gave  a 
description  of  Mongolian  basalts^-*  collected  at  several  widely 
separated  localities  by  tlie  celebrated  travellers,  M.  Potanin,  and 
General  PREjEVALSKy.  They  all  proved  to  be  plagioclase-basalts, 
and  no  mention  was  made  of  aiiy  feldspathoid  variety.  However, 
in  this  connection  it  may  be  of  special  interest  to  cite  from  the 
paper  referred  to,  the  occurrences  of  tachylite  and  limburgite,  with 
the  chemical  analyses  made  of  them  : 

t)  The  so-called  gneiss  that  built  np  the  core  of  the  Tsin-liny  Shan  range,  lying  to  the 
south  of  the  city  Si -nan  Fu  {M^M),  the  well-known  ancient  capital,  is  found,  on  microscopic 
exauaiuation,  to  be  a  sheared  modification  of  this  variety. 

1  )    R.  Daly,  loc.  cit.,  p.  103. 

2)  "  Geological  Researches  in  China,  Mongolia,  an  I  Japan."  Sviithsonian  Contribution 
Publication,  1886. 

3)  "China." 

4)  "  Les  roches  basaltiques  de  la  Mongolie."  Bulletin  de  la  société  BeUje  de  (jéoloyic  de 
ixdeontolO(jie  et  dliydroloaie,  Bruxelles,  tome  II.,  1888,  p.  441. 

5)  According  to  V.  A.  Obrutschew  ("  Central  Asia,"  I.),  there  is  a  large  basaltic  field, 
500  m  thick  near  Kaigan.  resting  on  a  great  thickness  of  loose  conglomerates  and  sandstones 
belonging  to  the  Gobi  series— the  fresh-water  Tertiary  with  Bhinoceros—-with.  trachyte  at  its 
base.    There  is  another  large  basaltic  area  in  Morgen  in  northern  Manchuria. 


On  Ncphclinc-basulfc  from  Yingc-mcn,  Manchuria. 


Tachylito  du  lac  Kyri/-iior'^), 

Liiiiburgi 

to  du  lac  Doloij-nor-'i, 

41°  lat.,  83°  long. 

43°  25 

lat.,  86°  30'  long. 

SiO, 

49.37 

41.69 

AlA 

17.67 

14.85 

Fe^Os 

6.28 

10.39 

FeO 

4.81 

5.43 

FeO. 

0.28 



CaO 

9.12 

11.20 

MgO 

5.02 

9.84 

NasO 

3.27      . 

3.71 

K2O 

1.41 

1.05 

H,0 

2. 15 

1.06 

99.38 

99.32 

Sp.  Gr. 

2.522 

2.851 

To  mention  the  only  remaining  locality,  an  occurrence  of  a 
nepheline-basalt  was  cited  by  A.  Lanick'^  from  Yami-slian^  whicli 
lies  to  the  west  of  the  town  of  Wei-lidcn  in  Kiau-clian,  It  is  an 
amygdaluidal  rock  in  which  nepheline  is  present  not  in  the  form 
of  crystals,  but  as  a  leptomorpliic  mesostasis.  So  much  for  our 
present  knowledge  of  the  alkaline  rocks  within  the  western  Pacific 
territory. 

Localities  of  the  MancUurian  NepJieline-hamlt. — It  may  not  be 
out  of  place  here  to  I'emark  briefly  on  the  region  in  which  the 
present  nephline-basalt  is  found.  So  far  as  I  know,  it  is  the  first 
occurrence  in  Manchuria  of  this  kind  of  rock  ever  recorded  in 
petrographical  literature.     As  the  region  is  entirely  unknown  to  the 


1)  Kir-)ior?  (lake),  2  degrees  west  of  Kahjan.     The  longitude   is  probal>ly  referred  to   a 
Russian  meridian. 

2)  Dalai-nor?  lying  to  the  X.X.E.  of  Dolon-nor  (Lauia-miau). 

3)  "  Beitrage  zur  Pétrographie  von   West-Schantung."    Inaugural-Dissertation,   Leipzig, 
1908,  S.  32. 


6  B.  Koto  : 

outside  world,  a  few  lines  may  be  proper  to  serve  as  an  orientation 
of  the  localities  of  my  find. 

During  my  seven  months' journey  last  winter  through  Man- 
churia and  Korea,  I  happened  to  pass  over  a  water-shed  (PI.  I. 
fig.  1)  of  the  Siuigari  and  the  Hiin-ho,  the  latter  Ijeing  a  tribuar}- 
of  the  well-known  Lian-ho.  I  struck  the  road'^  leading  southwest- 
wards  from  the  city  of  Kirin^^  to  Mopan-shan^\  and  then  to  the 
intermontane  plain  of  S]ian-chên(j-tzii^\  a  fertile  and  populous  flat 
on  one  of  the  upper  courses  of  the  Sungari,  drained  by  a  large 
tributary,  the  llui-fcv'\  Following  the  river  course  upstream  in 
Mat  land  and  diviating  from  the  high  road  (PI.  II.)  to  Kai-ijuaif>, 
I  rude  directly  south  to  a  very  low  and  lonely  snow-clad  water- 
.shed,  and  at  the  end  of  December  last,  came  down  to  the  source 
of  the  Hiui-ho'\  which  1  followed  downstream  southwestwards  as 
far  as  Miihden. 

On  the  south  of  the  above-mentioned  granitic  water-divide  on 
tlie  low  spur  of  a  hih,  called  the  Nien-yii-Ung^^  pass,  is  located 
the  noted  ancient  Gate  of  Yingc-mcii^^  in  the  long  palisade,  now 
ruined,  which  runs  through  the  heart  of  the  Manchurian  hinter- 
land.    It  is  200  kilometers  from  Kinn,  and  140  from  Miikden. 

The  "  Yingc-mcn  area  "  with  all  the  surrounding  districts  is  an 
elevated  granite  peneplain  of  490  ?».,  flanked  on  the  east  by  the 
overlying  volcanic  mesa  of  common  basalt  150  m.  thick,  and  limited 
on  the  west  by  hills  of  nepheline-hasalt  (PI.  I.  ßg.  1)  which  poured  out 
probably  at  the  junction  of  the  microcline-granite  and  the  Lower 


1)  In  Stielei's  Hand-Atlas,  No.  65,  and  Debcs'  Hand- Atlas,  No.  44,  the  region  is  very 
imperfectly  représente.!.  The  best  maps  ever  published  in  Europe  and  accessible  to  general 
readers  are  Karte  von  Ost-China,  scale  1 :  1,000,000,  Berlin,  Sheet  Mukden,  and  Paul  Langhan's 
Neuere  Tageskarte  von  Ost-Asien,  scale  1 : 5,000,000,  Gotha. 

2)  Properly  speaking  the  name  is  Chi-lin,  and  the  people  call  it  Cliuang-chang.  Kirin  is 
the  anglicized  name,  just  as  Mukden  stands  for  Fcng-tien. 

3)  m^\u     1)  iijj^iF     5)  Ht     ^)  mm     7)  mm 
8)  ¥mm     '-))  %mn 


On  Nepheline-basalt  from  Yingé-mt'n,  Mnnclunia.  7 

Cambrian  and  Tertiary  terranes.  The  present  paper  deah  with  thü 
neplieline-hamlt .  It  was  impossible  for  me  to  ascertain  exactly  the 
mode  of  occurrence  of  tlie  basalt  in  my  hasty  journey  through  the 
snow-covered  region  in  tlie  cold  Manchurian  winter  with  the 
mercury  at  -36°C;  and  moreover  the  presence  of  nepheline  in  the 
basalt  was  discovered  only  after  I  had  reached  home  and  was  able 
to  examine  slides  of  the  rocks  collected  during  my  trip.  The  age- 
relation  between  the  more  basic,  sodic  nepheline-basalt  on  the  west 
and  the  calcic  plagioclase-basalt  on  the  east  was  not  ascertained,  but 
the  former  is  probably  older  than  the  latter.  This  may  be  conjec- 
tured from  the  incised  character  of  tlie  topography  resulting  from 
denudation  ;  while  the  common  basalt  on  the  east  builds  up  a  long 
monotonous  mesa  (Cliang-hang^^)  with  sharp  escarpment.     (PI.  I. 

fig-  1.) 

One  specimen  was  struck  at  Tsao-shih-crr^  (PI.  II.)  in  the  upper 
Shan-sliêng-tzu  plain,  at  the  forking  of  the  road  to  Kai-yuan  and  Yingê- 
wên,  and  another  specimen  was  picked  up  quite  by  chance  by 
H.  Murakami  in  the  gravelly  bed  of  the  Hun-lio  river  near  the 
already-mentioned  Yingé-mên  gate.  The  geology  of  the  "  lingê-mên 
area  "  is  roughly  indicated  in  the  annexed  sketch  map  (PI.  IL). 

The  geologic  formations,  cartographically  represented,  are  as 
follows: 

1)  The  flesh-red,  coarse-aplitic,  microcline-rich  orthogneiss 
(^),  greatl^^  mylonitized,  forming  the  basement  of  the  overlying 
complex. 

2)  Diorite  (o),  probably  a  differentiation-product  of  an 
alkaline  granitic  magma,  No.  1.  It  is  a  grayish,  medium-grained 
quartz-diorite,  composed  of  short  prismoids  of  deep  greenish- 
In'own  hornblende,  and  bent  lamellae  of  oil-brown  biotite,  besides 
plagioclase  with  the  characteristic  zonal  structure.      Quartz  fills 

(I    ^Ig        2)    !^]fi^        lu  Plate  II.  it  is  erroneously  spelled  Tsaos/tKT, 


8  B.  Koto  : 

up  angular  spaces,  or  forms  myrmekitic  bodies  with  the  plagio- 
clase.     Locahty:   Nien-yii-ling. 

3)  The  Middle  Cambrian  limestone  (-Ga)  and  the  Lower 
Cambrian  red  breccia  (-G).  The  latter  is  a  rather  fine,  grayish  and 
reddish  variegated  breccia,  consisting  of  subangular  fragments  of 
microcline  and  quartz,  plagioclase  and  orthoclase,  and  lastly, 
melaphyre,  cemented  with  reddish  granitic  sand  and  calcareous 
matter.  The  breccia  contains  slightly  pinkish,  flattened  marly 
nodules  of  the  size  of  4  to  8  cm.  with  a  thickness  of  r/2  ^'"^  with 
no  trace  of   organic  structure.     Localitj^:      Hsiao-mai-pu-tzu. 

4)  The  Miocene  Tertiary  (t)  of  the  type  of  the  Mu-shun 
colliery,  composed  of  shales  and  medium-grained  gray  sandstone 
with  a  poor  seam  of  coal. 

5)  Plagioclase-basalt   (/9,). 

C)  Nepheline-basalt  (y9...).  Localit}^:  Yingc-men  and 
Tsaoshiherr. 

7)     Alluvium  (a). 

Nepheline-basalt 

CoiiiiMisition  :    Essential:   Augite,  nepheline,  olivine. 

Accessory  :  Magnetite,  titanomagnetite,  picotite. 

Accessory  part  :   Base. 
.^facrotextiire  :  Compact  with  minute  phenocryst  (less  than  1 

mm.)  of  olivine. 
ifiicro texture  :  Holocrystalline  with  a  few^  patches   of   brown 

base,  microporphyritic. 

Macroscopically,  the  rock  appears  uniformly    gi'ayish-black, 

and  is   heavy   and   aphanitic,  though,    strictly   speaking,    finely 

granular.     It  is  sometimes  crumbly,  falling  into  dull,  polygonal, 

incoherent  coccolitic  clods  on  a  slightly  weathered  portion,  which 


On  Xepbt'lir.e-)ias;vlt  from  \'ingé-uu>u,  Manchuria.  /■  9 

character  .seems  to  be  a  ^^pecial  feature  of  this  rock.  Tliere  occur  in 
the  general  mass  glittering  flecks  (0.9  by  0.:]  mm.)  of  oHvine  witli 
conchoidal  fracture  and  vitreous  luster,  seen  only  by  reflected  light. 
The  rock  weathers  into  an  ash-gray  earthy  mass  with  brown 
limonitic  spots  of  decomposed  olivine  projecting  from  the  general 
ground . 

Macroscopically,  the  rock  is  hjqîocrystalline,  varying  in  degree 
from  percr3^stal]ine  to  docrystalline  ;  and  microcrystalline  in  cry- 
stallinity  and  ranging  in  size  from  decimillimeter  to  micron  in 
granularity;  it  has  inequigranular,  prismoid  (augite)  and  equant 
(nepheline),  diverse  and  seriate  (augite,  nepheline)  fabric.  As 
in  all  basaltic  rocks,  the  olivine  is  of  a  relatively  large  size  as 
compared  with  the  other  constituents  of  tlie  groundmass.  So  the 
fabric  may  properly  be  called  seriate-porphyritic.  On  account  of 
the  isometric  habitus  of  the  microphenocrysts  of  both  the  augite'^ 
and  the  olivine,  the  texture  of  the  rock  is  orthophyric,  showing 
no  signs  of  fluidal  arrangement  of  components.  The  rock  pro- 
bably crystallized  out  from  an  undisturbed  magma.  (PL  I. 
ß(ß.   1  and  2.) 

Titanaiujite  is  a  dominant  ingredient  occurring  in  the  form  of 
rnicrohte  of  variable  size,  the  largest  being  0. 17  iinii.  long  and  0.037 
broad.  The  larger  ones,  rarety  seen  in  slides,  are  anhedral  and 
tabular  with  the  cleavage-plane  toward  (Oil);  the  smaller  ones,  on 
the  other  hand,  are  microlitic  and  euhedral.  The  extinction  of 
the  former  on  (010)  is  43°41'  toward  the  obtuse  angle.  The  color 
is  yellowish-brown  with  a  tinge  of  violet-green,  and  then  zonally 
colored,  the  interior  being  of  a  violet  shade;  non-pleochroic,  the 
polarization-color  being  a  grayish-yellow  of  a  low  order.  The 
crystals  are  often  transversely  cracked,   and  are  full  of  air-pores 

I)     'rhe  larger  anhedra  are  nut  seen  in  the  photomicrogra;)hs,  PI,  T.figs,  2  and  3. 


10  B.  Eotô  : 

and  granules  of  magnetite.  Next  in  abundcnice  is  iron-ore,  which 
occurs  in  octahedra  or  clumps,  periplierally  changing  into  leu- 
coxene  (titano-magnetite). 

Nepheline  is  the  characteristic  ingredient  occuri'ing  in  short 
prisms,  the  basal  section  of  which  is  hexagonal  and  the  longitudinal 
section  rectangular.  These  colorless  crystals  (the  largest  being 
0.13  by  0.11  mm.)  occur  in  large  quantities  (30%  of  the  volume) 
and  are  fresh  with  a  vitreous  lustre  ;  tliey  enclose  rounded  augite- 
microlites  arranged  parallel  to  the  contour  of  tlie  host — a  cliarac- 
teristic  habitus  by  Avhich  the  presence  of  nepheline  can  be  easily 
recognized.  Through  atmospheric  decomposition  the  mineral 
substance  becomes  parallel-fibrous  by  zeolitization  along  the 
vertical  axis,  and  in  basal  section  the  change  is  seen  advancing 
from  the  periphery.  Low  polarization-colors  and  other  optical 
behaviors  are  normal.  With  HCl  and  methylviolet  the  mineral  is 
easily  ascertained  by  the  staining  method. 

Olivine. — The  minutely  porphyritic  or  minophyric  phenocr^^sts 
of  olivine  of  variable  size  (usually  less  than  0.8  mm.  by  0.5)  occur 
abundantly  in  euhedral  or  corroded  subhedral  shape,  and  occasion- 
ally in  glomeratic  clusters.  It  often  changes  into  a  yellowish  or 
greenish  fibrous  substance,  from  which  it  may  be  inferred  that  it  is 
of  a  variety  rich  in  magnesia.  The  olivine  is  colorless  in  section, 
and  encloses  copious  octahedra  of  hroirn  spinel.  Polarization- 
colors  are  indigo-blue,  purple,  brown,  and  gray  according  to  the 
thickness  and  orientation  of  given  sections. 

No  sanidine  or  any  other  feldspars  are  pi-esent.  Apatite  is  also 
absent.  Sporadic  patches  of  a  brownish  basaltic  base,  granulated 
and  sometimes  fibrous,  fill  up  the  interspaces  left  between  the 
idiomorphic  nepheline'-*,    and   the   overcrowded  augite-microlites 

1)    For  example,  around  the  hexagonal  section  of  nepheline,  PI.  I.  Jig.  2. 


On  Nepheline-basalt  from  "i'ingc-mon,  Manchuria. 


11 


swim  in  this  scanty  base,  being  rudely  arranged  tangentially 
around  the  nearly  isometric  crystals  (see  PL  l.ßg.  3)  of  nepheline, 
thereby  producing  the  appearance  of  a  leucite-melilite  rock. 

The  rock  is  nearly  holocrystalline,  a  variety  typically  rich  in 
nepheline  and  simple  in  mineralogical  composition.  If  the 
nepheline  were  absent,  the  mineralogical  composition  of  the  rock 
would  correspond  to  a  limburgite,  to  which  the  texture  has  a  close 
resemblance,  as  may  be  seen  in  the  photomicrographs,  PI.  l.ßgs.  2 
and  3.  The  order  of  crystallization  of  the  rock-components  is 
shown  in  the  following  scheme: 

Relative  Duration 


Magnetite 

Olivine 

Nepheline^^ 

Augite'-'^ 

Base 

'S 

O 

'S 

o 

The  chemical  analysis  of  the  rock  from  Tsao-sluli-err  was 
undertaken  by  ^lessrs.  S.  Shimidzu  and  T.  Ohashi,  of  our  Geolog- 
ical Survey,  to  whom  I  would  acknowledge  my  indebtedness.  The 
result  is  given  below  : 


1)  Two  generations. 

2)  Two  generations  (not  seen  in  the  photomicrograph).  In  the  older  and  larger  ones 
crystals  of  nepheline  are  poikilitically  enclosed  in  the  substance  of  augite  (0.6  mm  long)  which 
sometimes  suffers  magmatic  corrosion.  The  augite  is  so  fully  stuffed  with  octahedra  of 
magnetite  that  the  whole  presents  the  appearance  of  some  rhonite  crystals,  though  the  substance 
of  the  host  is  here  pyroxenic 


12 


B.  Koto: 


SiO, 44.98% 

ALO3  ....15.50 
FeXX....   5.15 


FeO.... 

..   7.30 

MgO... 

0  01 
..    o.oi 

CaO.... 

..  9.20 

Na^O.. 

..  5.34 

KX).... 

..   1.29 

H,0.... 

. .  3.77 

TiO,.... 

..  2.89 

P.O5... 

..  0.43 

MnO.. 

..  0.23 

S 

.  0.04 

Norms. 

Orthoclase  (K.OAlAôSiOo)  ...  7.8 

Albite  (NaoOALOsßSiO.) 23. 1 

Anortliite  (CaOAL032Si02)....  14.G 

Nepheline  (Na20A]o032SiO,)...  11.9 

rCaOSiOO 

Diopside  iMgOSio.,> 93  0 

iFeOSiO,,')                     " 

rw     '          (2MoOSiO..)  -,     A 

Magnetite  (FeOFe.Oa) 7.4 

Ilmenite  (FeOTiOJ 5.5 

Apatite  (3CaOP.A\) ••.  1.0 


Total       99.49         Sp.  Gr.  2.947—2.950. 

From  the  ratios  expressed  b}'  the  above  norms,  om'  rock  findti 
its  final  position  in  the  C.I.P.W.  quantitative  system,  as  in  the 


following'^: 

Sal        57.4      5      3 
Fem  ""38.3  "^3  ^5 

Class  III.  Salfemane. 

L      11.9      3      1 
F      45.5^^5^7 

Order  0.  Portngare. 

K.O'  +  Na^O'     100^7^   5 
CaO'        ~  53  "^  1  ^  3 

Eang  2.  jMonchiquase. 

K2O'       14      3      1           5 

Na20'~8G^5''7''^'^3 

> 

0 

Subrang  3-4.  Shonkinose 
Monchiquose. 

In  sampling  the  material  for  the  chemical  analysis,  all  the 
necessary  precautions  were  taken  by  the  writer.  The  analytical 
result  shows  near  approach  to  that  of  the  basalt  from  Franklin 
Island,    Antarctic.       The    ]Manchurian    rock    has.     however,     a 


1)     Calculations  made  by  Kôzii. 


On  Nepheline-basalt  from  Yingé-mên,  Manchnria.  13 

specific  peculiarities  worthy  of  note,  showing  exceptionally  high 
percentages  in  CaO,  and  H2O.  Microscopic  volumetric  analysis 
made  with  J.  Hirschwald' s  planimeter-ocular  showed  30  per 
cent  of  nepheline  and  nearly  15  per  cent  of  olivine,  the  latter 
value  heing  only  approximative,  due  to  the  phenocrystic  habit 
and  irregular  distribution  of  the  crystals  in  the  microscopic  field 
(PL  I.ßgs.  2  and  3). 

The  presence  of  large  amounts  (45.5%)  of  feldspars,  as  they 
are  expressed  in  the  above  norms,  is  to  my  mind  a  paradox,  as 
basaltic  glass  is  scantily  present  in  the  rock  in  which  at  least  the 
feldspar  molecules  must  be  assumed  to  exist.  Otherwise  they  must 
be  looked  for  in  the  composition  of  nepheline. 

The  chemical  composition  of  nepheline  has  long  been  a  problem 
much  discussed  among  mineralogists.  Lately  Foote  and  Brad- 
ley'-* have  offered  an  explanation,  namely,  that  a  substance  on 
crystallizing  may  form  '  '  a  solid  homogeneous  solution  with  foreign 
matter,"  and  that  the  mineral  nepheline  consists  of  a  pure  com- 
pound, probably  NaAlSi04,  with  a  varying  amount  of  dissolved 
silica.  Very  recently,  W.T.  Schaller^^  has  proposed  still  another 
explanation,  viz.,  that  the  mineral  nepheline  is  an  isomorphous 
mixture  of  the  compounds  crystallizing  in  the  hexagonal  modifica- 
tion, which  are  AlNaSiOj  (essential  component),  AlKSiO*  (kalio- 
philite),  and  AlNaSisOg,  the  last  being  only  in  mixture  in 
nepheline,  and  being  best  known  in  its  triclinic  form  as  albite. 
He  says  "the  remarkable  fact  that  the  compound  KAlSi04  is 
always  present  to  the  extent  of  about  20  per  cent  has  as  yet 
received  no  adequate  explanation."  The  albite  molecule  in 
nepheline,   however,   varies  from   5.6   to   10.6   per  cent.     At  all 

1)  "On  solid  solution  in  minerals  with  special  reference  to  nephelite."     Amer.  Jour.  Sc/i., 
4tli  ser.,  31  :  25.  1911. 

2)  "  The  chemical  composition  of  nephelite."     Jour.  Washinftton  Acad.  Sei.,  Vol.  1.  No.  4. 
September,  1911,  pp.  109-112. 


14  B-  Koto  : 

events,  the  feldspars  in  the  norms  seem  to  enter  largely  into  the 
composition  of  our  nephehne.  A  portion  of  water  may  be  present 
in  combination  with  zeolitized  products  of  the  nepheline. 

The  present  rock  is  a  simple  unique  nepheline-basalt  in  regard 
to  its  mineralogical  components.  As  it  seems  to  me  the  present 
rock  is  neither  a  plutonic  nor  a  dyke-rock,  I  cannot  call  it  a 
shonkinite  nor  a  monchiquite.  Moreover,  the. essential  attribute  of 
alkah -feldspar  is  wanting  in  the  Manchurian  rock  as  to  be  classed 
among  the  former,  and  a  brown  biotite  and  barkevikitic  amphibole 
is  not  present  so  as  to  be  included  among  the  latter.  But  I  learn 
from  petrological  literatures  that  each  of  both  rocks  assumes 
various  phases  of  crystallinity  even  within  the  same  mass,  ex- 
tended researches  of  our  Manchurian  rock  are  necessary  in  regard 
to  its  geognostic  relation  and  chemical  composition  in  order  to 
give  the  final  decision  as  to  what  species  it  belongs.  Since,  how- 
ever, in  mineralogical  composition  and  in  appearance,  it  is  most 
closely  related  to  nepheline-basalts,  it  is  here  referred  to  that 
group.  It  is  a  noteworthy  fact,  especially  in  regard  to  the  Sino- 
Japanese  region,  that  up  to  the  present  time  no  leucite-rocks  have 
ever  been  recoi'ded  from  north  of  the  Philippine  islands. 

My  thanks  are  due  to  Mr.  Swift,  of  our  University,  for  read- 
ing through  my  Englisli  manuscript. 

Dcecuiher,  1911. 


B.  KOTO: 

NEPHELINE-BASALT  FROM  MANCHURIA. 


PLATE     I. 


PLATE  I. 

Tig.  1. — View  southwards  from  Tsaosbiér  (PI.  II.)  toward  the  low  granitic  water-slisd 
of  tlie  Nie a-yii -lying  pass,  wliicli  separates  the  waters  of  two  mighty  Asiatic 
rivers  ;  the  rivulet  running  toward  us  is  the  upper  course  of  the  Hui-fa  river, 
a  tributary  of  the  Kirin-ula,  which  ultimately  joins  the  great  Amur  ;  while 
the  valley  beyond  the  water-parting  is  the  source  of  the  Hun-ho,  which  runs 
south  westwards  to  meet  the  well-known  Liao-ho.  The  plain  in  the  fore- 
ground is  an  elevated  granite  peneplain  of  490  ;//,  flankeJl  on  the  east  (left) 
by  an  overlying  mesa  {Chanii-kani/)  of  common  basalt,  150  m.  thick,  and 
bounded  on  the  west  by  denuded  hills  of  the  nepheline-basalt  which  is  the 
subject  of  the  present  paper.     (Page  6.) 

Tig.  2. — Nepheline-basalt  from  Tsaoshièr,  magnified  260  diameters,  set  in  with  a  large 
anhedron  of  olivine  (on  the  left)  and  a  basal  hexagonal  section  of  nepheline 
with  zonally  arranged  prismoids  of  augite.  The  groundmass  is  a  plexus  of 
short  prisms  of  nephehne,  prismoids  of  augite,  and  minute  speck-like  crystals 
and  clumps  of  titanomagnetite,  floating  on  sporadic  brown  patches  (on  the 
left  of  the  hexagonal  section  of  nepheline)  of  basaltic  base.     (Pages  9-10.) 

Fig.  3. — The  same,  magnified  130  diameters,  showing  the  general  appearance  of  the 
diverse,  seriate  fabric  of  the  rock  under  weak  powers.     (Pages  9  and  11.) 


Kot 5  :  Nephel i  n c-bn sa  1 1 . 


Jour.  Sei.  Coli.  Vol.  XXXII.  Art.  6.  PI.  I. 


NiEN-YÜ-LING   PASS 


Fio-.  I. 


Author,    Photo. 


B.  KOTO: 

NEPHELINE-BASALT  FROM  MANCHURIA. 


PLATE    II. 


PLATE  II. 

Map  showing  the  general  distribution  of  the  geologic  formations  represented  in 
tlie  Yingé-mèn  area  (p.  7),  the  land-feature  being  seen  in  PI.  I.,  fig.  1.  As  to  its 
pétrographie  elements  the  Koreo-Manchurian  hinterland  is,  broadly  speaking,  built  up 
of  the  two  opposite  poles — the  pinkish  granites  and  the  black  basalt  ;  and  this  is 
typically  exemplified  in  this  small  limited  area. 

The  region  is  also  of  historical  interest,  as  the  Pohais,  the  Mancluis  and  other 
ancient  highlauders  of  eastern  Manchuria  took  the  road  in  the  intermontane  plain  of 
Shan-chéng-tzu  (see  Map)  for  their  expeditions  into  the  Manchurian  plain  with  bold 
intent  of  swollowing  gigantic  China.  With  this  aim,  these  peoples  marched  along  the 
high  road  from  Tsaoshitr  (see  Map)  either  westwards  through  the  Tukoutzu-mcn  gate 
to  Kai-yiian,  or  southwestwarcis  through  the  Yinge-min  gate  to  Mukden.  At  these 
two  gates  the  road  was  cut  by  the  loug  mound  of  the  ancient  palisade  (see  Map),  now 
ruiued,  which  was  built  to  keep  back  the  swarms  of  "  northern  barbarians."  The 
Russians  did  the  very  same  thing,  as  the  ancient  highlanders  did.  They  followed  the 
same  track  in  the  Piusso-Japanese  war  on  their  march  from,  and  retreat  to^  the' 
secluded  and  safe  cil  y  of  Kirin. 

This  region  is  geologically  interesting  as  well  ;  for  besides  the  03currence  of 
iiepheline-basalt  there  is  a  narrow  band  of  the  Cambrian  with  Tertiary  beds  fulded  in 
what  is  apparently  old  granite-gueiss  in  the  N.E.-S.W.  direction.  The  Tertiary  here 
is  the  north  end  of  the  well-known  coal  seam  series  of  Mushun.  The  above-mentioned 
trend  is  the  guide-line  of  the  geologic  structure  that  governs  the  whole  Manchurian 
hinterland.  The  diorite  region  marked  on  the  south  is  probably  a  difierentiation- 
product  of  an  alkaline  granitic  magma.  This  region  (the  Hei-niu  goldfield)  is  also 
one  of  the  richest  auriferous  areas  in  Manchuria. 


Koto  :  Nepheline-basalt. 


Jour.  Sei.  Coll.  Vol.  XXXII.  Art.  6.  PI.  II. 


GEOLOGIC    MAP    OF  THE   YINGEMEN   AREA 


HEI-\IU    GOLD    FIELD  s„u^^  «y //u„:h^ 


1  :    1<  «)<>.( M t  0 


SMHi  E]M1 


-t* 


ïï 


'."I 


JOURXAL  OF  THE  COLLEGE  OF  SCIENCE,  IMPERIAL  UNIVERSITY, 
TOKYO,  JAPAN. 

VOL.  XXXII.,  ARTICLE  7. 


The  Systematic  Motions  of  Sun-spots. 

BY 

S.   Hirayama,     Bigahuhalcusln 
Professor  of  Astronomy,  Scieuce  College,  Imperial  University,  Tokyo. 


With    3    plates. 


The  Systematic  Motions  of  Sun-spots. 

1.  The  object  of  this  paper  is  to  call  attention  to  the  possible 
existence  of  certain  systematic  motions  of  the  sun-spots,  which 
has  not,  so  far  as  I  know,  been  hitherto  noticed.  It  was 
suggested  to  me  by  ]\Ir.  and  Mrs.  Maunder' s  paper  ''The  Solar 
Rotation  Period  from  Greenwich  Sun-spot  Measures,"  (Monthly 
Notices,  vol.  LXV,  pp.  813-825)  in  which  they  summarized  various 
important  results  relating  to  the  solar  rotation  period.  Among 
others,  the  following  statement  is  made:  "The  rotation  periods 
given  by  different  spots  in  the  same  zone  of  latitude  differ  more 
widely  than  do  the  mean  rotation  periods  for  different  zones  of 
latitude."  This  is  based  on  their  Table  II,  which  shows  the 
number  of  the  spot-groups,  arranged  in  zones  of  latitude  5°  wide, 
giving  different  synodic  rotation  periods.  Graphical  study  of  that 
Table  led  me  to  think  that  there  is  a  tendency  in  the  spot-groups 
to  concentrate  upon  two  particular  periods.  I  was  also  able  to  find 
a  similar  tendency  in  Carrington's  and  Spörer's  observations  of 
the  sun-spots. 

2.  To  begin  with  the  Greenwich  sun-spots  observations 
(1879-1901),  I  simply  took  the  said  Table  II  prepared  by  Mr.  and 
Mrs.  Maunder,  and  combined  the  zones  of  the  same  latitudes  north 


12 


s.  Hirayaina. 


Table    I. 


Greenwich  Numbers  of  Spot-groups,  arranged  in  Zones  of  Latitudes  5"  wide, 

giving  Different  Rotation  Periods 

(1879-1901). 


Synodic 

Rotation 

Period 

0° 
to 
5° 

5« 
to 
10° 

10° 
to 
15° 

15° 

to 
20° 

20° 
to 
25° 

25° 
to 
30° 

30'' 
to 
35° 

>35° 

d 

23.4 

1 

G 

8 

24.0 

2 

2 

2 

3 

3 

4 

2 

2 

3 

1 

6 

3 

3 

4 

1 

1 

8 

2 

4 

5 

4 

^5.0 

5 

G 

5 

8 

3 

1 

2 

6 

10 

11 

4 

2 

4 

3 

G 

18 

8 

1 

1 

6 

4 

12 

17 

12 

5 

2 

8 

G 

20 

25 

18 

3 

1 

^6.0 

12 

22 

26 

15 

12 

1 

2 

6 

41 

39 

23 

11 

3 

4 

IG 

33 

45 

29 

9 

1 

6 

24 

33 

49 

29 

17 

5 

8 

16 

46 

61 

30 

10 

G 

27.0 

14 

59 

82 

48 

15 

1 

2 

5 

41 

76 

63 

25 

3 

4 

2 

27 

52 

51 

20 

6 

G 

2 

5 

20 

41 

26 

7 

1 

8 

1 

8 

13 

21 

24 

10 

28.0 

1 

G 

8 

6 

8 

11 

2 

2 

4 

5 

7 

12 

4 

4 

1 

2 

5 

5 

10 

G 

6 

1 

4 

1 

1 

2 

The  Systematic  Motions  of  Sun-spots. 


Synodic 

Rotation 

Period 

0'' 

to 
5° 

5° 

to 
10' 

10° 
to 
15° 

15° 
to 
20° 

20° 
to 
25' 

25° 
to 
30° 

30° 

to 
35° 

>35° 

8 

2 

G 

3 

2 

1 

29.0 

1 

1 

3 

o 

2 

1 

1 

2 

4 

1 

1 

1 

G 

1 

8 

1 

1 

m.o 

2 

1 

1 

2 

4 

Ü 

o 

8 

31.0 

2 

1 

Mean 

Rotation 

Period 

a 
2(3.36 

d 

26.59 

d 

20.73 

d 

2  ).89 

d 

27.22 

27.''48 

^and  south.  The  resuUing  numbers  are  given  in  Table  I,  which 
therefore  exhibits  the  distribution  of  the  spot-groups  in  two  ways, 
the  horizontal  lines  showing  the  numbers  of  groups  yielding 
■different  synodic  rotation  periods,  and  the  vertical  columns  the 
number  of  groups  in  each  zone  of  latitude  5°  Avide.  The  numbers 
of  Table  I  are  represented  graphically  in  Plate  T,  a  series  of  curves 
being  drawn,  one  for  each  zone.  Considering  now  these  curves 
shown  in  Plate  I,  each  small  circle  represents  the  number  of  spot- 
groups  corresponding  to  the  given  rotation  period.  They  are 
connected  by  straight  lines,  while  the  heavy  continuous  lines 
indicate  the  smoothed  values. 

This  Plate  shows  at  a  glance  that,  while  there  are  minor 
differences  between  the  curves  for  different  zones,  yet  the  main 
features  are  repeated  in  a  remarkable  way  in  all  six  series.     We 


4  -  s.  Hirayama. 

may  observe  that  there  is  a  tendency  in  the  spot-groups  to  concent- 
rate upon  one  particular  period,  but  each  curve  is  not  symmetrical 
about  the  ordinate  of  maximum  number.  Another  important 
point  is  the  existence  of  a  secondary  concentration.  At  first  I 
thought  that  if  there  existed  a  gi-eater  numljcr  of  recorded 
observations,  then  I  could  get  a  sort  of  probability  curve,  but 
actually  it  seems  likely  that  each  curve  will  be  a  combination  of  two 
probability  curves.  Because  of  the  impossibility  of  determining 
rigorously  the  values  of  several  constants  Avith  non-linear  equations 
by  the  method  of  least  squares,  and  moreover  because  of  the 
doubtful  nature  of  the  problem,  I  have  endeavoured  to  determine 
the  positions  of  the  principal  and  secondary  maxima  by  graphical 
method.  When  the  maximum  is  not  well  pronounced,  I  have 
derived  the  smoothed  curves  by  compounding  two  sjaiimetrical 
curves  about  the  directions  marked  I  and  II.  All  doubtful  cases 
(marked?  on  the  plates)  have  been  excluded.  I  have  thus  obtained 
the  following  results: 

Taele    II. 

Angular  Velocity  corresponding  to  Maximum  I. 


Heliographie 
Latitude 

Synodic 

Rotation 

Period 

Observed 
Ang.  Vel. 

Xumber  of 

Spots  in 
Maximum 

Smoothed 

Ang.  Vel. 

Ig 

Maunders 
Ang.  Vel. 

0 

2.5 

26.60 

14!52 

22 

14^45 

14^44 

7.5 

27.00 

14.32 

54 

14.88 

14.41 

12.5 

27.075 

14.28 

68 

14.29 

14.34 

17.5 

27.25 

14.20 

51 

14.19 

14.25 

22.5 

27.55 

14.05 

26 

14.05 

14.13 

27.5 

27.90 

13.88 

11 

13.88 

13.99 

The  Systematic  Motions  of  Sim-spots. 

Table    III. 

Angular  Velocity  CDrrasponding  t3  Maximum  II. 


Heliographie 
Latitude 

Synodic  Rotation 
Period 

Observed  Angular 
Velocity 

Number  of  Spots 
in  Maximum 

Smoothed  Ang. 
Velocity. 

IlQ 

0 

2.5 

? 

(J 

? 

9 

(14J2) 

7.5 

26.225 

\4.n 

28 

14.71 

12.5 

26.275 

14.69 

80 

14.69 

17.5 

26.35 

14.65 

20 

14.65 

22.5 

26.45 

14.60 

10 

14.59 

27.5 

26.70 

lAAl 

6 

14.49 

The  value  in  Ijrackets  is  one  obtained  by  extrapolation. 


Angular  Velocities  of  Maxima  I  and  II  for  Different  Zones  of  Latitude, 
deduc3d  from  Grsenwicli  Observations. 


Fig.  1 


30° 

1            1            1      1      ! 

'    '••-'■  '^ 

I       1    1 

- 

>o 

,-'           ^^^ 

- 

25° 

X 

,. 

y   ^0^ 

f 

p. 



o/jLo 

^Ic 

- 

10° 

' 

[ 

// 

5° 

.^ 

- 

1        1    i    1      /i 

1  1  1  1  1 

•1        1        1    1 

1        1 

.30 


25' 


ir.  (^ 


10' 


14°8   14°7   14°0    14''3    14°4    14°3   14°2    "'"1    14°0   rrn    ^?,'^H       ""-^   l-TG 

Angular  Velocity 


Figure  1  shows  diagvammatically  the  numerical  results  of 
Tables  II  and  III. 

In  the  last  column  of  Table  II,  I  have  added  Maunder' s 
values    of    daily    angular    velocity    derived    from     his    formula 


(3  S.  Hirayaira. 

^=866'. 6±  128'  sin^/,  where  ç  denotes  the  angular  velocity  and  ?^  the 
latitude.  TJiese  values  obtained  by  his  formula  are  generally 
greater  than  iny  results.  That  is,  they  do  not  correspond  exactly 
to  the  angular  velocities  for  my  Maximum  I.  As  to  the  second 
Maximum,  I  have  obtained  a  pretty  good  continuous  curve  by 
merely  connecting  the  successive  positions  of  Maximum  II, 
corresponding  to  different  zones. 

3.  Next  I  pass  on  to  examine  Carrington's  observations  of 
the  spots  on  the  sun  so  far  as  they  bear  on  the  question  of  the 
existence  of  a  secondary  maximum.  All  the  materials  from  which 
Carrington  deduced  his  formula  for  the  daily  motion  of  the  surface 
of  the  sun  in  different  solar  latitudes  are  condensed  in  his  "  Table 
of  Resulting  Diurnal  Motions"  contained  in  pp.  213-219  of  his 
work.  I  have  availed  myself  of  the  same  materials,  and  deduced 
Table  IV  which  shows  the  distribution  of  tlie  sun-spots  observed  by 
him  in  each  zone  of  latitude  5°  wide,   with  different  daily  drifts. 

Table   IV. 

Carrington's  Numbers  of  Sun-spots,  arranged  in  Sones  of  Latitudes  5°  wide, 
giving  Different  Diurnal  Motions  (1853-1861). 


Daily  Drift 

0° 

to 
5° 

G° 
to 
10" 

11° 

to 
15° 

16° 
to 
10° 

21° 
to 
25° 

26° 
to 
30° 

31° 
to 
35° 

From +  80' to -H  76' 

1 

„     +75  „  +71 

,     +70 

,+66 

,     +65 

,+61 

,     +60 

,+56 

1 

1 

1 

,     +55 

,+51 

1 

,     +50 

,+46 

1 

,     +45 

,+41 

1 

,     +40 

,  +36 

1 

2 

2 

,     +35 

,  +31 

2 

1 

2 

,     +30 

,+26 

6 

1 

1 

The  Systematic  Motions  of  Sun-spots. 


Daily  Drift 

0°     . 

k) 

5° 

6° 
to 
10" 

11° 
to 
15' 

16° 
to 
20' 

21° 
to 
25° 

26° 
to 
30° 

31° 
to 
35° 

From +25' to +  21' 

3 

4 

2 

1 

„     +20  „  +10 

i 

G 

6 

„     +15  „+11 

1 

8 

5 

4 

„     +10  „  +   6 

3 

11 

7 

7 

2 

,,     +   5  „  +    1 

1 

15 

7 

] 

0  „  -  4 

2 

5 

23 

0 

7 

1 

,.     -   5  „  -   9 

5 

0 

11 

4 

2 

„     -10  „  -14 

2 

0 

15 

10 

2 

1 

„     -15  „-19 

3 

7 

7 

4 

„     -20  „  -24 

5 

11 

4 

„     -25  „  -29 

2 

0 

2 

1 

„     -30  „-34 

1 

5 

10 

5 

„     -35  „  -39 

1 

2 

4 

„     -40  „  -44 

2 

4 

2 

„     -45  „  -49 

1 

„     -50  „-54 

2 

2 

„     -55  „-59 

1 

„     -60  „  -04 

1 

1 

,,     -05  „  -09 

„     -70  „  -74 

,.     -75  „-79 

1 

Mean  Daily  Drift 

+  17/7 

+  lO.'S 

+  3/3 

-3/9 

-1/0 

-29/1 

The  first  column  of  Table  IV  requires  further  explanation.  It 
represents  Carrington's  diurnal  motions.  According  to  his  words, 
"The  signs  prefixed  to  the  diurnal  motions  are  such  that  +  in 
longitude  indicates  rotation  faster  than  14°  11'  per  diem  (cor- 
responding to  the  assumed  period  of  25.  380  days)  and  —  rotation 
slower  than  that."     The  plotted  numbers  are  shoAvn  graphically  in 


8 


s.  Hirayama. 


Plate  II.  The  method  of  treatment  of  the  material  in  Plate  II  is 
just  the  same  as  before.  Here  again,  as  in  the  previous  case,  the 
secondary  maximum  is  clearly  brought  out  in  each  curve.  Graph- 
icalty  determined  maxima  may  be  tabulated  as  foUows  : 

Table   V. 

Angular  Velocity  corresponding  to  Maximum  I,  deduced  from 
Carrington's  Observations. 


Heliog. 
Latitude 


Observed  Angular  Velocity. 


Number  of 
Spots  in 
Maximum 


Heliog. 
Lati  ude 


Smoothed 

Ang.  Vel. 

Jc 


Carring- 
ton's Ang. 
Vel. 


8° 
13 
18 
23 

28 


851'+  5'=856'=14.''27 
„  -  2  849  14.15 
„  -11  840  14.00 
„  -20  831  13.85 
„    -33      818      13.63 


13 

19 

13 

9 

8 


2.5 
7.5 
12.5 
17.5 
22.5 
27.5 


(14.°36) 
14.27 
14.16 
14.02 
13.  b6 
13.67 


14.°42 
14.35 
14.21 
14.06 
13.90 
13.73 


Table   VI. 

Angular  Velocity  corresponding  to  Maximum  II,  deduced  from 
Carrington's  Observations. 


Heliog. 
Latitude 

Observed  Angular  Velocity 

Number  of 

Spots  in 

Maximum 

Heliog. 
Latitude 

Smoothed 

Ang.  Vel. 

lie 

8° 
13 
18 
23 

28 

85r  +  30'=881'=14.°68 
„    +18      869      14.48 
„    +10      861      14.35 
„    -   7      844      14.07 
„    -19      832      13.87 

4 

5 
6 

2 
4 

2.°5 
7.5 
12.5 
17.5 
22.  5 
27.5 

(14.°81) 

14.67 
14.51 
14.34 
14.13 

13.88 

'J'he  Systematic  Motions  of  Sun-spots. 

Angular  Veloc  ties  of  Maxima  I  and  II  far  Différent  Zon3s  of  Latitude, 
deduced  from  Carrington's  Observations. 

Fig.  2. 


-r.2.0 


ira       I'l  "7       14?6       14.6 


1-4  °4         14°3        14°2        14°  1        14°0 
Angular  Velocity 


l.'5°9        i;i°8        i;r7       13?6 


Carrington's  angular  velocity  given  in  the  last  column  of  Table 
V  is  derived  from  his  formula  for  the  angular  velocity,  ^=865'+ 
165' sin'/. 

4.  Similarly  I  have  examined  Spörer's  observations.  The 
data  for  our  discussion  of  the  spots  as  observed  by  him  have  been 
taken  from  the  same  source  as  that  from  which  he  deduced  his 
expression,  8°. 548 +  5°. 798  cos /Î,  for  the  daily  motion  of  the  sun- 
spots  in  different  solar  latitudes,  namely  his  "  Beobachtungen  der 
Sonnenflecken  zu  Anclam."^'^  The  observations  comprise  the 
period  1861-71,  the  number  of  the  spots  whose  angular  velocities 
are  computed  being  264.  To  increase  this  number,  I  have 
extended  the  discussion  of  his  observations^"^  up  to  the  end  of  the 
year  1878.  So  the  total  number  of  spots  employed  became  334. 
Similarly  to  the  others,  Table  VII  and  Plate  III  were  prepared. 
It  is  unfortunate  that  none  of  the  curves  in  Plate  III  show 
so  striking  a  secondary  maximum  as  in  the  previously  con- 
sidered   cases,     although    there    is    no    doubt    that    each  curve 


(1)  Publicationen.  der  Astronomischen  Gesellschaft.  XIII  1874  pp.  139—146. 

(2)  Publicationen  des  Astro-physikalis:heu  Observatoriums  zii  Potsdam.     Xr.  5.     p.  06. 


10 


s.  Hirayama. 


Table   VII. 

Sporer's  Wumbers  of  Sun-spots,  arranged  in  Zones  of  Latitudes  5°  wide, 
giving  Different  Diurnal  Motions.    (1861-1878) 


Diüly 
Motion 

0° 
to 
5" 

5° 
to 
10° 

10° 
to 
15° 

15' 
to 
20° 

20° 
to 
25' 

25° 
to 
30° 

VSM 

2 

13.5 

1 

1 

13.(5 

5 

3 

13.7 

7 

13.8 

7 

9 

5 

13.9 

4 

11 

12 

1 

14.0 

o 

17 

2i 

8 

2 

14.1 

1 

16 

41 

24 

4 

14.2 

3 

30 

35 

10 

5 

14.3 

5 

33 

20 

7 

1 

14.4 

11 

10 

8 

3 

14.5 

(■) 

6 

2 

1 

14. Ü 

3 

4 

;-5 

14.7 

3 

3 

1 

1 

14.8 

2 

2 

14.  U 

1 

1 

15.0 

2 

15.1 

1 

aJean  Daily 
Motion 

14°.43 

I4.°3l 

14.^18 

14.^07 

13.°94 

13.°71 

is  not  symmetrical  about  the  ordinate  corresponding  to  jNIaxi- 
nium  I.  In  fact,  the  sun-spots  selected  by  Sporer  are  best 
suited  for  finding  Maximum  I,  but  not  so  favourable  for  finding 
Maximum  II.  However,  so  far  as  Plate  III  shows,  I  can  deduce 
the  following  results.  Better  results  M'ould  perliaps  be  reached  by 
discussing  the  angular  velocities  of  all  the  spots  observed  by 
him. 


The  Systematic  Motions  of  Sim-spots. 


11 


Table   VI 11. 

Angular  Velocity  corresponding  to  Maximum  I. 


Heliog.  Latitude 

Observed  Ang. 
Vel. 

Number  of  Spots 
in  Maximum 

Smoothed  Ang. 
Vel. 

Sporer's  Ang. 
Vel. 

2.''5 
7.  5 
12.5 
17.5 
22.5 
27.5 

14.°40 
14.25 
14.14 
14.05 
13.  89 

10 
31 
38 
23 
11 

14.'3() 
14.28 
14.  17 

14.03 
13.  90 

(13.77) 

14.°34 
14.30 
14.21 

14.08 
13.  90 
13.69 

Table  IX. 

Angular  Velocity  corresponding  to  Maxmum  II. 


Heliog.  Latitude 

Observed  Ang.  Vel. 

Xumber  of  Spots  in 
Maximum 

Smoothed  Ang. 
Vel. 
lis 

2.*'5 
7.5 
12.5 
17.  5 
22.5 
27.5 

14.°71 

9 
14.35 
14.  30 
14.20 

1 

? 

6 

7 
4 

14.^6.-) 
14.53 
14.  42 

14.30 
14.18 

(14.07) 

Angular  Velocities  of  Maxima  I  and  II  for  Diiï'erent  Zones  of  Latitude, 
deduced  from  Spörer's  Observation. 


ira       Wl        14°c;         14"5        14°4 


14°3      U°2       j-ri 
-Vii.rnl.ir  Vi'l'tit.1 


o    « 


10  cc 


Il  II        l.Tf)        iXH        ]?.'l        13°6 


]2 


s.  Ilirayamn. 


5.     1  can   DOW  compare   Ihc  results  arrived  at    from    these 
various  observations  as  follows: 

Table    X. 

Angular  Velocity  corresponiiog  to  Maximum  I. 


Heliog. 
Latitude 

Deduced  from 
Greenwich 

Observations 

(1879-1901) 

Ig 

Deduced  from 

Spörer's 

Observations 

(1861-1878) 

Is 

Deduced  from 

Carrington's 

Observations 

(1853-1861) 

Ic 

Simple 

Mean 

I 

Mean  Ang.  Vel. 
coniputed  from 

Maunders, 

Spörer's,  and 

Carrington's 

Formulae 

2.^5 

14.°45 

14.°36 

(14.°36) 

14.°39 

14.°40 

7.5 

14.38 

14.28 

14.27 

14.31 

14.35 

12.5 

14.  29 

14.17 

14.  lu 

14.21 

14.25 

17.5 

14.  19 

14.03 

1J.02 

14.08 

14.13 

22.5 

14.05 

13  90 

13.86 

13-94 

13.98 

27.5 

13.88 

(13.77) 

13.67 

13.77 

13.80 

Table   XL 

Angular  Velocity  corresponding  to  Maximum  II. 


Heliog. 
Latitude 

Deduced  from 
Greenwich 

Observations 
(1879-1901) 

IlG 

Deduced  from 

Spörer's 

Observations 

(1861-187?) 

lis 

Deduced  from 

Carrington's 

Observations 

(1853-1861) 

lie 

Simple 

Mean 

II 

Mean  SynocU  c 

Rotation 

Period 

2.°5 

(14.°72) 

14.''65 

(14.°81) 

14.°73 

26.^9 

7.5 

14.71 

14.53 

14.67 

14.64 

26.  37 

12.5 

14.69 

14.42 

14.51 

14.54 

26.  56 

17.5 

14.65 

14.  30 

14.34 

14.43 

26.  78 

22.5 

14.59 

14.18 

14. 13 

14.30 

27.04 

27.5 

14.49 

(14.  07) 

13.88 

14.15 

27.  35 

The  values  in  brackets  have  been  obtained  by  extrapolation. 

The  mean  values  corresponding  to  Maximum  I  and  Maximum 
II  respectively  of  angular  velocities  in  zones  of  solar  latitude  each 
5°  wide,   are  shown  under  the  heading  'Simple  Mean'   of  Table 


The  Systematic  Motions  of  Sun-spots. 


13 


X  and  Table  XI.  These  velocities  are  plotted  gra2)liically  in  Fig. 
4.  Thé  dotted  curve  in  the  same  figure  represents  the  mean 
angular  velocity,  as  computed  from  Maunder' s,  SjoOrer's,  and  Car- 
rington' s  formula}.  Its  numerical  values  are  given  in  the  last 
column  of  Table  X.  For  the  sake  of  convenience,  the  two  drifts 
corresponding  to  Maximum  I  and  Maximum  II  will  in  the 
remainder  of  this  article  be  called.  Drift  I  and  Drift  II  respectively. 

Angular  Velocities  of  Drift  I  and  Drift  II. 

Fig.  4. 


10    ^ 


14°H       14° 


14° 


14°4 


\Td 


i3°a     'i'^'i     ^X(^ 


w?.     14°?.      ]4°i      wo 
Angular  Vtlocity 

6.  On  the  whole,  the  angular  velocities  of  Drift  I,  as 
represented  in  Table  X,  accord  very  Avell  with  the  mean  values  as 
computed  from  the  formuhie  of  the  three  investigators,  though 
there  is  a  small  systematic  différence  of  about  0°.04  on  the  average. 
In  fact,  the  values  obtained  by  these  authors  do  not  exactly  rep- 
resent the  angular  velocities  of  spot-groups  of  maximum  occur- 
rence. 

7.  As  to  Drift  II,  it  will  be  noticed  from  Table  XI  that  there 
is  a  rather  considerable  discrepancy  in  the  values  derived  from  the 
Greenwich  observations  and  from  the  other  two,  although  there  is 
no  great  difference  between  the  results  deduced  from  Spörer's  and 
Carrington's  observations.  It  is  evident  that  the  values  at 
latitudes  2°.5  and  27°.5  are  of  low  vreights,  the  first  being  based 


]  4  s.  Hirayama . 

upon  only  a  single  determination,  and  the  second  upon  but  two. 
A  comparison  of  Table  X  and  Table  XI  shows  a  faster  rate  of 
daily  angular  motion  for  Drift  II  than  for  Drift  I.  The  mean 
acceleration  of  the  angular  velocities  is  about  0°.35,  within  the 
limits  of  observation,  the  corresponding  acceleration  in  the  rota- 
tion period  being  O.'^Tl.  For  the  lower  latitudes,  the  rotation 
periods  of  the  spots  belonging  to  Drift  II  agree  approximately 
with  the  recent  spectroscopic  results  obtained  by  Messrs.  Storey 
and  Wilson. "^'^  Perhaps  certain  groups  of  spots  by  a  proper  motion 
of  their  own  come  to  the  same  level  as  the  reversing  layer  and 
attain  its  angular  velocity. 

As  to  the  peculiar  motions  of  sun-spots  Prof.  Sporer  sa3^s: 
"Die  Beobachtungen  haben  ergeben,  dass  im  östlichen  Theile 
einer  Gruppe  niemals  übergrosse  Rotationswinkel  vorkommen. 
Man  findet  daselbst  X'erkleinerung  der  Rotationswinkel,  aber  selten 
mit  bedeutenden  Betrage.  Uebergrosse  Rotationswinkel  kommen 
vor  an  der  Westgrenze  der  Gruppen  und  bei  neu  entstandenen 
Flecken."  He  then  mentions  6G  cases  of  great  angular  velocities 
during  the  period  1880-84.  This  excess  comes  out  to  be  about 
0°.5  on  the  average.  Perhaps  such  spots  and  those  of  short 
duration  niay  greatly  contribute  to  the  existence  of  Drift  II. 

8.  Since  the  ratio  of  the  amplitudes  of  the  two  drifts 
depends  on  the  choice  of  spots  by  each  investigator,  it  is  difficult 
to  find  the  true  ratio  from  the  investigations  hitherto  made. 
By  examining  the  tables  in  this  article,  it  will  be  found  that 
it  is  actually  different  for  different  investigators.  It  also  varies 
irregularly  in  different  zones.  The  mean  ratio  of  the  amplitudes 
of  the  two  drifts  is  about  J,  J,  and  7  in  Maunder' s,  Carrington's 
and  Sporer's  observations  respectively.  The  researches  of  Mr.  and 
Mrs.  Maunder  were  made  on  spots  in  general.  They  say:  ''  There 
has  been  no  selection  of  spots  because  they  seemed  to  be  steady  in 
motion  or  regular  in  shape,  no  rejection  because  of  unsteadiness  or 
irregularity.  The  only  criterion  for  the  inclusion  of  a  group  in  the 
discussion  has  been  that  it  lasted  for  six  consecutive  days  " .     This 


(I)    storey  and  Wilson,  Spectroscopic  Observations  of  the  Sun's  Rotation,  etc.     MonfMy 
Notices  LXXI  p.  674. 


The  Systematic  Motions  of  Sim-Spots. 


15 


fact  lias  led  me  to  assume  that  the  sun-spots  belonging  to  Drift  II 
are  about  ^  of  the  whole. 

9.     If  we   adopt  Faye's  empirical  formula  fur  solar  motion, 
then  our  results  may  be  expressed  as  follows: 
?  =  14°.37-2°.97  sin;.,     for  Drift  I. 
ç  =  14°.69-2°.65  sin;,     for  Drift  II. 
In  order  to   exhibit  how   these   formulœ  satisfy   the   obser- 
vations, I  give  the  following  table. 


Table   XII. 
Comparison  of  Computed  with  Observed  Angular  velocities. 


Drift  I. 

Drift  II. 

X 

Observed 

Weight 

Computed 

0-C 

Observed 

Weight 

Computed 

0-C 

2.°5 

1 4.°39 

o 
3" 

14."36 

+  0.°03 

14.°73 

1 
3 

14.°6S 

+  0.°05 

7.5 

14.81 

1 

14.32 

-       1 

14.64 

1 

14.04 

0 

12.5 

14.21 

1 

14.  23 

-       2 

14.  54 

1 

14.  57 

-       3 

17.5 

14.08 

1 

14.11 

-       3 

14.  43 

1 

14.45 

_       2 

22.5 

13.94 

1 

13.94 

0 

14.30 

1 

14.30 

0 

27.  5 

13.77 

2 
3 

13.75 

+       2 

14.  15 

3 

14.13 

+        2 

It  will  be  noticed  that  the  above  formula  for  Drift  I  nearly  coin- 
cides with  the  expression,  ^=14°.37-3°.10  sin';  deduced  by  Faye. 

10.  By  examining  all  the  spots  which  lasted  for  more  than 
six  consecutive  days,  Mr.  and  Mrs.  Maunder  (loc.  cit.  p.  818)  have 
deduced  another  formula,  ç=875'. 7=^164'  sin';,  which  nearly 
corresponds  to  the  mean  values  in  the  last  line  of  Table  I.  This 
formula  may  be  analysed  as  follows: 
i(ç  of  Drift  I)-l-|(,-  of  Drift  II) 

=Kl4°.37-2°.97  sin';0+l(l-i°ö9-2°.65  sin';) 
=  14°.61-2°.73  sin';=876'.6-163'.8  sin'A 

That   the   last    expression    practically    coincides    with    their 
formula  shows  that  the  latter  is  greatly  influenced  by  Drift  II. 


^ß  s.  Hirayama. 

The  mean  value  of  é,  (column  G  of  Table  X),  as  computed  from 
Carrington's,  Spörer's,  and  Maunder' s  formulœ,  is  nearly  expressed 
by  the  formula,  ç  =  14°.40-2°.83  sin';,  which  may  also  be  analysed 
as  follows  : 

r\,(c  of  Drift  I)  +  A(ç  of  Drift  II) 

=  ^(14°.37-2°.97  sim;0+TV(14°.69-2°.65  sin^;) 

=      14°.40-2°.94  sin';. 
11.     The  present  investigation,   though  cursory,  leads  meto 
conclude  that  there   are  two  apparent  drifts  in  the  motions  of  the 
sun-spots.     The  angular  velocity  of  Drift  I  is  represented  by 

.     ç  =  14°-37-2°.97  sin-/ 
and  that  of  Drift  II  by 

ç=14°.69-  2°.G5sin';-, 
showing  a  mean  rate  of  about  (/.35  greater  than  that  indicated 
by  Drift  I.  The  mean  ratio  of  the  number  of  sun-spots  in 
Maximum  II  to  those  in  Maximum  I  is  1 :  2.  This  hypothesis  of 
assuming  the  existence  of  a  secondary  drift  may  be  considered  a 
tentative  explanation  of  the  phenomena  of  the  distribution  of 
sun-spots  with  different  rotation  periods  in  any  particular  zone  of 
latitude,  and  I  do  not  claim  tliat  the  conclusion  I  have  arrived  at 
do  more  than  approximate  to  quantitative  precision. 

Tokyo  : 

1912,  January  20. 


Publ.  June  7th,  1912. 


Jour.  Sei.  Col  I., Vol.  XXX 1 1.,  Art.  7,  PI.  I. 

Greenwich  Number  of  Spot-groups  with  Different  Rotation  Periods 
in  Zones  of  Latitude  5°  wide. 


Jour.  Sei.  Coll., Vol.  XXXll.,Art.  7,  PI.  II. 

Carrington's  Number  of  Sun  spots,  arranged  in  Zones  of  Latitude  5^ 
wide,  giving  Different  Diurnal  Motions. 


-62-57'-5Z'-47'-42'-37'-32-Z'7'-Z2'-n'-JZ-T     -2'  *3'   *8'  *J3'  *1&'*Z3' *Z8' *J3*38'*43*46'*53' *B6' 


-6Z'-ST-32'-47-4Z'-37'-3Z'-ZT-2Z'-J7'-JZ'  -T  -2'    +3'    +8'  +73   +J8  ♦23'+28 +33'*38V43'+ 40 V53' 


Jour.  Sei.  Col  I., Vol.  XXX 1 1.,  Art.  7,  PI.  III. 


Spörer  s  Number  of  Sun  spots  with  Different  Angular 
Velocities  in  Zones  of  Latitude  5°  wide. 


13  5  13'6  13'7  13°ô  13°9  14"0  I4°l    14°2  14°3  I4!4    J4°5  14°6  14°7  I4°a  ]4?9  15°0  15°1 


T  ' 


JOUIiNAL  OF  THE  COLLEGE  OF  SCIENCK,  TOKYO  IMPERIAL  UNIVERSITY. 

VOL.  XXXII.,  ARTICLE  8. 


The  Metallogeny  of  the  Japanese  Islands. 

By 
C.  Iwasaki,  niqahushi. 


With  1  Map. 


1.    Magmalic  Emanations  and  their  Petrification. 

Since  SvanteArrhenius^^  made  public  the  results  of  his  studies 
on  the  chemical  properties  of  water  in  high  temperatures,  the 
world's  geologists  have  been  more  or  less  influenced  by  him,  their 
views  on  metallogeny  changing  gradually  from  the  hydrothermal 
to  the  magmatic  theory,  Heavj"  metals  were  formerly  supposed 
to  have  been  brought  up  by  hot  springs  in  the  form  of  mineral 
solution.  But  at  present  ore-deposits  are  believed  to  have  been 
deposited  by  gas,  or  by  a  mixture  of  gas  and  liquid,  or  by  liquid 
only,  emitted  from  magma  while  cooling.  This  is  asserted  by 
J.  H.  L.  VoGT,'^  who  calls  this  ''eruptive  after-action."  This  view 
naturally  leads  us  to  believe  that  where  ore-deposits  exist,  there 
must  be  found  igneous  rocks,  near  or  distant,  from  which  the 
materials  of  the  ore-deposits  were  emitted.  Prof.  Koto,^^  in 
his  recent  paper  entitled  "  Geology  anà  Ore-deposits  of  the  Holgol 
Mine,''  1910,  calls  these  rocks  '' ore-bringers." 

It  had  already  been  recognized  by  Matteucci,  Lacroix,*^ 
Gautier,^  and  others  that  a  great  many  elements  always  exist  in 
emanations.  When  the  latter  cool,  these  elements  form  different 
kinds  of  minerals.  It  is  in  this  way  that  petrification  of  emanations 
takes  place. 

1)  Svante  Arrhenius  :  "  Zur  Physik  des  Vulkanisms,"  Stockliolm,  1900. 

2)  The  Genesis  of  Ore  Deposits,"  p.  642. 

3)  B.  Koto  :  Jour.  Coll.  Sei.  Imp.  Univ.  Tokyo,  Vol.  XXVII.,  Art.  12,  1910. 

4)  Matteucci  and  Lacroix:  The  Digest  in  Economic  Geolor.u,  Vol  Tl.,  No.  3,  p.  25S,  1907. 

5)  Gautier  :  Economic  Geolomj,  Vol.  I.,  p.  690,  1906. 


Iwasaki 


The  petrification  may  often  be  seen  distinctly  in  recent  lavas. 
As  an  example,  let  me  explain  here  the  occurrence  of  tridymite  in 
Ishigami-yama,  a  hillock  in  the  environs  of  the  city  of  Kumamoto. 
This  hill  is  a  part  of  a  lava  flow  extruded  from  Kibo-san,  a  now 
extinct  volcano,  and  is  composed  of  amphibole-pyroxene  andésite. 
The  rock  is  fresh  in  composition,  whitish  in  color,  with  conspicuous 
phenocrysts  of  amphibole  converted  into  the  pseudomorphs 
of  magnetite  and  augite  grains  by  the  resorption  of  the 
original  crystals.  In  the  miarolitic  fissures  and  cavities  formed 
during  the  consolidation  of  the  lava,  several  minerals  are  found, 
such  as  tridymite,  phlogopite,  breislakite,^^  specular  iron,  and 
calcite,  all  forming  very  fine  crystals.  The  tridymite  is  in 
hexagonal  plates,  sometimes  attaining  5  mm.  in  diameter.  The 
mineral,  when  picked  out  of  the  rock  cavities,  is  transparent,  but 
very  soon  becomes  whitish  and  translucent  on  exposure  to  the  air. 
The  change  of  color  may  be  clearly  explained  by  microscopic  study. 
The  trid3aiiite  occurs  in  the  form  of  a  pile  of  thin  larainse,  and  has, 
when  it  is  picked  out  of  the  rock  cavities,  a  light-brownish  liquid 
in  the  interspaces  between  the  plates.  When  exposed  to  the  air, 
the  liquid  immediately  evaporates  and  the  tridymite  becomes 
whitish  by  total  reflection  of  light.  The  liquid  contained  in  the 
tridymite  is  supposed  to  be  what  is  left  of  juvenile  water  extruded 
from   the  cooling  lava. 

The  presence  of  calcite  as  an  emanation-product  in  recent 
lava  is,  so  far  as  known,  extremely  rare.  The  other  minerals  in 
the  miarolitic  cavities  are  also  supposed  to  be  all  of  juvenile  origin, 
and  not  of  the  vadose  formation  ;  in  other  words,  all  belong  to  the 
so-called  fossil  emanations  of  Lincoln."^  To  prove  it,  I  shall  give 
here  the  following  three  data:  (1)  the  side-wall  of  the  miarolitic 
cavities  and  fissures  presents  the  slaggy  aspect  usually  seen  on 
the  surface  of  lava,  (2)  the  cavities  and  fissures  are  perfectly  closed 
as  if  to  prevent  the  infiltration  of  vadose  water,  (3)  the  andésite 
in  which  the  cavities  and  fissures  are  found  is  quite  fresh,  showing 
that  the  minerals  in  question  are  not  decomposition-products. 

1)  It  was  so  determiaed  by  B.  Koto. 

2)  Liacoln  :  "  Economic  Geology,"  Vol  II..  No.  3,  p.  253,1907. 


-Metallos'ény  of  tho  Japaneso  Islands. 


2.    Classification  of  Ore-deposits. 

vox  Waldensteix^^  and  vox  Cotta'^  were  the  first  geologists 
to  try  (in  1824  and  in  1859)  to  classify  ore-deposits.  Since  then, 
many  methods  of  classification  liave  been  proposed  from  time  to 
time.  Groddeck's  system, "^^  which  takes  as  the  basis  of  classifica- 
tion the  origin  of  the  deposits,  is  perhaps  the  best  of  all.  He 
divided  ore-deposits  into  two  groups,  viz.,  original  and  fragment- 
::ary  deposits. 

Van  Hise*^  a  little  later  classified  them  into  three  groups, 
namely,  those  of  igneous  origin,  those  which  are  the  direct  result 
•of  sedimentation,  and  those  which  have  been  deposited  by  under- 
ground water.  A  classification  based  on  the  magmatic  theory  is 
perhaps  the  best  for  us,  when  looked  at  from  the  standpoint  of 
the  theory.  Tlie  greater  part  of  the  ore-deposits  in  the  case  of 
the  heavy  metals  is  of  igneous  origin,  and  also  since  there  are, 
in  my  opinion,  hardly  any  other  deposits  of  heavy  metals  found  in 
•Japan.  Accordingly  I  shall  here  classify  them  into  five  categories, 
based  on  the  magmatic  theory: 

a)  INIagmatic  segregations. 

b)  Contact  deposits. 

c)  Mineral  veins. 

d)  Replacement  deposits. 

e)  Impregnation  deposits. 

This  classification  has  been  made  quite  independently  by  me 
for  the  special  treatment  of  Japanese  ore-deposits.  I  am,  however, 
ver}^  glad  to  notice  its  close  resemblance  to  that  of  Richard  Beck, 
made  public  in  the  third  edition  of  his  '  '  Lehre  von  den  Erz- 
lagerstaetten,"  1909. 

Magmatic  segregations  are  heavy  metals  accumulated  in  a 
magmatic  body.     Contact  deposits  are  the   so-called  fossil  ema- 

1)  von  Waidenstein  :  "  Die  besonderen  Lagerstätten  der  nutzbaren  Mineralien,"  1S24. 

2)  von  Cotta  :  "'  Lehre  von  den  Erzlagerstätten,"  1859. 

3)  von  Groddeck  :"  Die  Lehre  von  den  Lagerstätten  der  Erze  "  1869. 

4)  Van  Hise  :  "  The  Genesis  of  Orj  Deposits,"  pp.  2S2-432. 

5)  Lincoln  :  Loc.  cit. 


4  Iwasaki  : 

nations  of  LI^x'OLx,^^  deposited  between  the  ore-bringers  and  the- 
preëxisting  rocks,  the  latter  of  which  obstructed  the  passage  of  the 
emanations  from  the  former.  When  the  emanations  force  them- 
selves into  the  fissm-es  of  the  rocks  and  deposit  heavy  metals  there, 
we  have  mineral  veins.  When  the  magmatic  emanations,  by  their 
strong  rush  and  chemical  action,  dissolve  part  of  rocks,  make 
cavities  of  various  forms,  and  deposit  heavy  metals  therein,  then 
we  have  replacement  deposits.  Impregnations  are  the  ore-bodies 
disseminated  in  the  rock-masses. 

3.    The  Order  of  Petrification  of  Emanations. 

Lincoln'^  has  classified  emanations  into  four  groups,  i.  e., 
actual,  fossil,  repressed,  and  potential  emanations.  '^  Actual 
emanations  may  be  observed  as  gases  and  vapours  from  lava 
streams  exi:>elled  from  volcanic  vents."  "Fossil  emanations  are 
the  more  or  less  well-preserved  remains  and  traces  of  actual  ema- 
nations. The  complete  preservation  of  past  emanations  is  often 
seen  as  inclusions  in  minerals,  while  partial  preservation  is  com- 
mon in  druses  and  in  lithophyses,  at  contacts  and  in  veins."  "  It 
is  well  to  remiember  in  this  connection  that  the  mineral  veins  and 
contacts  frequently  appear  to  be  in  whole  or  in  part  of  magmatic 
origin." 

As  the  emanations  begin  to  get  cool,  the  minerals  begin  to  be 
formed,  and  petrification  occurs.  From  frequent  observations,  I 
have  come  to  the  conclusion  that  the  order  of  petrification  of  ema- 
nations is  similar  to  that  of  the  formation  of  rock-forming  minerals 
in  magma;  for  in  both  cases,  the  falling  of  temperature  is  the  chief 
agent  in  forming  minerals  from  liquids  or  gases  at  high  tem- 
perature. The  following  list  shows  the  order  of  petrification  of 
emanations  observed  in  Japanese  ore-deposits:—!,  magnetite,  2. 
chromite,  o.  hematite,  4.  garnet,  5.  augite  and  hornblende,  6. 
scheelite,  7.  pyrite,  8.  cobaltite,  9.  chalcopyiite,  10.  barite,  11. 
argentite,  12.  gold,  13.  quartz,  14.  tetrahedrite  and  enargite, 
15.    calcite.      The  minerals  at  the  head   of  the  series  are  those 

1)  Lincoln  :  Economic  GcolOffij,  Vol.  II.,  No.  3,  ix  258,  1907. 

2)  Loc.  cit. 


Mot  illog'ony  of  the  Japanese  Islamls.  5 

formed  at  liigh  temperature,  and,  as  we  go  downwards,  the  tem- 
perature of  tlieir  formation  is  lower.  Minerals  having  a  high 
position  in  the  petrification  order  are  spoken  of  as  "  of  the  higher 
order  of  petrification."  Not  only  does  the  order  indicate  the  order 
of  the  formation  of  the  minerals,  but  also  their  position  in  certain 
deposits.  The  liigher  the  order  of  petrification,  the  lower  is  the 
position  of  the  minerals  in  the  ore-deposits,  for  it  is  natural  that  the 
temperature  of  emanations  in  rock  fissures  should  become  higher  the 
lower  we  go.  The  I'eason  of  the  constant  association  of  quartz  with 
.gold,  and  the  transition  of  quartzose  gold  ores  into  sulphides  in  the 
bottom  of  mineral  veins  may  be  readily  understood  in  the  order  of 
petrification  of  emanations  above  cited.  Lindgren^^  enumerated 
persistent  minerals  according  to  their  positions,  ranging  from  the 
contact  metamorphic  or  igneous  condition  to  the  surface  of  the 
earth,  as  shown  in  tlie  following  list,  viz: — pyrite,  chalcopyrite, 
bornite,  arsenopyrite,  galena,  zincblende,  molybdenite,  gold, 
quartz,  calcite,  etc.  The  reader  will  easily  recognize  the  essential 
coincidence  between  my  petrification  order  and  the  above  list. 

In  magmatic  segregations  found  in  Japan,  only  the  first  two 
minerals  in  my  petrification  order  occur,  of  Avhich  chromite  in 
serpentine  is  the  only  one  workable.  Contacts  constitute  the  best 
reservoirs  for  all  the  magmatic  emanations,  and  therefore  various 
minerals  are  found  there.  Of  these  minerals,  magnetite,  chalcopy- 
rite, cobaltite,  and  sometimes  gold  are  being  worked.  In  mineral 
veins  which  are  located  far  from  the  source  of  the  ore-bringer,  the 
temperature  must  be  lower  than  in  magmatic  segregations  and  con- 
tacts, and  initial  products  such  as  magnetite  and  chromite  can  not 
journey  through  so  long  a  passage.  Accordingly  there  occur  only 
pyrite  and  such  minerals  as  are  of  a  lower  order  than  it,  of  which 
the  copper  and  gold  ores  are  chiefly  being  Avorked.  Ores  occurring 
as  impregnations  and  replacements  do  not  present  many  points  of 
difference  from  those  in  veins,  but  are  very  complex  in  their 
composition;  for  all  elements  of  the  emanations  are  shut  up  in 
them  as  in  the  case  of  contacts.  This  is  especially  true  of  replace- 
ments, such  as  those  in  the  Kosaka  Mine. 

1)     Lindgren  :  Ezonomic.  Geology,  Vol.  II.,  p.  122,  1907. 


Iwaaaki 


4.    Metallogenetic  Provinces. 

In  Japan,  there  are  several  kinds  of  ore-briiigers,  such  as  granite^ 
diorite-porphyrite,  liparite,  and  andésite  ;  serpentine  is  also  suppos- 
ed sometimes  to  have  a  genetic  relation  to  ore-deposits.  All  these 
rocks  occupy  their  own  areas,  which  we  call  here  metallogenetic- 
provinces,  after  A.  M.  Finlaysox,^^  who  made  similar  divisions  of 
the  British  Isles.  Granite  is  very  extensivel}'  exposed  in  Japan, 
but  it  is  not  always  associated  with  ore-deposits.  As  an  ore-bringer 
it  is  most  frequent  in  Korea,  and  also  in  many  places  in  northern 
Kyûsliû  as  well  as  in  western  Honshu  (the  Main  Island).  Quartz- 
porphyry  associated  with  ore -deposits  in  central  Japan  is  also 
asserted  to  be  the  marginal  facies  of  this  particular  province  includ- 
ing Korea  and  other  regions.  Tliese  localities  therefore  may  be- 
called  the  Korean  Province. 

Diorite  is  not  scanty  in  Japan,  and  a  noteworthy  fact  is  that 
diorite-porphyrite  is  rather  better  suited  to  be  an  ore-bringer  tlian 
diorite  proper.  The  ore-deposits  formed  by  emanations  from 
diorite-porphyrite  are  chiefly  found  in  the  Paleozoic  formation  in 
the  outer  zone  of  North  Japan,  with  their  center  in  the  Kitahami 
Mountain-land,  which,  possessing  most  numerous  deposits  of  this 
sort,  may  give  the  name  to  this  Province.  Liparite  lava  is  not 
very  often  met  with  in  this  country,  but  the  rock  doing  the  function 
of  an  ore-bringer  is  found  more  frequently  in  the  form  of  dykes  or 
necks,  nearly  always  in  the  inner  zones  of  North  and  South  Japan, 
which  are  put  together  under  the  name  of  the  Kosaha  Province, 
the  Kosaka  Mine  being  its  exponent.  Besides  liparite,  propylite- 
is  somiCtimes  found  doing  the  work  of  an  ore-bringer  in  this 
province.  Andésite  is  the  volcanic  rock  of  widest  occurrence  in 
Japan,  but  it  is  not  always  associated  with  ore-deposits.  The  rock 
which  serves  as  an  ore-bringer  is  rather  the  older  rock  of  this  kind, 
recent  andésite  lava  being  always  barren  of  the  useful  heavy 
metals.  The  greater  part  of  the  andésite  that  acts  as  an  ore- 
bringer  is  supposed  to  have  erupted  in  the  later  period  of  the 
Tertiary  and  the  earlier  portion  of  the  Diluvial,  and  intrudes 
Tertiary  sediments,  forming  dykes,  necks,   or  denuded  volcanoes. 

1)    Finlayson  :  Quart.  Jour.  Geol.  Soc.  London,  p.  281,  1910. 


Metallogeny  of  the  Japanese  Islands.  7" 

The  andésite  of  this  kind  is  chiefly  fonnd  in  the  inner  zone  of  the 
Ryûkyû  (Loo-choo)  arc,  where  gohl  mines  are  very  hopeful, 
especially  in  Satsuma,  and  to  this  nietallogenetic  province  the  name 
Satsuma  is  given.  The  Sado  island,  famous  on  account  of  the 
rich  Sado  gold  mine,  is  also  supposed  to  helong  to  this  province, 
judging  from  the  properties  of  the  ores  from  the  mine. 

Pyrite  beds  in  Japan  have  for  a  long  time  been  supposed  to 
be  of  aqueous  origin,  but  at  present  they  arc  treated  as  bedded 
veins.  Their  ore-bringer  is  not  yet  definitely  known,  but  the 
author  believes  that  it  may  be  serpentine  or  a  like  rock,  just  as 
VoGT^^  explains  the  origin  of  the  pyrite  deposits  of  Norwaj^  as 
related  to  gabbro.  Such  beds  occur  in  the  outer  zones  of  South 
Japan  and  the  Eyû-Kyû  arc.  The  largest  of  the  kind  is  in  the 
Besshi  mine,  for  which  reason  the  author  calls  these  regions  the 
Besshi  Province. 

5.    The  Korean  Province. 

The  mineral  resources  of  the  Korean  Province  are  gold  and 
copper,  sometimes  with  cobalt,  zinc,  lead,  arsenic,  and  tungsten. 
The  origin  of  the  deposits  in  the  Korean  Province  is  most  clearly 
explained  in  "The  Geology  and  Ore-deposits  of  the  Holgol 
Mine,"  an  instructive  paper  by  Prof.  Koto."^  The  Holgol  mine 
is  situated  in  the  northeastern  portion  of  Hoan-haiclô  in  Korea. 
The  geology  of  Holgol  and  its  neighborhood  is  composed  of 
highly  metamorphosed  argillite,  calcareo-siliceous  slate,  limestone, 
porphyritic  granite,  and  basalt.  Prof.  Koto  describes  these  rocks 
in  a  most  elaborate  manner  ;  and,  from  various  facts  obtained  by 
this  study,  he  comes  to  "the  conclusion  that  the  gold  is  juvenile, 
and  must  have  come  from  deep  in  the  interior  as  an  exudation 
from  the  eutectic  mixture  of  the  granitic  magma." 

Ores  of  the  Korean  Province  occur  in  veins,  or  in  contacts. 
The  gold  ores  in  mineral  veins  are  always  quartzose.  The  quartz 
is  hard  and  translucent,  generally  being  very  poor  in  gold  content, 
except  when  sulphide  minerals  such  as  pyrite,  galena,   or  zinc- 

1)  Vogt  :  The  Digest  in  "  The  Genesis  ■  f  Ore  Deposits,"  p.  652. 

2)  Koto  :  Loc.  cit.,  p.  2. 


3  Iwasaki  : 

blende  are  present.  It  is  a  question  whether  the  gold  was  deposi- 
ted with  sulphide  minerals  as  a  primary  product,  or  has  been 
accumulated  around  the  sulphides  by  secondary  enrichment  in 
process  of  time. 

Copper  ores  in  this  province  are  often  found  in  contacts,  as  at 
Kapsan  in  Korea,  and  Naganobori  and  other  places  in  Japan  proper, 
where  gold  veins  of  the  Korean  type  are  very  scanty.  The  author 
studied  contacts  in  the  Naganobori  Copper  Mine  in  the  prefecture 
of  Yamaguchi.  This  mine  is  thirteen  miles  distant  from  the 
Ogori  railway  station,  near  the  western  extremities  of  Honshu. 
There  is  an  extensive  'karst\  called  Akiyoshi-dai.  Through  the 
limestone  a  small  granite  boss  ^bOOO  feet  long  and  1,700  feet  wide, 
is  exposed  forming  a  hillock  named  Hanano-yama.  All  around 
the  boss,  contacts  are  found,  Naganobori  being  one  of  them. 

The  deposits  of  tlie  Naganobori  Mine  are  30  feet  tliick,  the 
hanging-wall  l)oing  limestone  and  the  foot-wall  granite.  The 
greater  part  of  the  ores  consist  of  radially  aggregated  augite,  which 
is  either  mixed  with  garnet  crystals,  or  planted  upon  garnet 
nodules.  The  ground-mass  of  the  ore  is  a  mixture  of  quartz  and 
calcite,  in  which  eol)altite  crystals  and  chalcopyrite  masses  are 
imbedded.  The  chalcopyrite  is  always  amorphous,  but  the 
cobaltite  crystallizes  in  pentagonal  dodecahedrons,  showing 
cubical  cleavage.  Throughout  the  ore  body,  mineral  veins  with 
symmetrical  structure  are  frequently  met  with.  I'hese  consist  of 
quartz  in  the  salband,  calcite  in  the  middle,  and  tetrahedrite  on 
both  sides.  Branching  out  from  the  main  body  of  the  contact 
deposit,  veinlets  of  chalcopyrite  traverse  the  limestone.  In  my 
opinion,  the  ore  of  the  deposit  was  petrified  from  the  emanations 
emitted  fron^.  the  granite  magma.  First,  ferromagnesian  silicates 
have  crystallized  out  as  garnet  and  augite,  then  cobaltite, 
chalcopyrite,  quartz,  tetrahedrite  and  calcite  were  formed  one 
after  the  other. 

6.    The  Kitakami  Province. 

This  province  gets  its  name  from  the  fact  that  the  metal  mines 
in  it  are  most  flourishing  in  the  Kitakami  Mountain-land,  situated 


Metallogeny  of  the  Japiniese  Islands. 


between  the  Kitakami  Valley  and  the  Pacific  Ocean.  The  mines 
are  nearly  always  in  the  Paleozoic  formation  and  are  associated 
with  such  ore-bringers  asdiorite,  diorite-porphyrite,  and  sometimes 
granite.  The  diorite-porphyrite,  green  and  compact,  with 
phenocrysts  not  very  distinct  to  the  naked  eye,  is  the  most  favorable 
rock  for  ore-deposits.  Under  the  microscope,  the  felspar  is  seen  to 
be  kaolinized  or  to  form  epidote  in  combination  with  the  decom- 
position-products of  horn].)lende  turned  into  chloritic  matter. 
Magnetite  is  always  present,  l)eing  particularly  abundant  toward 
the  margin  of  the  eruptive  masses.  I  observed  a  very  interesting 
phenomenon  between  the  diorite-porphyrite  and  the  limestone  in 
the  Kamaishi  Iron  Mine.     This  mine  is  the  most   hopeful  one  in 


A  B  c 

Fig.  1. — Intrusion  of  diorite- 
porphyrite  into  limestone,  in 
Kamaishi  Mine.  A,  Diorite- 
l^orphyrite.  B,  Magnetite 
crystals.     C,  Limestone. 


Fig.  2. — Contact  of  diorite-porphyrite  with  limestone  in 

the  Kamaishi  Mine.     H,  Hornblende.     E,  Epidote.     CH, 

Chlorite.     L,  Limes^on"^.      M,  Magnetite. 


this  province,  and  is  well  known  as  the  only  private  iron-smelting 
work  in  Japan.  The  deposits  belong  either  to  the  contact  of  the 
<liorite-porphyrite  with  paleozoic  limestone,  or  to  that  of  the 
diorite-porphyrite  with  granite,  or  else  the  magnetite  is  wholly 
enclosed  in  granite  masses.      Where  the  diorite-porphyrite  is  in 


10 


Iwasaki  : 


contact  with  limestone,  magnetite  lamolke,  perhaps  flattened 
rhombic  dodecahedrons,  are  seen  projecting  from  the  porphyrite 
into  the  hmestone  (Fig.  1  and  Fig.  2).  While  the  porphyrite 
was  in  the  deep  as  fused  magma  at  high  temperature,  it  was  a 
eutectic  compound  with  the  iron  content  uniformly  diffused 
throughout.  But  Avhen  the  magma  Avas  erupted  and  came  into 
contact  with  the  limestone,  its  temperatm-e  fell  and  its  chemical 
properties  became  entirely  different  from  those  it  had  in  the  deep. 
Magmatic  differentiation  took  place  at  the  niargin  of  the  magma. 
Magnetite  was  driven  out  of  it  to  form  the  thick  deposit  between 
it  and  the  hmestone,  ^y[nch  finally  attained  a  thickness  of  30  feet. 

The  arrangement  of 
minerals  in  this  contact 
deposit  is  highly  instruc- 
tive as  to  the  order  of 
the  petrification  of  emana- 
tiojis  (Fig.  3).  Magnetite 
is  found  on  the  side  of  the 
diorite-porphyrite,  garnet 
constituting     the     central 

A  B  C  D  E  F  ,  T 

■^.    „    .  ,   ,  ,  .    ,  zone     comes     next,     and 

Jig.  3 — Arrangeirent  or  ii.merals  in  the  contact  deposit 

of  the  Kamaishi  Mine.     A,  Dioriteporpyrite.  quartZ  OU    the    sidc    of    tho 

B,  Magnetite.     C,   Garnet.     D,  Quartz.  limCStOUe.       lu  the  qUartz, 

E,  Cakite  with  gold.    F,  Limestone.  ...  ,,  t 

calcite  masses  are  scattered 
about,  and  in  them  the  gold  is  remarkably  rich.  The  gold  grains 
are  usually  microscopic,  being  sometimes  as  large  as  ]  cm. 
in  length,  and  0.2  cm.  in  diameter.  They  are  of  two  kinds. 
One  is  like  granulated  zinc  in  form,  and  is  supposed  to  have 
been  solidified  from  the  fused  drops  of  gold  in  the  liquid 
emanations  exuded  from  the  diorite  magma.  The  colour  of  the 
gold  is  very  fine  being  almost  like  that  of  pure  gold.  The  other 
kind  of  gold  is  always  in  long  prismatic  crystals,  acutely  pointed 
at  both  ends.  These  are  perhaps  rhombic  dodecahedrons,  elong- 
ated on  an  axis.  They  are  paler  in  colour,  are  found  in  the 
cleavage  of  calcite,  and  are  supposed  to  be  of  secondary  origin  in 
contrast  with  the  former,  which  are  of  primarN'  origin. 


Metallogeny  of  the  Japanese  Is-land?.  ]  ji 

Arguing  from  the  arrangement  of  the  minerals  in  the  ore- 
deposit,  I  have  come  to  the  conclusion  that  the  order  of  the- 
petrification  of  emanations  in  the  Kamaishi  Mine  must  have  been 
as  follows:  magnetite,  garnet,  quartz,  gold,  and  calcite. 

As  a  typical  example  of  the  mineral  veins  in  the  Kitakami 
Province,  I  shall  here  choose  the  Shikaori  Mine,  not  very  far  from 
the  Kamaishi  Mine.  The  deposits  of  this  mine  are  bedded  veins 
running  S.  5°  W.  along  the  stratification  of  the  Paleozoic  formation. 
As  in  the  case  of  the  Korean  Province,  the  ores  are  composed  of 
the  hard  translucent  quartz  of  a  whitish  colour,  Avhich  is  characte- 
ristic of  the  so-called  old  vein  of  Prof.  Vogt.  The  gold  content  in 
the  ore  is  very  variable.  As  a  whole  the  ore  is  not  very  rich,  l)ut 
big  nuggets  have  sometimes  been  found  in  the  veins.  A  nugget 
called  "  monster  "^^  consists  of  thick  plates  of  gold  in  the  cracks  of 
the  quartz  ore.  The  fineness  of  the  gold  is  estimated  at  882.844 
and  the  nuggets  910  grammes  in  weight.  From  the  specific- 
gravities  of  the  quartz,  the  gold,  and  the  nugget,  I  have  estimated 
its  value  at  950  yen.  This  is  one  of  the  largest  nuggets  ever  got 
from  mineral  veins  in  Japan,  and  is  considered  to  be  one  of  the- 
best  specimens  of  the  kind  in  the  world.  Gold  veins  containing 
coarse  grains  of  gold  disposed  in  an  irregular  manner  are  also  often 
found  in  this  province,  and  constitute  the  source  of  gold  placers. 

7.    The  Besshi  Province. 

This  metallogenetic  province  comprises  the  entire  outer  zones^ 
of  South  Japan  and  the  Ryûkyû  Arc,  and  the  southern  part  of  the 
outer  zone  of  North  Japan.  In  this  province,  the  pyrite  beds  are 
most  important — stibnite  and  gold  veins  as  well  as  manganese  beds 
being  of  rather  subordinate  value.  The  Besshi  Copper  Mine  con- 
tains the  best  of  the  pyrite  beds  in  this  province,  and  for  tliis- 
reason  it  is  called  the  Besshi  Province  in  this  i)aper. 

The  pyrite  beds  are  chieflj^  found  in  the  so-called  crystalline 

1)  The  description  of  "  Monster  "  is  given  in  detail  in  uiy  work  "  Gold,"  p.  284,  Tôkyô,. 
1910.  (Japanese) 


\  2  Iwasaki  : 

schists,  the  Sarnbagawan  Series  of  Prof.  Koto.^^  '^'hey  are  also 
sometimes  met  with  in  the  Paleozoic  formation.  As  mother 
rocks,  basic  rocks  such  as  chlorite  schist'^  and  graphite  schist  are 
the  most  favorable;  serpentine  is  often  found  near  the  beds. 
P^^'ite  beds  occur  in  the  crystalline  schists  in  nearly  concordant 
stratified  form,  cutting  them  crosswise  in  a  few  instances.  The 
pyrite  beds  consist  of  an  intimate  admixture  of  pyrite  and  chal- 
copyrite.  They  are  so  compact  that  these  two  minerals  can  only  be 
distinguished  from  each  other  under  the  microscope  after  polishing, 
•or  in  a  few  cases  by  the  naked  eye.  According  to  Sakawa,^^ 
the  pyrite  in  the  ore  is  usually  in  rounded  grains,  but  sometimes 
it  is  crystallized,  when  it  attains  0.7  cm.  in  diameter.  The 
interspaces  between  the  pyrite  grains  are  filled  with  massive 
■chalcopyrite,  which  often  enters  even  into  the  cracks  of  the 
former. 

Pyrite  beds  sometimes  form  lenticular  bodies  or  rounded 
nodules.  When  they  are  found  in  tlie  decidedly  younger  forma- 
tion, i.  e.,  the  Paleozoic,  they  are  usually  roundish.  The 
structure  of  the  pyrite  beds  in  the  so-called  crystalline  schists  is  of- 
ten very  complex.  In  the  central  portion,  there  are  very  rich 
■copper  ores  containing  rock  fragments,  the  outline  of  which  is 
either  rounded  or  indented.  On  one  or  both  sides  of  the  rich  zone 
of  the  pyrite  beds,  there  are  found  highly  contorted  ores,  with 
regularly  stratified  schists  on  their  outer  sides.  These  contorted 
■ores  constitute  the  "  shear  zone  "  of  Sakawa. 

From  the  facts  above  stated,  and  after  very  careful  examination 
•of  a  great  many  pyrite  beds  in  Japan,  he  comes  to  the  conclusion 
that  they  are  bedded  veins  of  epigenetic  origin.  In  my  opinion, 
however,  the  original  form  of  the  pyrite  beds  must  have  been  that 
of  replacements  Ijrought  up  in  different  successive  periods.  After 
their  deposition,  a  strong  mountain-making  force  flattened  them 

1)  Koto  "On  the  »o-called  Crystalline  Schists  of  Chichibu."  Jour.  Sei.  Coll.  Imp.  Univ. 
Tokyo,  Vol.  IL 

2)  It  is  said  that  greater  part  of  the  so-called  chlorite  schist  near  the  pyrite  lieds  is 
amphibole-schist. 

3)  Saka-v\'a  :  "Report  on  Cufriforous  Pyrite  Beds."  Bull.  Imp.  (leol.  Stin:  Japan,  Vol, 
XXII.,  Xo.  1.  (Japanese). 


Metallogen  y  of  the  Japanese  Islands.  13 

into  the  form  of  beds,  at  the  same  time  causing  regional  meta- 
moi'i^hism  of  the  country  rocks.  After  such  a  geological  change, 
a  fissure  was  formed  along  the  middle  line  of  the  bed,  and  the 
side-rock  masses  slipped  down,  producing  the  sliear  zone  on  the 
exterior  part  of  the  bed.  Finally,  a  secondary  enrichment  took 
place  around  the  faulted  rocks,  filling  up  the  interspaces  of  the 
fissure.  Thus  the  rock  fragments  in  the  beds  are  rounded  or 
indented  on  their  exterior  by  the  dissolving  action  of  the  vadose 
water.  This  explanation  will,  I  think,  solve  the  varied  structures 
of  the  pyrite  beds  in  a  very  natural  way.  S.  Ishikawa,^^  a  Mining 
Inspector,  enumerates  59  copper  mines  of  this  species  in  Japan, 
namely,  27  in  crystalline  schists,  and  32  in  the  Paleozoic 
formation. 

Quite  recently  very  interesting  gold  deposits  have  been  dis- 
covered in  central  Kijûslm,  which,  upon  investigation,  seem  to  l)e 
an  isolated  block  from  the  Besshi  Province.  The  region  is  billy 
Avith  an  altitude  of  about  400  to  2000  ft.  above  sea  level,  and  is 
geologically  composed  of  a  thick  complex  of  amphibolite  and 
phyllite  belonging  to  the  Lower  Paleozoic.  The  amphibolite 
is  a  pale  green  homogeneous  rock.  The  phyllite  is  a  highly 
contorted  lamellar  rock,  gray  to  black  in  colour,  showing  pearly 
luster  by  the  presence  of  the  abundant  quantity  of  mica.  The 
latter  rock  sometimes  contains  very  conspicuous  cubic  phenocrysts 
of  pyrite,  which  measure  up  to  0.5  cm.  in  diameter,  and  is  usually 
converted  into  limonite  pseudomorph.  Examined  under  the 
microscope,  the  amphibolite  consists  of  elongated  green  crystals  of 
amphibole  and  greenish -yellow  grains  of  epidote,  cemented  by  a 
transparent  quartz  matrix.  Phyllite  shows  beautiful  contortion, 
microscopically  as  well  as  macroscopically,  forming  alternate  layers 
of  quartz,  mica  and  graphite.  It  is  a  very  noticeable  feature  that, 
so  much  gold  exists  throughout  the  rocks  for  several  miles,  that 
sometimes  the}^  practically  become  gold  ore  themselves.  Now,  it  is 
a  question  whether  the  gold  is  a  primary  constituent  of  rocks,  or  it 
had  been  carried  into  them  at  the  time  of  the  formation  of  quartz 

1)     Isbikawa:  "  Geology  and  Ore-deposits  of  Oshima."     Jour.  Gcogr.  Soc.  Tohyo,  Xo.  260, 
1910  (Japanese). 


î)  4  Iwasaki  : 

veins,  which  traverse  the  rocks  everywhere.  The  quartz  veins  some- 
times produce  very  rich  gold,  but  it  is  most  hopeful  when  the  gold 
forms  placers.  The  placers  may  be  divided  into  two  distinct 
kinds,  original  and  alluvial.  The  former  is  seen  on  the  surface  of 
the  mountain  region,  forming  the  primary  soil.  The  latter  forms 
the  placer  beds  in  the  valley.  The  bedded  deposits  are  found  in  the 
form  of  successive  river  terraces,  the  highest  and  most  promising 
measuring  about  200  ft.  in  height;  the  next  is  on  a  hillside  and  is 
:about  80  ft.  high  ;  while  the  lowest  is  only  (>  ft.  from  the  valley 
level  and  forms  part  of  valley  ground.  These  terraces  cover  more 
than  several  hundred  acres,  and  form  horizontal  strata  of  gravels 
'Composed  of  pebbles  of  amphibolite,  ]:>hyllite,  quartz  and  andésite 
cemented  Ijy  reddish  clay,  sand  of  the  above  stated  rocks,  and 
limonite  pseudomorph  after  pyrite.  The  conglomerate  becLcontains  rOA-^A 
gold  in  the  high  proportion  of  from  3/1,000,000  to  9/100,000  or  " 

1/100,000  on  the  average  and  measures  from  6  to  120  ft.  in  thickness. 
The  gold  grains  are  similar  in  form  to  those  from  old  veins,  and 
may  be  classified  into  two  groups,  granular  and  crystallized.  The 
«colour  of  the  granular  gold  is  very  fine,  being  like  pure  gold,  and 
is  much  larger  in  size  usually  measuring  as  much  as  3  mm.  in 
diameter.  Gold  nuggets  weighing  131  gr.  and  67  gr.  were  once 
found  in  the  valley.  The  peculiarity  of  this  granular  gold  is  that  it 
is  of  a  flattened  form  with  cracks  in  it.  The  cr^^stallized  gold 
thought  to  be  of  a  secondary  origin  is  inferior  in  grade,  and  usually 
«mailer  in  size,  being  not  quite  0.5  mm.  in  diameter.  The  com- 
paratively large  size  and  flattened  form  of  the  granular  gold  is 
explained  by  the  fact  that  the  gold  was  formed  in  amphibolite  or 
other  crystalline  schists  in  old  time  and  subjected  to  intense 
mountain-making  force  during  the  metamorphosis  of  these  rocks. 
The  ore-bringer  of  this  gold  is  not  yet  definitely  known;  but  it  is 
probable  that  the  amphibolite  is  a  rock  metamorphosed  from  the 
diabase  or  like  rocks  which  brought  up  the  gold  from  the  interior 
<Â  the  earth,  thus  doing  the  work  of  the  ore-bringer  itself. 

8.    The  Kosaka  Province. 

This  is  the  region  containing  plagioliparite  and  propylite  as 


^letallogeny  of  the  Japanose  Tslantls.  15 

ore-bringers,  and  has  Kosaka  as  its  largest  and  most  important 
mine.  The  ore-deposits  fonnd  in  this  province  are  chiefly 
replacements,  bearing  the  so-called  "black  sulphide  ore";  but 
there  are  also  famous  veins  of  various  other  kinds.  It  is  a 
characteristic  of  this  province  that  abundant  sulphide  minerals  are 
always  present  in  the  ores.  The  presence  of  this  character  may  be 
•due  to  the  strong  acidity,  or  high  fusing  point  of  the  ore-bringer. 
Even  where  andésite  is  the  ore-bringer,  the  emanations  emitted 
from  the  rocks  must  have  been  at  a  high  temperature,  and  thus 
andésite  itself  was  changed  into  propylite.  Such  a  high  tem- 
perature compels  the  formation  of  sulphide  minerals,  which  belong 
to  the  liigher  order  of  petrification  of  emanations;  in  other  words, 
sulphides  were  formed  only  in  a  temperature  higher  than  that  in 
which  gold,  quartz  and  calcite  (all  common  minerals  in  the 
Satsuma  Province)  must  be  formed. 

The  deposits  in  the  Kosaka  Province  belong  to  the  younger 
veins,  formed  in  the  later  period  of  the  Tertiary  or  the  earlier  part 
of  the  Diluvium.  They  are  most  frequent  in  the  Tertiary  sediments 
or  in  the  ore-bringer  itself,  forming  replacements  or  veins.  The 
province  includes  nearly  the  whole  of  the  inner  zone  of  North 
Japan,  as  well  as  isolated  points  in  the  inner  zones  of  South  Japan 
and  the  Ryûkyû  Arc.  The  deposits  in  the  Kosaka  Province  may 
be  classified  into  mineral  veins,  black  sulphide  ores,  and  dis- 
seminations. 

Mineral  Veins: — These  veins  are  of  several  kinds.  They 
always  bear  a  greater  or  less  quantity  of  sulphides,  but  some  are 
composed  of  auriferous  quartz,  some  of  quartzose  copper  ore,  and 
some  rich  in  lead  or  zinc.  Generally  speaking,  the  deposits  in 
the  province  are  the  sulphide-rich,  ''younger"  veins  of  Vogt. 

As  a  type  of  the  auriferous  quartz  veins,  let  me  describe  the 
Hasami  Gold  Mine,  for  I  know  it  better  than  any  other  of  these 
veins.  It  is  situated  about  five  miles  to  the  south  of  the  Arita 
Station  on  the  Nagasaki  line  of  the  Kyushu  Railway.  The  mine 
was  discovered  only  fifteen  years  ago,  but  at  present  it  is  one  of 
the  most  important  and  hopeful  gold  mines  in  Japan.  The 
deposit  is  of  the  true    fissure-vein    type    traversing   the    Tertiary 


IG 


Iwasaki  ; 


80  X 

Fig.  4.— Qnartzose  gold  ore  of  the  Hasami  Mine.     Q,  Quartz 
containing  liquid  enclosures,  in  parallel  or  radial  arrange- 
ment.    P,  Pores  connected  by  passages  with 
a  lining  of  siderite. 


sandstone  and  sliale,  the 
former  being  the  more 
important  of  the  country 
-rocks.  Quartz-trachyte 
is  found  in  this  con- 
cession, and  is  supposed 
to  he  the  ore-bringer  of 
the  gold  A'eins.  Several 
veins  are  met  with  run- 
ning N.  45°  W.,  and  dip- 
ping 70°-80°  SW.  They 
are  mostly  simple  veins, 
about  5  feet  thick,  but 
sometimes  assembled 
together,  attaining  even 
100  feet  in  thickness. 
Usually       the       simple 


veins     are     reg 


ular 


m 


80X 

Fig.  5. — Colony  of  gold  in  the  quartzose  gold 
ore  of  the  Hasami  Mine. 


extent,  with  distinctly 
banded  or  brecciated 
structure.  The  ores 
now  being  worked  are 
stained  by  limonite, 
for  they  belong  to 
the  weathered  zone. 
When  the  working 
proceeds  deeper,  much 
sulphide  is  expected. 

Under  the  micro- 
scope, the  quartz  in 
the  gold  ore  shows  a 
granular  or  hyp- 
idiomorphic  struc- 
ture. (Fig.  4.)  The 
outline  of  the  quartz  is 
always      smooth,      in 


Metnlloft'ouy  of  tho  Tapanose  Islands.  ],',7 

contrast  witli  the  indented  grains  in  the  old  veins.  Ronndisi-i 
pores  with  narrow  connecting  passages  are  very  noticeable.  Both 
the  pores  and  passages  are  lined  with  brown  siderite.  These 
phenomena  show  that,  dining  the  formation  of  the  veins,  the 
emanations  from  tlie  qnartz-trachyte  (liparite)  contained  very  large 
quantities  of  gases  which  were  for  the  greater  part  carbonic 
acid  remaining  as  siderite  in  some  interspaces  between  quartz 
gi-ains.  This  is  surely  one  of  the  most  important  properties  of  the 
ores.  '.:^. 

The  gold  grains  from  this  mine  may  be  divided  into  two 
classes,  granular  and  crystallized.  Tlie  former  is  covered  with 
siderite,  while  the  surface  of  the  latter  is  fresh  and  brilliant.  The 
gold  grains  are  grouped  together  in  colonies  (Fig.  5). 

Blach  sulphide  ores:  — Ores  of  this  kind  are  found  in  the 
inner  zone  of  Nortli  Japan.  They  were  not  investigated  until  late 
years,  and  Hirabayashi, ^^  Geologist  to  the  Mining  Bureau,  was 
the  first  to  treat  their  origin  and  characteristics  in  detail.  Accord- 
ing to  him,  the  black  sulphide  ore  is  an  intimate  admixture  of 
galena,  zincblende,  and  barite.  It  occurs  most  frequently  in  the 
form  of  replacements,  but,  in  a  few  cases,  as  mineral  veins  or 
impregnations.  Tlie  ore-bringers  are  quartz-trachyte  or  andésite, 
the  latter  being  generally  altered  into  propylite.  Hirabayashi  gave 
a  single  example  of  basalt  taking  the  place  of  an  ore-bringer,  but  I 
believe  it  was  only  an  eruptive  succeeding  the  formation  of  the 
deposits,  as  in  the  case  of  the  Hol-gol  Gold  Mine,  described  by 
Prof.  KoTÔ.'^ 

The  occurrence  of  the  black  sulphide  ores  furnishes  us  with 
materials  for  making  clear  the  magmatic  theory.  According  to 
Hirabayashi,  there  are  43  mines  in  Japan,  in  which  these 
ores  are  worked.  They  are  all  in  Tertiary  beds,  and  always 
associated  with  younger  volcanic  rocks.  In  14  mines,  quartz- 
trachyte  is  found,  in  10  mines  andésite  or  propylite,  and  in  14 
mines  both  these  rocks  together.     Basalt  has  been  found  only  in 

1)  Hirabayashi:  "Report  on  Black  Sulphide  Ore-deposiis,"  I.  and  IT.,  Mining  Bureau, 
Tokyo,  1908  and  1910. 

2)  Loc.cit. 


18  Iwasaki: 

one  mine.  In  the  remaining  three,  their  existence  was  doubtful. 
When  the  ore-deposit  is  in  contact  with  the  volcanic  rocks,  the 
deposit  becomes  thinner  and  thinner  as  we  descend  changing  at 
last  into  a  network  or  dissemination,  and  passing  by  imperceptible 
changes  into  the  volcanic  rock  itself. 

Impregnations: — One  of  the  characteristics  of  the  ore-deposits  in 
the  Kosaka  Province  is  the  abundance  of  impregnations.  This 
type  of  ores  is  perhaps  evidence  of  the  intense  pressure  and  the 
high  temperature  of  the  emanations.  The  emanations  form 
massive  deposits  by  impregnation  in  the  igneous  rocks  such  as 
quartz-trachyte  or  andésite  ;  and  sometimes  they  produce  bedded 
deposits  by  dissemination  in  sandstone.  When  the  rocks  are 
traversed  by  numerous  veinlets,  the  result  of  emanation  is  em- 
bodied in  networks.  Sometimes  whole  masses  of  igneous  rock  are 
changed  into  metasomatic  ores.  From  these  deposits  gold  is 
usually  worked  ;  copper  and  iron  also  are  sometimes  got  from  such 
ores.  The  vein-stufïs  are  chiefly  quartz  and  clay;  besides,  pyrite, 
hematite,  chalcopyrite  and  barite  are  found  as  accessory  com- 
ponents. I  shall  take  the  Washinosu  Gold  Mine  as  an  example 
of  impregnation  in  the  Kosaka  Province. 

The  Washinosu  Gold  Mine^^  is  situated  in  the  prefecture  of 
Iwaté  in  the  inner  zone  of  North  Japan.  The  largest  part  of  the 
deposits  in  this  mine  consists  of  impregnations  in  quartz-trachyte 
(plagioliparite)  erupted  through  the  Tertiär}^  beds;  only  a  small 
portion  belongs  to  the  Tertiary  formation.  The  hill  of  quartz- 
trachyte  is  about  900  feet  above  the  lowest  water  level  in  the  con- 
cession. Veinlets  traversing  the  eruptive  are  usually  1  or  2  inches 
thick,  but  sometimes  they  become  as  much  as  one  foot  in  thick- 
ness, and  300  feet  in  length.  Where  the  veinlets  are  ver}^ 
densely  crowded,  the  impregnations  are  very  rich.  In  the  veinlets, 
quartz  and  chalcopyrite  are  most  abundant,  while  barite  and 
micaceous  iron  exist  in  small  amounts.  Gold  is  rich  in  the 
pyritic  quartzose  ore,  but  very  poor  in  the  chalcopyrite. 

There  are  numerous  gold  deposits  like  those  of  Washinosu 
in  the  environs  of  the  mine.      They  are  also  developed  in  the 

1^    Nishiwada  :  The  Digest  of  "  Report  on  Gold  and  Silver  Deposits,"  1907.  (Japanese) 


Metallogeny  uf  the  Jai>ancse  Islands.  19 

southern  extremity  of  the  Satsuma  Peninsula  in  Kyushu,  where 
there  is  an  extensive  lava  plain  of  loose  andésite,  through 
which  nine  independent  rocky  hills  project,  rising  from  500 
to  900  feet  above  the  sea  level.  They  are  composed  of  a  hard 
compact  rock  of  a  whitish  colour,  the  petrographical  properties 
of  which  are  not  definitely  known.  One  geologist  says  it  is  a 
quartz-trachyte,  but  others  treat  it  as  an  andésite  silicified.  Although 
all  of  the  rocks  are  not  quartz-trachyte,  at  least  a  part  of  the  hills 
belongs  toit;  besides,  the  properties  of  the  ore-deposits  are  like 
those  of  the  Washinosu  Mine.  Nearly  all  the  silicified  rock 
masses  of  the  hills  contain  a  trace  of  gold.  In  the  Kasuga  Gold 
Mine,  which  has  one  of  these  deposits,  the  whole  rock  mass  con- 
tains 0.0002%  gold,  but  in  the  cla^^ey  veins  running  through  it 
the  gold  content  is  richest  going  up  to  0.02%  and  even  more.  In 
this  ore-deposit,  I  recently  discovered  barite,  which  is  always 
present  in  black  sulphide  ores,  but  as  yet  has  not  been  found  in 
other  gold  mines  in  the  Satsuma  Province. 

9.    The  Satsuma  Province. 

The  ore-deposits  belonging  to  this  province  are  associated 
with  andésite  as  their  ore-bringer.  In  contrast  to  the  Kosaka 
Province,  they  are  characterized  by  a  scantiness  of  sulphides.  The 
ores  are  chiefly  composed  of  auriferous  quartz  and  calcite,  both 
belonging  to  the  loivesf^  order  of  petrification  and  are  generally 
worked  for  gold.  Transition  is  seen,  however,  between  the 
Satsuma  and  the  Kosaka  Province.  For  example,  in  the  Sado 
Gold  Mine,  which  is  supposed  to  belong  to  the  Satsuma  Province, 
quartzose  gold  ores  are  associated  with  some  sulphides,  and  in  the 
Kinkwaseki  Gold  Mine  in  Taiwan  (Formosa),  auriferous  enargite 
masses  occur  in  the  form  of  chimneys,  while  the  Tasei  Lode  in  the 
Ikuno  Mine,  which  is  supposed  to  belong  to  the  Kosaka  Province, 
is  a  true  quartz  vein.  Generally  speaking,  the  deposits  in  the 
Satsuma  Province  are  true  veins,  with  banded,  ring  or  brecciated 
ores.      The    ring    ores    are    most    beautifully    developed    in   the 

1)     Perhaps  gold  and  quartz  are  of  colloidal  origin. 


20  Iwasaki  : 

Serigano  and  other  gold  mines. 

First  of  all,  after  the  formation  of  fissures,  gold,  argentite  and 
pyrite  were  deposited  around  andésite  horses,  at  the  same  time 
that  the  quartz  veins  were  formed,  after  that,  secondary  enrichment 
took  place,  and  gold  together  witli  other  minerals  was  formed  upon 
the  andésite  nucleus  in  the  inner  side  of  the  first  ring.  Thus  double 
rings  were  produced  there.  Together  with  the  above-described 
processes,  silicification  also  took  place  in  the  andésite  horses,  which 
were  all  or  nearly  all  changed  into  quartz.  Although  such  ring 
ores  are  also  found  in  the  Kosaka  Province,  they  are  most  frequent 
in  the  Satsuma  Province.  This  is  perhaps  due  to  the  greater  basic 
property  of  andésite,  compared  with  quartz-trachyte,  shale,  and 
sandstone,  which  are  the  most  important  mother  rocks  in  the 
Kosaka  Province. 

The  so-called  replacement  veins  of  Lindgren^'  are  also  often 
found  in  the  Satsuma  Province,  and  are,  as  Emmons"'^  stated,  dis- 
tinguished by  their  unsymmetrical  structure,  variable  size,  complex 
arrangement  of  minerals,  and  the  preservation  in  the  ore  of  the 
microscopical  structure  of  the  original  rocks.  I  have  studied  the 
replacement  veins  in  the  Okuchi  and  Kushigino  Gold  Mines,  in 
the  Satsuma  Peninsula,  southern  Kyûsyû,  and  therefore  will  state 
here  the  results  of  my  observations. 

The  andésite,  which  is  the  mother  rock  of  the  mines,  seems 
to  have  erupted  in  the  Tertiar,y  or  the  Diluvial  epoch,  usually 
forming  low  undulating  hills  due  to  erosion.  The  rock  is  grey 
and  compact,  with  augite  and  felspar  phenocrysts.  The  former 
is  conspicuous  to  the  naked  eye,  but  the  felspar  is  not  so  distinct. 
Under  the  microscope,  the  felspar  is  seen  to  be  very  large,  usually 
twinned  in  the  Carlsbad  type,  enclosing  augite  and  magnetite. 
The  augite  is  monoclinic,  its  pleochroism  being  very  strong. 
Magnetite  is  so  abundant  that  the  ground-mass  often  seems  black 
and  opaque.  Brown  coloured  glass  is  abundant  in  the  ground- 
mass  containing  microlites  of  felspar  and  augite  in  the  fluidal- 
arrangement.     The  andésite  is  the  pyroxene-andesite,   very   com- 

1)  "  The  Genesis  of  Ore  De^josits,''  p.  517. 

2)  Loc.  cit.,  p.  517. 


IMetallogeny  of  the  Japanese  Islands.  21 

mon  in  Japan,  but  in  the  Satsuma  Province  it  is  frequently  rich 
in  the  precious  metals,  notwithstanding  the  fact  that  the  andésite 
of  other  provinces  is  generally  barren  of  these  metals.  I  made  a 
microscopic  study  of  the  ores  in  the  replacement  vein  in  the  and- 
ésite of  tliC  (Jkuchi  Gold  Mine. 

This  andésite  is  generally  fresh,  but  the  felspar  phenocrysts 
are  decomposed  from  the  central  portion  into  brownish  chloritic 
matter.  By  the  penetration  of  a  mineral  solution  into  the 
fissures  of  the  rock,  felspar  is  kaolinized,  and  augite  and  a  part 
of  the  felspar  are  chloritized.  The  groundmass  is  at  the  same 
time  transformed  into  a  mixture  of  chlorite,  kaolin,  and 
quartz.  Then  sihca,  separated  from  tlie  constitution  of  the  rock- 
forming  minerals,  is  deposited  in  the  rock  masses  in  the  shape 
of  veinlets  or  rounded  masses.  Magnetite  is  dissolved,  and  its 
iron  together  with  that  secreted  from  the  other  components  is 
deposited  again  as  pyrite  by  the  chemical  action  due  to  the 
introduction  of  hydrogen  sulphide.  The  auriferous  solution  is 
reduced  by  the  ferrous  oxide  formed  during  the  decomposition  of 
the  felspar  and  augite,  and  is  deposited  in  colonies  around  the 
chlorite  pseudomorph  after  the  augite  and  felspar,  or  is  imbedded 
in  tlie  kaolin-silica  mixture.  Thus  gold  in  rich  ores  exists  in 
colonies,  as  in  the  case  of  the  Hasami  Gold  Mine  already  referred 
tu.  In  ores  in  the  replacement  veins,  gold  is  never  found  in  tlie 
chlorite  pseudomorphs  or  in  the  quartz  veinlets  running  through 
the  decomposition-products  of  the  mother-rock.  It  will  be  seen, 
therefore,  that  the  introduction  of  gold  began  after  the  chloritiza- 
tion  of  the  rock-forming  minerals,  and  finished  before  the  forma- 
tion of  the  veinlets.  It  is,  however,  not  well  estabhshed  whether 
tlie  metasomatic  cliange  took  place  at  the  same  time  as  the  forma- 
tion of  the  mineral  veins,  or  after  the  completion  of  the  vein- 
making  and  during  the  period  of  its  secondary  enrichment.  I  am 
of  opinion  that  the  latter  is  the  more  probable  theory. 

10.    Summary. 

The  above  statements  may  be  summarized  as  follows:  — 

a.     During  the  consolidation  of  magma,   various  magmatic 


22 


Iwasaki  : 


emanations  are  exuded,  by  the  petrification  of  which  ore-deposits 
are  formed. 

h.  A  definite  process  of  deposition  of  emanations  determines 
the  order  of  petrification,  as  stated  in  Section  3,  p.  4. 

c.  The  ore-deposits  of  igneous  origin  may  be  classified  into 
five  divisions;  magmatic  segregations,  contacts,  replacements, 
mineral  veins  and  impregnations. 

d.  There  are  several  kinds  of  igneous  rocks  called  ore- 
bringers,  which  produced  the  ore-deposits  by  their  "eruptive 
aft  er- action." 

e.  The  Japanese  empire  is  divided  from  the  metallogenetic 
point  of  view  into  five  provinces,  according  to  the  difference  of  the 
ore-bringers.     They  are  as  follows:  — 


Frovinces. 

Ore-deposits. 

Ore-hringers. 

Korean. 

Contacts,  veins. 

Granite. 

Kitakami. 

Contacts,  veins. 

Diorite-porphyrite. 

Besshi. 

Beds. 

Serpentine  ? 

Kosaka. 

Replacements,  veins. 

Quartz-trachyte,  and 
propylite. 

Satsuma. 

Veins,  impregnations. 

Andésite. 

/.  The  minerals  in  the  ore-deposits  in  the  difïerent  provinces 
are  distributed  in  the  order  of  petrification  as  in  the  folloAving 
table  :  — 


Minerals. 
Provinces.     ^^ 

-4-1 

1 

s 

o 

g 
2 

O 

g 

0) 

i 

a 
o 

< 

-4-1 

'S 

(D 

o 
O 

•4-1 

'S 

P4 
O 

-4J 

< 

-g 

II 

1^ 

.2 

o 

Korean, 

1 

1 

Kitakami. 

Besshi. 

Kosaka. 

Satsuma. 

1        1        1        1 

Metallogeny  of  the  Japanosft  Islands.  23 

As  the  reader  will  see,  in  the  pertrification  order  all  the 
minerals  except  calcite  are  found  in  the  Provinces  of  Korea  and 
Kitakami.  This  is  because  they  all  belong  to  the  older  veins, 
which  means  that  the  formation  of  the  minerals  took  place  in  the 
deep,  where  the  temperature  of  the  emanations  was  very  high,  the 
petrification  continuing  until  the  emanations  got  entirely  cool. 
The  pyrite  beds  in  the  Besshi  Province  are  supposed  to  have  been 
emitted  from  the  most  basic  rocks  such  as  serpentine  or  gabbro, 
and  therefore  only  the  minerals  of  the  lower  position  in  the 
petrification  order  are  deposited  there.  The  scantiness  of  quartz  in 
the  pyritic  beds  is  explained  by  the  basic  character  of  the  supposed 
ore-bringers.  In  the  Provinces  of  Kosaka  and  Satsuma  only  the 
ore-deposits  of  younger  formation  are  distributed  and  therefore 
such  minerals  as  iron  oxides  or  ferromagnesian  silicates  are  never 
found  there.  Sulphides  are  more  abundant  in  the  Kosaka  Province 
than  in  the  Satsuma  Province,  for  the  fusing  point  of  quartz- 
trachyte  is  higher  than  that  of  andésite.  In  the  gold  veins  in  the 
Satsuma  Province,  sulphides  are  very  scanty,  but  if  we  go  deep 
down  we  shall  find  much  more  of  the  sulphides  which  are  of  a 
higher  petrification  order  than  quartz  and  calcite,  the  two  pre- 
dominating vein-stuffs  of  the  Satsuma  Province. 

Tokyo: 

1912,  October  20. 


Publ.  Nov.  3rd,  1912. 


JOURNAL  OF  THE  COLLEGE  OF  SCIENCE,  TOKYO  IMPERIAL  UNIVERSITY. 

Vol.  XXXII.,  Art.  9. 


Oogonium  Liberation 

and 

the  Embryogeny  of  Some  Fucaceous  Algae. 

By 

M.  Tahara,  Fdijakushi. 

With  o  Plates  and  5  Text  Fvjures. 

It  had  long  been  my  wish  to  make  some  biological  as  well  as 
cytological  observations  on  the  representative  members  of  Japanese 
Fucaceae,  as  om*  knowledge  of  this  family  especially  in  respect  of 
these  points  is  still  very  imperfect. 

Taking  the  opportunity  of  a  short  stay  at  the  Misaki  Marine 
Biological  Station  of  the  Tokyo  Imperial  University  in  the  winter 
of  1908-1909,  I  made  some  observations  on  Sargasswn,  especially 
on  the  liberation  phenomena  of  oogonia.  The  results  were 
published  in  a  preliminary  note  in  the  Botanical  Magazine,  Tokyo, 
Vol.  XXIII.  1909. 

To  carry  out  more  extensive  studies,  I  made  a  second  visit  to 
the  station  at  the  end  of  December,  1909  and  stayed  there  for 
about  three  months.  The  present  paper  presents  the  results  of 
that  visit.  The  substance  of  it  was  reported  on  the  28th  of  April, 
1910,  before  a  meeting  of  the  Tokyo  Botanical  Society^^;  and 
briefer  accounts  were  given  in  Japanese  in  the  Bot.  Mag.  Tokyo. 
VoL  XXV.  1911. 

According  to  Yendo's  well  known  monograph  on  Japanese 
Fucaceae'^  there  are  eight  genera  of  this  family  in  Japan,  namely, 
Fiicits,   Pelietia^,   Ishige^,    Cystoseira^  Cystophyllum,  Coccophora,  Turbi- 

1)  Proceedings  of  the  Tokyo  Bot.  Society.  Bot.  Mag.     Vol.  XXIV.,     P.  (246). 

2)  Yendo,  The  Fucaceae  of  Japan.  Journ.  Coll.  Sei.,  Imp.  Univ.  Tokyo.  Vol.  XXI., 
Art.  12. 


2  M.  Tahara. 

naria^  and  Sargasstmi^  ^  but  only  the  four  with  asterisk  are  found 
at  Misaki.  Of  these  genera  only  a  few  species  were  available  for 
my  studies,  as  the  other  species  did  not  come  to  maturity  during 
my  stay  there.  Sargassum  énerve,  Sargasswn  Horneri  and  Cystophyl- 
lum  sisymhrioides  furnished  the  principal  materials  of  my  investi- 
gation. All  these  three  species  are  dioeceous.  At  Misaki  Sarga- 
ssum énerve  grows  in  such  profusion  as  often  to  present  a  serious 
obstacle  to  the  navigation  of  smaller  craft.  The  liberation  of 
oogonia  begins  usually  at  the  end  of  December. 

Sargassum  Horneri  is  also  common  at  Misaki.  The  receptacle  of 
this  species  are  very  large  and  well-suited  for  investigation.  The 
liberation  of  oogonia  begins  at  the  end  of  December. 

Cystophyllvm  sisijmhrioides  is  not  so  common  at  Misaki  as  the 
former  two,  but  it  is  by  no  means  rare.  The  liberation  of  oogonia 
begins  at  about  the  middle  of  February. 

First  let  me  give  my  observations  on  oogonium  liberation. 

I.    Oogonium  Liberation. 

In  the  preliminary  paper  mentioned  above,  I  gave  the 
following  acount  of  my  observations  on  the  oogonium  liberiation 
of  Sargassum. — At  that  time  I  used  the  term  'Oosphère  liberation' 
intead  of  '  Oogonium  liberation.'  But  strictly  speaking,  the 
products  liberated  from  the  female  receptacle  on  the  day  of  the 
so-called  '  Oosphère  liberation  '  are  not  oosphères  but  oogonia 
containing  in  their  bodies  one  oosphère  initial. 

"On  the  24th  of  December  1908,  the  next  day  after  the  full 
moon,  almost  all  individuals  of  Sargassum  énerve  of  the  coast  dis- 
charged their  oosphères  simultaneously.  The  discharged  oosphères 
stayed  on  the  receptacle  for  about  three  days  and  then  dropped 
off  also  simultaneously,  so  that  on  the  28th  there  was  not  a 
single  stock  that  bore  the  sporelings  on  the  receptacle. 

''  After  a  fortnight,  i.  e.  on  the  day  of  new  moon,  the  next 
general  liberation  of  oosphères  occurred." 

These  facts  reminded  me  of  the  periodical  liberation  of  sexual 
products  in  Dictyota  dichotoma  and  led  me  to  the  conclusion  that, 
"The  liberation  of  oosphères  in  Sargassum  takes  place  simultane- 


Oogonium  Liberation  and  the  Embryogeny  of  Some  Fucaceous  Algae. 


ously,  not  only  for  a  given  plant,  but  also  for  all  plants  of  the 
same  locality.  This  simultaneous  liberation  proceeds  in  fort- 
nightly crops  on  a  particular  day  with  a  fixed  interval  after  the 
highest  spring  tide,  the  interval  varing  however  in  different 
species." 

As  this  conclusion  was  based  on  observations  made  during  a 
relatively  short  time,  naturally  entire  confidence  could  not  be  put 
in  its  validity.  My  second  visit  to  Misaki  was  made  mainly  to 
determine  this  point. 

I  arrived  at  Misaki  on  the  27th  of  December,  1909,  which 
was  just  the  day  before  the  highest  spring  tide.  To  my  disappoint- 
ment I  found  that  most  of  the  Sargassum  growing  in  that  locality 
had  not  yet  attained  their  maturity  and  for  some  time  no  general 
oogonium  liberation  could  be  observed. 

The  first  general  oogonium  liberation  of  Sargassum  énerve 
occurred  in  fact  on  the  12th  of  January,  the  next  day  after  the 
highest  spring  tide.  Three  days  later,  on  the  15th  of  January, 
the  first  general  oogonium  liberation  of  Sargassum  Horneri  also  took 
place. 

So  far  these  observations  confirmed  in  the  main  the  record  of 
the  preceding  year.  But  the  liberation  went  on  thereafter  quite 
irregularly,  without  showing  any  fixed  relation  to  the  highest 
spring  tide.  The  actual  state  of  things  is  shown  in  the  following 
table. 


Species^"^---^^^ 

Jan. 

Feb. 

*S'.  enerve 

12*  21     31 

11*  16 

S.  Horneri 

15     23 

2 

14     21 

The  numbers  in  the  table  denote  the  dates  of  general  oogo- 
nium liberations.  The  intervals  between  two  successsive  libera- 
tions is  quite  irregular,  for  example,  the  intervals  in  Sargassum 
enerve  are  5,  6,  9,  10  or  11  days.  But  the  intervals  between  the 
two  corresponding  liberations  in  Sargassum  enerve  and  Sargissum 
Horneri  are,  as  is  seen  from  the  table,  tolerably  constant;  namely, 


4  M.  Tahara. 

two  or  three  days  after  the  liberation  of  Sargasswn  énerve  occurs 
almost  always  the  liberation  of  Sargasswn  Horneri.  Tlie  same 
relation  is  given  in  my  record  of  the  previous  year. 

The  highest  spring  tides  occurred  on  the  11th  and  25tli  of 
January  and  on  the  10th  and  24th  of  Februar3^  The  dates  with 
asterisk  in  the  table  correspond  to  the  days  next  after  the  highest 
spring  tides. 

I  have  also  observed  such  oogonium  liberation  in  Sargassum 
Kjellmanianum,  Sargasswn  tortil,  and  CystophjUiim  sisymbrioides.  In 
these  algae  too,  the  successive  liberations  do  not  seem  to  show  a 
fixed  relation  to  the  highest  spring  tide;  for  example,  in  S. 
Kjellmanianum  the  liberations  took  place  on» the  7th,  15th  and  23rd 
of  February  and  in  Cyst,  sisymbrioides  on  the  17th  of  February  and 
the  ord  and  20th  of  March. 

All  these  facts  led  me  to  the  conclusion  that  the  liberation  of 
oogonia  in  Sargasswn  and  CystophyUum  takes  place  periodically  and 
simultaneously  among  individuals  of  the  same  species  growing  in 
the  same  locality  ;  but  the  intervals  between  two  successive  libera- 
tion vary  in  an  irregular  manner,  without  having  at  least  any  fixed 
relation  to  the  highest  spring  tide.  Thus  the  liberation  phe- 
nomena in  our  plants  are  somewhat  different  from  those  in  Dictyota 
dichotonia^  the  periodicity  of  which  was  studied  first  by  Williams'-* 
and  recently  by  Hoyt"'-*  and  Lewis.  ^^  Probably  owing  to  differences 
in  local  conditions,  the  result  of  the  observations  of  these  authors 
do  not  agree  in  details,  but  all  of  them  agree  that  the  liberation 
phenomena  of  this  alga  have  a  certain  relation  to  the  highest 
spring  tide. 

To  ascertain  how  the  liberation  of  oogonia  proceeds  under 
artificial  conditions,  I  kept  some  branches  of  S.  Humeri  and  Cyst. 
sisymbrioides  in  a  glass  vessel  filled  with  sea- water.  To  prevent 
contamination  the  culture  Avater  was  renewed  almost  every  day. 
The  liberation  did  not  occur  as  in  nature  at  all;   but  after  about  a 


1)  Williams,  Studies  in  the  Dictyotaceae.     Ann.  Bot.,  XIX.  1905. 

2)  Hott,  Periodicity  in  the  production  of  the  sexual  cells  of  Dictyota  dichotom:i.     Bot. 
Gaz.,  XLIIL  1907. 

3)  Lewis,  Periodicity  in  Dictyota  at  Naples.     Bot.  Gaz.  L.  1910. 


Oogonium  Liljeration  and  the  Emljryogeny  of  Some  Fucaceous  Algae.  5 

month's  culture  the   branches  Kberated   the   oogonia  in    a   very 
feeble  manner. 


I  often  observed  the  actual  mode  of  oogonium  liberation  in 
Sargassimi  and  Cijstoplujllum  in  common  sea-water  under  the  micro- 
scope. But  when  I  collected  some  small  l)ranches  on  the  day 
before  the  day  of  an  oogonium  liberation  and  put  them  into  a 
mixture  of  sea-water  and  fresh  water  (one  volume  of  sea-water  to 
four  of  fresh  water  proved  to  be  the  best  proportion).  I  could 
quite  easily  observe  the  mode  of  the  oogonium  liberation.  I 
studied  the  mode  of  oogonium  liberation  in  S.  Hörnen  more 
minutely  than  in  other  species,  and  the  following  description  refer 
to  this  plant. 

Generally  speaking,  the  paraphyses  of  Sargassum  do  not 
protrude  from  the  conceptacle  as  they  do  in  Fucm.  In  5'.  Horneri 
they  compose  a  disklike  plug  at  the  opening  of  the  conceptacle. 
In  a  fcAv  seconds  after  immersion  in  the  mixtui'e  medium  above 
mentioned,  the  plug  comes  out  slowly,  with  some  broken  pieces 
of  paraphyses  on  its  inner  surface  (Text.   Fig.   1);   and  then  the 

conceptacle  commences  to  discharge 
its  oogonia  one  after  another.  The 
discharged  oogonium  has  a  thick 
outer  layer  of  gelatinous  substauce, 
which  trails  out  a  tail  fastened  to 
the  interior  of  the  conceptacle 
(Text  Fig.  1).  In  one  or  two  days 
the  gelatinous  substance  becomes 
less  rigid  and  the  tail  can  no  more 
be  recognized  with  certainty,  but 
still  for  a  time  the  discharged  oogo- 
nium remains  attached  to  the  recep- 
tacle. 

As  already  stated  in  my  preliminary  note,  the  oogonia  in  one 
receptacle  are  not  liberated  at  one  time,  but  in  succession,  one 
zone  after  another  in  acropetal  order.     Fig.  1  of  Plate  I  shows  the 


Fig.  I.     X  140 


6  M.  Tahara. 

receptacle  in  its  first  oogonium  liberation:  the  black  spots  on  the 
receptacle  are  oogonia.  The  area  in  which  the  discharged  oogonia 
have  already  dropped  off  shows  a  remarkable  contrast  to  the  area 
in  which  the  oogonia  are  in  situ  in  the  conceptacle  (Fig.  2,  PL. 
I).  Fig.  3,  PL.  I  shows  the  second  oogonium  liberation.  The 
older  part  of  the  receptacle  becomes  more  slender  and  the  surface 
has  the  appearance  of  a  young  pine  cone  (Figs.  2,  3,  4,  PL.  I). 

The  mode  of  oogonium  liberation  in  Ci/stojûylhim  sisymhrioides 
differs  considerably  from  that  of  S.  Hoineri  and  deserves  special 
mention.  The  receptacle  of  Cijst.  sisymhrioides  is  rather  small  and 
slender  and  what  is  remarkable  is  that  it  shows  no  trace  of 
the  conceptacle  opening  on  its  outer  surface.  As  above  stated 
S.  Horned  has  something  like  a  plug  at  the  conceptacle  opening, 
but  the  outline  of  the  conceptacle  opening  on  its  outer  surface  can 
be  seen  under  the  microscope  without  any  difficulty;  this  however, 
is  not  the  case  with  Cyst,  sisymhrioides. 

On  the  day  before  the  oogonium  liberation,  paraphyses  in 
this  alga  begin  to  grow  very  rapidly.  As  there  is  no  opening, 
paraphyses  must  make  their  way  through  the  outer  wall  of  the 
conceptacle.  The  tips  of  paraphyses  peeping  out  through  the 
cracks  on  the  outer  surface  of  the  receptacle  are  disposed  quite 
regularly  in  a  line  parallel  to  the  main  axis  of  the  receptacle. 
And  in  the  mean  time,  the  oogonium  liberation  takes  place  in  the 
same  way  as  in  Sargassim-,  paraphyses  continue  to  grow  and 
entangle  with  one  another  about  the  oogonia,  giving  an  appearance 
of  a  mycelium  growing  on  a  nutritive  substratum  (Fig.  1,  PL. 
III).  As  in  the  case  of  Sargasswn,  the  liberation  at  one  time  is 
always  restricted  to  a  zone  of  receptacle,  the  liberation  proceeding 
acropetally.  The  discharged  oogonia  remain  attached  to  the 
surface  of  the  receptacle  for  about  five  days  and  then  drop  off, 
when  the  paraphyses  protruded  from  the  conceptacle  and  entangled 
about  the  oogonia  are  also  cleared  away  from  the  surface  of  the 
receptacle.* 


*  The  protruded  paraphyses  are  at  first  relatively  stiff,  but  become  gradually  slushy. 


Oogonium  Lil»eration  and  th9  Embryogeuy  of  Some  Fucaceous  Algae.  7 

I  often  came  across  some  detached  branches  of  Ciist.  Tumeric 
floating  on  the  sea  near  the  Marine  Biological  Station  ;  and  the 
receptacle  on  the  branches  bore  many  discharged  oogonia  within 
the  mass  of  entangled  paraphyses.  Thus  the  rapid  growth  of 
paraphyses  at  the  time  of  the  oogonium  liberation  appears  to  be  a 
characteristic  of  the  genus  CystopliiiUum. 


II.    Early  Stages  of  Embryogeny. 

Since  the  appearance  of  Oltmanns'  classical  work  on 
Fucaceae'\  it  has  been  generally  beheved  that  in  tiie  Fucaceae  in 
general  the  three  successive  nuclear  divisions  in  the  oogonium 
take  place  before  the  formation  of  oosphères,  no  matter  how  many 
oosphères  come  to  function  in  one  oogonium. 

At  this  stage  of  the  discussion  Miss  E.  B.  Simons'  paper  on 
Sargasswn  filipemlula-^  attracted  the  attetion  of  plant  morphologists. 
According  to  her,  in  this  alga  the  three  successive  nuclear  divisions 
before  the  oosphère  formation  are  entirely  suppressed  and  the  one 
nucleus  of  the  oogonium  initial  remains  in  a  resting  condition 
throughout  the  entire  period  of  growth  of  the  oogonium  and 
becomes  directly  the  nucleus  of  the  single  oosphère, 

During  my  stay  at  Misaki,  I  paid  special  attention  to  this 
point  and  was  so  fortunate  as  to  be  able  to  observe  the  successive 
developmental  stages  of  the  oosphère  in  the  oogonium  of  Sargas- 
swn and  Cystophyllmn.  The  result  of  my  observations  differs, 
however,  remarkabley  from  that  of  Miss  Simons;  namely  the 
oogonium  development  in  these  algae  is  quite  normal,  showing 
the  usual  three  successive  nuclear  divisions  in  the  oogonium. 

As  a  matter  of  course,  the  periodical  oogonium  liberation  is 
accompanied  by  the  periodical  development  of  the  oogonium. 
All  the  oogonia  in  one  conceptacle  are  liberated  at  one  time,  so 
the  developmental  stage  of  all  the  oogonia  in  one  conceptacle  is 
always  the  same.      In  other  words,  differing  from  the  other  cases 

1)  Oltmanns,  Beiträge  zur  Kenntnis  der  Fucaceen.     Bibl.  Bot.  Cassel,  1898. 

2)  Simons,  A  morphological  study  of  Sargasmm ßlipendnla.     Bot.  Gaz.  XVIX.  1906. 


8  M.  Tahara.    • 

in  Fucaceae^^  in  Sargassum  and  CystophyUwn  one  can  not  observe 
several  developmental  stages  of  the  oogonium  in  the  same  con- 
ceptacle.  Not  only  the  same  conceptacle,  but  also  all  the  materials 
collected  on  the  same  place,  on  the  same  day,  do  not  show  in 
general  the  several  developmental  stages  of  the  oogonium. 

After  the  occurrence  of  an  oogonium  liberation,  the  single 
nucleus  of  the  oogonium,  which  is  to  be  liberated  for  the  next 
period,  remains  in  a  resting  condition  for  a  time  and  for  the  first 
time  on  the  day  before  the  day  of  the  next  oogonium  liberation, 
the  nucleus  begins  to  divide  to  form  the  nuclei  of  oosphères,  and 
the  oogonium  attains  the  di-  or  tetra-uucleate  condition.  On  the 
day  of  the  liberation,  the  oogonium  contains  eight  nuclei  evenly 
distributed  in  its  substance;  the  dense  mass  of  chromatophores 
assembled  around  each  nucleus  facilitates  the  counting  of  the 
number  of  the  nuclei.  Fig.  1,  PL.  II,  Fig.  3,  PL.  Ill  show  this 
condition  of  the  oogonium  in  S.  Horncri  and  Cyst,  sisymhrioidcs  re- 
spectively. Nienburg' s  recently  published  paper  on  C y stoseira  and 
Sargasswir^  states  also  that  three  successive  nuclear  divisions  take 
place  in  the  oogonium  before  the  formation  of  oosphères.  While 
the  result  of  my  observations  is  based  on  living  materials, 
Nienburg  made  his  study  on  the  microtome-sections  of  fixed 
materials.  At  aii}^  rate  the  occurrence  of  the  three  successive 
nuclear  divisions  in  the  oogonium  development  seems  to  be  general 
in  Fucaceae.  The  eight  nuclei  formed  in  one  oogonium  have  at 
first  the  same  appearance.  But  Sargassum  and  CystophyUwn  develop 
only  one  egg  in  one  oogonium,  so  ouly  one  of  the  eight  nuclei 
becomes  the  functional  nucleus  of  the  oosphères  and  the  others  are 
destined  to  degenerate  in  the  course  of  future  development.  Fig. 
4,  PL  III  shows  a  stage  in  which  some  of  the  eight  nuclei  are 
about  to  degenerate.  The  degeneration  of  the  seven  nuclei  does 
not  proceed  simultaneouly. 

In  other  Fucaceae  the  superfluous  nuclei  are  regularl}^  thrown 
out  into  the  space  between  the  oogonium  wall  and  the  oosphère. 
But  it  seems  to  me  that  this  is  not  the  case  in   Sargassum  and 

1)  Oltmanns,  Beiträge  zur  Kenntnis  der  Fucaceae,  Bibl,  Bot.  1889.  p.  84. 

2)  Nienburg,  Die  Oogonentwicklung  bei  Cystoseira  und  Sargassum.     Flora  Bd.  I.  1910. 


Oogonium  Literation  and  the  Embryogeuy  of  Some  Fucaceous  Algae.  9 

Gystophylhnn^  for  in  living  materials  I  liave  very  often  had  the 
chances  to  observe  the  stage  of  oogonium  development,  in  which 
such  cast  off  plasma-masses  would  likely  be  found,  had  they  ever 
been  present,  but  I  failed  to  find  any  trace  of  such  a  body. 

It  is  rather  a  curious  fact  that  none  of  the  foregoing  observers 
have  succeeded  in  observing  not  only  the  fertilization  but  even 
the  moving  spermatozoids  in  Sargasswii  and  Cijstophyllwn.  Miss 
Simons  writes  in  her  paper  already  cited  that,  "A  study  of 
fertilization  in  Sargassum  is  surrounded  by  serious  technical 
difficulties,  because  both  eggs  and  sperms  develop  upon  the  same 
plant,  thus  making  it  difficult  to  isolate  the  sexual  cells."  As 
above  described,  the  common  species  of  Sargassum  and  CystopkyUum 
in  our  coast  are  all  dioeceous,  so  the  difficulty  pointed  out  by  Miss 
Simons  does  not  occur  in  our  materials.  Nevertheless  the  study 
of  the  fertilization  of  these  algae,  both  in  living  and  fixed  materials, 
is  not  an  easy  task.  I  have  never  succeeded  in  observing  even 
the  spermatozoid  itself.  From  the  phenomena  of  the  periodical 
development  of  the  oosphères,  one  naturally  presupposes  the 
periodical  development  of  spermatozoids,  but  I  could  find  no  sign 
of  such  a  phenomena  in  the  antheridia  of  these  plants. 


Now  let  me  give  my  observations  on  the  development  of  the 
sporelings  of  Sargassitm  and  Cystophylluin.  The  early  development 
of  these  algae  goes  on  within  the  oogonia  which  after  having  been 
discharged  from  the  conceptacle,  are  attached  to  the  surface  of  the 
receptacle:  this  condition  fascilitates  the  investigation  in  no  small 
degree.  The  later  development  may  also  be  studied  easily  in 
materials  cultured  in  a  glass  basin.  The  method  is  very  simple. 
I  collected  some  small  branches  of  these  algae  which  carried  many 
hundreds  of  attached  sporelings  on  their  receptacles  and  cultured 
them  in  natural  sea-water,  some  of  the  sporelings  may  in  the 
course  of  development  fall  to  the  bottom  of  the  glass  basin  but 
many  remain  attached  to  surface  of  the  receptacle  for  a  long  time 
and  still  persue  the  normal  course  of  development.     The  detached 


]  0  M.  Tahara. 

sporeliiîgs  become  fastened  to  the  bottom  of  glass  basin  and  may 
also  be  used  for  investigation. 

Generally  speaking,  the  oosphères  of  S.  Honieri,  S.  énerve  and 
Cyst,  sisymhrioides  are  equally  oval  or  elliptical  and  common  to  all 
of  them  the  first  segmentation  wall  runs  perpendicular  and  about 
midway  to  the  long  axis  of  the  oosphère  (Fig.  2,  PL.  Ill;  Fig.  5, 
PL.  Ill):  the  second  wall  runs  parallel  to  the  first,  cutting  ofï  a 
small  lens-shaped  cell  at  one  end  of  the  sporeling.  Nienburg 
states  in  his  paper,  "  Die  erste  Wand  steht  senkrecht  zur  Läng- 
sachse. Die  zweite  steht  senkrecht  auf  der  ersten  und  teilt  das 
Vorderende  in  zwei  gleiche  Hälften.  Darauf  wird  von  der  unteren 
Spitze  durch  eine  Wand,  die  der  ersten  parallel  ist,  eine  schmale 
Rhizoidzelle  abgeschnitten."  Thus  his  observations  do  not  agree 
with  mine. 

Further  development  differs  in  Sargasswn  and  Cystophyllum, 
and  would  better  be  described  separately. 

Sargassum. 

The  lens-shaped  cell,  cut  off  by  the  cecond  segmentation  wall 
will  hereafter  be  called  for  the  sake  of  convenience  the  '  Rhizoid 
cell.'  This  cell  divides  simultaneously  with  the  segmentations  of 
the  other  cells,  until  the  eight-celled  stage  is  reached.  The  segmen- 
tations of  the  rhizoid  cell  are  quite  regular  and  the  segmentation 
wall  are  all  perpendicular  to  the  outer  surface  of  the  sporelings. 
(Figs.  5,  8,  9,  PL.  II)  clearly  show  this  regularity.  While  the 
segmentation  of  the  other  cells  proceeds  further,  the  rhizoidal 
portion  remains  in  the  eight-celled  stage,  and  in  the  mean  time 
the  rhizoid  formation  begins.  At  the  outset,  we  see  the  papilla- 
like protuberances,  eight  in  number  (Fig.  11,  PL.  II).  These 
protuberances  grow  gradually  and  become  a  group  of  rhizoids 
arranged  in  a  circle.  But  later  there  arises  in  the  central  region 
another  group  of  rhizoids  which  elongate  with  greater  rapidity  so 
that  they  become  longer  than  those  of  the  first  and  outer  group 
(Figs.  13  and  14,  PL.  II).  This  difference  in  length  becomes, 
however,  obscure  in  the  further  development  of  the  rhizoids,  the 
outer  group  of  rhizoids  also  growing  rapidly  (Fig.  15,  PL.  II). 


Oogonium  Liberation  and  the  Embryogeny  oï  Some  Fucaceous  Algae.  H 

To  ascertain,  if  possible,  the  origin  of  this  central  group  of 
rhizoid  I  made  some  microtome-sections  of  the  sporehngs  in  this 
stage.  Text-Fig.  2  was  drawn  from  one  of  such 
sections.  The  two  rhizoids  in  shade  are  the 
descendants  of  the  rhizoid  cell  formed  by  the 
second  segmentation  wall.  Other  rhizoids  situated 
in  the  central  portion  seem  to  have  been  derived 
from  the  body  cells,  without  having  any  direct 
relation  to  the  above  mentioned  rhizoid  cell.  In 
T.-  o  ^^Af^  the  later  development  the  number  of  rhizoids 
gradually  increases.  The  rhizoids  are  at  first 
unicellular  but  later  we  find  several  partitions  here  and  there  (Fig. 
16,  PL.  II).  Under  a  strong  magnification  we  see  pecuhar  thicken- 
ings on  the  cell  walls  of  the  rhizoids,  which  remind  us  of  the 
spiral  thickenings  of  tracheids  of  higher  plants  (Fig.  16  b,  PL.  II). 
The  cells  contain  small  granules  of  what  seems  to  be  a  fat-like 
substance,  so  far  as  can  be  seen  by  the  reactions  of  osmic  acid  and 
Sudan  III.  The  same  substance  is  found  not  only  in  the  rhizoids 
but  also  very  abundantly  in  the  body  cells  in  general. 

For  a  long  time,  the  shape  of  the  sporelings  remains  oval  or 
elliptical,  but  later  becomes  like  a  flask  and  at  last  at  the  boundary 
between  the  slender  and  the  swollen  part  of  the  sporeling,  there 
grows  a  neAv  branch  (Figs.  17-20,  PL.  II).  This  is  the  last  stage 
that  I  observed  at  Misaki. 

Cystophyllum. 

The  segmentation  process  in  the  rhizoid 
cell  of  this  plant  difïers  considerably  from 
that  above  described  ;  that  is,  in  this  plant 
before  the  rhizoid  formation  begins,  the 
rhizoid  cell  is  already  divided  into  about  30 
small  cells.  Text-fig.  3  is  the  surface  view 
of  the  rhizoidal  potion  of  the  sporeling  and 
text-fig.  4  presents  a  median  longitudinal 
section  of  the  same.  In  the  latter  figure, 
we  can  see  the  two-storied  arrangement,   a  Fig.  3.    x24o 


12 


M.  Tahaw. 


Fig.  4.    X140 


state  which  is  never  found  in  Sargassum. 
The  rhizoid  cells  elongate  to  form  a  numebr 
of  rhizoids;  similar  to  what  has  been 
described  in  the  case  of  Sargassum,  the 
rhizoids  originating  from  the  central  region 
grow  more  rapidly  than  those  originating 
from  the  outer  region  (Fig.  8,  PL.  III). 
But  in  this  case  the  central  group  of 
rhizoids  are  also  derived  from  the  rhizoid 
cell  formed  by  the  second  segmentation 
wall,  although  in  later  development  there 
may  be  some  rhizoids  derived  from  body 
cells  as  in  Sargassum. 


^t\ 


\ 


\ 


Wß 


V 


Fig.  5.     X 175 


As  already  stated,  the  early  development  of  sporelings,  both 
in  Sargassum  and  CijstophyUum^  is  carried  out  while  they  are 
enveloped  in  the  oogonium  wall.     But  in  the  course  of  develop- 


Oogonium  Liberation  and  the  Embryogony  of  Some  Fucaceous  Algae.  13 

ment  the  wall  ruptures  at  one  end  by  the  pressure  of  the  growing 
rhizoids  (Fig.  14,  PL.  II)  and  the  sporeling  becomes  free 
thereafter. 

During  the  study  of  the  sporeling-development  I  often  met 
with  different  abnormities  and  some  which  are  interesing  are 
shown  in  text-fig.  5.  These  abnormities  are  often  found,  especially 
in  *S.  encrvc,  and  all  these  figures  were  sketched  from  the  sporelings 
of  this  alga.  Fig.  5,«  represents  an  abnormity  in  which  the  first 
segmentation  wall  is  oblique  to  the  long  axis,  fig.  5, />  is  the  later 
stage  of  this  abnormity,  c  and  d  of  the  same  figure  show  something 
like  '  Simultan-dreier'  and  'Simultan- vierer.'  Superfluous  nuclei 
in  the  oo;_onium  often  fail  to  degenerate,  and  this  results  in  the 
formation  of  a  curious  abnormity  in  the  later  development  (Fig- 
5/). 

In  conclusion  I  wish  to  express  my  hearty  thanks  to  Professor 
K.  Fujii  for  his  v^'Juable  suggestions  and  assistance  given  me 
during  the  progress  of  this  work  and  to  Professor  Iijima,  the 
director  of  the  Misaki  Marine  Biological  Station  through  whose 
kindness  many  facilities  were  afforded  me  in  the  course  of  my 
investigations. 

Botanical  Institiitc, 

College  of  Science, 

Tokyo  Imperial  University. 


Publ.  March.  31st,  1913. 


31.  TAHAltA. 

OOGONIUM  LIBERATION  AND  THE  EMBRYOGENY  OF 
SOME  FÜCACEOUS  ALGAE. 


Plate    I. 


Explanalion  of  PL  I. 


Female  receptacle  of  Sarçiassinn  Horneri.  All  figures  were  drawu 
with  the  aid  of  camera  Incida  from  living  materials.  Magnification:  ca. 
10  times. 

Fig.  1.     Fiist  oogonium  liberation. 

Fig.  2.  After  the  oogonia  discharged  in  the  first  liberation  had 
dropped  off. 

Fig.  3.     The  second  oogonium  liberation. 

Fig.  4.  After  the  oogonia  discbarged  in  the  first  and  second 
liberations  had  dropped  off. 


Jour.  Sei.  Coll.  Vol.  XXXII.,  Art.  9. 


PI.  I. 


M.^1 


M.  Tahara  del. 


■I 


]>I,  TAHAKA. 

OOGONIUM  LIBERATION  AND  THE  EMBRYOGENY  OF 
SOME  FUCÄCEOUS  ALGAE. 


Plate    II. 


Explanation  of  PI.  11. 

Sporeling-development  of  Saninssnui,  Ilonieri.  All  Figures  were 
drawn  with  tlic  aid  of  camera  lucida  from  living  materials.  Magni- 
fications :  liigs.  1-lGa  ca.  140  times;  figs.  17-20  ca.  50  times;  fig. 
l()b  ca.  500  times. 

Fig.  1.  Oogouinin  with  eight  nuclei,  chromatophores  grouped 
around  the  nuclei. 

Fief.  2.     First  segmentation. 

Fief.  3.  Second  segmentation,  forming  the  rhizoid  cell  at  one 
•end. 

Fief.  4.     Beginning  of  the  third  segmentation. 

Fiö.  5-     Completion  of  the  third  segmentation. 

Fig.  6,  a.     Side  view  at  the  hegimiing  of  the  4th  -iegmeatation. 

Fief.  6,  b.     Tolar  view  of  the  same. 

Fig.  7-     First  segmentation  of  the  rhizoid  cell. 

Fig.  8.     Second  segmentation  of  the  same. 

Fig.  9.     lîhizoid  cell  in  the  eight-celled  stage. 

Fig.  10.  Further  segmentation  of  body  cells.  Khizoid  cell 
Temains  in  the  eight-celled  stage,  a,  side  view  ;  b,  polar  view. 

Fiö".  11.     Beginning  of  rhizoid-formation. 

Fig.  12.     Khizoids  somewhat  elongated. 

Figs.  13-16  a.     Further  development  of  rhizoids. 

Fig.  16  b.     '-f  he  terminal  portion  of  a  rhizoid. 

Figs.  17-20.     Development  of  the  body  of  a  sporeling. 


Jour.  sa.  Coll.  Vol.  XXXII.,  Art.  9. 


PI.  II. 


M.  Tahara  del. 


I 


.•»I.    TAIIVUV. 

OOGONIUM  LIBERATION  AND  THE  EMBRYOGENY  OF 
SOME  FUCACEOUS  ALGAE. 


Plate    III. 


Explanation  of  PI.  III. 

b'emale  receptacles  aiul  sporeliug  development  of  Cißtoj)hi/Uii)ii 
sisi/ijibriniiles.  All  figures  were  drawn  with  the  aid  of  camera  lucida 
from  living  materials.  Magnifications  :  figs.  1  and  2.  ca.  10  times  ; 
figs.  2-0  ca.  170  times. 

Fig.  1.  Female  receptacle  in  the  first  oogonium  liherations. 
Paraphyses  protruding  from  the  conceptacle  and  forming  a  mass 
resembling  a  mycelium. 

Fig.  2.  After  tlie  oogonia  discharged  in  the  first  liberation 
had  dropped  oô'. 

Fig.  3.     Oogonium  with  eight  nuclei. 

Fig.  4.  Some  of  the  eight  nuclei  in  the  oogonium  about  to 
degenerate. 

Fig.  5.     "The  first  segmentation. 

Fig.  6.     The  second  segmentation. 

Fiö"S.  7-9.     Later  stages  of  the  sporeling  development. 


i 


Jour.  Sel.  Coll.  Vol.  XXXII.,  Art.  9. 


PI.  m. 


M.  Tahara  deL 


I 


JOURNAL  OF  THE  COLLEGE  OF  SCIENCE,  TOKYO  IMPERIAL  UNIVERSITY. 

VOL.  XXXCI.,  ART.  10. 


Beiträge  zur  Kenntnis  der 
Morphologie  und  Stammesgeschichte  der  Gorgoniden. 

Von 

Kumao  Kinoshita,  lUgakushi. 

{Zoolopisches  Institut  der  Kaiserlichen  Universität  zu  Tokio). 


Mit  13  Textfiquren. 


I.     Einleitung. 

Die  Gorgoniden  waren  früher  für  eine  ganz  natürliche  Gruppe 
gehalten  worden.  Heutzutage  werden  sie  jedoch  auf  Grund  der 
Forschungen  von  v.  Koch,  Studer  und  vielen  anderen  allgemein 
als  diphyletisch  angesehen.  Sie  zerfallen  nämlich  in  die  Ordnung 
Gorgonacea,  Axifera  v.  Koch  oder  Holaxonia  Studee,  bei  welchen 
die  Skeletachse,  wie  bei  Primnoa,  Isis,  Gorgonia,  etc.,  lamellöse 
Struktur  zeigt,  und  in  die  Ordnung  Pseudaxonia  v.  Koch  oder 
Scleraxonia  Studer,  bei  welchen  das  Innere  der  Kolonie  zur 
Skeletachse,  die  in  sich  spikuläre  Struktur  erkennen  lässt, 
differenziert  ist. 

Während  die  Trennung  in  diese  zwei  Gruppen  fast 
allgemeine  Anerkennung  gefunden  zu  haben  scheint,  ist  doch  die 
morphologische  Bedeutung  der  Kolonie  von  den  verschiedenen 
Forschern  ungleichartig  angedeutet,  und  infolge  dessen  sind 
die  Ausgangspunkte  dieser  beiden  Gruppen  bei  verschiedenen 
Formen  gesucht  Avorden.  Und  es  dürfter  wohl  klar  sein,  wenn 
man  einmal  die  Litteratur  durchsieht,  dass  diese  Verhältnisse  noch 
von  niemand  ganz  endgültig  aufgeklärt  worden  sind. 

In  dieser  Abhandlung  nun  beabsichtige  ich  also  einige 
Tatsachen  zu  schildern,    welche  geeignet  sein  dürften  auf  diese 


2  K.  Kinosbita  :  Beiträge  zur  Kenntnis  der 

Fragen  Antwort  zu  geben.  Hier  möchte  icli  mir  erlauben,  auch  an 
dieser  Stelle,  den  Herren  Professoren  Ijima  und  Goto  für  ihre  viel- 
fachen Ratschläge  und  ihr  stetiges  Interesse  während  der  Ausfüh- 
rung dieser  Arbeit,  meinen  herzlichsten  Dank  auszusprechen. 

Milne-Edwards  (1)  1857  schloss  in  der  Familie  Gorgonidae 
diejenigen  festsitzenden  Alcyonarien  zusammen,  bei  denen  das 
Innere  der  Kolonie  durch  die  Skeletachse,  welche  entweder 
homogene  oder  lamellöse  Struktur  zeigt,  eingenommen  ist.  Als 
ein  anderes  Hauptmerkmal  der  Familie  hob  er  besonders  hervor, 
dass  alle  Polypen  kurze  Magenhöhlen  besitzen,  wodurch  sie 
sich  von  denjenigen  der  Alcyonaceen,  welche  stark  verlängert 
und  nach  der  Basis  der  Kolonie  allmählich  verschmälert  sind, 
deutlich  auszeichnen. 

Diese  Familie  teilte  Milne-Edwaeds  in  drei  Unterfamilien, 
Gorgoninae,  Isidinae  und  Corallinae,  ein. 

Die  Skeletachse  der  Gorgonaceen  nun  hielt  er,  ebenso  wie 
Lamarck  (1),  Ehrenberg  (1)  und  Dana  (1,2),  für  ein  Sekret  des 
Ektoderms  der  Basis.  Dieselbe  tritt  nun  nach  ihm  anfänglich  an  der 
Kontaktfläche  des  Ektoderms  mit  dem  Substratum  in  der  Form 
einer  dünnen  Lamelle  auf;  sie  wird  jedoch  durch  wiederholte 
Ansätze  neuer  Lamellen  allmählich  erhoben,  sodass  endlich  eine 
einfache  oder  verzweigte  Skeletachse  zustande  kommt. 

Neben  den  typischen  Formen,  welche  die  Skeletachse  eben 
erwähnter  Art  besitzen,  führte  er  in  die  Familie  Gorgonidae  auch 
diejenigen  Formen  ein,  deren  Skeletachse  suberöse  Struktur 
erkennen  lässt  {Briarcum^  Solanderia,  Faragorgia),  und  dazu  auch 
noch  eine  solche  Form,  bei  welcher  Stamm  und  Zweige  von  einer 
zentralen  weiten  Höhlung  durchsetzt  sind  (Goeloqorgia).  Alle  diese 
einbezogenen  Formen  gruppierte  er  in  "Briarées,"  welche  eine 
"  Agèle  "  der  Gorgoninae  bildeten.  Nun,  wenn  man  nach  seiner 
Bemerkung  über  die  Gattung  Faragorgia  schliesst,  welche  ich 
folgends  anführe,  scheint  er  diese  Gruppe  als  die  primitivste 
der  Gorgoniden  betrachtet  zu  haben. 

"Ce  genre  établit  le  passage  entre  les  Gorgoniens  et  les 
Alcyoniens.     Il  tient  de  ces  derniers  par  la  texture  de  son  axe 


Morphologie  u.  Stammesgeschichte  der  Gorgoniden.  3 

qui  paraît  être  formé  par  un  coenenchyme  très  développé,  et 
non  par  un  tissu  sclérobasique,  comme  chez  les  Gorgones,  le 
Corail,  etc.  ;  mais  il  se  rattache  à  ceux-ci  par  la  conformation  des 
polypes,  dont  la  cavité  viscérale  ne  se  prolonge  pas  inferieurement 
dans  le  coenenchyme,  et  se  termine  brusquement  en  forme  de 
cul-de-sac  arrondi." 

Lacaze-Duthiers  (1)  1863  wies  nach,  dass  die  Skeletachse 
von  CoralUum  rubrum  nicht  durch  die  Ausscheidung  eines  Epithels, 
sondern  durch  die  Aggregation  der  Skleriten  im  Innern  des 
Coenenchyms  gebildet  wird. 

Eine  ähnliche  Entwicklungsweise  der  Skeletachse  wurde  von 
KöLLiKER  (1)  1866  bei  der  Skeletachse  der  Sderogorgia  und  bei 
den  harten  Achsengliedern  von  Melühaea  und  Mopsea  mit  aller 
Bestimmtheit  bewiesen.  Weiter  wandte  derselbe  Autor  die  Idee, 
dass  die  Skeletachse  bei  diesen  Gorgoniden  im  Innern  des 
Coenenchyms  entsteht,  auch  auf  diejenigen  Formen  der  Gor- 
goniden an,  bei  welchen  die  Skeletachse  eine  einfach  lamellöse 
Struktur  erkennen  lässt.  Kölliker  betrachtete  also  die  Gor- 
goniden ganz  als  eine  einheitliche  Gruppe. 

Er  klassifizierte  die  Alcyonarien  folgendermassen: 

Ordo:  Alcyonaria  M.  E. 
Farn.  I.     Alcyonidae  M.  E.     Festsitzende  Alcyonarien  mit 
langen  Leibeshöhlen. 
Subfam.  I.     Cornularidae  M.  E. 
Subfam.  IL     Alcyonidae. 
Fam.  IL     Pennatulidae  M.  E.    Freie  Alcyonarien  mit  langen 

Leibeshöhlen. 
Fam.  III.     Gorgonidae  M.  E.    Festsitzende  Alcyonarien  mit 
kurzen  Leibeshöhlen. 
Subfam.  I.     Gorgonidae  M.  E.     Mit  ungegliederter  hor- 
niger   oder    verkalkter   Axe,    die    eine    Aus- 
scheidung des  Parenchyms  ist. 
Subfam.  IL     Isidinae.     Axe   gegliedert,    aus   hornigen 
und     verkalkten     Stücken    zusammengesetzt, 
von  denen  die  letzteren  einen  lamehösen  Bau 


4  K,  Kinoshita  :  Beiträge  zur  Kenntnis  der 

besitzen  und  nach  dem  Ausziehen  der   Salze 
in  ihrer  Form  sich  erhalten. 
Subfam.  III.     Briareaceae  M.  E.     Gorgoniden,    deren 
Inneres   aus   verschmolzenen   Spicula  besteht, 
die   zum  Theil   eine   ziemlich    gut    begrenzte 
Axe  bilden. 
Subfam.  IV.     Sclerogorgiaceae  Köll.    Gorgoniden  mit 
ungegliederter  Axe,  die  aus  Horn  Substanz  und 
verschmolzenen  Kalkkörpern  besteht.  Coenen- 
chym  wie  bei  Gorgonia. 
Subfam.  V.       Melithaeaceae.        Axe   gegliedert.       Die 
weichen     Glieder     bestehen    aus     getrennten 
Kalknadeln,  umgeben  von  Hornsubstanz  und 
Bindegewebe,    die    harten    Glieder    aus    ver- 
schmolzenen Kalkkörpern. 
Subfam.  VI.     Corallinae  M.  E.     Axe  ungegliedert  aus 
krystallinischer  Kalkmasse  und  mit  derselben 
verschmolzenen     Kalkkörpern     gebildet,     die 
beim  Auflösen  der  Erdsalze  in  der  Form  sich 
nicht  erhält. 
In  dieser  Klassifikation  ist  besonders  zu  beachten,  dass  Kölli- 
KEE,  unter  den  Charakteristiken  der  Familie  Gorgouidae  den  gröss- 
ten  Wert  darauf  legte,  dass  die  Polypen  alle  kurze  Magenhöhlen 
besitzen,  und  auch  dass  er  in  die  Unterfamilie  Briareaceae  die  krie- 
chenden Formen  wie  Sympodium  und  Enjthropodium ,  welche  sonst 
in  die  Familie  Alcyonidae  gestellt  worden  waren,  hineinbrachte.'^ 
Dies  scheint  die  Ansicht  Kölliker's  über  die  Abstammung  der 
Gorgoniden  anzuzeigen,  dass  die  baumförmigen  Briareaceae  direkt 
von  den  kriechenden  Briareaceae  ohne  irgend  eine  Zwischenform 
in  der  Alcyonidae  abgestammt  seien. 

Später  jedoch  beschrieb Kölliker  (3)  1870  seine  neue  Gattung 
Siphonogorgia,  die  er,  seiner  früheren  Meinung  Avidersprechend, 
für  eine  Zwischenform  zwischen  den  Gorgoniden  und  Alcyoniden 
hielt.     Darüber  sagt  Kölliker  Seite  22  folgendes  : 

1)  Vergl.  auch  Kölliker  (2). 


Morphologie  u.  Stamme sgesctiichte  der  GorgoaLioti.  5 

"Erwägen  wir  nun  nach  der  Beschreibung  von  Siplionogorgia 
ihre   Stellung  im  Systeme  und  ihre  Verwandtschaften,    so  ergibt 
sich,    dass    dieselbe    weder    den    Gorgoniden,     noch    auch    den 
Alcyoniden  eingereiht  werden  kann,  vielmehr  eine  Zwischenform 
zwischen  diesen  beiden  grossen   Abtheilungen   der   achtarmigen 
Polypen     oder     Alc^^onarien     darstellt.      Mit     den     Gorgoniden 
und  zwar  mit  der  Unterfamilie  der  Briareaceen  stimmt  dieselbe 
durch  den  Gesammthabitus  so  überein,   dass  sicherlich  jeder,  der 
Siplionogorgia    zuerst    sieht,     diese    Form    den    Gattungen    Para- 
gorgia    oder    Semperina    Köll.     anreihen    und    selbst    im    Zweifel 
sein  wird,  ob  sie  nicht  zu  der  einen  oder  andern  dieser  Gattungen 
gehört.     Wie    bei    den    Gorgoniden  ist  aucli  das  Sarcosoma  von 
Siplionogorgia    durch    Kalknadeln    sehr   hart  und   zerfällt   wie  bei 
den  Paragorgiaceae  mihi  in  eine  Rinden- und  in  eine  Kernsubstanz. 
Ganz  abweichend    von  den  Gorgoniden  ist  auf  der  anderen  Seite, 
dass  die  Darmhöhlen   (Leibeshöhlen  der  Aelteren)  der  Polypen 
von  Siplionogorgia  nicht  kurz  sind,    sondern    wie    bei   den  Alcyo- 
niden in  lange   Kanäle,   die  von  mir   sogenannten    Darmröhren 
auslaufen,  die  durch  den  ganzen  Stock  sich  erstrecken,  und  kommt 
dieser  Thatsache   die  grösste  Wichtigkeit   zu,    die    nur  in    etwas 
dadurch  abgeschwächt  wird,   dass  von  den  acht  Septa  der  Polypen 
nur  4  in  dieses  Röhrensystem  übergehen,  und  nicht  alle  acht,  wie 
bei  den  Alcyoniden.      Ausserdem  ist  der  grosse  Reichthum  des 
Sarcosoma  an  Ernährungsgefässen  und  an  Bindesubstanzzellen  auch 
etwas,    das  bei  den   Gorgoniden  selten  sich   findet  (bei  einigen 
Briareaceen),  bei  den  Alcyoniden  dagegen  sehr  häufig  ist." 

Auf  diese  Gattung  errichtete  Kölliker  eine  besondere 
Unterfamilie,  Siphonogorgiaceae,  welche  er  in  die  Familie 
Gorgonidae,  und  zwar  neben  die  Briareaceae  stellte. 

Klunzinger  (1)  1877  schloss  sich  über  die  Ableitung  der 
Gorgoniden  gänzlich  Kölliker  an.  Auch  er  meinte  nämlich  dass 
die  Gattung  Siplionogorgia  den  Uebergang  von  den  Alcyoniden 
zu  den  Gorgoniden,  und  zwar  spezieller  von  den  Spongodes  zu 
den  Briareaceen,  bilde.  Die  Briareaceen  soll  Klunzinger  nur  auf 
die  Paragorgiaceae  Köll.   (Köll.  2,  S.  11)  beschränkt  haben,  da  er 


g  K.  Kinoshita  :  Beiträge  zur  Keuntnis  der 

SijmpocUum^^  aus  den   Briareaceen    ausschied    und   in    die    Cornu- 
lariden  stellte. 

Die  bis  jetzt  angeführten  Forscher  betrachten  die  Gor- 
goniden  als  eine  einheitliche  Gruppe,  indem  sie  diese  Formen 
nach  den  Beschaffenheiten  der  Skeletachse,  welche  sie  bei  allen 
Gorgoniden  für  morphologisch  homolog  hielten,  wenn  sie  auch 
dieselbe  als  mesogloeales  oder  ektodermales  Sekret  ansahen,  nur 
in  einige  Untergruppen  teilten. 

V.  Koch  (2)  1878  wies  jedoch  bei  Gorgonia  verrucosa  zwischen 
dem  Coenenchym  und  der  Skeletachse  eine  Epithelschicht  nach, 
die  er  Achsenepithel  nannte  und  für  eine  solche  hielt,  welche 
zweifellos  die  Skeletachse  ausgeschieden  hatte,  da  die  Bindesub- 
stanz überhaupt  gar  nicht  mit  der  Skeletachse  in  Berührung 
stand. 

Die  Existenz  einer  solchen  Epithelschicht  konnte  er  (3)  auch 
später  in  demselben  Jahre  noch  bei  einigen  anderen  Gorgoniden 
konstatieren.     Da  er  sich  nun  durch  diese  Untersuchung  davon 
überzeugte,  dass  die  Skeletachse  bei  diesen  Gorgoniden  von   der- 
selben bei  3ïeUtho  les,  Coralliwn,  etc.,  welch  letztere  spikuläre  Struk- 
tur zeigt,  morphologisch  gründlich  verschieden  ist,  so  nahm  er  die 
Gorgoniden  als  eine  diphyletische  Gruppe  an,  und  demgemäss  teilte 
er  dieselbe  in  zwei  Familien  ein  :  namentlich  Axifera  und  Pseud- 
axonia,   von  welchen  die  erstere  den   Unterfamilien   Gorgonidae 
und    Isidinae,    die   letztere  den    Unterfamilien   Sclerogorgiaceae, 
Melithaeaceae  und  Corallina  in  der  Klassifikation  von  Kölllker 
(1)  entspricht.     Ueber  die  Ableitung  der  Pseudaxonia  schloss  sich 
V.  Koch  den  Ansichten  von  Kölliker  (-3)  und    Klunzinger  (1) 
an.    Die  Axifera  aber  wollte  er  von  den  Cornulariden  oder  von  den 
Alcyoniden  ableiten.     Darüber  sagt  er  Seite  476  folgendermassen: 
"  Vni.     Die    Berechtigang   dieser   Familie    (Axifera)   liisst 
sich    wohl    nach    der    vorhergegangenen    Auseinandersetzungen 
nicht  mehr  bezweifeln.     Ihre  näheren  Verwandtschaftsverhältnisse 
zu  den  übrigen  Gruppen  lassen    sich  vor  der   Hand    nicht    mit 
Genauigkeit  angeben,  doch  ist  zu  vermuthen,  dass  sich  entweder 

1)  Eri/tJtropodium  ist  hiervon  Klunzinger  nicht  behandelt. 


Morphologie  n.  Stammesgeschichte  der  Gorgoniden.  7 

bei  den  Cornularien  oder  bei  den  Alcyoniden  (im  letzteren  Falle 
Avolil  in  Verbindung  mit  den  Pennatuliden)  Anknüpfungspunkte 
finden  lassen." 

V.  Koch  (5)  1882  berichtete  vorläufigerweise  die  Resultate 
seiner  Untersuchung  über  die  Entwicklung  von  Gorgonia  cavolini, 
die  er  (6)  später  1887  in  einer  Monographie  der  Zoologischen 
Station  zu  Neapel  ausführlich  beschrieb. 

In  derselben  Untersuchung  konnte  er  feststellen,  dass  das 
Achsenepithel,  welches  er  vormals  bei  den  verschiedenen  Gorgoni- 
den  nachgewiesen  hatte,  in  den  früheren  Stadien  der  Entwicklung 
sich  von  dem  Ektoderm  der  Fussscheibe  des  Primärpolypen 
ableitet.  Unter  den  übrigen  Resultaten,  welche  v.  Koch  in  dieser 
Untersuchung  erlangte,  ist  auch  besonders  zu  beachten,  dass  die 
Skeletachse  bei  ihrem  Wachstum  in  die  Magenhöhle  des  Primär- 
polypen hineintritt,  und  dass  der  Stamm  selbst  eine  modifizierte 
aborale  Hälfte  des  so  von  der  Skeletachse  durchgezogenen 
Primärpolypen  darstellt.  Was  die  Verhältnisse  des  Polypen  und 
der  Skeletachse  anbelangt,  so  fasse  ich  hier  aus  den  Schilderungen 
V.  Koch's  folgendes  zusammen: 

Das  Ektoderm  der  Fussscheibe  des  fertig  gebildeten  Primär- 
polypen scheidet  die  Hornlamellen  wiederholt  aufeinander  aus, 
welche  allmählich  einen  kleinen  Höcker,  den  Anfang  der  Ske- 
letachse, bilden.  Der  letztere  dringt  in  die  Magenhöhle  des 
Primärpolypen  ein,  indem  er  die  drei  primären  Gewebsschichten 
vor  sich  treibt.  Der  so  entstandene  Hügel  liegt  immer  exzentrisch 
in  der  Pol^^penhöhle  und  verschmilzt  teilweise  mit  der  Körperwand, 
teilweise  mit  den  benachbarten  Mesenterien.  Ob  die  Lage  der 
Skeletachse  in  Bezug  auf  die  Mesenterien  eine  konstante  ist, 
wurde  nicht  festgestellt,  doch  fiel  sie  nicht  in  die  Sagittalachse. 
Ehe  der  Hügel  die  Höhe  der  Mundscheibe  des  Polypen  erreicht, 
beginnen  die  um  ihn  liegenden  Teile  der  ursprünglichen 
Polypenhöhle  mehr  Selbständigkeit  zu  zeigen  und  sich  als 
Anfänge  der  späteren  Ernährungskanäle  zu  dokumentieren.  Im 
Vei'laufe  des  weiteren  Wachstums  werden  die  Ernährungskanäle 
noch  selbständiger  und  erhalten  auf  der  dem  Schlundrohr  entgegen- 
gesetzten Seite  der  Skeletachse  schon  nahezu  die  Anordnung  wie 


8  K^.  KinosHta  :  Beiträge  zur  Kenntnis  der 

an  einem  älteren  Busche.  Später  grenzt  sich  der  Polyp  von  dem  die 
Skeletachse  umschliessenden  Teile  durch  eine  Furche  ab,  und 
wird,  da  er  bald  im  Wachstum  hinter  dieser  zurückbleibt,  und 
diese  geradeaus  wächst,  auf  die  Seite  gedrängt  und  erscheint  bald 
nur  als  ein  Anhängsel  des  wachsenden  Stammes.  Der  zweite 
Polyp  bildet  sich  aus  einer  Erweiterung  eines  Ernährungskanals, 
in  die  gewöhnlich  noch  zwei  benachbarte  Kanäle  münden,  auf  der 
dem  ersten  Polypen,  oder  eigenthch  nur  oraler  Hälfte  desselben, 
entgegengesetzten  Seite.  Auf  ähnlicher  Weise  entstehen  weitere 
Polypen.  Der  Stamm  und  auch  die  Zweige  der  Gorgonidenkolonie 
also  stellen  je  die  aborale  Hälfte  der  Axialpolypen  dar.  In  den 
Zweigen,  wie  bei  Muricea  in  welcher  die  acht  Stammkanäle 
gleichmässig  um  die  Skeletachse  angeordnet  sind,  ist  der  Bautypus 
der  Polypen  selbst  sehr  gut  beibehalten. 

Obgleich  V.  Koch  nun  sich  damit  die  Gorgonacea  von  der 
Rhizoxenia  ausgegangen  zu  sein  dachte,  ergibt  sich  doch  aus 
seinen  Schilderungen  dass  die  Gorgonidenkolonie  gerade  einen 
Bautj^pus  der  Telesto  besitzt,  und  auch  dass  es  nicht  Rhizoxenia^ 
sondern  Telesto  ist,  welche  als  die  Ahnenform  angenonmien 
werden  darf. 

In  derselben  Abhandlung  äusserte  v.  Koch  auch  ausführlicher- 
weise seine  Ansicht  über  die  Ableitung  der  Pseudaxonia,  welche 
er  vorher  (3)  in  Kürze  ankündigte.  Er  schloss  sich  nämlich 
KöLLiKER  (3),  Klunzinger  (1)  uud  HrcKsox  (1)  gänzlich  an. 
Er  sagt  darüber  Seite  7  folgendes  : 

"In  derselben  Richtung  schreitet  dann  die  Ausbildung  der 
Gestalt  weiter  fort,  begleitet  von  einer  Regularisierung  der 
Polypenhöhlungen  und  ihrer  sie  verbindenden  Gefässe,  deren 
Veränderungen  in  der  Lage  der  Skelettheile  zur  Seite  gehen, 
und  es  entstehen  Buschformen,  die  sich  denen  der  echten  Gorgo- 
nien  nähern.  Von  solchen  sind  anzuführen  Siplionoijorgia  und 
Verwandte.  Wird  durch  weitere  Vermehrung  oder  durch  Ver- 
schmelzung der  Skelettheile  die  Colonie  immer  mehr  geeignet,  sich 
selber  zu  tragen,  so  werden  die  A  este  schlanker  und  ihre 
Verzweigung  reicher  und  die  Gefässe  werden  immer  regehnässiger 
in  ihrem  Verlaufe,  so  dass  sie  zuletzt  einen  regelmässigen  Zylinder 


Morphologie  u.  Stammesgeschichte  der  Gorgoniden.  9 

(Kreis  in  Querschnitt)  bilden,  welcher  den  Axentheil  von  der 
Peripherie  trennt.  Ersterer  behält  dann  bloss  die  Function  des 
Tragens,  letzterer  dient  als  Schutz  für  die  Polypen  und  ihre 
Verbindungskanäle.  Eine  Anschauung  von  dieser  allmählichen 
Umwandlung  geben  am  besten  die  in  Figur  19-21  dargestellten 
Querschnitte  von  Siphomgorgia,  CoraUium  (Zweigspitze,  weiter 
unten  ähnlich  wie  bei  folgender  Gattung),  Mopsea  (Hornglied). 
Von  anderen  dieser  Reihe  angehörenden  Formen  sind  noch  zu 
erwähnen  Briareiim,  Seniperina,  Sclerogorgia,  3Ielitliaea  etc." 

Diese  Ansicht  v.  Koch's  über  die  Struktur  der  Kolonie  der 
Scleraxonia  und  über  die  Stellung  derselben  in  den  Alcyonacea 
ist  in  der  folgenden  Synopsis  der  Familien,  welche  er  (7)  1890  in 
seinen  vorläufigen  Mitteilungen  über  die  Alcyonaceen  des  Golfes 
von  Neapel  gab,  bestimmt  ausgesprochen: 

I.  Polypen  miteiuander  durch  basale  Stolonen  oder  Stolonenplatten  verbunden. 
Die  Länge  der  vollständig  ausgebildeten  Polypen  einer  Colonie  nahezu  gleich. 

Farn.  Coriiularidae. 

II.  Polypen  mit  einander  durch  verästelte  Eöhren  verbunden,  die  in  verschie- 
dener Höhe  über  der  Basis  einer  Kolonie  verlaufen  und  deren  Wände  zu  einer 
gemeinsamen  Masse  verschmolzen  sind.  Die  Länge  der  vollständig  ausgebildeten 
Polypen  kann  sehr  ungleich  sein  land  ist  dann  von  deren  relativem  Alter  abhängig. 

1.  Spicula  von  einander  getrennt.  Fam.  Alcyonidae. 

2.  Spicula  entweder  durch  Hornsubstanz  oder  durch  krystallinische  Kalk- 
ausscheidung zu  zusammenhängenden  Skeletten  vereinigt. 

Fam.  Scleraxonidae. 

V.  Koch  meinte  nämlich  dass  die  Pseudaxonia  den  Alcyoni- 
den  ähnlich  gebaut  seien,  indem  die  Pol3^pen,  wenn  vollständig 
gewachsen,  Langendifïerenz  zeigen  und  also  in  verschiedenen 
Ordnungen  stehen. 

Etwas  früher  als  die  Veröffentlichungen  der  letzteren  zwei 
Arbeiten  v.  Koch's,  hat  Hickson  (1)  1883  mit  besonderer 
Berücksichtigung  auf  die  Schlundrinne,  Siphonoglyph,  einen 
Versuch  gemacht,  die  Alc^^onarien  systematisch  zu  klassifizieren. 
Was  nun  die  Stammesgeschichte  der  Gorgoniden  anbelangt,  so 
stand  er  auch  Kölliker  (3)  und  Klunzinger  (1)  nahe.  Er  nahm 
die  Siphonogorgia  als  die  gemeinsame  Ahnenform  von  Paragorgia 
und  CoraUium  an.     Von  einer  hypothetischen  Alcyonium-shnlichen 


10  K.  Kinoshita  :   Beiträge  zur  Kenntnis  der 

Ahnenform  (x^)  wollte  er  Briarcns  ableiten,  von  welchem  er  wieder 
Siplionogorgia,  Coelogorgia  etc.,  und  anch  die  übrigen  Gorgoniden 
abgestammt  dachte. 

Gorgoniden 
Coelogorgia,  etc. Briarcus  — 


^ 

C-i- 

"-^ 

:-^ 

^ 

rCt 

^ 

O 

r  - 

/-^ 

^ 

*  "^ 

f"-! 

■O 

Stüder  (2)  veröffentlichte  1887  seinen  Versuch  eines  Systèmes 
der  Alcyonarien,  welches  auf  das  umfangreiche  Material  des  Chal- 
lenger basierte.  Im  demselben  hielt  er  die  Gorgoniden  ebenso  wie 
V.  Koch  für  eine  diphyletische  Gruppe  und  teilte  sie  ein  in  zwei 
Sektionen,  Scleraxonia  und  Holaxonia,  welche  resp.  beinahe  den 
Familien  Pseudaxonia  und  Axifera  v.  Koch' s  entsprechen.  Was 
die  Ableitung  dieser  Gruppen  anbelangt,  so  war  er  jedoch  der 
Meinung,  dieselben  von  den  anderen  Ahnenformen,  als  denjenigen 
in  der  Auffassung  v.  Koch's,  abzuleiten 

Als  den  Bautypus  der  Holaxonia  stellt  sich  Stucer  eine  Tdcsio- 
Kolonie  vor,  deren  Stamm  und  Zweige  je  einen  verlängerten 
Axialpolypen  darstellen  und  zahlreiche  Seiten poly pen  mit  kurzen 
Magenhöhlen  tragen.  Die  Gorgonienkolonie,  meint  er,  lasse  sich 
von  der  Telesto-ls^oioniQ  leicht  ableiten,  wenn  man  bei  der  letzteren 
die  zentrale  Höhlung  der  Axialpolypen  von  unten  her  von  einer 
zur  Skeletachse  sich  differenzierenden  Mesodermwucherung,  welche 
sich  am  besten  mit  der  Kolumella  der  jMadreporarier  vergleichen 
lässt  und  natürlich  das  Entoderm  der  Axialpolypen — Achsenepithel 
— vor  sich  herschiebt,  ausgefüllt  denkt.  Die  Radialfächer  der 
Verdauungshöhle  der  Axialpolypen  könnten  dabei  direkt  zu  in 
Achtzahl  auftretenden  Stammkanälen  umwandeln. 

Diese  Auffassung,  dass  die  Stämme  und  Zweige  je  die  Axial- 
polypen darstellen,  stimmt  gewissermassen  mit  den  Schilderungen 
V.  Koch' s  (5, 6)  überein.  Der  Unterschied  zwischen  den  Ansichten 
beider  Autoren  liegt  nur  darin,  dass  die  Axialpolypen  nach  Stüder 


MorphoJoyie  u.  Stammosgeschichto  der  GorgonicU-n.  ^^ 

meist  nicht  mehr  vollständig  geformte,  sondern  reduzierte,  rein 
vegetative  Individuen  sind,  Avährend  dieselben  nach  v.  Koch  sich 
den  gewöhnlichen  Polypen  vergleichen  lassen. 

Betreffs  der  Herkunft  der  Skleraxonier  äusserte  Studer  die 
folgende  Idee: 

"Eine  solche  Anordung  ist  aber  nur  von  Vortheil 
unter  günstigen  Nahrangsverhältnissen,  unter  denen  nicht 
nur  den  Randpolypen,  sondern  auch  den  im  Centrum  der 
Colonie  befindlichen  gleichmässig  die  Beute  zugeführt  wird, 
oder  wo  die  Colonie  mannigfach  gestaltete  Körper  überzieht,  welche 
bewirken,  dass  die  Einzelthiere  in  verschiedene  Lage  zu  einander 
kommen,  und  verschiedene  Wasserschichten  beherrschen.  Immer 
wird  dieses  abhängig  sein  von  dem  Vorkommen  der  Fremdkörper, 
welche  als  Unterlage  dienen.  Der  Vortheil  der  Vertheilung  der 
Einzelthiere  wird  jedoch  bei  höheren  Formen  in  anderer  Weise 
erreicht.  Die  Colonie,  statt  Fremdkörper  zu  überziehen,  erhebt 
sich  von  einer  Basis,  welche  sich  überrindend  an  fremde  Körper 
anheftet,  frei  in  die  Höhe  in  Form  eines  Blattes,  dessen  eine 
Fläche  die  Polypen  einnehmen,  während  die  andere  der  Basalseite 
der  Rasenkolonie  entspricht.  Aus  statischen  Gründen  bleibt  aber 
die  Colonie  nicht  fläche nh aft,  sondern  rollt  sich  röhrenförmig 
zusammen,  so  dass  die  polypentragende  Seite  nach  aussen  kommt, 
die  frühere  Basis  die  Innenw^and  der  Röhre  darstellt.  Zugleich 
gruppiren  sich  im  Coenenchym  besonders  differenzirte  Spikula 
dicht  an  einander,  um  eine  aus  Spicula  gebildete  stützende  Axe 
darzustellen.  Diese  Verhältnisse  zeigen  noch  niedere  Briareiden, 
so  Solenocaulon,  bei  höheren  Typen  ist  diese  Axe  mehr  ent- 
wickelt, rückt  ins  Innere  der  Colonie  und  bildet  einen  cylind- 
rischen  Stab,  der  rings  von  polypentragendem  Coenenchym 
umgeben  ist.  In  dieser  Weise  können  wir  uns  die  Ableitung  der 
Scleraxonia,  deren  höchste  Form  CoraUktm  darstellt,  entwickelt 
denken." 

KÜKENTHAL  (1,2)  schloss  sicli  Über  die  Ableitung  sowohl  der 
Gorgonacea  als  auch  der  Scleraxonia  beinahe  gänzlich  der  ilnsicht 
Studek's  an.  Die  Gattungen  Tdesto  und  Enj ihr opo ilium  nahm 
er  resp.   als  die  Ahnenformen  der  Gorgonacea  und  Scleraxonia 


12 


K.  Kinoshita  :  Beiträge  7,ur  Kenntnis  der 


an.      Diese    Auffassung  ist  in  dem  folgenden    Schema,  welches 
er  Seite  99  gab,  leicht  zu  ersehen  : 


Scleraxonia 


L'njthropodiwn 


Pennatulacea 

Holaxonia 


Telesto 


Coelogorgia 


Pseudogorgia 


Sclerantlielia 


Anthelia 


Cormdana 


Wenn  er  Seite  52  bemerkt,  dass  Alcijonium  (Enjthropo- 
diiim)  contortum,  welches  einen  durch  die  Kutikula  gefüt- 
terten basalen  Hohlraum  zeigt,  die  Zwischenform  zwischen  den 
Alc3^onaceen  und  den  Gorgonaceen  bildet,  bezog  sich  dies 
wahrscheinlich  auf  die  Scleraxonia  und  nicht  auf  die  Axifera  hin, 
da  er  (3)  später  eine  ähnliche  Umbildung  bei  Solenocaidon  stechet 
berichtete,  von  der  er  glaubte,  dass  sie  für  die  Solenocaulon- 
Hypothese  eine  neue  Stütze  liefere. 

Im  Gegensatz  zu  den  übrigen  neueren  Forschern  betrachtete 
HicKSON  (3)  1906  die  Gorgoniden,  welche  nach  ihm  sich  von  den 
Alcj^onacea  nur  durch  die  von  den  Polypenhöhlen  nicht  durchzo- 
gene Skeletachse  unterscheiden,  als  eine  monophyletische  Gruppe. 
Er  fasste  also  die  Pseudaxonia  und  die  Gorgonacea  als  Unter- 
ordnungen in  der  Ordnung  Gorgonacea  zusammen.  Die  Ansicht 
Hickson's  über  die  Abstammung  dieser  beiden  Gruppen  ist  in 
den  Sätzen,  welche  ich  folgends  wörtlich  anführe,  wohl  ersichtlich  : 


Morphologie  u.  Stainmesgeschichte  der  Gorgoniden.  13 

*  '  It  has  been  suggested  that  as  the  Axifera  represent  a  line 
of  descent  distinct  from  that  of  the  Pseudaxonia  they  should  be 
placed  in  a  separate  order.  Apart  from  the  character  of  the  axis, 
however,  the  two  suborders  show  so  man}^  affinities  in  their  general 
anatomy  that  it  is  better  to  regard  the  two  lines  of  descent  as  united 
within  the  Gorgonacean  limit.  It  is  very  improbable  that  the  two 
groups  sprung  independent^  from  a  stoloniferous  ancestor." 

Wie  schon  angedeutet,  scheinen  die  neueren  Forscher  mit 
alleiniger  Ausnahme  von  Hickson  alle  darin  übereinzustimmen, 
die  Gorgoniden  als  diphyletisch  zu  betrachten.  Während  diese 
Trennung  wohl  ausser  Zweifel  gesetzt  zu  sein  scheint,  bleibt 
jedoch  die  Frage,  ob  bei  den  beiden  Gruppen  die  Axialpolypen 
vorkommen  oder  nicht,  ganz  unentschieden.  Uebrigens  ist  die 
morphologische  Bedeutung  der  Skeletachse  beider  Gruppen  von 
den  verschiedenen  Forschern  verschieden  beantwortet,  so  dass 
ihre  Ableitung  auch  verschiedenartig  versucht  worden  ist. 

In  dem  folgenden  also  möchte  ich  nunmehr  über  die  morpho- 
logische Bedeutung  der  Kolonie  und  über  die  Stammesgeschichte 
der  Gorgoniden,  besonders  von  dem  Standpunkte  meiner 
Auffassung  aus,  diskutieren. 

II.     Gorgonacea. 

Wie  im  vorigen  Kapitel  angeführt,  ist  die  morphologische 
Bedeutung  der  Gorgonidenkolonie  sowie  die  Abstammung 
der  Gorgonacea  von  den  verschiedenen  Forschern  abweichend 
aufgefasst  worden. 

KöLLiKER  (3)  und  Hickson  (1),  welche  beide  die  Gorgoni- 
den als  einheitlich  betrachteten,  meinten  dass  die  Gorgonacea, 
durch  Zwischenformen  wie  Siplionogorgia  oder  Briareum,  von 
den  Alcyoniden  abzuleiten  seien.  Diese  Auffassung  beruht 
jedoch  auf  die  Voraussetzung  dass  alle  unter  den  Gorgoniden 
zusammengefassten  Formen  phyletisch  einheitlich  seien,  v. 
Koch  (2,  3,  5,  6)  jedoch  stellte  fest,  dass  die  echten  Gor- 
goniden einen  anderen  Entwicklungszweig  als  die  Skleraxonier 
darstellen  ;      er    glaubte     dass    dieselben     von    der    Rhizoxenia 


24  ^-  Kinoshita  :   Heiträge  zur  Kenntnis  der 

ausgegangeil  seien.  Studer  (2),  der  die  Gorgoniden  auch  als 
diphyletisch  betrachtete,  war  der  Ansicht,  dass  dieselben,  ganz 
wie  es  sich  aus  den  Schilderungen  v.  Koch's  (6)  ergibt,  von 
den  Telestiden  abgeleitet  werden  müssen.  Während  man  Grund 
hat,  die  Auffassung  der  ersteren  zwei  Forscher  als  veraltete 
anzunehmen,  scheint  doch  die  Idee,  dass  die  Gorgonacea  aus 
den  Telestiden  ausgegangen  seien,  jetzt  allgemein  zu  herrschen. 
Der  Grund,  weshalb  diese  so  allgemeine  Annerkennung 
fand,  liegt  wahrscheinlich  darin,  dass  v.  Koch  und  Studer, 
obgleich  sie  die  Skeletachse  niorphologisch  ganz  verschieden 
andeuteten,  doch  gleich  übereinstimmend  die  Existenz  der  Axial- 
polypen behaupteten.  Diese  hier  in  Bezug  genommene  Idee 
scheint  mir  jedoch,  wie  ich  im  folgenden  zeigen  möchte,  nicht 
ganz  statthaft  zu  sein. 

Beim  Studium  der  Stammesgeschichte  der  Gorgonacea  nun 
ist  es  dringend  wichtig  die  folgenden  zwei  Punkte  voraus 
festzustellen:  1)  Die  morphologische  Bedeutung  der  Skeletachse; 
und  2)  die  Art  und  Weise,  wie  die  Polypen  sich  an  der 
Kolonienbildung  beteiligen,  oder  in  welcher  die  Polypen  und  das 
Coenenchym  im  Zusammenhange  stehen.  In  dem  folgenden 
beziehe  ich  mich  nun  auf  diese  Fragen. 

i.      3Io)'p}iologisc]ic  Bedeutung  der  Sheletachse. 

Lamarck  (1),  Ehrenberg  (1),  Dana  (1,2)  und  Milne- 
Edwards  (1)  sahen  die  Skeletachse  der  Gorgoniden  als  ein  epithe- 
liales Sekret  an.  Diese  Annahme  war  jedoch  nicht  auf  wirkliche 
Beobachtung  gestützt.  Kölliker  (1)  konstatierte  die  Tatsache, 
dass  die  Teilnahme  der  Skleriten  an  der  Ausbildung  der  Skelet- 
achse, was  schon  bei  Coralliwn  ruhrum  von  Lacaze-Duthiers 
(1)  nachgewiesen  worden  war,  auch  bei  denjenigen  Formen  der 
Gorgoniden,  welche  man  jetzt  in  den  Melitodiden  und  Suberogorgi- 
iden  zusammenfasst,  der  Fall  ist.  Da  es  ihm  aber  nicht  gelungen 
war,  bei  den  echten  Gorgoniden  ein  Epithel  zwischen  der  Skelet- 
achse und  Mesogloea  nachzuweisen,  sah  er  diese  Skeletachse  als 
ein  mesogloeales  Sekret  an. 


Morphologie  u.  Stammesgescliichte  der  Gorgoniden.  25 

Als  Stütze  dieser  Annaliiiie  führte  Kölliker  die  folgenden 
fünf  Gründe  an: 

"  1)  Manche  Axen  dieser  Abtheilung  schliessen,  wenn  auch 
nur  zufällig,  im  Innern  vereinzelte  Kalkkörper  des  Coenenchyms 
ein,  was  zu  beweisen  sclieint,  dass  der  Zusammenhang  zwischen 
Coenenchvm  und  Axe  ein  viel  grösserer  ist,  als  man  bisher 
anzunehmen  geneigt  war. 

"  2)  In  der  That  habe  ich  auch  nirgends  als  Begrenzung  des 
Coenenchyms  gegen  die  Axe  eine  Epithelschicht  gefunden,  wie 
sie  doch  dasein  müsste,  wenn  die  gang  und  gäbe  Auffassung  der 
Axen  die  richtige  wäre. 

"  3)  Scheinen  die  netzförmigen  Verbindungen,  die  die  Axen 
vieler  Gorgonien  eingehen  (Eliipidogorgia  etc.)  zu  beweisen,  dass 
die  Axen  innere  Productionen  des  Coenenchyms  sind.  Wenn 
nämlich  Aeste  verschmelzen,  so  verschmilzt  erst  das  Coenenchym 
derselben  und  erst  dann  bildet  sich  eine  Vereinigung  der  Axen 
auf  Kosten  des  Coenenhyms,  wie  man  am  besten  daraus  sieht,  dass 
diese  Axentheile  häufig  viele  Kalkkörper  einschliessen. 

'  '  4)  Der  Bau  der  fraglichen  Axen  ist  derart,  class  sie  viel 
mehr  an  Bindesubstanz  als  an  Cuticularbildungen  sich  schliessen, 
und  erinnere  ich  vor  Allem  1)  an  die  feinen  Fasernetze  im 
Central  strange  und  dem  Schwammgewebe  der  Rinde  bei  vielen 
Gattungen  mit  hornigen  Axen  und  2)  an  den  Bau  der  Weichtheile 
der  Pennatulidenaxen  mit  ihren  feinen  Fäserchen  und  sie 
durchsetzenden  Radialfasern. 

'*  5)  Endlich  erwähne  ich  noch  eine  Thatsache,  die  im 
Allgemeinen  zeigt,  dass  auch  Hornsubstanz  für  sich  allein  im 
Innern  des  Coenenchyms  sich  bilden  kann.  Bei  Alcijonium 
pabnatuni  fand  ich  in  Einem  Falle  in  den  oberen  Theilen  des 
Stammes  eine  kurze  Axe  aus  lamellöser  Hornsubstanz,  rings 
umgeben  von  der  gewöhnlichen  Bindesubstanz  des  Coenenchyms, 
eine  Bildung,  die  sicher  nicht  auf  eine  Epithelialausscheidung 
zurückzuführen  ist." 

Kölliker  verglich  somit  die  Skeletachse  mit  den  Kalk- 
körpern. Darüber  drückte  er  Seite  167  aus:  ''Zum  Schlüsse 
kann  ich  die  Bemerkung  nicht  unterdrücken,  dass  die  Bildung 


2^g  K.  Kinoshita  :  Beiträge  zur  Kenntnis  der 

einer  Gorgoniden-  und  Pennatulidenaxe,  sofern  sie  aus  homogener 
Substanz  besteht,  im  Grossen  das  wiederholt,  was  ein  spindel- 
förmiger Kalkkörper  im  Kleinen  zeigt." 

Stüder  (1  y^  trat  dann  an  die  Seite  von  Kölliker.  Nach  ihm 
bildet  sich  die  Skeletachse  folgendermassen  :  Die  Spicula  dringen 
aus  dem  Coenenchym  in  die  Tiefe,  dienen  als  Centra  für  die 
Hornablagerung  ;  sie  werden  aber  dort  später  resorbirt  und  an  ihre 
Stelle  treten  mit  spongiöser  Substanz  ausgefüllte  Lücken;  sie 
bleiben  nur  in  seltenen  Fällen  erhalten. 

V.  Koch  (2,  3)  jedoch  zeigte,  dass  diese  Auffassung  von 
Kölliker  und  Stüder  ein  Irrtum  ist.  Er  fand  nämlich  bei  einigen 
Gorgonien  das  Achsenepithel  um  die  Skeletachse,  welches  er  (5, 
6)  nachher  embryologisch  als  vom  Ektoderm  der  Fussscheibe  des 
Primärpolypen  herstammend  feststellte. 

Stüder  (2)  aber  mass  diesem  Befunde  keinen  Glauben  bei. 
Obgleich  er  auch  das  Vorkommen  eines  Epithels  um  die  Skelet- 
achse bestätigte,  nahm  er  dasselbe  doch  nicht  als  das  Ektoderm 
an.  Diese  Auffassung  Stüder' s  wurde  1905  von  Schneider  (1) 
abermals  durch  wirkliche  Beobachtungen  zu  stützen  versucht. 
Wenn  er  jedoch  sagt,  dass  das  Achsenepithel  die  Zellauskleidung 
des  axialen  Hohlraumes  sei,  so  ist  ihm  hierin  keinen  Glauben 
zu  schenken. 

Diese  Frage  nun  glaube  ich  durch  meine  Untersuchungen 
über  die  Entwicklung  von  Antliopkxaura  dimorpha  beinahe 
entschieden  zu  haben. -^  Hier  ist  das  Achsenepithel,  ganz  wie  die 
Befunde  v.  Koch's,  entschieden  ein  Derivat  vom  Ektoderm  der 
Fussscheibe,  mit  welcher  der  junge  Primärpolyp  an  dem  Substratum 
anwächst.  Da  beim  Wachsen  des  jungen  Primärpolypen  die 
Erweiterung  der  Fussscheibe  an  der  Peripherie  nicht  gleichmässig 
stattfindet,  so  rückt  die  Stelle,  wo  das  Achsenepithel  und  das  Sekret 
desselben,  die  Skeletachse,  sich  befinden,  nach  einer  Seite  zu,  was 
auch  bei  den  v.  Koch' sehen  Exemplaren  als  allgemein  angegeben 
worden  ist. 


1)  Diese  Arbeit  Studeb's  stand  mir  leider  nicht  zur  Gehote,  darum  habe  ich  mich  mit  den 
Zitaten  in  den  Abhandhingen  von  v.  Koch  (6),  Studeb  (2)  und  Schneider  (1)  begnügen  müssen. 

2)  Kinoshita  (1). 


Morphologie  ii.  Stammesgeschicht'^  der  Gorgoniden.  17 

Das  Achsenepithel  ist  von  der  Form  eines  blinden  Sackes 
and  setzt  sich  weder  in  das  Entoderm  der  Solenia,  welche  sich  in 
der  Mesogloea  um  das  Achsenepithel  reichlich  entwickeln,  noch  in 
dasjenige  der  Magenhöhle  des  Primärpolypen  fort.  Es  ist  also 
ganz  unmöglich,  dass  dieses  Achsenepithel  entodermal  ist,  wie 
dies  von  Studer  und  Schneider  behauptet  wurde.  Auch  kann 
dasselbe  bei  den  Zweigspitzen  der  erwachsenen  Kolonien,  falls 
das  Achsenepithel  dort  nachweisbar  ist,  ebenso  mit  Sicherheit 
behauptet  werden. 

Das  Achsenepithel  scheidet  die  dünnen  Hornlamellen  wieder- 
holt  aus,    Avelche  nur  in  ihrem  zentralen  Abschnitt  durch  die  gal- 
lertartige Substanz  getrennt,  doch  durch  ein  diese  durchziehendes 
Fasernetzwerk,    verbunden   sind.      Eine    gute    Vorstellung    von 
der  Skeletachse  gewinnt  man,  wenn  man  diese  Hornlamellen  mit 
Probiergläsern    vergleicht,    welche     derart      tief      in      einander 
gesteckt    worden    sind,    dass    an    den    Boden    derselben     enge 
Zwischenräume  zurückbleiben.     Aus  dem  Verhalten  des  Fasernetz- 
werks   in   der  interlamellären  Gallerte   könnte   man   schliessen, 
dass   die    Hornlamellen   zuerst    dicht    auf    den    vorhergehenden 
ausgeschieden,  und  erst  nachher    durch    die   Anschwellung    der 
dazwischen    vorhandenen  Gallerte  auseinander  getrennt  worden 
sind.     Die  Tatsache,  dass  die  Skeletachse,  soweit  meine  Erfahrung 
reicht,   immer  mit  der  Hornlamelle,  aber   nie   mit  der   Gallerte 
endet,  macht  diese  Vermutung  wahrscheinlich. 

Es  ist  nun  beinahe  nicht  mehr  zu  bezweifeln,  dass  bei  den 
gewöhnlichen  Gorgoniden  die  Skeletachse,  welche  die  lamellöse 
Struktur  besitzt,  das  Sekret  des  Achsenepithels  ist.  Dieses  letztere 
ist  jedoch  nicht  immer  gleich  leicht  nachweisbar,  da  dasselbe 
nicht  selten  Umbildung  zu  erleiden  scheint. 

V.  Koch  (2,  3,  6)  beschreibt  dass  das  Achsenepithel  am 
Zweigende  im  allgemeinen  aus  den  höheren  Zellen  besteht,  die 
aber  proximalwärts  immer  niedriger  werden,  um  endlich  ganz 
flach  zu  werden.  Nach  meinen  Erfahrungen  scheint  das  Achsen- 
epithel auch  in  den  unteren  Abschnitten  der  Zweige  sich  ähnlich 
zu  verhalten.  Darin  dass  das  derart  membranartig  gewordene 
Epithel    noch    die    Sekretionsfähigkeit    beibehält,     darf    einiger 


J^g  K.  Kinosliita  :  Beiträge  zur  Kenntnis  der 

Zweifel  gesetzt  werden.  Dieser  nimmt  im  Grade  zu,  wenn  man 
hierbei  das  Verhalten  der  Skeletachse  von  Keroeidcs^^  in  Betracht 
zieht.  Diese  Gattung,  obgleich  sie  diagnostisch  von  den  ge- 
wöhnlichen Gorgonaceen  sehr  weit  abweicht,  scheint  doch  eine 
allgemeine  Tendenz  zur  Reduktion  des  Achsenepithels  zu  zeigen. 
Bei  Keroeides  verhält  sich  die  Skeletachse  folgendermassen: 

Dieselbe  besteht  aus  zwei  Komponenten  gänzlich 
differenten  Ursprungs.  Es  sind  nämlich  der  hornig  lamellöse 
Zentralstrang  und  die  sklerogorgische  Rindenlage. 

Der  erstere,  welchen  schon  Hiles  (1)  1899  gesehen,  aber 
nicht  näher  studiert  hat,  zeigt  deutlich  die  eigentümliche  Struktur, 
welche  bei  den  gewöhnlichen  Gorgonaceen  ganz  allgemein  zu 
sehen  ist.  Die  Rindenlage,  die  den  ersteren  dicht  und  direkt 
umschliesst,  ist  von  ganz  typisch  sklerogorgischer  Struktur, 
sodass  Weight  und  Studer  diese  Gattung  in  die  Suberogorgiidae 
hineinstellte,  da  die  Existenz  des  Zentralstranges  ihnen  un- 
bekannt blieb.  Sie  besteht  aus  den  spindelförmigen  Skleriten 
und  der  dieselben  verkittenden  Hornsubstanz,  welch  letztere 
allmählich  ohne  Grenze  in  die  Substanz  der  Mesogloea  übergeht. 
Verfolgt  man  nun  die  Skeletachse  nach  der  Zweigspitze  hin, 
so  sieht  man  sie  die  Scheidewand  hindurchziehen,  welche  die 
Magenhöhlen  der  meist  gegenüber  stehenden  Apikaipolypen 
von  einander  trennt,  und  gerade  unterhalb  der  oberflächlichen 
Skleritenschicht  des  Zweigapex  terminieren.  Dieselbe  besteht 
auf  einige  Länge  nur  aus  dem  Zentralstrange  und  ist,  soweit 
es  nackt  bleibt,  von  dem  fingerhutförmigen  Achsenepithel 
bekleidet.  Die  spindelförmigen  Skleriten,  welche  die  Rindenlage 
aufbauen,  sieht  man  schon  in  der  Mesogloea  der  oben  erwähnten 
Scheidewand,  wohin  das  Achsenepithel  noch  reicht,  sich 
ausbilden.  Unten  aber  werden  diese  Skleriten  um  den  Zentral- 
strang durch  die  Hornsubstanz  gebunden,  wobei  die  Zellen  des 
Achsenepithels  in  die  Mesogloea  sich  verlieren  und  wahr- 
scheinlich schliesslich  zugrunde  gehen. 


1)   KiNOSHITA  (2). 


Morphologie  ii.  Stammesgeschichte  der  Gorgoniden.  19 

Durch  Betrachtung  der  erwähnten  Verhältnisse  der  Keroeides- 
Skeletachse  wird  man  gewahr  von  der  Analogie  welche  zu 
bestehen  scheint  zwischen  diesen  und  der  Tatsache  dass  bei 
manchen  Gorgoniden,  in  der  Rindenlage  der  Skeletachse  im 
unteren  Abschnitte  der  Zweige,  zahlreiche  Rindenskleriten  sich 
eingebettet  zeigen,  was  wohl  einer  anderen  Erklärung  zu  bedür- 
fen scheint,  als  dass  sie  nur  aus  mechanischen  Gründen  in  die 
Rindenlage  hineingekommen  sind. 

Ein  etwas  differenter,  doch  ähnlicher  Fall  der  Umbildung  der 
Skeletachse  wurde  neuerdings  von  Müller  (1)  bei  Corallium 
berichtet.  Diese  Gattung  ist  nach  ihm,  wie  schon  v.  Koch  (3) 
vermutete,  eine  echte  Gorgonie  mit  einem  vollkommenen 
Achsenepithel,  welches  die  Kittsubstanz  ausscheidet,  während  die 
Skleriten,  welche  die  Skeletachse  aufbauen,  anfangs  im  Coen- 
enchym  gebildet  werden  und  dann  durch  das  Achsenepithel  in 
den  Abschnitt  der  Skeletachse  hineinkommen. 

2.      Morphologische  Bedeutung  der  Polypen. 

KöLLiKER  (1)  gab  der  Familie  die  Charakterisierung,  dass 
die  Polypen  kurze  Magenhöhlen  besitzen,  und  er  brachte  in  diese 
Familie  die  kriechenden  Formen,  Symjwdnun  mid  Enjtliropodium, 
hinein.  Köllikee  also  scheint  gemeint  zu  haben,  dass  die 
Polypen  der  Gorgonidenkolonie  alle  gleichwertig  sind,  und  dass 
die  Stämme  und  Zweige  alleinig  aus  dem  Coenenchym  bestehen, 
ohne  Beteiligung  der  Polypen.  Diese  Auffassung  hat  er  jedoch 
später  fallen  lassen.  Er  nahm  nämlich  die  Sijihonogorgia,  wegen 
der  vier  bleibenden  Mesenterien  in  den  unteren  Abschnitten  der 
Polypenhöhlen  (Stammkanälen)  als  eine  Zwischenform  zwischen 
den  Gorgoniden  und  den  Alcyoniden  an.  Nach  dieser  Auffassung 
stellen  die  Stämme  und  Zweige  deutlich  nicht  das  eigentliche 
Coenenchym,  sondern  die  Bündel  verlängerter  Polypen  (Axial- 
polypen) dar. 

V.  Koch  (5,  6)  erlangte  in  seiner  Untersuchung  über  die 
Entwicklung  von  Gorgonia  cavolini  unter  anderm  ein  wichti- 
ges  Resultat,    dass   nämlich   die   wachsende    Skeletachse    in  das 


20  ^-  Kinoshita  :  Beiträge  zur  Kenntnis  der 

Coelenteron  des  Primärpolypen  eindringt.  Obgleich  v.  Koch 
selbst  die  Gorgonien  nicht  von  den  Telestiden,  sondern  von  der 
Rhizoxenia  ableiten  wollte,  ergibt  sich  doch  offenbar  aus  diesem 
Resultate,  dass  der  Aufbau  der  Gorgonienkolonie,  ganz  wie  Studer 
behauptet,  dem  Telesto-Ty^ius  angehört. 

Diese  Idee  hatte  wahrscheinlich  auf  Studer  (2)  Einfluss, 
wenn  er,  auf  das  Challenger-Material  basierend,  seine  Hypothese 
vorschlug,  dass  die  Gorgonien  in  ihrem  Aufbau  der  Kolonie  vom 
Telesto-TyTpus  seien  und  demnach  von  der  Telesto  abzuleiten 
seien.  Die  Stämme  und  Zweige  stellen  nach  ihm  je  die  Axial- 
polypen dar.  In  normalen  Fällen  werden  die  acht  Radialfächer 
der  Magenhöhle  der  Axialpolypen  direkt  zu  den  acht  Stamm- 
kanälen. Die  Fälle,  in  welchen  die  letzteren  weniger  als  acht 
zählen,  hielt  Studer  für  Umbildungen.  Obgleich  v.  Koch  sich 
darauf  beschränkte,  die  Axialpolypen  nur  bei  den  jungen  Kolonien 
zu  beschreiben,  versuchte  Studer  dieselben  der  Zweigspitzen 
bei  erwachsenen  Kolonien  zu  schildern.  Er  sagt  nämlich  Seite 
37: 

"Fernere  Analogieen  mit  den  Pennatuliden  zeigen  sich  darin, 
dass  der  als  axialer  Polyp  aufgefasste  Stamm  selten  an  seinem 
Ende  in  einen  Polypen  mit  Mund  und  Magenrohr  ausläuft  sondern 
ein  rein  vegetatives  Individuum  darstellt,  das  am  Ende  blind 
geschlossen  ist,  resp.  immer  auf  dem  Stadium  eines  Stolonen 
bleibt.  Selten  sind  übrigens  die  axialen  Polypen  radiär,  wie  in 
den  erwähnten  Fällen,  meist  macht  sich  eine  Tendenz  zur  bilate- 
ralen oder  biradiären  Symmetrie  geltend.  So  können  sich,  wie  bei 
DasTjgorgia,  die  Längskanäle  auf  zwei  reduciren,  welche  auf  zwei 
Seiten  der  Axe  verlaufen,  oder  es  können,  wie  bei  den  meisten 
Gorgonelliden,  zwei  Hauptkanäle  vorhanden  sein,  welche  an  zwei 
Seiten  verlaufen,  während  zwischen  diesen  eine  Anzahl  kleinerer 
entwickelt  sind.  Mit  diesem  primär  biradiären  Bau  des  axialen 
Stammpolypen  steht  dann  gewöhnlich  auch  ein  nach  demselben 
Princip  angeordnetes  System  von  Aesten  und  Zweigen  im  Zusam- 
menhang." 

Diese  Annahme  Studer' s,  dass  rein  vegetative  Axial- 
polypen   an    den    Zweigenden    auftreten,     ist    nichts    als    eine 


Morphologie  u.  Stammesgeschichte  der  Gorgoniden.  21 

Hypothese.  Das  ist  auch  klar  zu  schhessen  aus  dem  Satze 
Schneider's  (1),  welcher  folgendermassen  lautet  :  "Dieser 
Befund  stimmt  in  geradezu  frappierender  Weise  mit  der 
auf  rein  theoretischen  Erwägungen  beruhenden  Darstel- 
lung der  Achsenbildung  Studer's  überein,  die  er  in 
seinem  Versuch  eines  Systèmes  der  Alcyonaria  niedergelegt 
hat." 

Schneider  (1)  beschrieb  bei  Eunicella  cavolini  einen  am 
Stammapex  vorhandenen  axialen  Hohlraum  als  einen  Axi- 
alpolypen. Den  Grund,  weshalb  er  diesen  Hohlraum  als  einen 
solchen  ansieht,  gab  er  aber  niemals  an.  Menneking  (1)  beschrieb 
auch  Zweige  gewisser  Primnoiden  als  umgewandelte  Axialpolypen. 
Es  ist  jedoch  auch  nicht  angegeben,  warum  er  die  Scheidewände 
der  Stammkanäle  als  die  Mesenterien  der  Axialpolypen,  ansieht. 
Das  Vorhandensein  eines  solcherweise  reduzierten  Polypen  kann 
nur  durch  die  Untersuchungen  derselben  im  Anfange  der 
Knospung,  im  Zusammenhang  mit  der  Knospung  von  den 
Zweigen,  bewiesen  werden. 

Kükenthal  (1,  2)  schloss  sich  Studer  auch  darin  an,  dass 
er  die  Gorgonienkolonie  als  dem  Telesto -Typus  angehörend 
ansah. 

Wie  angeführt,  stimmen  die  neueren  Forscher  alle  merkwür- 
digerweise darin  überein,  dass  sie  in  der  Gorgonienkolonie 
die  Existenz  der  Axialpolypen  behaupten.  Die  Schilderung 
V.  Koch's,  dass  die  wachsende  Skeletachse  in  das  Coelenteron  des 
Primärpolypen  eindringt,  lag  wahrscheinlich  den  Auffassungen 
der  späteren  Autoren  zugrunde.  Es  könnte  sein  dass  v.  Koch 
seine  Schilderung  auf  direkte  Beobachtungen  basierte;  gleich- 
wohl scheint  es  unläugbar  zu  sein  dass  seine  Angabe  mit  vielen 
anatomischen  Verhältnissen  leider  kaum  in  Einklang  steht. 
1.  Kommen  wirklich  Axialpolypen  vor,  welche  nicht  wie  nach 
Studer  rein  vegetative,  aber  wie  nach  v.  Koch  normal  gebaute  sind, 
so  ist  kaum  erklärbar,  dass  die  Mesenterien  der  oberst  gelegenen 
Polypen  sich  ganz  wie  bei  den  gewöhnlichen  Polypen  verhalten 
und  niemals  in  die  Scheidewände  der  Stammkanäle  übergehen. 


22 


K.  Kinoshita  :  Beiträge  zur  Kenntnis  der 


A-bbildnng  1. 
Primnoa  pacißca   Kinoshita. 


2.  Sind  die  Stammkanäle  in  der  Tat  direkte  Fortsetzungen  der 

Radialfächer  von  den  Magenhöh- 
len der  Polypen,  so  sollen  einige 
Mesenterien  ebenfalls  in  den 
Stammkanälen,  wenn  auch  in 
zufälliger  Weise,  gefunden  wer- 
den, falls  natürlich  die  lezteren 
in  weniger  Anzahl  als  acht  vor- 
kommen. Das  ist  jedoch  nicht 
der  Fall,  soweit  berichtet  ist 
oder  meine  eignen  Erfahrungen 

Querschnitt  eines  Zweiges  gerade  ober-      hinreichen.      3.    DaSS  bei  PlimnOa 
halb    des  obersten  Polypen.   Ektoderm  und  ,  , 

Entoderm    mit    dicker    Linie,    Lacunae   der     paClflCCl    dlC     mclir     apikalen     Ab- 

^^^:t^r^^Z^S:^^'^SSê:rl    schnitte  der  Zweige  als  bei  dem 
^^^'S^^^ISStîl^    obersten     Polypen     auch     acht 

genden  Mesenteric  zu  sehen,  welch   erster        g^ammkauäle  Zeigen  (Abbildung 
in  eins  der  acht  Hauptkanale  sich  tortsetzt.  o         v  o 

X20.  1),  darf  nur  entweder  durch  die 

Annahme    Stüder  s  dass  die  Axialpolypen  rein  vegetative  Indivi- 
duen darstellen,   oder  durch  eine  andere  Annahme  dass  der  acht- 
strahlige  Bau  die  Eigenschaft  der  Zweige  selbst  ist,  aber  nicht  durch 
die  Annahme  v.  Koch's,   aufgeklärt  werden.      4.  Der  Einwand, 
welchen   Versluys    (1)     1907    gegen   die    Hypothese    Studer's 
anführte,  ist  auch  durch  die  v.  Koch' sehe  Annahme  nicht  leichter 
zu   beseitigen,   als  durch   diejenige   von    Stüder.      Der   hier   in 
Betracht  genommene  Einwand  lautet  nämlich  folgendermassen: 
"Auch  stehen  bei  den  Telestidae  die  Magenhöhlen  der  Seiten- 
polypen   immer    nur    mittels    Solenia    in    Verbindung    mit   der 
Magenhöhle  der   Mutterpolypen,   aus  denen  sie  sich   entwickelt 
haben.     Bei  den  Gorgoniden   dagegen  können  die  Magenhöhlen 
der  Polypen  unmittelbar  übergehen  in   die  geräumigen  Rinden- 
kanäle,    welche    nach    Stüder    den    jMagenhöhlen    der   Mutter- 
polypen der  Telestidae  entsprechen.     Diese   deutet   aber   darauf 
hin,  dass  die  Rindenkanäle  der  Gorgoniden  selbst  Solenia  sind." 

Während  die  Annahme  v.  Koch's,  ungeachtet  dessen  dass 
sie  auf  die  direkte  Beobachtung  basiert  sein  dürfte,  doch  kaum 
hinreicht,  die  angegebenen  anatomischen  Verhältnisse  zu  erklären, 


Morphologie  u.  Stammesgeschichte  der  Gorgoniden. 


23 


giebt  das  Resultat  meiner  Untersuchung  über  die  Entwicklung 
von  Anthoplcxaura  dimorpha,  wie  es  mir  scheint,  genügend  klare 
Hinweisung  zur  Beseitigung  der  angeführten  Einwände. 


Abbildung  2. 
Anthoj>le.vaura  diinorplia  Kükenthal. 

Rekonstruierte  Abbildung  von  einer  Schnittserie  einer  jungen 
Kolonie.  Ein  Primärpolyp  mit  7,wei  auf  Stolonen  gebildeten  Sekundär- 
polvpen  ;  die  distale  Hälfte  der  Polypen  nicht  gez?ichnet;  Achsenskelet 
schraffiert.      X30. 


Die  Skeletachse,  welche,  wie  auch  bei  der  v.  Koch' sehen 
Untersuchung,  in  der  Basis  des  Primärpolypen  exzentrisch  liegt, 
ragt  nicht  ins  Coelenteron  desselben  hinein,  sondern  bleibt 
immer  unterhalb  der  Fussscheibe,  indem  dabei  ihre  Umgebung 
mehr  vaskulär  wird,  je  mehr  sie  in  die  Höhe  wächst.  Die  Spitze 
derselben  also  bleibt  immer  ausserhalb  der  Polypenhöhle  und 
richtet  sich  schief  dazu  (Abbildung  2).  Obgleich  das  obige  Ver- 
hältnis bei  meiner  Untersuchung  nur  durch  ein  einziges  Exemplar 
dargestellt  worden  war,  liegt  doch  kein  Grund  vor,  dieses  Ver- 
hältnis als  anormales  anzusehen,  denn  es  fehlte  nicht  an  Fällen 
der  Skeletachse,  welche,  wenn  auch  in  jügeren  Ausbildungsgraden, 
in  Zustimmung  mit  jener  Beobachtung  standen. 

Durch  die  Annahme,  dass  die  Skeletachse  immer  ausserhalb 
des  Primärpolypen  bleibt  indem  die  Stämme  und  Zweige  beson- 
ders ausgebildetes  Coenenchym  darstellen,  dürfen  die  bisher 
angeführten  Einwände  gar  leicht  entfernt  werden.  Die  Tatsache, 
dass  die  Hauptkanäle  in  Zweigen  an  der  Zahl  nach  den  Spezies 
bestimmt   sind,    möchte   ich   vor   allem   auf  die  Eigenschaft  der 


24  ^-  îfinoshita  :  Beiträge  zur  Kenntnis  der 

Zweige  zurückführen.  Dahin  weist  die  Tatsache,  class  bei  Primnoa 
pacißca  in  den  mehr  apikalen  Abschnitten  der  Zweige  als  bei  dem 
obersten  Polypen  auch  acht  Kanäle  zu  finden  sind. 

Schlussbemerkung  zu  den  Gorgonacea. 

In  den  vorgehenden  Paragraphen  habe  ich  dargetan,  erstens 
dass  die  Skeletachse  der  Gorgonien  nicht,  wie  Kölliker  (1)  und 
Studer(1,2)  es  behaupten,  im  Innern  des  Coenenchyms.  entsteht, 
sondern  wie  v.  Koch  (2,  3,  6)  angibt,  ein  Ausscheidungsprodukt 
des  Achsenepithels  ist  ;  und  zweitens,  dass  die  Polypen  sämmtlich 
gleichw^ertig  sind,  und  unter  sich  nicht  Differenz  in  Länge  oder 
in  Ordnung  zeigen,  sodass  die  Stämme  und  Zweige,  wie  v.  Koch 
(6)  und  Studer  (2)  es  schilderten,  durch  die  Umbildung  der 
Axialpolypen  gebildet  worden  sind.  Diese  zwei  Resultate  künden 
offenbar  dass  der  Aufbau  der  Gorgonienkolonie  nicht,  weder  wie 
Stüder  (2)  und  Kükenthal  (1 ,  2)  meinen  noch  wie  es  aus  den 
Schildeiiingen  v.  Koch' s  (6)  hervorgeht,  vom  Telesto-Typus  ist, 
und  also  auch  dass  man  die  Gattung  Telesto  nicht  als  die  Ahnenform 
betrachten  kann,  aus  welcher  die  Gorgonien  entsprungen  sind. 

Sieht  man  nun  in  der  Gorgonienkolonie  von  der  Skeletachse 
ab,  welche  das  Sekret  des  Ektoderms  ist  und  also  zur 
Organisation  des  Tierkörpers  von  unwesentlicher  Bedeutung  ist, 
so  darf  man  die  Gorgonienkolonie  direkt  mit  den  auf  dem 
Boden  flach  kriechenden  Kolonien  vergleichen.  Wenn  man 
jedoch  noch  einsieht  dass  die  Rinde  der  niederen  Gorgonien  meist 
ein  einfaches  Kanalsystem  besitzt,  so  wird  man  sich  gerechtfertigt 
finden,  die  Gorgonien,  wie  v.  Koch  dieselben  von  Rhizoxenia 
abzuleiten  versuchte,  von  den  Stolonifera  hergestammt  zu 
denken. 

III.     Scleraxonia. 

Wie  in  den  vorigen  Kapiteln  angeführt,  ist  die  Gruppe 
der  echten  Gorgoniden  von  v.  Koch  (3,  6)  unwiderlegbar 
nachgewiesen    als      einem    phyletisch     besonderen    Stamm    der 


Morphologie  u.  Stammofgescliichto  der  Gorgoniden.  25 

Entwicklung  angehörend,  als  die  Gruppe  der  Scleraxonia 
oder  Pseudaxonia.  Studer  (2)  stimmt  auch  darin  mit  v.  Koch 
überein,  dass  er  die  Gorgoniden  als  diphyletischen  Ursprungs 
ansieht. 

Ueber  die  Herkunft  der  Gruppe  der  Scleraxonia  scheint  nun 
KöLLiKER  (1 ,  2)  der  Meinung  gewesen  zu  sein,  dass  die  Paragorgia- 
ceae  mit  den  Sympodiidae,  welche  er  auf  den  Gattungen  Sijmpo- 
diiim  und  Erijtlirojmdiuiii  errichtete,  in  direkter  Verwandtschafts- 
beziehung standen,  da  er  diese  beiden  Gruppen  in  eine  Unter- 
familie, Briareaceae,  zusammenstellte.  In  einer  späteren  Abhand- 
lung aber  äusserte  er  seine  neue  Ansicht  über  die  Ableitung  der 
Scler-axonia.  Er  hielt  nämlich  die  Siphonogorgia  für  die  Zwischen- 
form zwischen  den  Alc^^oniden  und  den  Paragorgiaceae,  was 
sich  auf  die  Gründe  stützt,  dass  die  Siphonogorgia  im  äusseren 
Habitus  den  Gorgoniden  sehr  stark  ähnelt,  und  dass  die  Mesen- 
terien bei  dieser  Gattung  in  den  Stammkanälen,  welche  die 
proximalen  Abschnitte  der  Polypenhöhlen  darstellen,  bis  zu 
vier  reduziert  sind,  während  sie  bei  den  Alcyoniden  in  Achtzahl 
erhalten  bleiben. 

Diese  Ansicht  aber  bedarf  noch  vielfacher  Beweisführungen, 
da  zwischen  der  Siphonogorgia  und  den  Briareiden  eine  zu  weite 
Lücke  vorhanden  ist,  während  darin,  dass  die  Siphonogorgia  sich 
aus  den  Nephthyiden  entwickelt  haben,  kein  Zweifel  mehr  zu 
setzen  ist. 

Studer  (2)  machte  1887  einen  Versuch,  die  Scleraxonia  von 
einer  kriechenden  Ahnenform  wie  SympOilium  und  Erythro- 
podium  durch  die  Erhebung  des  Coenenchymrandes  und  durch  die 
Einrollung  und  Solidifizierung  der  erhobenen  Abschnitte  abzu- 
leiten. Diese  Ansicht,  welcher  Kükenthal  (1 ,  2)  sich  anschloss, 
ist  von  derjenigen  von  Kölliker  (3)  gründlich  verschieden. 
Wenn  man  nämlich  bei  den  kriechenden  Stammformen  wie 
Sijmpodium  oder  Erythropodiwu,  von  welchen  die  Skleraxo- 
nier  allerdings  direkt  oder  indirekt  hergestammt  vsind,  die 
Hauptachse  des  Coenenchyms  in  vertikaler  Richtung  auf 
dem  horizontal  ausgebreiteten  Coenenchym  annimmt,  so  stimmt 
diese    Hauptachse,    nach    der  Auffassung  Kölliker' s,     mit    der 


26  ^'  Kinoshita  :  Beiträge  zur  Kenntnis  der 

Starnmachse    überein,     während     sie    nach    dem     Studer' sehen 
Bauplan  mit  der  letzteren  senkrecht  zusammenstösst. 

Die  Hypothese  von  Studer  basiert  sich  nur  auf  die  Tatsache, 
dass  die  Gattung  Solenocaidon  neben  den  kompakten,  auch  noch 
rinnen-  sowie  röhrenförmige  Zweige  aufweist.  Obgleich  Küken- 
thal (2,  3)  von  selten  des  Erythrojjodium  zu  dieser  Hypothese  eine 
weitere  Stütze  liefert,  doch  ist  sie  nicht  endgültig  bewiesen, 
sodass  die  folgenden  Einwände  dadurch  noch  kaum  überwindbar 
erscheinen. 

HiCKSON  (2)  gibt  nämlich  an,  dass  es  sich  bei  den  rinnen-  oder 
röhrenförmigen  Zweigen  der  Gattung  Solcnocaulon  um  eine 
pathologische  Umformung  handele,  welche  durch  parasitische 
Krebstiere  von  der  Gattung  Alphciis  verursacht  würde,  und  auch 
dass  solche  Formen  kaum  als  die  Grundform  der  höheren  Skler- 
axonier  angenommen  werden  können. 

Die  Frage,  ob  diese  Umformung  der  Zweige  eine  eigen- 
tümliche Charakteristik  von  Solenocaidon  odei*  eine  pathologi- 
sche Erscheinung  ist,  kann  nur  durch  entweder  eine  genauei'e 
anatomische  oder  eine  expérimentale  Untersuchung  auf  leben- 
den Exemplaren  entschieden  werden.  Wenn  man  jedoch 
das  folgende  Verhältnis  in  Betracht  zieht,  so  wird  man  nicht 
im  geringsten  zögern,  diese  Umformung  mit  Hickson  auf 
eine  pathologische  Erscheinung  zurückzuführen,  welche  für  die 
Kolonienbildung  von  keiner  Bedeutung  wäre. 

Die  Zweigneubildung  in  Solenocaidon,  wenn  diese  Form 
wirklich  nach  dem  Studer' sehen  Bauplan  aufgebaut  ist,  kann  nur 
an  den  Zweigspitzen  oder  an  den  Rändern  der  flachen  rinnenför- 
migen  Zweige  stattfinden,  da  nach  ihm  das  Wachstum  der  Zweige 
die  Ausdehnung  des  Coenenchyms  an  den  Randsäu- 
men bedeuten  und  diese  letzteren  bei  Solenocaidon  nur  an  den 
gegebenen  Stellen  anzutreffen  sind.  Die  Aeste  von  Solenocaidon 
tortosum  und  cervicorne,  wie  sie  von  Jaxower  (1,  PI.  VIL,  Fig.  1, 
PL  VIIL,  Fig.  4,  5)  dargestellt  wurden,  scheinen  dieses  Verhältnis 
wahrscheinlich  zu  machen.  Aber  in  den  Abbildungen  von  den 
Solenocaidon- Arten,  welche  in  den  Schriften  von  Germanos  (1), 
und  besonders  von   Nuttixg    (1,   PI.  L,   Fig.    1)    gegeben    sind, 


Morphologie  u.  Stainmesgoscliichte  der  Gorgoniden.  27 

findet  man  Aeste,  welche  zweifellos  durch  Knospung  gebildet 
worden  zu  sein  scheinen.  Da  man  denn  unter  Knospung  allgemein 
die  Verdickung  eines  Coenenchymabschnittes  in  der 
senkrechten  Richtung  zur  Oberfläche  versteht,  so  ist  zwi- 
schen den  durch  die  marginale  Extension  des  Coenenchyms 
gebildeten  und  den  durch  Knospung  entstandenen  Zweigen  ein 
gründlicher  Unterschied  zu  erwarten.  Es  ist  nun  kaum  denkbar, 
dass  in  einer  selben  Kolonie  nach  zwei  solcherweise  gründlich 
verschiedenen  Bauplänen  gebildete  Stämme  und  Zweige  vorkom- 
men. Aus  diesem  Widerspruche  schliesst  man  wohl  mit  Recht, 
dass  die  Kolonienbildung  bei  Solenocaidon  nur  durch  die  Verdickung 
des  Coenenchyms  in  der  Richtung  der  Hauptachse  vor  sich  geht, 
und  dass  die  gegebenen  Umformungen  der  Zweige  ganz  auf  eine 
pathologische  Erscheinung  zurückführbar  sind. 

Was  nun  die  Resultate  meiner  Untersuchung  anbelangt,  so 
scheinen  die  Skleraxonier,  wie  schon  angegeben,  durch  die  Ver- 
dickung des  Coenenchyms  in  ihrer  Hauptachse  von  einer  kriechen- 
den Stammform  wie  ErytJiropodium  ableitbar  zu  sein,  was  ganz 
anders  zutrifft  als  die  Hypothese  von  Siuder,  aber  mit  derjenigen 
von  KöLLiKER  sehr  viel  Uebereinstimmung  zeigt.  Meine  Befunde, 
dass  bei  den  Gattungen  Briarewn.  und  Paragorgia  die  Terminal- 
polypen an  ihrer  Basis  je  zu  einem  Längskanal  führen,  welcher 
durch  den  Zentralstrang  der  Skeletachse  hindurch  bis  zum  Aus- 
gangspunkt der  betreffenden  Zweige  reicht,  scheinen  uns  gewiss  zu 
der  Auffassung  zu  führen,  dass  die  Stämme  und  Zweige  dieser 
Formen,  ähnlich  wie  bei  der  Gattung  Siplionogorgia,  je  eine 
Syndete^-*  darstellen.  Die  Lücke  zwischen  der  Gattung  Sipliono- 
gorgia  und  diesen  Formen,  ist  jedoch  immer  noch  zu  weit,  um 
die  erstere  als  eine  Ahnenform  der  letzteren  anzunehmen.  Es 
wird  nun  noch  beim  jetztigem  Stande  der  Kenntnis  gerechtfertigt 
sein,  die  Skleraxonier  aus  den  Erythroiioclium-i\hi\\iQ\\e\\  niederen 
Alcyonaceen,  parallel  mit  den  höheren  Formen  der  letzteren, 
abgestammt  zu  denken. 

In  dem  folgenden  gebe  ich  nun  die  Beschreibung  derjenigen 

1)      Siehe  Bourne  (1). 


28  ^-  Kinoshita  :  Beiträge  zur  Kenntnis  der 

anatomischen  Verhältnisse  einiger  Skleraxonier,  welche  haupt- 
sächlich den  theoretischen  Betrachtungen  über  ihre  Stellung  im 
Systeme  wichtig  zu  sein  scheinen. 

Briareidae. 

Diese  Famihe  umfasst  die  niedersten  Formen  der  Scler- 
axonier,  bei  denen  die  Differenzierung  der  Rinden-  und  Ach- 
senteile in  einem  noch  sehr  primitiven  Zustande  verharrt.  Die 
Mitglieder  dieser  Familie  werden  darnach  in  zwei  Unterfamilien 
eingeteilt,  ob  der  Achsenabschnitt  von  Solenia  durchzogen  ist 
(Briareinae),  oder  derselben  entbehrt  (Spongioderminae). 

Von  dieser  Familie  habe  ich  nur  zwei  Gattungen  Briareum 
und  Paragorgia,  welche  beide  der  Unterfamilie  Briareinae 
angehören,  untersucht.  Leider  habe  ich  jedoch  keine  Formen  aus 
der  Unterfamilie  Spongioderminae  untersuchen  können. 

Briareum   B L AIK VILLE.  ^-^ 

Der  Typus  dieser  Gattung  ist  Gorgonia  hriareiis  Ellis  & 
Solander.  Studer  (2)  gab  1887  dieser  Gattung  eine  erneuerte 
Charakterisierung,  welche  folgendermassen  lautet: 

'^  Bei  Briareum  endlich,  das  unregelmässig  lappige,  aufrechte 
Colonie  bildet,  ist  die  von  Kanälen  durchzogene  Axe  wenig 
begrenzt,  die  regelrecht  am  Stamme  verteilten  Polypen  sind  ohne 
Kelche  und  ganz  in  das  Coenenchyma  zurückziehbar." 

Wie  ein  Vergleich  mit  dem  Originale  von  Blainville  (1) 
zeigt,  deckt  diese  Diagnose  nicht  den  Typus  von  Ellis  und  So- 
lander. Es  ist  auch  nicht  sicher  ob  Briareum  hriareum.  eine  von 
den  Solenia  durchzogene  Skeletachse  besitzt,  was  im  Sinne 
von  Studer  als  die  Haupteigentümlichkeit  der  Gattung  Bria- 
reum  gilt.  Daneben  scheint  die  Differenzierung  der  Skelet- 
achse bei  B.  hriareum,  B.  suherosum  und  B.  ßelei,  von  welchen 
die  letztere  bis  jetzt  die  einzige  genau  beschriebene  Art  ist, 
weiter    gerückt    zu  sein,   als  Studer  in  seiner   Diagnose   angab. 

1)  Briareum  Blainville,  183i  ;  Dana,  1846  ;  Milne-Edwakds,  1865;  Briarea  Düchassaing 
et  MiCHELOTTi,  1860  ;  Briareum  Kölliker,  1866  ;  Studer,  1S87  :Wright  &  Studer,  1889. 


Morphologie  u.  Stammesgeschichte  der  Gorgoniden. 


29 


Abbildung  3. 
Briareiim    asbesthim    (Pallas) 
Aeusserer  Habitus.     Nat.  Gr. 


Wenn  man  diese  Tatsachen  in 
Betracht  zieht,  dürfte  man  wohl 
sagen,  dass  diese  Diagnose  alleinig 
auf  B.  ashestinum  (Pallas)  basiert 
ist. 

Von  dieser  Gattung  nun  steht 
mir  ein  Exemplar,  welches  aus 
den  Bahamas  stammt  und  den 
Namen  Briar eum  ashestinum  trägt, 
zur  Verfügung.  Die  Richtigkeit 
der  Bestimmung  dieses  Exem- 
plares  ist  mir  nicht  positiv 
sicher,  da  die  Beschreibungen  der 
Autoren  alle  dazu  kaum  hinrei- 
chend sind.  Es  liegt  jedoch  kein 
genügender  Grund  vor,  jene  Be- 
stimmung zu  verwerfen. 

Briar  eum  ashestinum  (PallAs). 

Das  vorliegende  Exemplar  weist 
zwei  verschieden  grosse  Stämme  auf, 
welche  von  einer  gemeinsamen 
Basalausbreitung  emporsteigen.  Sie 
sind  einfach,  fingerförmig,  im  Quer- 
schnitte gerundet,  zeigen  jedoch 
einige  leichte  Anschwellungen,  von 
welchen  die  eine,  die  im  Drittel 
vom  Apex  beim  grösseren  Stamme 
auftritt,  offenbar  als  ein  hervortre- 
tender Ast  anzusehen  ist. 

Die  monomorphischen  Polypen 
sind  auf  die  ganze  Oberfläche  der 
Kolonie  uniform  dicht  verteilt.  Ihr 
vorstülpbarer  Abschnitt  entbehrt 
durchaus    der   Skleritenbewehrung, 


30 


K.  Kinoshita  :  Beiträge  zur  Kenntnis  der 


\ll 


und  im  zurückgezogenen  Zustande  werden  da  auf  der  glatten 
Coenenchymoberfläche  kleine  gerundete  oder  längsspaltförmige 
Oeffnungen  gelassen,  weil  eben  da  die  Kelche  gar  nicht  oder 
ganz  rudimentär  entwickelt  sind. 

Die  Magenhöhlen  sind,  von  dem  Niveau  der  Rindenoberfläche 
gemessen,  3  Mm.  tief,  und  sind  etwas  schräg  nach  unten  gerichtet. 
Sie  enden  nicht  am  Boden  blind  abgerundet,  sondern  führen  je  zu 
einem  schmalen  Kanal,  welcher,  bald  sich  verschmälernd,  gerade 
hinunter  läuft  und  sich  in  das  Netzwerk  der  Solenia  verschwindet. 

Dieses  Verhalten  ist  bei  dem  am  Stammapex 
befindlichen  Polypen  viel  stärker  betont.   Die 
Magenhöhlen  bei  diesen  Pol^^pen  sind  näm- 
lich viel  tiefer  als   bei   den    Lateralpolypen. 
^^^^m\M  A  \       und  gehen  in  den  Kanalabschnitt,  der  den 
^^^«    I  1^^     Stamm  hinunter  durchläuft,   ganz  allmälich 
j^^M    11    ^\    über  (Abbildung  4). 

j^^l  11  W^  Der    Stamm    oder    das     Coenenchym, 

welcher  durch  seinen  plumpen  Bau  ausge- 
zeichnet ist,  weist  in  sich  zwei  Abschnitte, 
den  Rinden-  und  den  Achsenabschnitt,  auf, 
welche  beide  auch  wieder  in  zwei  Schichten 
zerfallen  (Abbildung  5).     Es  sind  nämlich  : 

Î^.    ^       1     1     .    CAeussere Rinde 
RmdenabschnittX^.  „       ^.    . 
(iieiere  Kmde 
AchsenabschnittH'^^^^^^^^^^S® 

(Zentralstrang. 

Die  Differenzierung  in  diese  vier  Abschnitte  bleibt  aber  in 
einem  sehr  primitiven  Zustande.  Die  Grenzen  dieser  Schichten 
sind  nämlich  nicht  ganz  deutlich  und  die  Skleriten  stimmen, 
zwar  nicht  in  der  Farbe  und  Grösse,  doch  in  der  Form  und 
Skulptur  beinahe  überein. 

Aeussere  Pdndenschiclit. — Diese  die  ganze  Kolonie  ununter- 
brochen überziehende  Schicht  beträgt  durchschnittlich  0.5  Mm. 
in  der  Dicke  und  ist  von  einem  Netzwerke  feiner  Solenia  durch- 
zogen.     Die    Skleriten,     welche    diese    Schicht    aufbauen,    sind 


/' 


Abbildung  4. 
Briarem  ashestinum  (Pallas). 
Medianer  Längsschnitt  des 


Mori)hologie  u.  Stauimesgeschichte  der  Gorgoniden. 


31 


Abbildung    5.    Briareum    ayhcstinum    (Pallas). 
Querschnitt  des    Stammes.     Polypenhöhlen,    Lüngskaniile    und 
Solenia  schwarz  gemalt  ;  gefärbte  Skleriten  schraffiert  ;   a   iiussere, 
b  tiefere  Kindenschicht,  c  Kindenlage  der  Achse,  d  Zentralstrang  der 
AchsL^     X  20. 


32  ^'  Kinoshita  :  Beiträge  zur  Kenntnis  der 

klein  und  in  der  Mehrheit  tief  jDurpurn  gefärbt.  Sie  sind 
parallel  der  Rindenoberfläche,  sonst  unregelmässig  gelagert. 
Ihre  Grösse  nimmt  in  der  Regel  je  nach  der  Tiefe  zu. 

Tiefere  Piindenschiclit . — Diese  Schicht  ist  dick  und  durch 
dünne  sowie  mittelweite  Solenia  durchzogen.  Die  Polypen- 
höhlen erreichen  beinahe  die  innere  Grenze  der  Schicht. 
Die  Skleriten  sind  gross  und  meist  farblos,  selten  leicht 
purpur.  Sie  sind  den  Wandungen  der  Polypenhöhlen  parallel, 
also  in  der  Mehrheit  senkrecht  zur  Stammachse  gelagert.  In 
diesem  Falle  auch  nimmt  ihre  Grösse  durchschnittlich  von  aussen 
nach  innen  zu. 

Die  beiden  Rindenschichten  jedoch  sind  am  Apex  des 
Stammes  wenig  differenziert,  und  die  dort  tief  gelagerten  Skleriten 
sind  auch  tief  purpurn  gefärbt. 

Rindenlage  der  Achse. — Dieser  Abschnitt  der  Achse  ist  von 
den  grossen  Skleriten,  welche  in  harten  Längsbündeln  ge- 
lagert sind,  aufgebaut.  Er  enthält  nicht  das  Netzwerk  der 
Solenia,  sondern  eine  Anzahl  einfach  längs  verlaufender 
Längskanäle,  welche  distalwärts  in  die  tiefere  Rindenschicht 
hineintreten. 

Zentralstrang  der  Achse. — Das  Zentrum  der  Achse  ist  durchaus 
von  Skleriten  anderer  Art  besetzt.  Wie  in  der  Rindenlage 
greifen  dieselben  aneinander  und  so  bilden  die  harten  Längs- 
bündel, aber  zeichnen  sie  sich  dadurch  aus,  dass  sie  viel 
kleiner  und  in  der  Mehrheit  tief  purpurn  gefärbt  sind.  Die 
Längskanäle,  welche  auch  diesen  Abschnitt  durchlaufen,  besitzen 
meist  weiteren  Kaliber  wie  diejenigen  der  Rinde.  Verfolgt  man 
nun  diesen  Abschnitt  nach  den  Apex,  so  findet  man,  dass  er 
dort  zur  Rindenschicht  übergeht,  indem  die  Skleritenbündel  loser 
werden.  Dass  die  Längskanäle  je  zu  einer  Polypenhöhle  sich 
fortsetzen,  habe  ich  schon  angegeben. 

rtiragcrgia  Milne-EdwARDS. 

Von  dieser  Gattung  stehen  mir  einige  Exemplare  zur  Ver- 


■Morphologie  u.  St  inimesgesohichte  der  Gorgoniden. 


33 


>: 


fügung,  Avelche  aus  der  Sagami-Bai  stammen,  und  in  drei  oder 
vier  Arten  oder  Varietäten  teilbar  zu  sein  scheinen.  Von  diesen 
Formen  scheint  die  eine  zu  Pamgorgia  arhorea  (L.),  die  zweite 
zu  Pamgorgia  nodosa  Korejst  &  Danielssen  einbezogen  werden 
zu  dürfen,  während  die  anderen  neu  zu  sein  scheinen.  Die 
eine  von  diesen  möchte  ich  hier  provisorisch  P.  granulosa,  die 
andere  P.  tenuis  nennen. 

Ueber  die  Struktur  der  Kolonie  bemerke  ich  folgendes. 

Siphonozooiden. — Stüder  (2)  bemerkt  in  seiner  Diagnose 
der  Gattung,  dass  '  '  neben  den  Polypen  bei  Paragorgia  nodosa 
auch  noch  Siphonozooiden  ohne 
Tentakel  vorkommen."  Es  könnte 
dieser  Satz  zur  Annahme  misleiten, 
dass  Paragorgia  arhorea  (L),  die 
Typenart  der  Gattung  und  die  einzige 
von  ihm  anerkannte  Art  neben  P. 
nodosa,  der  Siphonozooiden  entbehrt. 
Dieses  Misverständnis  beruht  wahr- 
scheinlich darauf  dass  die  letzteren  bei 
P.  arhorea  nicht  über  die  Rindenober- 
fläche hervorragen.  Das  weist  deut- 
lich die  Abbildung  (PI.  IX.,  fig.  5) 
eines    Schnittes   von    P.    arhorea  auf, 

welche     Koren     und     DanIELSSEN    (1)     zahlreichen  siphonozooiden.  X  6 

zum  Vergleich  mit  ihrer  Art  gaben.  Dasselbe  ist  auch  der  Fall 
bei  meinen  Exemplaren,  P.  afïînis  arhorea  (Abbildung  6)  und 
P.  tenuis.  Also  die  Tatsache  dass  die  Polypen  bei  allen  Arten 
dieser  Gattung,  soweit  bekannt,  dimorphisch  sind,  scheint  nicht 
mehr  zu  bezweifeln. 

Autozooiden. — Milne-Edwards  (1)  schon  bemerkte  dass  die 
Polypen  bei  Paragorgia  Magenhöhlen  besitzen,  welche  kurz  und 
blind  terminieren.  Dies  bezieht  sich  jedoch  nur  auf  die 
Seitenpolypen.  Die  Polypen,  welche  sich  auf  den  Zweigspitzen 
finden  und  Köpfe  bilden,  führen,  wie  wir  es  bei  Briareum 
ashestinum  gesehen  haben,  direkt  je  in  die  Längskanäle  welche 
durch  den  Zentralstrang  bis  zur  Ausgangsstelle  der  betrefïenden 


Abbildung  6. 
Paragorgia  äff.  arhorea   (L). 
Vordere    Ansicht    eines     Zweigab- 
schnittes  mit  fünf  Autozooiden    und 


il 


34 


K,  Kinoshita  :  Beitrüge  zur  Kenntnis  der 


Abbildung   7. 
Parafjorgia  temiis,  n.  sp. 
Medianer     Läng^chnitt 
eines  Zweigapex.   X  5. 


Zweige  hinunterlaufen  (Abbildung  7).  Hierbei  auch  bleiben  die 

Mesenterien  kurz  uud  gehen  nicht  in  die  Kanalabschnitte  ein. 

Das  Coenenchym  ist  auch  hier  wie  in 
Briareuin  ashestimim  in  vier  Schichten  diffe- 
renziert, was  aber  in  viel  deutlicher  Weise 
vor  sich  geht  (Abbildung  8). 

Aeiissere  Fàndensch'clit. — Diese,  die  ganze 
Kolonie,  sogar  die  Polypenkelche,  über- 
ziehende äussere  Schicht  ist  kaum  0.15  Mm. 
dick.  Sie  entbehrt  in  sich  gänzlich  der 
Solenia.  Diese,  welche  von  geringerem 
Kaliber  sind  und  im  ganzen  der  Länge  nach 
verlaufen,  bilden  ein  einfaches  Netzwerk 
und  trennen  die  in  Rede  stehende  Schicht 
von    der    tieferen  Rindenschicht   ab.      Die 

Skleriten,   welche  dieser  Schicht  eigentüm- 

hch    sind,    sind    kleine   Achter'^   von    roter 

oder  gelbroter  Farbe. 

Tiefere     Rinde  nschiclit.  —  Diese     Schicht 

weicht  in  der  Struktur  und  Spikulation  von 

der  letzteren  bedeutend  ab.     In  ihr  ist  die 

proximale  Hälfte    der   Polypenhöhlen    auf- 
genommen.    Darum  hangt  die  Ausbildung 

der  Schicht  sehr  viel  von  der  Verteilung  der 

Polypen  ab.      Sie  ist  nämlich  in  der  Proxi- 

mität  der  Polypen,   also  an  der  Vorderseite 

.  -  T  rr        '  • .  ^  Querscîhnitt  eines  Zweiges. 

der  Kolonie  und  an  den  Zweigspitzen,   sehr    zwei     Autozooiden      und 

,.,,.         ^nT  i'iij.  "1  1        •         ïtint  Siphonozooiden    gerade 

dick-bis    2  Mm. -entwickelt,     wahrend    sie 

sich  in  den  polypenlosen  Stellen  sehr  wenig 

ausbildet,     sodass    oft    die    Solenia   gerade 

unterhalb      der      äusseren      Rindenschicht 

beinahe  den  Achsenabschnitt  zu  berühren   scheinen.      Von  den 

vorliegenden  Exemplaren  ist  bei  P.  granulosa  diese  Rindenschicht 

sehr  gering  entwickelt,   infolge  dessen  die  Polypen  hoch  konisch 


■M 


Abbildung  8. 
ParagOTßia  äff.  arborea  (L.). 


getroffen ,"  Polypenhöhlen, 
Längskanäle  und  Solenia 
schwarzangegeben  ;  Areae,  wo 
die  gefärbten  Skleriten  sich 
befinden,  jjunktiert.   X  6. 


1)  KöLLiKER  (1),  Seite  122. 


Morphologie  u.  Stammesgesjhichte  der  Gorgoniden.  35 

hervorragen,  während  dieselben  bei  den  andern  Exemplaren  meist 
niedrig  bis  beinahe  flach  bleiben. 

Diese  Rindenschicht  ist  von  den  weiten,  Netzwerk  bildenden 
Solenia  durchzogen.  Die  Skleriten  lassen  sich  in  zwei  Typen 
unterscheiden:  1)  Diejenigen  Skleriten,  welche  von  den  der 
äusseren  Rindenschicht  eigentümlichen  Achtern  durch  Weiter- 
entwicklung der  Achsensprossen  abzuleiten  sind.  Obgleich  sie 
sehr  oft  beträchtlich  verlängert  sein  können,  sind  sie  doch  an 
ihren  medialen  sechs  Warzen,  welche  am  Scheitel  pilzhutförmig 
ausgebreitet  sind,  erkennbar.  2)  Diejenigen  Skleriten,  welche 
typisch  spindelförmig,  mit  zerstreuten  Stacheln  besetzt  und  an 
den  beiden  Enden  zugespitzt  sind.  Oft  jedoch  zeigen  die  Stacheln 
eine  Andeutung  von  mehr  oder  weniger  regelmässigen  Gürteln. 
Was  die  Färbung  dieser  Skleriten  betrifft,  so  sind  sie  farblos  oder 
schwach  rötlich. 

Von  diesen  beiden  Arten  der  Skleriten  ist  nur  die  erste  als 
die  eigentlichen  Rindenskleriten  anzusehen.  Die  zweite  betrachte 
ich  als  die  eigentlich  der  Rindenlage  des  Achsenabschnitts 
angehörige,  da  jene  Lage  für  die  grössten  Teile  durch  die  Skleriten 
dieses  Typus  gebildet  ist. 

Pdndenlage  des  Achsenahschiütes. — Diese  Schicht  stellt  wie 
bei  Briareuni  ashestiiuim  den  eigentlichen  Achsenabschnitt  dar. 
Sie  besteht  aus  den  in  festen  Längsbündeln  gelagerten  Skleriten, 
welche  in  der  grossen  Mehrheit  farblos,  sonst  sehr  schwach 
rötlich  gefärbt  sind,  und  denjenigen  des  zweiten  Typus  in 
der  tieferen  Rindenschicht  ganz  gleichen.  Neben  diesen  Skleriten 
kommen  auch  diejenigen  Skleriten,  welche  ich  ebenda  als  die 
der  tieferen  Rindenschicht  eingentümlichen  angegeben  habe, 
vereinzelt  oder  meist  in  konzentrischen  unregelmässigen  Grup- 
pen vor  (Abbildung  8).  Diese  Skleriten  sind  mit  Recht  als  solche 
anzunehmen,  welche  sich  an  der  Peripherie  dieses  Abschnittes 
zufällig  hinzugefügt  haben.  Diese  Eindringung  geht  bei  P.  granu- 
losa auch  noch  rege  vor  sich,  sodass  die  Achter,  welche  der 
äusseren  Rinde  angehören,  auch  manchmal  in  diesem  Abschnitte 
angetroffen  werden.  Das  Verhältnis,  dass  die  äussere  Rinde  bei 
dieser  Art  wegen  der  schwachen  Ausbildung  der  tieferen  Rinden- 


Qg  K.  Kinoshita  :  Beiträge  zur  Kenntnis  der 

Schicht   beinahe    die  Achsen  berührt,   wird  diese  Tatsache  leicht 
verständlich  machen. 

Dieser  Achsenabschnitt  ist  von  den  rnittelweiten  Längska- 
nälen, welche  vereinzelt  oder  meist  in  mehreren  konzentrischen 
Reihen  sich  anordnen,  dm'chzogen. 

Zentrahtrang. — Der  Zentralstrang  zieht  sich  inmitten  des 
Achsenabschnittes  der  Stämme  mid  Zweige  hindurch.  Es  ist 
jedoch  an  der  Basis  jedes  Zweiges  von  dem  Zentralstrang  des 
Stammes  oder  des  Zweiges  höherer  Ordnung  durch  die  Rinden- 
lage getrennt,  wie  es  durch  die  Wandung  der  Axialpolypen  bei 
Teksto-Kolonie  mit  Axialliöhlen  geschieht,  da  bei  der  Zweig- 
bildung nicht  die  Gabelung  wie  bei  den  Melitodiden,  sondern 
allein  Knospung  stattfindet. 

Der  Zentralstrang  weist  eine  ganz  gleiche  Struktur  wie  die 
Rindenlage  auf.  Der  Unterschied  liegt  nur  darin,  dass  die  Skleri- 
ten  Rindenskleriten  sind,  und  dass  die  Längskanäle,  welche 
einen  weiteren  Kaliber  besitzen,  dichter  gedrängt  vorkommen. 
In  jeder  Schnittfläche  quer  durch  die  Zweige  ist  der  Zentralstrang 
an  den  gewöhnlich  als  '  'Hauptkanäle'  '  bezeichneten  Längskanälen 
und  der  roten  Area,  welche  sich  von  der  Umgebung  deutlich 
unterscheiden  lässt,  zu  erkennen  (Abbildung  8).  An  dem  Zweig- 
apex, Avo  eine  Anzahl  Polypen  den  Kopf  bilden,  geht  der 
Zentralstrang  ganz  vollkommen  zur  Rindenschicht  über.  Die 
Rindenlage  oder  der  eigentliche  Achsenabschnitt  kommt  erst 
unterhalb  des  Kopfes  in  der  Form  einer  Röhre  zum  Vorschein. 

Bemerliungcn  über  die  Gattungen  Briarcum  und  Paragorgia. 

Wie  es  in  den  vorhergegangenen  Beschreibungen  ersichtlich 
sein  dürfte,  stimmen  Briarewn  und  Faragorgia  im  wesentli- 
chen Aufbau  der  Kolonie  mit  einander  ganz  überein,  sodass  es 
keinem  Zweifel  unterliegt,  dass  diese  beiden  Gattungen  in 
einer  sehr  innigen  Verwandtschaftsbeziehung  stehen. 

Die  Kolonien  dieser  Gattungen  zeigen  nun  unter  anderm 
zwei  Verhältnisse,  welche  besonders  beachtenswert  sind.  Nämlich, 
erstens  dass  im  Zentrum  der  Skeletachse  ein  Zentralstrang,  welcher 


Morphologie  u.  Staumiesgeschiclite  der  Gorgoniden.  37 

aus  den  Rindenskleriten  besteht  und  am  Zweigapex  in  die  Rinde 
übergeht,  vorhanden  ist  (Abbildungen  5  und  8),  und  zweitens 
dass  die  Terminalpolypen  an  ihrer  Basis  je  zu  einem  den  Zentral- 
strang durchlaufenden  Kanal,  der  bisher  nur  als  "Hauptkanal" 
bezeichnet  worden  war,  führen  (Abbildungen  4  und  7). 

In  den  Kolonien  beider  Gattungen  darf  die  Rindenlage  der 
Skeletachse  als  der  eigentliche  Achsenabschnitt  angesehen  werden, 
da  der  Zentralstrang  seiner  Struktur  nach  der  Rinde  zuzurechnen 
ist.  Die  Form  der  Skeletachse  bei  diesem  Typus  ist  dann  ein 
Hohlzylinder. 

In  der  »S'o/c'/i06"t<»/o;i- Hypothese  von  Stüdek  ist  jedoch  die 
Skeletachse  der  höheren  Skleraxonier  nur  als  "  ein  zylindrischer 
Stab"  vorgestellt,  welcher  im  Zentrum  der  Zweige  liegt.  Dies 
trifft  aber  derselben  beim  Biiareiim-Pamgonjia-Typus  durchaus 
nicht  zu.  Da  jedoch  dieselbe  in  der  Studer  sehen  Hypothese  als 
eine  solche  gedaclit  ist,  welche  sich  einst  im  Stadium  von  Solcno- 
caulon  an  der  Innenseite  der  rinnen-  oder  röhrenförmigen  Zweige 
in  derselben  Form  wie  die  Zweige  selbst  ausgebildet  haben  sollte, 
so  dürfte  man  auch  wohl  annehmen,  dass  der  als  zylindrisch 
angesehene  Stab  in  Wirklichkeit  ein  Hohlzylinder  ist.  Dann 
scheint  der  Briareiim-Pamgorgia-Ty pus  dieser  Hypothese  nicht  zu 
Avidersprechen.  Wenn  man  jedoch  die  Längskanäle  im  Innern 
des  Zentralstranges  in  Betracht  zieht,  welche  je  zu  einem 
Terminalpolypen  führen,  wird  man  kaum  den  Zentralstrang  mit 
der  Rinde  der  Innenseite  der  hohlen  Zweige  bei  Solcnocanlon 
homologisieren  können.  Diese  Kanäle,  nämlich,  möchte  ich 
doch  nicht  einfach  als  die  Solenia,  sondern  als  einen  Abschnitt 
der  Polypenhöhlen  ansehen,  obgleich  sie  niemals  Mesenterien 
enthalten.  Der  Grund  liegt  darin,  dass  bei  Paragorgia  diejenigen 
Lateralpolypen,  welche  noch  nicht  in  die  Länge  gewachsen  sind, 
keine  mit  ihnen  zusammenhängende  Kanal  Verlängerung  aufweisen, 
und  dass  bei  Briareiim  asbestinum  anderseits,  welches  phj^letisch  in 
einem  jüngeren  Stadium  als  Paragorgia  zu  stehen  scheint,  die 
Lateralpolypen  auch  basalwärts  verlängert  sein  können,  eine 
starke  Stütze  zu  liefern. 

W'ie  schon  angedeutet  ist  es  kaum  denkbar  dass  in  einer  und 


38 


K.  KinosLita  :  Btiträge  zur  Kenntnis  tier 


derselben  Kolonie  die  Stamme  und  Zweige  nach  zwei  verschie- 
denen Bautypen  aufgebaut  sind.  Aus  diesem  Grunde  können 
wir  den  Modus  der  Kolonienbildung  im  allgemeinen  durch  das 
Verhältnis,  in  welchem  die  Zweige  sich  bilden,  klar  stellen.  Die 
Zweigbildung  geht,  wie  schon  angegeben,  durch  Knospung  vor 
sich.  Bei  der  Knospung  hebt  sich  eine  Gruppe  der  Lateral- 
polypen, [deren  kurze  Magenhöhlen 
proximal  abgerundet  enden,  und  bildet 
den  Kopf,  indem  unterhalb  desselben 
ein  Stiel  allmählich  zutage  tritt.  Beim 
Erheben  bleibt  das  proximale  End  jeder 
Polypenhöhle  in  der  Form  eines  Kanales 
zurück,  da  die  Zone  des  Längenwachstums 
nicht  unterhalb  der  Polypenbasis,  sondern 
etwas  oberhalb  derselben  liegt.  Darauf 
im  Stiele  und  zwar  um  das  Bündel  dieser 
Kanäle  wird  die  eigentliche  Skeletachse  in 
der  Form  eines  Hohlzylinders  von  unten 
her  gebildet.  Bei  einer  jungen,  kaum  1 
Cm.  hohen  Kolonie  (Abbildung,  9),  welche 
wahrscheinlich  J'arcKjorgia  nodosa  angehört,  konnte  ich  auch 
konstatieren,  dass  die  Terminalpolypen  sich  gleich  wie  bei  den 
Zweigen  erwachsener  Kolonien  verhalten. 


Abbildung  9. 
Farafjorgia  äff.   nodosa 

KoB,  &  Dan. 
Aeusserer  Habitus  einer 
jungen  Kolonie.  X  6. 


Melitodidae. 


Die  Melitodidenfamilie  umfasst  deutlich  ausgezeichnete  For- 
men, deren  Stamm  und  Aeste  die  Skeletachse  aufweisen,  welche 
aus  den  durch  Verschmelzung  besonderer  Spikula  entstandenen 
harten,  und  den  die  Nodien  bildenden  weichen  Gliedern  besteht. 
Die  Mitglieder  dieser  Familie  stimmen  in  Avesentlichen  Eigen- 
schaften mit  einander  überein.  Nur  l'ansis  weicht  nicht 
unerheblich  von  den  übrigen  ab,  indem  bei  ihr  der  Zentralstrang 
beinahe  reduziert  ist  und  die  Verzweigung,  im  Gegensatz  zu  den 
übrigen  Gattungen,  nur  in  den  harten  Gliedern  stattfindet. 

Die  Anatomie  der  Melitodiden  war  schon  vor  längerer  Zeit 


Morphologie  n.  Stammesgeschichte  der  Gorgoniden.  39 

ziemlich  klar  gestellt  worden.  Kölltker  (1)  1866  wies  nämlich 
nach,  dass  die  harten  Achsenglieder  hei  Mopsea  und  Melithaea  aus 
verschmolzener  Skleriten  bestehen,  und  dass  ein  Zentralstrang, 
welcher  aus  den  Coenenchymskleriten  vergleichbaren  Skleriten 
besteht,  sich  durch  die  Skeletachse  ganz  hindurch  zieht.  Auf 
diese  neuen  Befunde  basierend,  stellte  sich  Küllikee,  gegen  die 
Hypothese  von  Milne-Edwards  vor,  dass  die  Skeletachse  der 
Gorgoniden  ein  ektodermales  Ausscheidungsprodukt  sei,  und 
hob  besonders  hervor,  dass  dieselbe  im  Coenenchym  entsteht. 
Er  hat  aber  die  Frage  nicht  verfolgt,  wie  diese  kompliziert 
gebaute  Skeletachse  im  Coenenchym  entsteht,  oder  was  der 
Zentralstrang  morphologisch  bedeutet. 

Von  dieser  Familie  habe  ich  nun  die  Gattungen  Mopsella, 
Melitodcs^  Acaharia  und  Parisis  untersuchen  können.  Zuerst  gebe 
ich  die  Resultate  der  Untersuchung  an  Mclüoles,  Mojjsella  und 
Acaharia,  und  zuletzt  dieselben  an  Parisis  an. 

Verzu-eKjimg.— Bei  den  Gattungen  Mclitodes,  Mopsella  und 
Acaharia  werden  die  Zweige  gewöhnlich  durch  typische  Spitzen- 
dichotomie  gebildet.  Die  Zweigbildung  durch  Knospung  findet 
aber  nur  ausnahmsweise  statt. 

An  den  Apikalal)schnitten  der  Zweige  stehen  die  Polypen 
in  zwei  gegenseitigen  Streifen,  und  zwar  bei  den  typischen  Fällen 
gewöhnlich  beinahe  in  Paaren.  In  der  Regel  stehen  die  Paaren 
von  kleineren  und  grösseren  Polypen  in  Abwechselung,  und  das 
oberste  Paar  ist  immer  von  den  vöUig  erwachsenen,  grossen, 
sogar  oft  in  der  Grösse  die  allen  übrigen  weit  übertreffenden 
Polypen  gebildet.  Durch  die  angedeutete  Verteilung  der  jungen 
Polypen  kann  man  schliessen  dass  das  Längemvachstum  der 
Rinde  nicht  am  Scheitel,  sondern  in  einer  Zone  unterhalb  der 
Apikaipolypen  stattfindet,  indem  neue  Polypen  in  den  erweiter- 
ten Zwischenräumen  zwischen  den  schon  vorhandenen  Polypen 
ausknospen.  Bei  den  Formen  mit  gedrängt  stehenden  Polypen, 
z.  B.  bei  Melitoiies  fiahellifera,  ist  diese  regelmässige  Abwechselung 
mehr  oder  minder  undeutlich.  Dennoch  lässt  es  sich  immer 
einblicken  dass  auch  hier  dasselbe  Verhältnis  in  gewissem  Masse 
Geltung  hat. 


40 


K.  Kinoshita  :  Beiträge  zur  Kenntnis  der 


Abbildung  10. 
Acabaria  sp. 
Neu  gegabelter  Zweig. 
X5. 


Der  Apex  der  Zweige  ragt  gewöhnlieli  ein  wenig  über  die 
Apikaipolypen-  bei  Acabaria  8p.  (Abbildung  10),  die  in  der 
Nachbarschaft  der  Biologischen  Station  zu  Misaki  ziemlich 
allgemein  vorkommt,  oft  bis  1  Mm.-  hervor.  Der  Apex  weist 
in  der  Mitte  der  Spitze  eine  kleine,  mehr  oder  minder  seichte, 
kraterförmige  Vertiefung  auf,  welche  uns  an 
einen  zurückgezogenen  Polypen  erinnert. 
Beim  Verzweigen  gabelt  sich  der  Zweig  an  der 
Spitze  in  zwei.  Die  zwei  obersten  Polypen 
schreiten  je  an  die  äussere  Seite  der  neu  gebil- 
deten Zweige  über,  und  darnach  wird,  an  der 
Innenseite  gerade  ihnen  gegenüber,  je  ein 
neuer  Polyp  gebildet,  welcher  bald  aufwächst 
und  den  an  der  Aussenseite  gelegenen  alten 
Polypen  gleich  kommt.  Diese  zwei  neuen 
Polypenpaare  an  den  Zweigspitzen  behalten 
ihre  endständige  Stellung  bei,  während  alle  nachträglich  sich 
bildenden  Polypen  unter  ihnen  auftreten. 

Zentralstrang. — Wie  es  in  den  Abbildungen  von  Köllikeu  (1) 
und  Kükenthal  (4)  deutlich  angegeben  ist,  ist  der  Zentralstrang 
bei  den  Gattungen  Melitodcs,  Mojisdki  und  Acabaria  immer  sehr  gut 
ausgebildet.  Er  läuft  die  ganze  Länge  der  Stämme  und  Zweige 
ununterbrochen  mit  gleicher  Dicke  durch. 

Die  Skleriten,  welche  diesen  Zentralstrang  aufbauen,  sind 
natürlich  von  denjenigen  der  Rindenlage  verschieden.  Bei 
Melitodcs  und  Acabaria  sind  sie  bedornte  Spindeln,  welche  denen 
der  Rinde  ganz  gleichen.  Bei  Mopse  la  jedoch  werden  auch 
Stachelkeulen,  untermengt  in  den  Spindeln,  sehr  häufig  ange- 
troffen (Abbildung  11).  Wie  man  deutlich  nach  der  Form 
schliessen  kann,  gehören  diese  Skleriten  eigentlich  zur  Rinden- 
schicht, wo  sie  einst  an  der  Oberfläche  derart  angeordnet  waren, 
dass  die  Stacheln  sich  nach  aussen  richteten  und  so  der  Rinde 
Schutz  bieteten.  Von  da  aus  sollen  sie  zufällig  in  den  Zentral- 
strang hineingekommen  sein. 

Der  Zentralstrang  nun  ist  in  der  Nähe  der  Zweigspitze  ganz 
ähnlich  gebaut  wie  die  Rinde.       Die  Rinde  an   der  Zweigspitze 


Morphologie  u.  Stammesgeschichte  der  Gorgoniden. 


41 


ist  von  einem  Netzwerke  dünner  Solenia  durchzogen  und  weist 
meist    kleine,     nicht    ganz    ausgebildete,     aber    dicht    gedrängt 

auftretende  junge  Skleriten  auf. 
Im  obersten  Abschnitte  lässt  sich 
der  Zentralstrang  kaum  von  diesem 
Rindengewebe  unterscheiden,  aber 
nach  abwärts  fangen  die  Skleriten 
allmählich  an,  durch  kalkige  Ab- 
lagerung fest  an  einander  verkittet 
zu  werden,  indem  die  Solenia 
degenerieren  und  sich  schliesslich 
spurlos  verschwinden. 

Diese  Verkittung  findet  in  den 
weichen  Gliedern,  wie  schon  Köl- 
LTKER  bemerkt,  nicht  statt. 

Die      harten       Achsenglieder 
werden      gerade      unterhalb      der 
Rinde  der  Zweigspitze   durch   An- 
sätze   neuer    Skleriten,     und    zwar 
von  Anfang  an  in  der  Form  einer 
Röhre,  gebildet.      Die  Röhrenform 
kommt  daraus  zustande  dass  die  Rinde,  welche  jene  Achsenglieder 
von  aussen  umgibt,   an  die  Zweigspitze  in  die  letzteren  als  den 
Zentralstrang  hinein  dringt. 

Von  der  Gattung  Pansis  habe  ich  Parisis  fruticosa  und  Parisis 
affinis  minor  untersuchen  können.  Diese  Formen  weichen  in 
mancher  Hinsicht  von  den  übrigen  Mitgliedern  der  Familie  ab. 
Es  liegt  jedoch  kein  positiver  Beweis  vor,  diese  Gattung  als  solche 
zu  betrachten,  welche  von  den  letzteren  ph^letisch  sehr  fern 
steht. 

Der  Zentralabschnitt  der  harten  Achsenglieder  von  Parisis 
besteht  aus  dornigen  Skleriten,  welche  von  denjenigen  der 
Umgebung  in  der  Form  unterscheidbar  sind.  Ich  habe  jedoch 
nicht  genau  ermitteln  können,  welche  Form  dieselben  besitzen, 
da  sie  wie  die  letzteren  ganz  durchsichtig  sind.  In  den  meisten 
angefertigten    Schliffen    habe   ich    gefunden,    dass   zwischen    den 


Abbildung  11. 
Schematischer  Längsschnitt  eines 
Zweigapex  von  einer  Mojj.'^el la-Kolonie. 
Polypenhöhlen  und  Solenia  schwarz  ; 
obertiächliche  Rindensklpriten  mit  dicken 
Konturen,  tiefere  Eindenskleriten  mit 
dünnen  Kontaren  gezeichnet;  Kiudenlage 
der  Achse  längs  schraffiert. 


42  ^'  Kinoshita  :  Beiträge  zur  Kenntnis  der 

Skleriten  zahlreiche  kleine  Luft  enthaltende  Lücken  existieren. 
Der  Zentralstrang,  wenn  man  auch  hier  diese  Bezeichnung 
brauchen  darf,  scheint  in  den  weichen  Gliedern  unterbrochen  zu 
sein,  denn  das  Zentrum  derselben  ist  auch  durch  die  den  weichen 
Gliedern  eigentümlichen  Skleriten  besetzt,  welche  so  dicht  vor- 
kommen dass  der  Abschnitt  sich  von  der  Umgebung  sehr  leicht 
unterscheiden  lässt. 


BemcrhiüKj  über  die  MeUtodidae. 

In  einem  vorhergehenden  Kapitel,  in  welchem  ich  über  die 
Gattungen  Briareum  und  Paragorgia  Bemerkungen  gegeben  habe, 
habe  ich  folgende  drei  Verhältnisse  als  Einwände  gegen  Studer'  - 
sehe  Hypothese  hervorgehoben:  1)  dass  die  Skeletachse  im 
Zentrum  den  aus  Rindenskleriten  bestehenden  Zentralstrang 
aufweist;  2)  dass  die  sogenannten  Hauptkanäle  in  diesem 
Zentralstrange  nicht  als  Solenia,  sondern  als  Verlängerungen  der 
Polypenhöhlen  anzusehen  sind;  3)  dass  die  Neubildung  der 
Zweige  bei  Pamgorgia  (wohl  auch  bei  Solenocaulon)  alleinig  durch 
Knospung  unterhalb  der  Zweigspitze  stattfindet.  Von  diesen  drei 
Verhältnissen  wurden  die  zwei  letzteren  als  besonders  unüber- 
windbare  Einwände  gegen  Studer' sehe  Hypothese  angegeben. 
Bei  den  Melitodiden  jedoch  zeigen  sich  dieselben  etwas  anders, 
da  der  Zentralstrang  nämlich  ganzlicli  der  Hauptkanäle  entbehrt, 
und  die  Zweigbildung  gewölnilich  (mit  einer  Ausnahme  von 
Parisis)  durch  Spitzen  gabelung,  nur  ausnalnnsweise  durch  Knos- 
pung, vor  sich  geht. 

Dass  bei  der  Zweigbildung  Spitzengabelung  vorherrscht, 
scheint  für  die  Solenocaulon-Hyjyothese  eine  gewisse  Stütze  zu 
liefern,  da  nach  dieser  Hypothese  die  Randsäume  des  eigentlichen 
flachen  Coenenchyms  nur  an  der  Zweigspitze  ausgesteht  sind,  und 
das  Wachstum  des  Coenenchyms  nur  an  den  Randsäumen 
stattfinden  sollte.  Jedoch  falls  die  Zweige,  wenn  auch  ausnahms- 
weise, unterhalb  der  Zweigspitze  durch  Knospung  gebildet 
Averden,  so  kann  man  diese  Zweige,  bezw.  die  Kolonie,   auch  als 


Morphologie  n.  Stammesgeschichte  der  Gorgoniden. 


43 


M 


i 


Abbildung  12. 
Meli  indes  sp. 
Hartes  Achsenglied   mit 
einem   sekundären   Zweig. 
Zentralstrang  punktiert. 


solche  annehmen,   welche  durch  Verdickung  des  Coenenchjmis 
in  der  Richtung  der  Hauptachse  ausgebildet  sind. 

Die  durch  Knospung  gebildeten  Zweige  wurden  manchmal 
bei  verschiedenen  Mclitodes-  und  ilIo^)se//a-Arten  angetroffen. 
Sie  sind  alle  dünn  und  kurz,  aber  tragen  Polypen  in  gewöhn- 
hcher  Weise.  Dass  diese  Zweige  durch  Knospung  gebildet 
worden  sind,  ist  daran  deutlich  zu  erkennen 
dass  der  Zentralstrang  der  harten  Achsen- 
glieder gerade  hinweggeht  und  von  demselben 
des  Zweiges  durch  eine  dicke  Rindenlage 
getrennt  sind  (Abbildung  12).  Solche  Zweige 
weisen  in  ihrem  Basalabschnitte  eine  Menge 
in  den  Achsenskleriten  untermengt  auftre- 
tender Skleriten  auf,  welche  bald  in  die 
Mittellinie  der  Achsen  dicht  ansammeln, 
um  endlich  einen  typischen  Zentralstrang  zu 
bilden.  An  der  Spitze  des  Zweiges  ist  auch 
sogar  die  Apikaivertiefung  zu  finden.  Also 
ist  der  Apikaiabschnitt  der  Zweige  ganz  normal  gebaut. 

Aus  dieser  Tatsache  ist  zu  ersehen  dass  der  Zentralstrang  bei 
den  Melitodiden  nicht  mit  der  Rinde,  welche  nach  der  Solenocaulon- 
Hypothese  die  Innenwand  der  rinnen-  oder  röhrenförmigen 
Zweige  bilden  soll,  sondern  mit  dem  Zentralstrange  des  Briareum- 
Paragorgia-Typus  zu  vergleichen  ist;  denn  dieser  Zentralstrang  ist 
auch  in  normaler  Weise  in  den  Skeletachsen  derjenigen  Zweige 
entwickelt,  welche  nicht  durch  Spitzengabelung,  sondern  durch 
Knospung  gebildet  worden  sind. 

Die  Tatsache,  dass  die  Hauptkanäle  im  Zentralstrang,  welche 
als  proximale  Verlängerungen  der  Magenhöhlen  der  Terminal- 
polypen anzusehen  sind,  bei  den  Melitodiden  nicht  erhalten 
bleiben,  scheint  allerdings  nicht  völlig  mit  dieser  Homologisierung 
zu  stimmen.  Aber  wenn  man  in  Betracht  zieht,  dass  die 
Terminalpolypen  etwas  seitwärts  von  der  Zweigspitze  gerückt 
sind,  und  wenn  man  weiter  annimmt,  dass  bei  diesen  Terminal- 
polypen die  Kanalabschnitte  sich  infolge  der  Ortveränderung 
reduziert  haben,    aber  der  Zentralstrang  zurückgeblieben  ist,    so 


44  K'  Kinoshiti  :  Beitrüge  zur  Kenntuis  der 

kann  nmn  wohl  vermuten  dass  die  Melitodideu  demselben  Typus 
wie  Briareiim  und  Paragorgia  angehören.  Jedenfalls  ist  es  nicht 
vernünftig,  den  Bautypus  der  Melitodidenkolonie  durch  die 
Soleiiocaidon-JlypoÜ-ieeQ  zu  erklären,  solange  diese  als  gar  statthaft 
nicht  gestützt  wird. 

Suberogorgiidae. 

Von  dieser  Familie  sind  drei  Gattungen,  Siiherogorgia,  Keroeides 
und  Dendrogorgia^   bekannt. 

Wie  ich  (2)  schon  dargetan  habe,  ist  die  Gattung  Kcrueldes 
keine  Skleraxonie,  sondern  eine  aberrante  Gorgonie,  bei  welcher 
nur  der  Zentialstrang  der  Skeletachse  von  einem  denselben  an 
der  Spitze  fingerhutförmig  bekleidenden  Achsenepithel  ausge- 
schieden wird,  aber  die  Rindenlage  derselben  von  mesogloealen 
Skleriten  und  Hornsubstanz  aufgebaut  wird. 

In  demselben  Jahre,  aber  etwas  früher  als  ich  die  obige 
Ansicht  veröffentlichte,  gab  Simpsox  (1)  die  Beschreibung  seiner 
neuen  Gattung  Dendrogorgia^  welche  er,  wegen  der  sklerogorgi- 
schen  Struktur  der  Skeletachse,  auf  eine  früher  als  JunceUa  capeitsis 
bezeichnete  Spezies  errichtete.  Üeber  den  Zentralstrang  der 
Skeletachse  von  dieser  Form  nun  machte  Simpson  keine  Bemer- 
kung. Aber  soweit  man  nach  den  von  ihm  gegebenen  Abbil- 
dungen schliessen  kann,  scheint  die  Skeletachse  einen  deutlichen 
Zentralstrang  zu  besitzen.  Es  ist  jedoch  nicht  sicher,  ob  der 
Zentralstrang  bei  seinem  Exemplare  demselben  Typus  Avie  bei 
Keroeides  oder  bei  den  Melitodiden  und  Siiherogorgia  angehört. 

Subevogorgia 

Der  Tj^pus  dieser  Gattung  ist  Pterogorgia  suherosa.  Kölliker 
(1)  wies  nach,  dass  die  Skeletachse  dieser  Art  eine  sklerogorgische 
Struktur  zeigt,  und  demnach  errichtete  er  auf  diese  Form  eine 
neue  Gattung  Sderogorgia  die  in  seiner  Klassifikation  eine 
besondere  Unterfamilie  bildete.  Dieser  Gattungsname  soll  aber 
mit  dem  älteren  Namen  Siiherogorgia  von  Gray  ersetzt  werden. 


Morphologie  u.  Stammrsgeschicbte  der  Gorgonick-n. 


45 


Von  dieser  Gattung  nun   habe  ich   zwei  Arten,    Suberugorgia 
hüUikeri  und  Subcrogorgia  affinis  appressa  untersuchen  können. 

Die  Skeletachse  bei  diesen 
Formen  besteht  aus  glatten  Spin- 
dehi,  welche  dicht  mit  einander 
verschmelzen,  sodass  meistens  ihre 
originale  Form  verloren  geht.  Sie 
zeigen  in  Querschnitten  der  Achse 
deutliche  konzentrische  Anordnung. 
KüLLiKEE,  (1)  schilderte  schon  das 
Vorkommen  der  Rindenskleriten  in 
der  Skeletachse.  Darüber  sagt  er: 
'  '  Auch  die  Axe  der  Scierogorgia 
enthalt  wie  die  der  Melithaeaceen 
vereinzelt  oder  in  kleinen  Nestern 
warzige  Kalkkörper  von  der  Form,  Grösse  und  Farbe  derer  des 
Coenenchyms."  Bei  meinen  Exemplaren  habe  ich  auch  in  der 
Mitte  der  Skeletachse  beinahe  unfehlbar  eine  geringe  Anzahl 
von  Rindenskleriten  finden  können  (Abbildung  13).  Diese 
Skleritenreihe  möchte  ich  mit  dem  sehr  wohl  entwickelten 
Zentralstrang  der  Melitodiden  und  auch  der  Briarewn  und 
Paragorgia  vergleichen,  denn  diese  Skleriten  sind  es,  welche  auch 
von  der  Spitze  in  die  Skeletachse  hineingelangt  sind. 

Bei  den  vorliegenden  zwei  Formen  dieser  Gattung  scheint 
die  Verzweigung  nur  durch  Knospung  vor  sich  zu  gehen. 

Wenn  man  die  oben  erwähnten  Verhältnisse  in  Betracht 
zieht,  so  wird  man  finden  dass  diese  Gattung  auch  demselben 
Typus  wie  die  Melitodiden,  sowie  auch  wie  die  Gattungen  Briareum 
und  Paragorgia  angehört. 


Abbildung  13. 

Subcrogorgia  äff.  appressa  Xutt. 

Querschnitt  eines  Zweiges.  Poly- 
penhöhlen und  Stammkanäle 
schwarz  ;  Rindenskleriten  mit  dik- 
ken  (Konturen,  Achsenskleriten  mit 
dünnen  Konturen  gezeichnet.   X20. 


Schlussbemerkung  zu  den  Scieraxonra. 

In  den  vorhergehenden  Kapiteln  habe  ich  geschildert,  dass 
die  untersuchten  Skleraxonier  sich  in  drei  Typen  zerfallen,  welche 
resp.  von  Briarewn,  Melitodes  und  Svôerogorgia  repräsentiert  werden, 
und    Avelche    ph3detisch    in    einer    direkten    Verwandtschaft    mit 


46  ^'  Kinosliita  :  Beiträge  zur  Kenntnis  der 

einander  stehen  würden.  Den  Bnareiim-Pamgorgia-Typus,  welcher 
den  anderen  T3q3en  zugrunde  liegt,  möchte  ich,  wie  schon 
angeführt,  nicht  durch  die  Solenocaidon-'RyYiOtheëe  von  Studer, 
sondern  durch  die  Verdickung  des  Coenenchyms  in  der  Richtung 
der  Hauptachse  erklären.  Diese  Auffassung  scheint  mir  beinahe 
unwiderlegbar  zu  sein,  da  die  Verhältnisse,  welche  die  Stämme 
und  Zweige  bei  dieseii  beiden  Gattungen  zeigen,  nur  erst  dadurch 
erklärt  werden  können. 

Von  den  ICigentümlichkeiten,  wodurch  sich  der  Briarewn- 
Paragorgia-Typus  auszeichnet,  gilt  die  Existenz  der  sogenannten 
"  Hauptkanäle  "  im  Innern  des  Zentralstranges  als  die  wichtigste 
für  Erwägung  der  systematischen  Stellung  der  Gattungen  Briareum 
und  Pamgorgia,  oder  vielmehr  der  ganzen  Skleraxonier.  Dass 
es  sich  dabei  nicht  um  Solenia  handelt,  habe  ich  schon  erwähnt. 
Wenn  man  nun  die  Hauptkanäle  als  proximale  Abschnitte  der 
Terminalpolypen  betrachtet,  so  muss  man  die  Stämme  und  Zweige 
natürlich  als  die  Bündel  der  Axialpolypen  ansehen.  Diese 
Eigenschaft  der  Axialpolypen,  d.  h.  dass  dieselben  Syndete^-* 
bilden,  bildet  eine  Eigentümlichkeit  der  Ordnung  Alcyonacea. 
Da  die  Scleraxonia  nach  den  bisherigen  Auffassungen^^  als  solche, 
bei  denen  alle  Polypen,  wie  bei  den  Gorgonacea,  kurze  Magen- 
höhlen besitzen,  charakterisiert  worden  sind,  so  sollen  die  beiden 
Briareidengattungen  theoretisch  aus  der  Ordnung  Scleraxonia 
entnommen  und  in  die  Ordnung  Alcyonacea  gebracht  werden. 
Aber  in  der  letzteren  Ordnung  sind  keine  Formen  bekannt,  bei 
welchen  die  Anthostele^^  gänzlich  der  Mesenterien  entbehren. 
Weiter  es  existiert  zwischen  dem  Briarewn-Paragorgia-Typus  und 
der  Siplionogorgia.  welch  letztere  Kölliker  (3)  wegen  der  nur  in 
Vierzahl  bleibenden  Mesenterien  in  den  "Stammkanälen"  als 
eine  Zwischenform  zwischen  den  Alcyoniden  und  den  Paragor- 
giaceae  annahm,  noch  eine  zu  weite  Lücke,  um  die  beiden 
Briareiden    in    die    direkte    Nähe  von   Siplionogorgia    zu    bringen. 


1)  Siehe  Bourne  (1). 

2)  V.  Koch  (7)   schildert  deutlich  die   Differenz   der  Polypen   in  der   Länge,   also  in    der 

Ordnung  der  Polypen  bei  Skleraxoniern.    Es  ist  jedoch  nicht  klar  genug,  auf  welche 
Tatsachen  sich  diese  Auffassung  stützt.     Siehe  Seite  9. 

3)  Siehe  Bourne  (1). 


Morphologie  u.  Stauiinesgeschichto  der  Gorgoniden.  47 

Schliesslich  ist  noch  eine  Möglichkeit  nicht  ausgeschlossen,  dass 
dieselhen  durch  einen  mit  der  Entwicklungsbahn  der  höheren 
Alc3"onacea  parallel  gehenden  Weg  aus  den  Enjtliropodium- 
ähnlichen,  niederen  Alcyoniden  abstammten,  da  Briarewn 
asbestinum  und  wahrscheinlich  auch  noch  viele  andere  niederen 
Skleraxonier,  in  den  Beschaffenheiten  der  Skleriten  sowie  im 
Modus  der  Polj^Denverteilung,  mehr  den  niederen  als  den 
höheren  Alcj^oniden  verwandt  sind. 


48 


Lilteratur. 

Blainville  {1)  :   Maunel  cV  Actiuologie  ou  de  Zoopbytologie.  1834. 

Bourne  (1)  :  On  the  Genus  Lemnalia  Gbay  with  an  Account  of  tbe  Branching 
System  of  the  Order  Alcyonacea.     Transact.  Linn.  Soc.  London,  Vol.  vii,  1900. 

Bourne  (2)  :  Anthozoa.     Treatise  on  Zoology,  Part  ii,  1900. 

Dana(l)  :  Report  ou  the  Zoophytes  of  tbe  United  States  Exploring  Expedition.  184G. 

Dana  (2)  :  Structure  and  Classification  of  Zoophytes.   1816. 

Ehrenberg  (1)  :  Beiträge  zur  Physiologischen  Kenntniss  der  Korallentbiere  im  All- 
gemeinen, nnd  besonders  des  Pioten  Meeres,  nebst  einem  Versuche  zur  Physiolo- 
gischen Systematik  derselben.  Abb.  d.  K.  Ak.  d.  AViss.  z.  Berlin  a.  d.  Jahre 
1832,  I.  Tbeib  1834. 

Qenth  (1)  :  Ueber  Solenogorgia  tubulosa  (eine  Neue  Gattung  der  Gorgoniden). 
Zeitschr.  d.  Wiss.  Zool.,  xvii.  Bd  ,  1867. 

Qermanos(l):  Gorgonaceen  von  Ternate.  Abhandl.  d.  Scnckenb.  Nat.  Gesell., 
xxiii.  Bd.,  i,  1896. 

Hickson  (1)  :  On  tbe  Ciliated  Groove  (Sipbonoglyphe)  in  the  Stomodaeiim  of  the 
Alcyonarians.     Phil.  Trans.  Pioy.  Soc.  London,  Part  iii,  1883. 

Hickson  (2)  :  The  Alcyonaria  of  the  Maldives.  I.  Genera  Xenia,  Telesto,  Spongodes, 
Nephthya,  Paraspongodes,  Chyronephtbya,  Siphonogorgia,  Solenocaulon,  and 
Melitodes.  The  Fauna  and  Geography  of  the  Maldive  and  Laccadive  Archi- 
pelagoes, Vol.  ii,  pt.  i,  1903. 

Hickson  (3)  :  Coelentera     and     Ctenophora.      The    Cambridge     Natural    History, 

Vol.  i,  1906. 
Hiles  (1)  :    Gorgonacea    collected    by    Dr.  Willey.     Willey's  Zool.  Fiesults,  Part 

ii,  1899. 
Janover  (1)  :  Die  Gattung  Solenocaulon.     Inaug.  Dissert.,  Bern,  1904. 
Kinoshita  (1)  :  Ueber  die  Postcmbryonale  Entwicklung  von  Anthoplexaura  dimorpha 

IvtJKENTH.AL.     Jour.  Coll.  Sci.  Imp.  Univ.,  Tokio,  Vol.  xxvii.  Art.  xiv,  1910. 
Kinoshita  (2)  :   On  the  Keroeididae,  a  New  Family  of  Gorgonacea,  and  Some  Notes 

on  tiie  Suberogorgiidae.     Annot.  Zool.  Jap.,  Vol.  vii,  Part  iv,  1910. 
Klunzinger  (1)  :     Die  Koralltbieie  des  Boten  Meeres.     I.    Theil.   Alcyonarien  und 

Malacodermen.  1877. 
V.  Koch(lj:  Anatomie  von  Isis  neapolitana.     Morph.  Jahrb.,  iv.  Bd.,  1878. 
V.  Koch  (2)  :    Mittheilungen   über   Gorgonia  verrucosa.     Morph.     Jahrb.,  iv.    Bd., 

1878. 
V.  Koch  (3}  :  Das  Skelett  der  Alcyonarien.     Morph.  Jahrb.,  iv.  Bd.,  1878. 
V.  Koch  (^4)  :   Die  Morphologische  Bedeutung  des  Koralleuskeletts.     Biol.  Zentralb., 

ii.  Bd.,  1882. 
V.  Koch  (5)  :   Vorlaufige  Mittheiluugen  über  die  Gorgonien  (Gorgoria  axifera)  von 

Neapel  und  über  die  Entwicklung  der  Gorgonia  verrucosa.     Mitth.  a.  d.  Zool. 

Stat.  zu  Neapel,  iii.  Bd.,  1882. 
V.  Koch  (6)  :   Die  Gorgoniden  des  Golfes  von    Neapel  und  der  angrenzenden  Meeres- 


49 

iibscluùttc.     Fauna  nnd  Flora  des  Golfes  von  Neapel  und  der  aiip;venzenden  Mee- 

resabsclniitte,  15,   1887. 
V.  Koch  (7)  :  Die  Alcyonaceeii  des  Golfes  von   Neapel.     Mittli.  a.  d.  Zool.     Stat.  z. 

Neapel,  ix.  Bd.,  iv.  Heft,  1890. 
V.  Koch  (8)  :     Kleinere    Mittlieilungen    über  Antliozoen.     7)    Ueber    Kolonien    von 

Bebryce  mollis  Phil.,  welches  Cornulariden  iihnlicli  sind.     Morph.  Jahrb.,   xviii. 

Bd.,  i.  Heftf,  1896. 
Kölliker(l)  :    Icônes  Histiologicae  oder  Atlas  der  Vergleichenden  Gewebelehre.  II. 

Abtheilung,  1860. 
Kölliker  (2)  :  Beiträge  zur  Kenntnis   der  Polypen.     A^erh.    d.    Phys.  Med.     Gesell., 

Neue  Folge,  iL  Ed.,  i.  Heft,  1870. 
K'  Iliker  (3):    Die   Pennatulide   Unibellula  und   zwei  Neue  Typen   der  Alcyonarien. 

Festscbr.     z.     Feier     d.     fiinfundzwanzigjährigen    Bestehens     d.     Phy.     Med. 

Gesellsch.,     1875. 
Koren  og  Danielssen  (1)  :  Nye  Alcyonider,  Gorgonider  og  Pennatalider,  tilhörende 

Norges  Fauna.     Bergens  Museum,  1883. 
Kükenthal  (1)  :    Die  Stammesgeschichte  und  die  Geographische  Verbreitung   der 

Alcyonaceen.     Verh.  d.  leut.  Zeel.  Gesellsch.,  1906. 
Kükenthal(2)  :     Alcyouacea.     Wiss.     Ergeb.     d.   Deut.     Tiefsee  Exped.  xiii.    Bd., 

1906. 
Kukenthal  (3)  :     Diagnosen    Neuer   Gorgoniden   (4.    Mitteilung).      Zool.     Anzeig., 

xxxiii.  Bd.,  Nr.  i,  1908. 
Kükenthal  (4)  :   Japanische  Gorgoniden.  11.  Die  Familien  der  Plexauriden,  Chryso- 

gorgiden  und  Melitodiden.     Doflein's  Beiträge  z.  Naturgesch.   Ostasieus,  1909. 
Lamarck  (1)  -.  Histoire  Naturehe  des  Animaux  sans  Vertèbres.   1816. 
Lacaze=Duthiers  (1):  Histoire  Naturelle  du  Corail.   18f)4. 

Lacaze=Duthiers  (2)  :  Polypiers  des  Gorgones.     Acad.    d.   Sei.  Nat.  T.  iii,  1808. 
Menneking  (1)  :    lieber    die   Anordnung  der  Schuppen  und  das   Kanalsystem   bei 

Stachyodes  ambigua  (Siuder),  &c.     Arch.  f.  Nat.  Gesch.,  Lxsi.  Bd ,  i.  Heft,  1905. 
Miîne=Edwerds  (1)  :  Histoire  Naturelle  des  Coralliaires  ou  Polypes  proprement  dits. 

1857. 
Müller  (1)  :  Über    die   Bildung  des  Achsenshelets  von  Corallium.     Mitth.  a.  d.  Zool. 

Stat.  Neapel,  xx.  Bd.,  i.  Heft,  1910. 
Nutting  (1)  :  The  Gorgonacea  of  the  Siboga  Expedition.  VIII.   Scleraxonia.     Siboga 

Expeditie,  XIII.  b,  5,  1911. 
Pallas  (1)  :  Elenchus  Zoophytorum  (Deutsche  Uebersetzung).   1760. 
Schneider  (1)  :  Das  Achsenskelet  der  Gorgoniden.     Arch.  f.  Nat.  Geseh  ,  Lxxi.   Bd., 

i.   Heft,  1905. 
Simpson  (1):    On  a  New  Pseudaxonid  Genus  Dendrogorgia.    Proc.   Pioy.   Phys.  Soc. 

Edinburgh,  Vol.  xviii,  No.  i,   1910. 
Studer  (1)  :    Ueber  Bau  und  Entwicklung  der  Achse  von  Gorgonia  Beitoloni  Lamx 

Mitth.  d.  Naturforsch.  Geseh.  Bern,  1874. 
Studer  (2)  :    Versuch    eines    Systèmes   der   Alcyonaria.   Arch.   f.   Naturgesch.,   LÜi. 


50 


Bd.,  i.  Heffc,   1887. 
Studer  (3)  :  La  Squelette  axiale  des  Gorgonacea.  Arch.  Sei.  Nat,  T.  xx,  1905. 
Versluys  (1)  :   Pseudociadochonas  hicksoni  n.g.,  n.sp.     Siboga    Expeditie,    XIIIc, 

ii.  Teil,  1907. 
Wright  and  Studer  (1)  :   Report  on  the  Alcyonaria  collected  by  the    Challenger. 

1889. 


Published  Mar.  31st,  1913. 


JOURNAL  OF  'niK  COLLEGE  OP  SCIENCE,  TOXYO  IMPERIAL  UNIVERSITY, 
TOKYO,  JAPAN. 

VOL.  XXXII. ,  ART.  11. 


Ueber  die  charakteristische  Mannigfaltigkeit 

der  partiellen  Differentialgleichungen 

erster  Ordnung. 

Von 
T.  Yoshiye,  liU/aJcushi, 

Professor  der  Mathei)iatik,  Tokyo  Kaiserl.   Universität. 

Die  vorliegeDcle  Arbeit  ist  eigentlich  als  eine  Fortsetzung 
meiner  Arbeit*  in  den  "Mathematischen  Annalen,"  Bd.  57, 
anzusehen.  Es  wird  jetzt  durch  die  Variationsmethode  die  charak- 
teristische Mannigfaltigkeit  eines  Systems  partieller  Differential- 
gleichungen erster  Ordnung  mit  einer  einzigen  abhängigen 
Variabein  hergeleitet. 

I. 

Es  sei 

(1) 

das    vorgelegte    System    partieller   Differentialgleichungen   erster 
Ordnung,    wobei  z  die  zu  bestimmende  Funktion  von  x^,  ...,  x„ 

und  Pi  ihre  Ableitung  -^ —  bezeichnet.     Man  kann  offenbar  fest- 

yf  p"     ...    TP  \ 

setzen,    dass  die   Determinante  ~\r^ — ' —  .   nicht  identisch   ver- 
schwindet,    denn  mindestens  eine  Determinante  von  der  Form 

7i(  TP   ...    TP  ') 

-^^ — '    "'  ■  nicht  identisch  verschwindet,  sobald  die  Gleichunsren 

*     „  Anwendungen  der  Variationsrechnung  auf  partielle  Differentialgleichungen  mit  zwei 
unabhängigen  Variabein." 


2  T,  Yoshiye  : 

von  (1)   von  einander   verschieden   sind:    wir  nehmen   sogar  an 

Das  System  (1)  definirt  eine  Schar  von  00-'^-"'+^  Flächenele- 
mente (z,  oTi,  •-.,  Xn,  Pi,  ■■■,  Pn)  im  n+1  dimensionalen  Räume. 
Wir  mögen  nun  den  eindimensionalen  Elementverein  aus  dieser 
Schar  finden,  nämlich  das  Funktionensystem  z,  x^^  ...,  a:„,  p^,---,  p„ 
einer  einzigen  Variabein  t,  welches  die  Beziehung 

^'-2>,a-/=0  (2) 

befriedigt,     z',  x-    stehen  für  die  Ableitungen  von  z^  Xi  nach  t. 
Zu  diesem  Zwecke  bilden  wir  das  Integral 


y"  f  n  m 

«0  i=i  ^=1 


wobei  ?.  und  //j,  •••,//«  beliebige  Funktionen  von  t,  und  ^0,  h  ge- 
wisse bestimmte  Werte  von  t  bezeichnen. 

Dieses  Integral  verschwindet  offenbar,  wofern  die  Glei- 
chungen (1)  und  (2)  erfüllt  sind.  Daraus  folgt  unmittelbar  die 
Gleichung 

§  f  \X  (/_  2>^  x!)  +  2  ^,  F,]  dt=0, 

J  ta  i  =  i  k  =  l 

sobald  die  Variationen  den  Bedingungen 

oV-2o>,a;/)=0, 

i  =  l 

genügen. 

Diese  Gleichung  und  die  Bedingungsgleichungen  können  wir 
folgendermassen  umschreiben: 

K-!  t^^  'OQ  U/^  ^  -^']  ''  ''  (3) 

oV-2>,  oV  +  a:/ o>,)=0 


^dz+î^dx,+  i^dp,=0     ik  =  h2,  ■■■m) 

ÙZ  i=i     dXi  i=l    opi  ' 


(4) 


Charakt.  Mannigfaltigkeit  d.  part.  Differentialgleichungen.  ^ 

Man  erhält  ein  Funktionensystem  z,  x,  p,  welche  die  Glei- 
chung (3)  befriedigt,  wenn  man  0,  x,  p  so  bestimmen  kann,  dass 
jedes  Glied  der  Gleichung  (3)  identisch  verschwindet.* 

Wir  unterscheiden  nun  zwei  Fälle,  nKzn  und  m=n. 

Der  Fall  m<^n.  Könnte  man  nun  jeden  Ausdruck  unter  dem 
Integralzeichen  aus  dem  letzten  Glied  von  (3)  verschwinden 
lassen,  dann  müssen  die  Grössen  z^  x,  ii  die  folgenden  n—m  Glei- 
chungen befriedigen  : 


aF„ 


s  Fl 

aF, 

Ip^ 

^Pi 

S  Fl 

3  F., 

^P'i 

Ip., 

SF, 

3F, 

^Pm 

^P,n 

3Fi 

SF, 

3^1 
^P-i 


X; 


^Pi 


^F„ 

^Pr. 

IF,,, 
^Pi 


:0  (i  =  W+l,--w) 


(5) 


Wir  setzen  also  diese  n—7ii  Gleichungen  (5)  fest  als  die 
Bedingungsgleichungen  für  die  Grössen  z,  x,  p.  Dann  sind  nur 
m  Gleichungen  aus  den  folgenden  ?i  Gleichungen 

i  =  l  àpi 

Avesentlich . 

Wir  betrachten  also  die  ?i  +  l  Gleichungen 


I 


IFk 
SF.. 


•A'  =  0, 


2>,-^  -;.x/=0         (^=],  2,- ••70 

als  die  Bestimmungsgleichungen  von  m  +  1  Grössen  ?.,  ^i,  •••,//, 
Nach  den  Beziehungen  (5)  sind  die  m  + 1  Ausdrücke 


(6) 


rr/opi+- ••+j:-„'^P„, 


fc  ,^^    ,^       ,     ^F,   .^^^         Ck  =  \ß,...m) 


Ipi  ^Pn 


*    Da  (1)  eine   Differentialgleichung  enthält,   kann   man  nicht  ohne  weiteres,   wie   im 
gewöhnlichen  Variationsproblenie,  das  Verschwinden  jedes  Gliedes  von  (3)  scbliessen. 


T.  Yoshiye  : 


in  den  Bedingungen  (4)  nicht  von  einander  linear  unabhängig, 
und  daher  folgt  die  Gleichung: 


dz'—I'p^dx! 

i  =  l 

cz  ,=1  dr.- 


W^ 


Ipi  S/>2 


}F\_ 


OZ  +  -i   -.^ ~  OX; 


Iz 


^X; 


^F^ 


IF^ 


(7) 


Denkt  man  sich  die  Variationen  oXi,--,dxn  gegeben,  dann 
ist  (7)  die  Bestimmungsgleichung  von  dz,  welche  die  Form  der 
linearen  Differentialgleichung 

dz'  +  Xdz+X^=0 

besitzt.  Wie  man  leicht  sieht,  ist  À  gleich  e  ,  und  daher 
kann  die  Integrationskonstante  von  oz  nicht  immer  so  gewählt 
werden,  dass  [/o2;]Jj  gleich  Null  wird. 

Wir  nehmen  also  an,  dass  00:1,  •••,  dx„  an  einem  Endpunkte 
to  verscliAvinden.  Man  kann  offenbar  die  Integrationskonstante 
von  dz  so  bestimmen,  dass  Xdz  am  Punkte  ^0  verschwindet.  Am 
anderen  Endpunkte  ti  nehmen  wir  an:  eins  aus  dx^^  •••,  ox„,  etwa 
^Xj,  wird  immer  so  eingerichtet,  dass  der  Wert  [dz—pjdxj]t^  gleich 
Null  wird,  während  alle  anderen  an  diesem  Punkte  verschwinden. 
Dann  verschwinden  die  von  Integralzeichen  freien  Glieder  von 
(3).  Diese  Annahme  für  oxj  ist  immer  möglich,  sobald  Pj  am 
Punkte  ti  nicht  verschwindet,  und  da  der  Parameterwert  ^1  beliebig 
gewählt  Averden  kann,  ist  diese  Annahme  immer  möglich,  wenn 
Pj  nicht  identisch  Null  ist.  Die  Schwierigkeit  tritt  also  nur  dann 
ein,  wenn  alle  p  identisch  Null  sind.  Im  letzten  Falle  wird  z, 
wegen  der  Beziehung  (2),  gleich  konstant.  Diesen  Fall  schliess- 
en  wir  aus. 

Die  Gleichung  (0)  nimmt  nun  die  Form  an: 

Alle  Variationen  ux,-^   ausser  ox,,   verschwinden  an  den  Punk- 


Charakt.  Mannigfaltigkeit  cl.  part.  Differentialgleichungen.  5 

ten  ^)  und  ^i,  sonst  willkürlich;  die  Variation  àxj  verschwindet  am 
Punkte  ^0,  ^^ei"  nicht  am  Punkte  ^.  Wenn  man  alle  o.r.,  ausser 
^.f ,  identisch  gleich  Null  denkt,  dann  erhält  man 


./."[l'''^+'^^')']"^''"=°- 


Obgleich  dXj  am  Punkte  t:  einen  von  Null  verschiedenen  Wert 
annimmt,  doch  kann  man  aus  der  letzten  Gleichung,  in  ähnlicher 
Weise  wie  bei  der  Herleitung  der  Eulerschen  Gleichung  in  der 
Variationsrechnung,  schliessen 

Das  Verschwinden  von  den  anderen  Integranden    ^ /^i- -s— ^  +  te)' 

°  k=i         aXi 

kann  gewöhnlicherweise  bewiesen  werden.     Wir  haben  nämlich 

^>^-47^+^^^^)-ö        (*=l,'2,--n) 

i  =  l  OX: 


welche  sich  durch  die  Gleichung 
in  die  folgenden 


umschreiben  lassen.      Das  Symbol  (  ~v^  )  steht  für  den  Ausdruck 


3F,  ^^,  SF, 


+Pi 


IXi      ^'    Iz 

Setzt  man  die  durch  (G)  bestimmten  Werte  von  A,  ^i,  •••,//„.  in 
die  letzten  Gleichungen  ein,  so  erhalten  wir  n  Gleichungen, 
welche  mit  (5)  zusammen  2n—m  Bedingungen  für  die  Funktio- 
nensystem z,  a'i,  ■••,  ar„,  pi,  ...,  p„  bilden. 

Dieselben  Bedingungen  erhalten  wir  offenbar  von  den  Glei- 
chungen 

fc  =  l  àXi 


6  T.';,Yoshiye  : 

durch  Elimination  von  k,  fx^,  •••,/^,„.     Wir  haben  nämlich 


'^Pl 

^Pi 

w. 

ÔF2 

^P,n 

^P,n 

^F, 

SF., 

^Pi 

^Pi 

èFi 

3F, 

^Pl 


Ô/J1 


Ipl 


IF 

■^  III 

^Pi 

OF 

¥1 


=  0         (î  =  W+l,"-7t)* 


(8) 


_aFi 


'^Pm 


IF.„ 
Ml. 


m)  m   m  --' 


=0  (z  =  l,2, --w) 


Die  Gleichungen  (8)  sind  hinreichend  fürs  Bestehen  der 
Gleichung  (3),  wenn  -^^  und /^  geeignet  bestimmt  sind;  die  Glei- 
chungen (1)  und  (2)  folgen  aber  nicht  notwendig  daraus.  Wir 
dürfen  also  zu  den  Gleichungen  (9)  noch  die  Gleichung 


z'-Ip,x/=0 


(2) 


oder  was  dasselbe  heisst: 

öFi  IF.2 


IF\_ 


IFi 


i=i    '  "àp,      i^i^'  Ip 


IF, 

^Pin 

^t       IF. 

^Pi     ^ 


IF,„, 

IF„, 
^Pii, 


=0 


hinzufügen. 

Wir  gehen  nun  über  in  den 

Fall  iii=zn.  Das  Gleichungensystem  (4)  besteht  aus  «  +  1 
Gleichungen.  Von  den  n  letzten  Bedingungen  lassen  sich  die  n 
Variationen  dp^,  ...,  op,,  als  lineare  Funktionen  von  ^z,  ox^,  ...,  ôx„ 


Die  Gleichungen  (5)  selbst. 


Charakt.  Mannigfaltigkeit  d.  part.  Differentialgleichungen.  7 

ausdrücken,  da  wir  am  Anfang  die  Determinante    ^,  \      V   nicht 

gleich  Null  vorausgesetzt  haben.  Setzt  man  diese  Ausdrücke  in 
die  erste  ein,  dann  erhält  man  fin-  oz  eine  lineare  Differentialglei- 
chung erster  Ordnung  wie  früher.  Man  kann  daher  die  von 
Integi-alzeichen  freien  Glieder  von  (3)  als  verschwunden  anneh- 
men. 

Die  Bestimmungsgleichungen  von  ^^  und  fi  lauten  hier 


-//=:0 


n  ■\-ni 

Wir  brauchen  keine  Bedingungsgleichung  wie  (5). 

Wegen  der   Willkür  von  ^.r,,  •••,  dx„  folgen  die  Gleichungen 

i/^lt' +(^^') -^       (t=l,2,.-.n) 

Durch   die  Elimination  von  -^  und  ft  ergiebt  sich  das  Gleichungen- 
system 


^Pl 


^Pl 


IF., 

^Pn 


^F„ 

^Pa 


m  (t)  ■  ■  m  - 


=0     (i  =  l,2,--n)     (8') 


In  diesem  Falle  haben  wir  zu  bemerken,  dass  das  System  (8') 
die  notwendige  Folge  von  (1)  und  (2)  ist,  welches  für  m<zn  nicht 
der  Fall  war. 


II. 


Bezeichnet  man,  wie  gewöhnlich,  den  Ausdruck 
^  r  IF,  (^FA_  C^^FA^FA 


3  T.  Yoshiye  : 

mit  dem  Symbol  [Fi,  Ff,],  dann  folgt  aus (2)  und  (8)  die  Beziehung 


IFi 


IF, 


[F,F,]     [F,F,] 


[F,„F^]       F/ 


=0 


Soll  nun  der  Ausdruck  F^  für  die  Lösungen  des  Systems 
der  Gleichungen  (2)  und  (8)  einen  konstanten  Wert  annehmen,  so 
muss  die  folgende  Beziehung  identisch  bestehen. 


2>Fi 
2>Pi 


IF, 


IF„., 
^Pi 


^Fy 
^P,n 


^F„„ 

^Pr. 

X 

[F„F,-] 

0 

=  0 


[F,F^     [F,F,] 


Nach  den  Gleichungen  (8)  sind  a-„/+i,  •••,  x„',  p^',  •••,  p„' 
lineare  Funktionen  von  a-/,  •••,  xj  und  die  letzteren  Grössen  sind 
von  einander  unabhängig,  und  daher  können  sie  sich  willkürlich 
verändern.  Daher  müssen  die  Koeffizienten  von  x^\  •••,  xj  in 
der  letzten  Determinante  identisch  verschwinden.  Wir  haben  also 
m  Gleichungen 

an  \f,f;\  +  4.  [f.f;\  +  •  •  •  +  J,„,  [f,„f;\  =o 

4i  [F,F:\  +  J.«  [F,F,]  +  •  •  •  +  J,,.  [i^.F,]  =0 


^„IFif;\  +  â„,if,f;\  +•■  +  j,„„[f„fj  =o, 


"^F,  . 


wobei  Jik  das  algebraische  Komplement  des  Elementes  y^  in  der 
Determinante 


Clurakt.  Mannigfaltigkeit  d.  part.  Differentialgleichungen. 


J  = 


IFi 


3F., 


bezeichnet. 

Da  nun  die  Determinante  der  Koeffizienten  der  letzteren 
Gleichungen  gerade  //'"'  gleich  ist  und  daher  nicht  verschwindet, 
so  folgt,  dass  jeder  von  den  KlammerausdrUcken  [Fj  i^',],  [F^,  FJ 
...,  [F„„  FJ  verschwindet. 

Umgekehrt,  wenn  alle  Klammerausdrücke  [Fi  F;,]  für  ein 
Lösungssystem  von  (2)  und  (8)  identisch  verschwinden,  so  ist  un- 
mittelbar ersichtlich,  dass  alle  F/  identisch  verschwinden  müssen, 
d.h.  dass  dann  F^,  ■•■,  F,„  konstante  Werte  erhalten. 

Dieser  Fall  tritt  offenbar  ein,  wenn  alle  Ausdrücke  [F^  F,,] 
als  Funktionen  der  Veränderlichen  -T],  •••,  .t,„  2^1,  •••,  p„  identisch 
verschwinden,  d.h.  wenn  die  vorgelegten  Gleichungen  (1)  ein 
Involutionssystem  bilden. 

Im  folgenden  setzen  wir  also  fest,  dass  das  System  (1)  zuerst 
zu  einem  Involutionssystem  gebracht  worden  ist. 

Wählt  man  nun,  bei  der  Integration  vom  System  (2)  (8),  die 
Anfangswerte  so,  dass  dafür  alle  Ausdrücke  -F,,  ••,  F,„  gleich  Null 
werden,  dann  befriedigen  die  Lösungen  dieses  Systems  das 
vorgelegte  Involutionssystem  (1).  Nennen  wir  ein  solches 
Lösungssystem  eine  charakteristische  Mannigfaltigkeit,  so  haben 
wir  im  Ganzen  2n—2m  +  l  fach  unendlichviele  charakteristische 
Mannigfaltigkeiten. 

Die  gesammten  go^""'"^'  Elemente,  welche  dem  Involutions- 
system (1)  entsprechen,  schliessen  sich  zu  cc-""^'"^^  charakteris- 
tische Mannigfaltigkeiten  zusammen. 

Da  wir  die  simultanen  Gleichungen  (2),  (8)  nach 

'^m  +1»    ■  ■  ■  >   -^n  »   ■^  '  Pi  >    '  ■  *>  Pn 

auflösen   können,    wobei   die  rechten  Seiten  eindeutig  bestimmt 
werden,  können  wir  folgendermassen  schliessen: 


10 


T.  Yoshiye  ; 


Jede  charakteristische  Mannigfaltigkeit  wird  durch  2n—2m  +  l 
Anfangswerte  von  a'„,+,,  •••,  rr„,  z,  pj,  ■■-,  Pn  eindeutig  bestimmt. 

Wir  möchten  nun  zeigen,  dass  die  so  definirte  charakteris- 
tische Mannigfaltigkeit  mit  der  gebräuchlichen  übereinstimmt. 

Bezeichnet  man  die  2n—2m  +  l  von  einander  und  von  jPj,  ..., 
F^  verschiedenen  Lösungen  des  Gleichungensystems 


[F„^I>]=0,  [F,,^l>]=0,  ■■■,  [i^„„<I>]=0 


(9) 


durch  •!>„  •••,  *2„_2m+i,  so  ist  die  charakteristische  Mannigfaltigkeit 
gewöhnlich  durch  die  Gleichungen 


F,  =  0,    ■'■,  F„=0,  a>i  =  Ci.    •••,   'K,-:,„.+l  =  C-M-2m  +  U 


(10) 


wobei  Cj,  •••,  6'2„_2,„+i  Konstanten  bezeichnen,  gegeben.* 

Es  sei   *  (z,  x^,  •••,  a-„,  p„   •••,  i)„)  irgend    eine    Lösung   des 

Systems  (9). 

Wenn  n:ian  irgend  ein  Lösungssystem  z,  .r,  p  vom  System  (2) 

(8)  in  die  Funktion  *  einsetzt,  dann  gilt  identisch  die  Gleichung 


IFi         

^P,n  ^P,n 

[Fl*]       [F,<ï>] 


IF. 


SF,„ 

î>F,„ 

^Pm 

[F,„*]      «!>' 


=  0 


Da  aber  alle  Klammerausdrücke  in  der  Determinante  ver- 
schwinden, so  folgt  unmittelbar  die  Beziehung 

«Ii=Konst. 

Diese  letzte  Beziehung  zeigt  uns,  dass  jede  Lösung  *  von  (9) 
für  jedes  Lösungssystem  z,  x,  p  von  (2)  (8)  konstant  wird,  d.h. 
$i=Cf(«=l,  2,  ■••,  2/i— 2»i+l)  sind  Lösungen  vom  System  (2)  (8), 
und  daher  bildet  das  System  (10),    welches  im  Ganzen  2«— ?»+l 


*     Goursat,    Leçons   sur  l'intégration   des  équations  aux  dérivées   partielles  du    premier 
ordre.  §94. 


Charakt.  Mannigfaltigkeit  d.  part.  Differentialgleichungen.  Jl 

Gleichungen  mit  2n—2m+l  !  Integrationskonstanten  enthält, 
dieselbe  charakteristischen  Mannigfaltigkeiten,  welche  durch  das 
System  (2)  (8)  definirt  sind. 

Da  die   Gleichungen  (8)  nichts  anders  als  die  Gleichungen 
für  Extremalen  des  Variationsproblems 


sind,  bekommen  wir  den  folgenden  Satz: 

Die  charakteristischen  Mannigfaltigkeiten  eines 
Involutionssystems  der  partiellen  Differentialgleichun- 
gen erster  Ordnung 

Fi(z,  xi,  ■■■,  x„,p„  ...,2)„)  =  0 


FJ^>  Xly   ■'•,  X„,Pi,   ■■■,p„)  =  0 


sind  die  Extremalen  des   Variationsproblems,    dass  das 
Integral 

r'Ä{z'-Ip,xncU 

ZU  einem    Extremum    zu    machen,    dabei    die    zulässigen 
Funktionen  den  Nebenbedingungen 

unterworfen  sind,   und  ^  eine  geeignet  zu  bestimmende, 
nicht  mitvarierende  Funktion  von  t  ist.* 


III. 

Von  den  charakteristischen  Mannigfaltigkeiten  kann   man, 

*    Von  den  GHeichungen  (6)  und  Sju-  (-JT^)  +  ^Pi  =0  erhält  man  sofort  : 

àFu 


L=l       (  °~'  i  =  l  ) 


daher  folgt,  wegen  der  Nebenbedingungen,  z'  —  ^pi  xî  =0. 


J^2  T.  Yoshiye  : 

nach  Caucby  und  Lie,  die  allgemeine  Lösung  des  vorgelegten 
luvolutionssystems  ableiten.  * 

Wir  nehmen  nämlich  an,  dass  ein  Litegralgebilde  des  vor- 
gelegten Involutionssystems  (1)  für  die  Anfangswerte  x°^  •••,  x^ 
von  x^,  •••,  x„,  sich  auf  das  Gebilde 

"7=  *  (a:;,„+i,  •  •  •  a-J 

reduzirt.       Das   genannte    Integralgebilde    ist    dann    umgekehrt 
durch  das  letzte  Gebilde  im  Allgemeinen  eindeutig  bestimmt.** 
Für  dieses  Wertsystem  x°,  ■■■,  xj"  erhalten  wir 

die  Werte  ^^i,  ••■,'p,,,  lassen  sich  durch  die  Beziehungen 

F,=0,  F,=0,  •••,  F„=0 

bestimmen. 

Hierdurch  sind  längs  des  Gebildes  V=  $  alle  diese  Elemente 

L'^i  >  ■■■>  a%„ ,  x„,^i,  ■■■,  x„,  z,  pi,  "•",  2^„j 

an  einander  gereiht  und  wir  machen  jetzt  jedes  einzelne  Element 
zum  Ausgangspunkt  für  die  Konstruktion  einer  charakteristischen 
Mannigfaltigkeit. 

Alle  so  konstruirten  charakteristischen  Mannigfaltigkeiten 
bilden  zusammen  wieder  das  ganannte  Integralgebilde. 

Wenn  man  die  Funktion  «I»  als  willkürlich  betrachtet,  so 
erhält  man  die  allgemeine  Lösung  des  vorgelegten  Involutions- 
systems. 

Wir  haben  nun  zu  verifizieren,  ob  die  partiellen  Ableitungen 
der  eben  gewonnenen  Lösung  wirklich  mit  pi  der  charakteris- 
tischen Mannigfaltigkeiten  übereinstimmen.  Den  Beweis  dafür 
kann  man  ahnlicherweise  wie  in  Goursat's  ,, Leçons"  §50  führen. 

*     Goursat,  Leçons.  §  95. 
**    Goursat,  Leçons.  §  71. 


Charakt.  îlannigfaltjgkeit  d.  part.  Differentialgleichungen. 


13 


Braucht  man,   nämlich,   die  dort  benutzten  Bezeichnungen, 
dann  wird 

U  =  OZ—pi  OXi  —    ■■■    —  Pn  0X„ 

dU  =  fl{dz)  -  ^„„+1  d{dx,„^{)  -  ■■■  -  p„  d{dx„) 
—  dpi .  oxi  —  •  •  •  —  dp,,  '  o jr„ 

n 

=  2"  (opi,  '  dXi  —  dpi  •  oxj. 
Aus  den  Gleichungen  (8)  leitet  man  leicht  die  Beziehung  ab: 


Ip, 


dxi 


U^    u 


IF., 


u 


^F„ 

^P>. 
IF 

~ä7 


dx, 


-dU 


=  0, 


welche  sich  in  der  folgenden  Form  schreiben  lässt: 
dU 


ü 
Man  erhält  daher 


=  Zi  dxi  +  Zn  '  dx.,  +  •••  +  Z„,  dx. 


U=U, 


i  =  \J  0 


Falls  alle  Zi  endlich  bleiben,  kann  man,  da  U^  für  unsere  An- 
fangswerte gleich  Null  sein  muss,  daraus  schliessen 

C7=0, 
was  zu  beweisen  war. 

Tt'  F  •  ■  -F  'i 

Sollte  aber  weiterhin  durch  VerschAvinden  von  -~ H-  sich 

^{Pvp,n) 

eine  Schwierigkeit  ergeben,  so  bemerke  man,  dass  man  bei  In- 
tegration längs  einer  charakteristischen  Mannigfaltigkeit  statt 
(^1,  '",  ^m)  irgend  m  der  Variabelen 


^>  ^l>    ■■■>  S"«,  Pi,    •••,  Pn 


14 


T.  TosHye  : 


gerade  so  gut  als  unabhängig  ansehen  kann.  Eine  Schwierig- 
keit kann  also  nur  dann  eintreten,  wenn  auf  dem  gerade  betrach- 
teten Gebilde  eine  Stelle  erreicht  wird,  an  welcher  alle  Determi- 
nanten m-ieY  Ordnung  aus  der  Matrix 


im 


^'"û) 


verschwinden. 

Die  bisherige  Methode  gibt  also  kein  Integral,  für  welches 
alle  diese  Determinanten  gleichzeitig  verschwinden.  Ein  solches 
bezeichnet  man  als  singulare  Lösung. 

Wählt  man  jetzt  für  $  eine  bestimmte  Funktionsform,  so 
ergibt  sich  ein  bestimmtes  Integral gebilde.  Eine  Schar  von 
c»"""'  charakteristischen  Mannigfaltigkeiten  bildet  ein  Integi'alge- 
bilde.  Da  es  nun  im  Ganzen  00""-^'"+^  charakteristische  Mannig- 
faltigkeiten gibt,  so  haben  wir  Scharen  von  cc"""'"^^  Integralgebil- 
den, welche  man  die  vollständige  Lösung  nennt. 

Um  also  eine  vollständige  Lösung  zu  erhalten,  wählen  wir  für 
*  eine  bestimmte  Funktionsform  mit  n—m  +  1  wesentlichen  Para- 
metern. Die  so  gewonnene  Lösung  mit  n—m-\-l  Parametern  ist 
eine  vollständige  Lösung. 

Wir  wollen  nun  ein  Beispiel  geben,  welches  sich  leicht  durch 
diese  Methode  integrieren  lässt. 

Beispiel.  *     Es  sei  das  System 

p.2Pi—xiXi=0 


Goursat,  Leçons  p.  15ö. 


Charakt.  Mannigfaltigkeit  d.  p.xrt.  Differentialgleichungen.  15 


vorgelegt;   wir  bilden  daraus  das  Involutionssystem 

i>4 


Pi  -  ^^^^  =0, 


X.y 


p,-^  =0. 


I 


Das  Gleich ungensy stem  (2)  (8)  besitzt,  ausser  dieser  drei  die  drei 
folgenden  Integi'ale: 

PlPs-X2X^  =  ß, 

mit  den  Integrationskonstanten  «,  ß,  y. 

Nehmen  wir  für  die  Anfangswerte  die  folgenden: 

x,^=~l,        x:=l,        0^4°= 0, 

dann  erhalten  wir  die  Beziehungen  zwischen  den  Konstanten 

a  =  z  —  '2iXiPi, 

ß=—Xi, 

r--=x,{pj. 

Die    Gleichungen    für    charakteristische    Mannigfaltigkeiten 
lassen  sich  in  der  folgenden  Form  schreiben: 

^_      1        XiX<,  +  Xi(pi)- 


z  ^  z  + 

Pl= 
p 


2^1  a^s 


Pi 

Pi 

_  x^x^-\-Xi{p,Y 


Vi  •  X-z 


X,   ' 

Pi 

Pi=Pi  372 


-[ß  T.  Yoshiye . 

Um  die  allgemeine   Lösung  zu  erhalten  haben   wir   nur  zu 
setzen  : 

wo  <I>  eine  willkürliche  Funktion  und  *'  ihre  Ableitung  nach  Ici 
bezeichnet,   und  dann  sind  die  zwei  ersten  Gleichungen  aus  dem 
letzten  Systeme  als  die  allgemeine  Lösung  zu  betrachten,    wenn 
man  a\  als  Parameter  denkt. 
Setzt  man  speziell 

z  —  a  Xi  -r  b, 

SO   erhält   man  gerade  diejenige  vollständige  Lösung  wie  man  im 
,,Goursat's  Le<^ons"  p.  157  findet,  Avelche  lautet: 

z  =  — i— ^  +  a  X.,  Xi  +  b. 
a 


IV. 

Ein   spezieller    Fall   ist   die    Integration    eines   Involutions- 
systems linearer  partieller  Differentialgleichungen  erster  Ordnung.* 
Jede  Gleichung  ist  von  der  Form 

worin  Pj,  Pg,  •",  -P»,  P^  Funktionen  von  z,  x^,  ...,  x„  sind. 

Das    Gleichungensystem    von    (2)    und    (8)    sind    offenbar 
Beziehungen  zwischen  z,  x^,  ■■■  x„,  und  enthält  kein  p  drin. 

Schreibt  man  die  Integrale  dieses  Systems  in  der  Form 


Z   =  (p{Xi,   ■•■,Xn,z), 

und  setzt  man,  wie  früher, 


*    Vgl.  Gouisat,  Leçons.  §97.  U. 


Charakt.  Mainiij-faltigkeit  «1.  parr.  DilTiToutialg-leichiiugen.  jy 

dann  erhält  man  die  allgemeine  L(')^ung  des  vorgelegten   Systems 
in  der  Form 

Wenn   man,    im   x'\llgemeinen,    irgend    welclie  n  —  m  +  l    von 
einander  unabhängigen  Lösungen 


mit  den  Integrationskonstanten  c,,  ••.,  6'„_„,+,  findet,   dann  liefert, 
wie  man  leicht  sieht,  die  Gleichung 

*(f^,•■•,^^-..-.l)=0 

die  allgemeine  Lösung,  wobei  ^t  auch  eine  willkürliche   Funktion 
bezeichnet. 

Beisj^iel.    1.      Das  vorgelegte  S3^stem  sei 

+  Xo  *T~  JLa 

. .      P-i  =       ..      , 

,  ^1  "p  X--) 

Diese  beiden  Gleichungen  bilden  ein  Livolutionss^^stem.      Da  hier 

-^^^ — l-identi.-ch   verschwindet,   nehmen  wir  -Tg,  .Tg,   anstatt  .r,,  a-^, 

für  unabhängige  Variabein  an. 

Man   findet  leicht  die  folgenden  drei  Litegrale  vom  System 
(2)  und  (8):      ' 

iTj  —  a-.^  =  Konst. 
ir.,  —  .r^  =  Konst. 

-^  —  i\  .-r^  —  .r^  .T.,  =  Konst. 
Die  allgemeine  Lösung  des  vorgelegten  Sj'stems  lautet  also: 

2 

—s~  ~  ^i  ^i  ~  ^--i  ^'-i  —  *!■*  (•'^'i~"  A',  x.  —  x^). 
Beispiel  2.      Das  vorgelegte  System  sei 


18 


T.  Yoshiye  : 


Durch  Hinzufüguiig  der  Gleichung 

erhalten   wir  ein   vollständiges    System.     Löst    man    diese    Glei- 
chungen  nach  7^1,  ;>2,  ih  '"^^if,   dann  bekommt  man   das  Jacobische 

System  : 

Ih  +  (-ys  +  '^.Ti")  2^4=0, 
Vi  +  a:'2  2^4  =0, 
i?s  +  ^iP^  =0. 

Man  findet  leicht  die  zwei  Integrale 

z  =  Konst. 

Xi  +  -;^  ■\-  Xy_x^  —  Xi  =  Konst. 

z 

Die  allgemeine  Lösung  unseres  Systems  lautet  daher: 

X  ^ 

Z  =  *^{x^+^^-X^X^-X^). 

Ich  fühle  mich  Herrn  Geheimrat  Hubert  in  Göttingen  für  die 
wertvollsten  Ratschläge  zu  grossem  Danke  verpflichtet. 

*    Goursat,  Leçons,  p.  G2. 


Publ.  Aug.  25th,  1913. 


JOUENAL  OF  THE  COLLEGE  OF  SOENOE,  TOKYO  IMPERIAL  UNIVERSITY. 

VOL.  XXXII.,  ABTICLE  12 


On  Cyathocormus  mirabilis  nov.  gen.,  nov.  sp., 

the  Type  of  a  New  Family  of  Compound 

Ascidians  from  Japan. 

By 
Dr.  Asajiro  Oka. 


With  3  Plates  and  6  Text-figures. 


Although  the  Ascidiae  Salpaeformes  or  Luciae  have  all  along 
been  regarded  as  a  subdivision  of  the  Order  Ascidiacea,  the 
difference  between  the  free  swimming  Pyrosoma  and  an  ordinary- 
fixed  compound  ascidian  is  certainly  very  striking.  No  form  has 
as  yet  been  made  known  which  might  be  considered  intermediate 
between  these  two  groups.  Even  the  curious  unattached  colony, 
Coelocormus  huxleiji,  which  w^as  believed  by  Prof.  Herdman  to  be 
a  transition  form  between  them,  has  in  reality  nothing  in  common 
with  Pyrosoma  beyond  the  tubular  shape  of  the  colony.  On  the 
other  hand,  there  are  some  recent  investigators  who  regard 
Pyrosoma  as  more  closely  related  to  Salpa  and  Doliolum  than  to  the 
ascidians.  Neumann  ('09),  for  instance,  in  Beonn' s  Klassen  und 
Ordnungen  des  Tierreichs,  has  transferred  the  family  Pyrosomidae 
to  the  Thaliacea  in  which  it  constitutes  the  group  Synthaliacea  ; 
and  Paeker  and  Haswell  ('97  &  '10),  in  their  excellent  text- 
book, have  also  adopted  a  similar  view. 

The  new  genus  of  compound  ascidians  described  in  the 
present  paper  is  quite  unique  in  combining  the  characters  of  the 
compound  ascidians  and  the  Ascidiae  Salpaeformes.  In  shape  and 
general  appearance  of  the  colony  it  very  much  resembles  the 
latter,  while  it  agrees  with  some  of  the  former  in  the  possession  of 


2  Dr.  Asajiro  Oka  : 

a  peduncle  by  which  it  is  attached  to  some  foreign  body.  The 
colony  proper  or  head,  which  alone  is  composed  of  zooids,  has  the 
form  of  a  short  hollow  cylinder,  about  as  long  as  it  is  wide  and 
closed  at  one  end  where  it  joins  the  stalk.  Unlike  Fyrosoma  there 
is  no  diaphragm  at  the  open  end,  so  that  the  terminal  aperture  is 
of  the  same  width  as  the  central  cavity.  The  peduncle  is  short, 
columnar,  and  dilated  at  the  lower  extremity  to  form  a  base  of 
attachment.  The  entire  animal,  in  consequence,  is  so  perfect- 
ly cup-like  in  appearance  that  I  could  think  of  no  better 
generic  name  than  the  one  given  in  the  title  of  this  paper. 

The  zooids  forming  the  wall  of  the  hollow  cylinder  are 
arranged  in  vertical  lines  which  run  distinctly  in  pairs.  Looked 
at  from  inside  each  double  row  of  zooids  with  their  common 
investing  mass  is  found  to  form  a  cushion-like  longitudinal  ridge 
projecting  into  the  central  cavity  and  separated  from  its  neigh- 
bours on  both  sides  by  narrow  deep  furrows.  The  zooids  are 
imbedded  in  the  common  test  in  such  a  manner  that  the  branchial 
apertures  all  open  on  the  outer  surface  of  the  colony,  and  the 
atrial  cavities  all  communicate  directly  with  the  longitudinal 
furrows  just  mentioned,  which,  of  course,  are  nothing  but  the 
peripheral  portions  of  the  central  cavity.  There  are  no  definite 
atrial  apertures  present,  since  the  wall  of  the  peribranchial  cavity 
is  wanting  in  the  greater  part  of  the  thoracic  region  and  the 
stigmata  of  the  branchial  sac  as  well  as  the  anal  and  genital 
apertures  open  immediately  into  the  adjacent  longitudinal  furrow. 
Thus  the  central  cavity  serves,  as  in  the  case  of  Fyrosoma^  as  the 
common  cloaca  of  the  whole  colony  which  may  be  regarded  as 
consisting  of  a  single  system  of  zooids,  but  this  S3^stem  is,  in  the 
present  case,  divided  into  groups  of  zooids  whose  atrial  cavities 
are  partly  fused  together  to  form  a  large  continuous  peribranchial 
space. 

It  is  perfectly  obvious  that  a  compound  ascidian  with  such 
characters  as  those  mentioned  above  can  not  be  included  in  any 
of  the  recognized  families,  and  I  consider  myself  fully  justified  in 
forming  a  new  family  for  its  reception.  Since,  how^ever,  only  one 
genus  of  this  family  is  at  present  known,  I  believe  it  would  be  of 


Cyathocormus  mirabilis  n.  g, ,  n.  sp.  3 

no  use  to  consider  the  diagnostic  characters  of  the  family  and  the 
genus  separately.  They  are  naturally  the  same  and  may  be 
summed  up  as  follows  : 


Cyathocormus  nov.  gen. 

Colony  fixed,  stalked;  the  head  having  the  form  of  a 
short  hollow  cylinder  closed  at  one  end  and  open  at  the 
other. 

Systems — only  one  present,  the  terminal  aperture  being  the 
common  cloacal  opening,  and  the  central  cavity  the 
common  cloaca. 

Zooids  placed  in  a  single  layer  with  their  anterior  ends 
external  and  their  posterior  ends  internal.  Body 
divided  into  two  regions,  thorax  and  abdomen,  the  latter 
provided  with  a  long  vascular  appendage.  Branchial 
apertures  opening  on  tlie  outer  surface  of  the  colony,  not 
lobed.  Peribranchial  wall  imperfectly  developed,  stig- 
mata of  the  branchial  sac  as  well  as  the  anal  and 
genital  apertures  opening  directly  into  the  centrally 
placed  common  cloaca. 

Test  soft,  gelatinous,  and  transparent;  no  calcareous  spicules, 
bladder  cells  very  numerous. 

Branchial  sac  large  and  well  developed,  with  four  rows  of 
stigmata  ;  no  internal  longitudinal  vessels  present. 
Stigmata  very  long  and  narrow. 

Tentacles  simple. 

Dorsal  lamina  represented  by  a  series  of  languets. 

Alimentary  canal  forming  a  simple  loop  placed  posteriorly  to 
the  branchial  sac.  Stomach  ellipsoidal  and  smooth 
walled,  no  coecum.  Duodenal  portion  of  intestine  dis- 
tinctly bounded,  fine  tubular  intestinal  gland  Avell 
developed. 

Reproductive  organs  not  conspicuous.  Incubatory  pouch  pre- 
sent.    Larva  tailed. 


4  Dr.  Asajiro  Oka: 

It  will  be  seen  at  once  from  the  above  diagnosis  that  the 
present  genus  exhibits  in  certain  essential  points  a  close  relation- 
ship to  the  Pyrosomidae,  while  in  other  characters  no  less 
important  it  markedly  approaches  the  Distomidae  (Polycitoridae) 
among  the  compound  ascidians.  It  agrees  with  Pyrosoma,  on  the 
one  hand,  not  only  in  the  form  and  general  appearance  of  the 
colony  already  referred  to,  but  also  in  the  structure  of  the  alimentary 
canal,  which  is  practically  the  same  in  both  forms.  On  the  other 
hand,  the  presence  of  numerous  bladder  cells  in  the  test  naturally 
reminds  of  the  condition  found  in  the  genera  Colella  (=:Sycozoa) 
and  DistapUa  (^Holozoci)  among  the  Distomidae  (Polycitoridae), 
in  some  members  of  which  these  cells  form  by  far  the  greater  part 
of  the  investing  mass.  The  stalk,  too,  much  resembles  that  of 
some  species  of  the  genus  Colella.  e.g.  C.  quoyi  ;  but  this  is  unim- 
portant, as  similarly  stalked  forms  are  also  met  with  in  some  other 
families. 

One  of  the  most  notable  jDoints  about  this  form  is  without 
doubt  the  partial  fusion  of  the  peribranchial  cavities  in  a  number 
of  zooids.  So  far  as  I  am  aware,  there  is  but  one  member  of  the 
Ascidiae  compositae  in  which  all  the  zooids  exhibit  a  similar 
structure.  According  to  Lahille  ('90),  the  zooids  of  Diplosomoides 
lacazei  (= Polysyncraton  /.)  have  their  peribranchial  walls  very 
imperfectly  developed  so  that  the  greater  part  of  the  branchial 
sac  is  uncovered.  Since  other  species  of  the  genus  do  not 
present  the  same  condition,  this  character  ought  to  be  looked 
upon  as  simply  specific.  In  the  present  family,  however,  of 
which  no  other  genera  or  species  are  known  yet,  it  must  be  left 
doubtful  whether  this  peculiarity  should  be  regarded  as  a  family 
character  or  as  of  only  generic  or  even  specific  importance. 

This  genus  contains  the  single  species  C.  miraUlis. 

Cyathocormus  mirabilis  nov.  sp. 

The  colony  is  cup-shaped  and  is  attached  by  the  base  of  a 
short  peduncle.  The  head  has  the  form  of  a  short  hollow  cylinder 
nearly  as  long  as  it  is  wide.     The  outer  surface  shows  a  number  of 


Cyathocormus  mirabilis  n.  g.,  n.  sp.  5 

very  shallow  longitudinal  depressions  separating  as  many  slightly 
elevated  longitudinal  zones  upon  which  double  rows  of  very  short 
branchial  siphons  are  placed.  The  inner  surface  has  a  number  of 
deep  longitudinal  furrows  corresponding  to  the  shallow  depres- 
sions on  the  outer  surface.  In  the  ridges  separated  by  them  the 
dark  coloured  contents  of  the  intestines  look  conspicuously 
through  the  test  and  bod}^  wall.  The  stalk  is  shorter  than  the 
head,  columnar  in  shape,  and  is  dilated  at  the  lower  end  to  form 
the  base  of  attachment. 

Size — head  18 — 22  mm.  in  length  and  15 — '21  mm.  in  width; 
the  stalk  about  12  mm.  in  length  and  4 — 5  mm.  in  diameter. 

The  test  is  soft,  gelatinous,  and  perfectly  transparent  in  the 
head,  and  considerably  harder  and  more  or  less  opaque  in  the 
peduncle. 

The  zooids  are  of  moderate  size  and  are  entirely  imbedded  in 
the  common  test.  They  are  3 — 4  mm.  in  length  and  about  IJ 
mm.  in  greatest  breadth.  The  body  is  divided,  though  not  very 
distinctly,  into  thorax  and  abdomen,  the  latter  provided  with  a 
long  vascular  appendage. 

T]ie  mantle  is  very  thin,  muscle  fibres  being  found  only  in  the 
anterior  region  where  they  form  a  sort  of  sphincter  around  the 
branchial  aperture. 

The  Irœnchial  sac  is  well  developed  but  delicate.  The  trans- 
verse vessels  are  narrow  and  are  all  of  the  same  size.  The  stig- 
mata are  very  long  and  narrow  with  rounded  ends. 

The  endostijle  is  consj^icuous.  It  is  plicated  dorso-ventrally 
through  the  greater  part  of  its  length. 

The  dorsal  lamina  is  represented  by  three  short  pointed 
languets  projecting  from  the  dorsal  parts  of  the  transverse 
vessels. 

The  tentacles  are  usually  sixteen  in  number.  Four  are  long 
and  meet  in  the  centre  of  the  branchial  aperture  when  laid  flat; 
four  others  alternating  with  these  are  somewhat  shorter,  while 
the  remaining  eight  are  onl}^  half  as  long  as  the  former. 

The  dorsal  tubercle  is  very  prominent,  being  unusually  large 
in  proportion  to  the  size  of  the  branchial  sac.     The   aperture   is   a 


ß  Dr.  Asajiro  Oka: 

simple  oval  slit  placed  transversely  on  the  anterior  wall  of  the 
tubercle. 

The  alimentary  canal  forms  a  simple  loop  lying  almost  entirely 
behind  the  branchial  sac.  The  stomach  is  only  moderately  large 
and  is  rather  thick  walled.  The  duodenal  portion  of  intestine  is 
somewhat  pyramidal  in  shape  with  the  base  next  the  stomach,  its 
wall  is  very  thin.     Anal  aperture  without  lobes. 

The  reprochictive  organs  are  mostly  atrophied  in  fully  grown 
individuals.  A  large  oval  incubatory  pouch  containing  a  single 
larva  is  seen  alongside  the  branchial  region. 

There  are  at  present  only  two  examples  of  this  species  known 
to  me,  both  attached  to  a  colony  of  a  white  coral.  According  to 
the  statement  of  the  collector,  K.  Aoki,  they  were  obtained  in 
the  deeper  parts  of  the  Sagami  Sea  together  with  Euplectella, 
Metacrinus,  etc.,  but  the  label  having  been  lost,  the  exact  date  and 
locality  of  their  capture  can  not  be  ascertained.  A  brief  account 
of  these  colonies  with  remarks  upon  the  systematic  position  of 
the  species  was  given  by  me  at  the  meeting  of  the  Tokyo 
Zoological  Society  in  January  of  the  year  1907.  The  type  spec- 
imens, partly  injured  by  dissection,  are  deposited  in  the 
zoological  collections  of  the  College  of  Science,  Imperial  Univer- 
sity, Tokyo.  They  are  preserved  in  a  mixture  of  alcohol, 
glycerin,  and  water,  and  seem  not  to  have  lost  much  of  the 
original  transparency. 

Description  of  the  Colony. 

General  appearance.  The  colony  has  exactly  the  shape  of  a 
drinking  cup  (PI.  I.,  fig.  1.)  and  may,  for  convenience  of  descrip- 
tion, be  divided  into  the  stalk  and  the  head.  The  stalk  is  attached 
by  its  lower  extremity  to  some  foreign  body,  and  bears  the  head 
on  its  upper  end.  It  is  short  and  rather  thick.  It  is  of  equal  width 
throughout  from  the  point  of  junction  with  the  head  to  within  a 
short  distance  of  the  lower  end,  where  it  expands  to  form  a  base 
of  attachment.  The  diameter  of  the  base  measures  8 — 12  mm., 
that  is,  nearly  as  much  as  the  length  of  the  stalk  and  more  than 
twice  its  diameter  in  the  upper  part.     Its  surface  is  quite  smooth. 


Cyathocormus  uair.ibilis  n.  g.,  n.  sp.  Y 

showing  neither  longitudinal  striae  nor  transverse  wrinkles.  It  is 
almost  transparent  except  the  axial  portion  which  is  more  or  less 
opaque  and  whitish  with  a  tinge  of  pale  yellow,  especially  near 
the  lower  end.  The  stalk  is  made  up  entirely  of  the  test  sub- 
stance, and  no  zooids,  whether  adult  or  in  developmental  stages, 
are  found  imbedded  in  its  tissue. 

The  head  has  the  form  of  a  hollow  cylinder  open  at  one  end 
and  closed  at  the  other.  Viewed  from  the  side,  the  lateral  walls  of  the 
cylinder  are  nearly  straight.     At  the  open  end  the  wall  terminates 


I 


^/ 


^m-i. 


Text-fig.  1.  Text-fig.  2. 

Diagram  showing  the  longitudlinal         Diagram  showing  the  cross 
section  of  a  colong.  section,  of  a  colong. 

quite  abruptly,  there  being  no  thinning  out  of  the  investing 
mass  to  form  a  particular  marginal  zone  (text-fig.  1).  The 
bottom  of  the  cylinder  is  somewhat  lens-shaped,  convex  on  both 
outer  and  inner  surfaces.  It  is  at  the  centre  of  its  outer 
or  lower  surface  that  the  peduncle  joins  the  head.  Like 
the  stalk  this  part  is  entirely  composed  of  the  common 
test  and  contains  no  zooids.  Near  the  periphery  of  the  basal 
surface,  however,  a  number  of  very  small  zooids,  apparently  in 
early  stages  of  development,  are  to  be  seen.  The  side  wall  of  the 
cylinder  shows  on  its  external  surface  a  number  (12  and  16  in  the 
two  colonies  examined)  of  very  shallow  longitudinal  depressions 
separating  as  many  slightly  elevated  longitudinal  zones.  Under 
a  low  power  of  microscope  these  zones  are  found  each  to  be 
composed  of  a  double  row  of  hexagonal  areas  representing  the 


3  Br.  Asajiro  Oka: 

anterior  ends  of  zooids  with  a  short  Ijranchial  siphon  at  the 
centre.  Looked  at  from  inside,  each  double  row  of  zooids  forms 
a  cushion-like  ridge  projecting  into  the  central  cavity  and 
separated  from  its  neighbours  by  narrow  deep  furrows.  These 
furrows  correspond  in  position  to  the  shallow  longitudinal 
depressions  on  the  outer  surface  of  the  colony.  A  cross  section 
through  the  middle  of  the  head  has  therefore  the  shape  repre- 
sented in  text-fig.  2.  Jt  consists  of  thick  and  thin  portions  placed 
alternately  in  a  circle,  the  former  being  each  a  mass  of  the 
investing  substance  containing  two  rows  of  zooids,  while  the 
latter  are  nothing  but  a  thin  layer  of  the  common  test  forming 
the  bottom  of  the  deep  longitudinal  furrows  just  referred  to.  As 
the  test  and  body-wall  are  transparent  the  intestines  with  their 
dark  coloured  contents  stand  out  very  conspicuously  on  the  inner 
surface  as  dark  brownish  loops  (see  PI.  I.,  fig.  1).  Under  a  lens 
the  stomach  is  also  very  clearly  visible. 

Arrangement  of  zooids.  As  stated  before,  the  zooids  are 
arranged  in  longitudinal  rows  which  run  distinctly  in  pairs.  In 
one  of  the  two  colonies  examined  by  me  there  are  sixteen  of  such 
double  rows  present,  while  the  other  one,  which  is  somewhat 
smaller,  though  not  much  shorter,  has  only  twelve.  Each  row 
begins  at  the  very  margin  of  the  lateral  wall  of  the  head  and  ends 
on  the  basal  surface  at  some  distance  from  the  top  of  the  peduncle. 
The  number  of  zooids  which  form  a  longitudinal  row  is 
generally  twelve  or  thirteen,  not  counting  a  few  small  buds 
situated  on  the  basal  surface  of  the  head.  The  zooids  in  a  pair  of 
lines  are  placed  alternately,  and  as  the  amount  of  the  test  sub- 
stance intervening  between  them  is  proportionally  very  small,  the 
anterior  ends  of  the  zooids  are  visible  externally  as  hexagonal 
areas  with  the  branchial  siphon  projecting  from  the  centre.  The 
whole  external  surface  of  the  head  therefore  looks  like  a 
pavement  composed  of  equal  sized  polygons  (PI.  I.,  fig.  2)  in 
which,  however,  the  longitudinal  rows  are  grouped  in  pairs 
separated  from  one  another  by  narrow  zones  free  of  zooids. 

The  thoracic  regions  of  the  zooids  are  as  a  rule  placed  nearly 
at  right  angles  to  the  outer  surface  of  the  colony.      At  the  bound- 


CyathocormuS  mirabilis  n.  g.,  n.  sp.  0 

aiy  between  the  thorax  and  abdomen  the  body  is  generally  bent 
downwards,  i.e.,  towards  the  closed  end  of  the  cylinder,  in 
consequence  of  which  the  individuals  overlap,  so  that  the 
abdomen  of  one  zooid  covers  the  thorax  of  its  neighbour  below  on 
the  inner  surface  (PL  I.,  fig.  4).  For  the  same  reason  a  cross 
section  passing  through  the  branchial  sac  of  one  individual  cuts 
also  the  alimentary  canal  of  one  or  even  two  individuals  lying  above 
it  (PL  I. ,  fig.  3).  Moreover,  the  median  plane  of  each  zooid  does 
not  lie  parallel  to  the  main  axis  of  the  colony,  but,  as  shown  in 
PL  I.,  fig.  2,  the  zooids  in  a  pair  of  rows  have  their  endostyles 
turned  toward  one  another,  while  their  dorsal  edges  are  next  the 
space  between  that  row  and  the  adjacent  one.  In  this  respect  the 
arrangement  of  zooids  in  the  present  form  exactly  agrees  with  that 
found  in  some  species  of  Colella  (=Sijcozoa,  Less),  e.g.  C.  jmklira 
('86,  Herdman,  Challenger  Reports,  Part  2,  PI.  XV.,  fig.  6)  and 
G.  temiicaulis  ('99,  Herdman,  Australian  Museum  Catalogue,  PL 
Dist.  I.,  fig.  3).  As  will  be  described  further  on,  the  course  of  the 
intestinal  loop  is  not  the  same  in  all  the  individuals,  but  is 
different  according  to  the  position  of  the  zooid  in  a  double  row. 
Those  of  the  left  side  have  their  intestines  opening  into  the 
longitudinal  furrow  lying  on  the  left  side,  while  those  belonging 
to  the  right  hand  row  have  their  anal  apertures  looking  to  right, 
although  in  both  cases  the  intestine  is  bent  invariably  to  the  left 
side  of  the  stomach  (see  PL,  I.  fig. 3). 

Test.  The  test  is  only  weakly  developed  in  the  side  walls  of 
the  head.  Here  the  bodies  of  the  zooids  themselves  form  by  far 
the  greater  part  of  the  mass,  while  the  test  exists  only  in  the  form 
of  thin  covering  round  each  zooid.  In  some  places  it  forms  an 
exceedingly  thin  partition  between  the  zooids,  so  that  the  latter 
seem  to  be  separated  only  by  a  delicate  membrane.  Along  the 
median  line  of  the  longitudinal  ridges  formed  by  double  rows  of 
zooids,  however,  the  test  substance  is  found  to  take  the  shape  of 
more  or  less  thickened  masses.  On  the  outer  surface  of  the  head 
the  test  forms  a  thin  layer  of  nearly  uniform  thickness  (0,035- 
0,045  mm.)  covering  the  anterior  ends  of  the  zooids. 

The  extent  and  relative  amount  of  the  investing  mass  in  the 


10  ßr.  Asajiro  Oka  : 

wall  of  the  head  may  be  most  clearly  understood  by  examining 
the  transverse  and  longitudinal  sections  of  this  region.  In  PI.  I., 
fig.  3,  representing  a  part  of  cross  section  of  the  head,  the  test  is 
found  to  comprise  the  following  parts:  1)  a  superficial  layer  cover- 
ing the  outer  surface  of  the  head,  2)  thin  partitions  separating  the 
thoracic  regions  of  various  zooids,  and  3)  more  or  less  thickened 
masses  in  which  their  abdominal  regions  are  imbedded.  The 
superficial  layer,  which  is  nearly  of  the  same  thickness  through- 
out, not  only  covers  the  external  surface  of  the  anterior  ends  of 
the  zooids,  but  is  turned  inwards  at  the  branchial  aperture  and 
reaches  the  bases  of  the  tentacles  so  as  to  form  an  internal  lining 
of  the  branchial  siphon.  The  walls  of  test  substance  forming 
partitions  between  the  thoracic  regions  of  the  zooids  are  in  some 
places  much  thicker  than  the  superficial  layer,  but  in  other  places 
they  are  exceedingly  thin.  Their  thickness  depends  on  the 
configuration  of  the  zooids  the  interstices  between  which  they  fill 
up,  but  on  the  whole  the  test  is  not  particularly  well  developed 
in  these  parts.  It  is  in  the  inner  half  of  the  longitudinal  ridges 
that  the  common  test  attains  its  highest  development.  Here  it  is 
found  to  form  masses  of  more  or  less  considerable  thick- 
ness in  which  the  abdominal  parts  of  the  zooids  are  completely 
imbedded.  The  masses  also  contain  blood  vessels  to  be  described 
further  on.  The  extent  of  the  common  investing  mass  in  the 
longitudinal  ridges  is  perhaps  better  shown  in  PI.  I.,  fig.  4,  which 
represents  a  part  of  longitudinal  section  through  the  side  wall  of 
the  head.  From  this  figure  it  will  be  seen  that  the  posteror 
halves  of  the  abdomens  are  imbedded  in  a  continuous  mass  of 
test  substance  running  longitudinally,  Avhile  the  anterior  halves 
containing  the  oesophagus  and  stomach  are  covered  separately  by 
a  thin  layer  of  test.  Where  the  peribranchial  wall  is  wanting, 
as  shown  in  the  figure,  the  branchial  sacs  as  well  as  the 
anterior  portions  of  abdomens  are  separated  from  one  another 
simply  by  the  common  peribranchial  space  (p'"&.)  directly 
continuous  with  the  adjacent  longitudinal  furrow.  The  trans- 
verse walls  of  test  intervening  between  the  anterior  regions  of  the 
branchial   sacs  are   as  a  rule   of  the  same  thickness  as  the  super- 


Cyathocormus  mirabilis  n.  g.,  n.  sp.  2^1 

iicial  layer,  and  terminate  quite  abruptly  at  the  level  where  the 
first  row  of  stigmata  begins  to  appear.  As  already  motioned,  the 
basal  part  of  the  head  as  well  as  the  whole  peduncle  is  entirely 
composed  of  test  substance  which  is  somewhat  firmer  and  a  little 
more  opaque  than  that  of  the  side  w^alls  of  the  head. 

The  test  consists  of  a  hyaline  homogeneous  matrix  in  which 
two  kinds  of  cells  are  imbedded,  the  bladder  cells  and  the  test 
cells.  The  former  are  very  large,  oval,  spherical,  ellipsoidal,  or 
polygonal  in  shape,  and  have  only  a  thin  layer  of  protoplasm 
lining  the  inner  wall,  against  which  the  nucleus  lies  (PI.  I.,  fig.  5, 
h.c).  The  nuclei  are  veiy  distinct,  oval  or  fusiform  in  shape,  and 
are  always  situated  close  to  the  inner  wall  of  the  cell.  There  is 
often  a  little  mass  of  protoplasm  at  the  point  where  the  nucleus 
lies.  The  bladder  cells  are  exceedingly  numerous  throughout, 
forming  by  far  the  greater  part  of  the  test  tissue.  In  some  places 
they  are  so  numerous  that  the  matrix  exists  only  in  the  form  of 
delicate  membranes  surrounding  the  large  vesicles  which  have 
become  polygonal  by  mutual  pressure.  The  test  cells  are  found 
wherever  a  small  patch  of  matrix  can  be  seen  (PL  I.,  fig.  5,  t.c). 
They  are  small  (0.005 — 0.006  mm.  in  diameter),  oval  or  ellipsoidal 
in  shape,  and  consist  of  a  large  nucleus  surrounded  by  a  small 
quantity  of  protoplasm. 

In  the  superficial  layer  of  test  forming  the  external  covering 
of  the  head  the  tissue  consists  mainly  of  bladder  cells  generally 
arranged  in  a  single  layer  with  onty  a  little  mass  of  matrix  about 
them  (PL  I.,  figs.  7,9,10,11).  This  layer,  therefore,  is  throughout 
its  whole  extent  only  just  as  thick  as  the  greater  diameter  of  the 
bladder  cells.  The  walls  of  test  substance  intervening  between 
the  branchial  regions  of  the  zooids  are  in  some  places  much 
thicker  than  the  surface  layer  and  may  contain  a  large  number  of 
bladder  cells  grouped  irregularly,  but  in  other  places  they  are 
extremely  thin  and  are  entirely  composed  of  matrix,  being  in  fact 
too  thin  to  contain  a  bladder  celL  In  the  thickened  masses  of 
common  test  enclosing  the  abdominal  regions  of  the  zooids  the 
bladder  cells  are  also  abundant,  but  small  patches  of  matrix  are 
found  here  and  there,  and  in  such  places  the  test  cells  are  very 
clearly  visible  (PL  I. ,  fig.  5). 


12  Dr.  Asajiro  Oka: 

Discription  of  the  Zooid. 

Form.  As  the  zooids  are  attached  to  the  common  test  rather 
firmly  at  the  branchial  apertm'e  and  at  certain  regions  of  the 
thorax,  it  is  almost  impossible  to  dissect  out  a  zooid  uninjured 
from  the  investing  mass.  It  invariably  breaks  in  the  middle  of 
the  branchial  region  where  it  is  weakest,  since  the  body  wall  is 
here  very  imperfectly  developed  and  the  delicate  longitudinal 
vessels  of  the  branchial  sac  lie  exposed  on  the  outer  surface.  At 
other  places,  however,  the  body  wall  seems  to  be  only  loosely 
attached  to  the  common  test,  since  in  sections  of  preserved  speci- 
mens empty  spaces  are  commonly  found  surrounding  the  zooids, 
apparently  in  consequence  of  contraction  of  the  latter. 

The  body  of  the  zooid  may  be  divided  into  two  regions,  the 
thorax  and  the  abdomen.  They  are  about  of  the  same  size,  and 
the  boundary  between  them  is  not  very  distinct.  The  thorax 
comprises  chiefly  the  branchial  sac  and  its  accessory  organs,  while 
the  abdomen  is  composed  of  the  remaining  parts  of  the  alimentary 
canal  together  with  the  heart  and  the  reproductive  organs.  A 
vascular  appendage  springing  from  the  posterior  end  of  abdomen 
traverses  the  common  test  towards  the  base  of  the  colony.  The 
zooids,  without  the  appendage,  are  3 — 4  mm.  in  length  and 
nearly  half  so  much  in  breadth  ;  when  fresh  and  fully  expanded 
they  measured,  of  course,  somewhat  more. 

The  thorax  is  roughly  cylindrical  in  shape.  The  anterior  end 
is  only  slightly  convex  and  hexagonal  in  outline,  with  the  bran- 
chial siphon  placed  at  the  centre.  The  side  walls  of  the  thorax 
are  very  incomplete,  as  the  external  wall  of  the  peribranchial 
space  is  developed  only  on  the  ventral  side.  Dorsally  as  well  as 
laterally  there  is  no  body  wall,  and  consequently  the  branchial 
sac  is  quite  exposed  on  these  sides  (Pl.  I.,  fig.  3).  This  condition 
may  perhaps  be  considered  as  the  result  of  the  atrial  aperture 
having  become  disproportion  ally  widened  so  as  to  occupy  the 
gi'eater  part  of  the  surface  of  the  thoracic  region.  As  mentioned 
before,  there  is  only  one  species  of  compound  ascidians  in  which  a 
similar  condition  is  met  with.  Lahille,  in  his  "Recherches 
sur  les  Tuniciers  "  ('30)  describes  the  thoracic  region  of  the  zooids 


Cj-athocormus  mirabilis  n.  g.,  n.  sp.  13 

of  Diplosomoides  lacazel  Giard  in  following  words  :  "La  paroi 
péribranchiale,  fort  réduite,  laisse  la  branchie  presque  entièrement 
à  découvert  ",  and  this  is  exactly  what  takes  place  in  our  species, 
though  very  probably  we  have  here  a  case  of  convergence  rather 
than  that  of  a  direct  phylogenetical  affinity.  The  posterior  limit 
of  the  thorax  is  marked  for  the  most  part  by  the  bottom  of  the 
branchial  sac,  at  one  point  of  which  the  funnel-shaped  oesophagus 
commences.  The  terminal  portion  of  intestine  lying  just  beneath 
the  branchial  sac  is  also  comprised  in  the  thorax  (PL  II.,  fig.  8). 

The  abdomen  is  irregularly  ellipsoidal  and  somewhat  flattened 
from  side  to  side.  Its  shape  is  determined  principally  by  that  of 
the  intestinal  loop,  since  the  heart  and  other  organs  contained  in 
the  abdomen  are  all  placed  between  the  stomach  and  ascending 
part  of  the  intestine.  As  the  investing  mass  as  well  as  the  thin 
body-wall  is  perfectly  transparent  and  colourless,  the  external 
form  of  the  abdomen  is  hardly  recognizable  to  the  naked  eye, 
while  the  intestine  itself  is  rendered  very  conspicuous  by  its  dark 
coloured  contents.  The  vascular  appendage,  whose  lumen  is 
continuous  with  the  blood  spaces  of  the  zooid,  starts  from  the 
posterior  end  of  the  abdomen  and  runs,  as  will  be  more  minutely 
described  hereafter,  inwards  and  dow^nw^ards  through  the  common 
investing  mass  to  the  base  of  the  colony. 

Body  ivall.  The  body  w'all  or  mantle  forms  the  outer  covering 
of  zooids  and  lies  immediately  under  the  test,  with  which,  however, 
it  is  in  the  greater  part  of  its  extent  not  closely  connected.  In 
preserved  specimens  a  cavity  in  which  the  zooid  lies  is  distinctly 
visible  (PI.  I.,  fig.  3;  PL  IL,  figs.  7,  8;  PL  IIL,  fig.  20,  sp.).  In 
the  living  animal,  however,  when  the  contractile  tissues  were 
relaxed  and  the  viscera  were  filled  with  water,  the  body  Avail  was 
doubtless  in  contact  with  the  test  throughout,  and  there  was 
therefore  no  cavity  around  the  zooid.  In  those  places  where  the 
test  ends  with  a  free  margin,  the  body  w^all  is  more  firmly 
attached  to  the  test  and  remains  closely  connected  with  it  even 
when  preserved.  Such  places  are  found  at  the  base  of  the  branchial 
siphon  and  along  the  border  of  the  body  wall  in  the  thoracic 
region. 


1  ^  Dr.  Asajiro  Oka  : 

The  body  wall  is  composed  inainl}^  of  three  elements — the 
ectodermal  epithelium,  the  connective  tissue,  and  the  muscular 
fibres.  The  ectodermal  epithelium  is  present  throughout  the 
whole  extent  of  the  mantle  and  is  in  direct  contact  with  the  test. 
The  cells  of  this  layer  are  generally  flattened  (PL  II.,  fig.  7,  ect.), 
especially  so  in  the  walls  of  abdomen  where  they  form  a  thin 
pavement  epithelium,  but  at  the  margin  of  the  branchial  siphon 
they  are  considerably  thicker  and  almost  cubical  in  shape  (PL  II., 
fig.  9,  ect.).  The  connective  tissue  is  present  in  the  form  of  a 
hyaline  homogeneous  mass  containing  scattered  cells  and  per- 
forated by  blood  spaces.  The  cells  are  fusiform  or  stellate  in  form 
and  have  distinct  nuclei.  The  stellate  cells  are  found  generally  in 
such  places  where  the  layer  of  connective  tissue  is  rather  thick, 
and  are  then  scattered  equally  all  over.  Where  the  connective 
tissue  forms  a  thin  layer  more  fusiform  cells  are  met  with,  some- 
times with  their  longer  axes  perpendicular  to  the  plane  of  the 
adjacent  ectodermal  epithelium. 

The  musculature  is  on  the  whole  very  feebly  developed. 
Over  the  abdomen  not  a  fibre  is  visible,  and  on  the  thorax  the 
body  wall  has  only  a  few  scattered  fibres  here  and  there  except  on 
the  anterior  surface  where  they  are  more  numerous.  In  the  wall 
of  the  branchial  siphon  the  fibres  are  placed  transversely,  forming 
a  sort  of  sphincter  round  the  external  opening.  Longitudinal 
fibres  are  also  present,  but  are  much  less  numerous.  Compact 
bands  of  muscle  fibres,  such  as  are  frequently  met  with  in  other 
genera,  do  not  occur  in  any  part  of  the  body  wall. 

Branchial  sac.  This  organ  occupies  tlie  greater  part  of  the 
thorax,  and  is  only  partly  covered  by  the  body  wall.  When  fully 
expanded  it  is  somewhat  barrel-shaped  and  extends  from  the 
anterior  wall  to  the  basal  part  of  the  thorax  (PL  I.,  fig.  3,  hr.  s.) 
In  most  zooids  examined  by  me  the  branchial  sac  was  strongly 
contracted,  and  it  was  with  great  difficulty  that  the  exact  form  and 
structure  of  the  organ  could  be  made  out.  In  a  few  exceptional 
cases,  however,  it  could  be  satifactorily  observed  under  a  dissect- 
ing microscope. 

The  endostyle  runs  along  the  ventral,  and  the  row  of  languets 


Cyathocornius  mirabilis  n.  g.,  n.  sp.  15 

along  the  dorsal  edge  of  the  branchial  sac,  while  the  circlet  of  tenta- 
cles, the  dorsal  tubercle,  and  the  peripharygeal  band  are  situated 
at  its  anterior  extremity  (PL  II.,  fig.  8).  All  these  organs  will  be 
described  further  on.  Very  often  one  or  two  small  copepod  crus- 
taceans were  found  living  in  the  branchial  cavity  as  commensals. 

The  stigmata  occur  over  the  whole  extent  of  the  sac,  from 
the  peripharyngeal  band  anteriorly  to  the  base  where  the  oeso- 
phagus opens  posteriorly,  with  the  exception  of  a  narrow  band 
along  the  dorsal  and  ventral  edges,  where  the  median  dorsal  vessel 
and  endostyle  are  placed.  They  are  arranged  in  horizontal 
(transverse  to  the  aiitero-posterior  axis)  rows,  and  are  separated  by 
transverse  and  longitudinal  vessels  (PL  II.,  fig.  12).  There  are 
four  such  rows  and  more  than  twelve  stigmata  in  a  row  on  each 
side.  In  shape  they  are  elongated  slits  with  parallel  sides  and 
rounded  ends.  The  width  of  the  stigmata  is  about  equal  to  that 
of  the  fine  longitudinal  vessels  separating  them. 

Examined  under  a  moderate  power  of  microscope,  each 
longitudinal  vessel  is  found  to  have  the  form  of  a  rectangular 
pillar,  the  sides  of  which  are  made  of  two  kinds  of  cells.  On  the 
sides  facing  the  branchial  and  atrial  spaces  the  wall  is  very  thin 
and  is  composed  of  flattened  polygonal  cells.  On  the  sides  sur- 
rounding the  stigmata,  on  the  contrary,  the  cells  are  of  consi- 
derable height.  Seen  from  the  interior  or  exterior  of  the 
branchial  sac  each  of  the  stigmata  appears  to  be  surrounded  by  a 
single  row  of  ciliated  cells  (PL  II.,  fig.  12).  A  transverse  section  of 
the  longitudinal  vessels,  however,  shows  most  clearly  that  what 
looked  like  a  single  cell  is  in  reality  a  group  of  very  narrow  cells 
placed  side  by  side  (PI.  II.  fig.  13),  and  consequently  what 
appeared  as  a  single  row  of  cells  is  in  reality  a  longitudinal  band 
of  ciliated  epithelium  extending  down  the  side  of  the  vessel.  The 
ciliated  cells  vary  somewhat  in  shape,  being  taller  and  more 
columnar  at  the  ends  of  the  stigmata.  The  most  common  form  is 
nearly  semicirculer,  the  flat  face  being  next  the  vessel  and  the 
curved  one  next  the  stigma.  As  they  are  grouped  regularly  in 
transverse  rows  they  look  somewhat  like  a  pile  of  coins  cut  in 
halves.     The  nuclei   are   distinct    and   placed   rather   below   the 


lg  Dr.  Asajiro  Oka: 

middle  of  the  cells.  The  cilia  are  long  and  delicate.  They  are 
from  five  to  ten  on  each  cell,  attached  to  the  more  or  less  convex 
outer  edge,  and  when  fully  extended  they  are  more  than  twice 
the  height  of  the  cells. 

The  transverse  vessels  are  three  in  number  and  are  all  of  the 
same  size.  They  are  about  twice  as  wide  as  the  fine  longitudinal 
vessels  and  contain  delicate  muscle  fibres  (PL  II.,  fig.  12,  tr.  v.). 
When  contracted  the  branchial  sac  is  always  strongly  constricted 
along  these  vessels.  The  zone  devoid  of  stigmata  along  the 
dorsal  edge  of  the  branchial  sac  is  very  narrow,  so  that  the  rows  of 
stigmata  of  one  side  appear  to  pass  over  to  the  opposite  side 
without  much  interruption.  The  median  dorsal  vessel  also 
contains  muscle  fibres  in  its  walls  and  is  found  strongly  contracted 
in  most  zooids. 

Endostyle.  The  endostyle  runs  along  the  ventral  edge  of  the 
branchial  sac  (PL  IL,  fig.  8).  It  begins  anteriorly  at  the  base  of 
the  branchial  siphon  immediately  behind  the  peripharygeal  band 
and  runs  backwards  to  the  base  of  the  sac,  ending  at  a  short 
distance  from  the  oesophageal  aperture.  The  ends  are  bluntly 
conical,  otherwise  it  is  of  the  same  width  throughout.  In  the 
greater  part  of  its  course  it  forms  a  series  of  minute  dorso- 
ventral  undulations.  Whether  or  not  this  is  a  result  of  contrac- 
tion of  the  branchial  sac  I  am  not  able  to  say.  At  any  rate,  in  all 
the  zooids  examined  by  me  the  endostyle  invariably  showed  this 
character.  The  anterior  end  of  the  endostyle  is  turned  inw^ards 
forming  a  sort  of  blind  sac  at  the  point  of  bending  (PL  IL,  fig.  7, 
end.),  so  that  in  certain  sections  passing  through  this  region  the 
endostyle  appears  as  a  canal. 

The  endostyle  is  a  groove  with  greatly  thickened  sides 
formed  of  columnar  epithelium,  w^hile  the  base  is  covered  with 
somewhat  shorter  cells  (PI.  IL,  fig.  14).  The  summits  of  the 
edges  are  continued  up  as  lip-like  folds.  A  tract  along  each 
side  and  the  base  are  richly  ciliated.  As  shown  in  the  figure 
several  kinds  of  cells  can  be  distinguislied  in  the  wall  of  the 
endostyle.  Those  forming  the  base  stain  deeply  with  haemalaun 
and  are  provided  with  very  long  cilia,    reaching,   if  not  exceed- 


Cyathocormus  mirabilis  n.  g.,  n.  sp.  \'J 

ing,  the  free  edge  of  the  lateral  lips.  On  each  side  of  the 
base  there  is  a  zone  of  large  columnar  cells  without  cilia  which 
stain  only  faintly  and  seem  to  be  glandular  in  nature.  Beyond 
this  glandular  zone  the  cells  are  again  ciliated,  but  the  cilia 
are  very  short,  and  the  cells  bearing  them  become  successively 
shorter  as  they  are  traced  away  from  the  base.  On  the  inner 
surface  of  the  lips  there  is  again  a  zone  of  ciliated  epithelium 
composed  of  cubical  cells.  The  cilia  are  very  short.  On  tracing 
these  zones  anteriorly  they  are  found  to  be  continuous  with  the 
band  of  ciliated  epithelium  covering  the  peripharyngeal  ridge  to 
be  described  below.  Seen  from  the  dorsal  or  ventral  aspect, 
the  endostyle  shows  a  pair  of  thick  semi-opaque  bands  separated 
by  a  more  translucent  area  (PI.  I.,  fig.  6,  end.).  The  opaque 
bands  are  caused  by  the  thickened  sides,  separated  by  the  less 
massive  floor  of  the  groove.  The  narrow  dark  lines  seen  along 
both  sides  of  the  endostyle  are  the  zones  of  ciliated  epithelium 
on  the  lateral  lips  of  that  organ. 

Dorsal  lamina.  The  dorsal  lamina  is  represented  by  a  series 
of  three  short  triangular  languets  springing  from  the  dorsal  edge 
of  the  branchial  sac.  Each  languet  is  placed  at  the  intersection 
with  a  transverse  vessel  and  is  ciliated  at  the  sides.  They  seem 
to  arise  directly  from  the  inner  wall  of  the  vessel,  there  being 
no  ridge  or  membrane  present  to  unite  their  bases. 

Tentacles.  The  tentacles  are  placed  in  a  circle  round  the  base 
of  the  branchial  siphon  just  at  the  entrance  of  the  branchial  sac. 
The  branchial  siphon  is  a  short  funnel  with  the  rim  turned  out- 
wards, and  when  wide  open  its  diameter  is  about  double  its  height. 
Its  inner  surface  is  lined  by  an  invagination  of  the  superficial  layer 
of  test,  which  is  nearly  of  the  same  thickness  as  that  layer  and 
extends  as  far  doAvn  as  the  tentacular  circlet. 

The  tentacles  are  simple  and  rather  stout.  When  directed  up- 
Avards  they  project  beyond  the  external  opening  of  the  branchial 
siphon  (PL  IL,  fig.  8,  10).  They  are  generally  sixteen  in  number 
and  are  of  three  different  sizes.  Four  are  long  and  meet  in  the 
centre  of  the  branchial  aperture  even  when  the  latter  is  wide  open. 
Four  others  alternating  with  these   are   somewhat   shorter,    while 


IQ  Dr.  Asajiro  Oka  : 

the  remaining  eight  are  only  half  as  long  as  the  longest  ones.  Two 
of  the  longest  tentacles  are  placed  in  the  median  plane  of  the  zooid, 
the  other  two  making  right  angles  with  them.  At  the  level  of 
their  attachment  a  sphincter  composed  of  loose  circular  muscle 
fibres  encircles  the  base  of  the  branchial  siphon  (PL  II.,  figs.  9  & 
10). 

Each  tentacle  is  attached  separately  and  has  a  round  tapering 

and  generally  curved  stem  and  a  rather  blunt  apex.  The  wall  of 
tentacle  is  formed  externally  by  an  epithelium  of  flattened  or 
cubical  cells  and  is  nowhere  ciliated.  The  interior  of  the  tentacle 
is  filled  with  the  homogeneous  matrix  of  connective  tissue  with 
some  scattered  cells.  A  few  delicate  longitudinal  muscle  fibres  are 
sometimes  distinguishable  under  the  epithelial  layer.  Whether 
the  blood  spaces  are  continued  into  the  interior  of  the  tentacles, 
though  very  probable,  could  not  be  ascertained. 

Periphanjngeal  hand.  In  the  present  species  the  peripharyn- 
geal band  is  remarkably  well  developed,  resembling  in  many 
respects  that  of  simple  ascidians.  It  consists  of  two  parts,  a  thin 
membrane  and  a  ciliated  ridge  (PL  IL,  fig.  9,  piyh.  m.,  prph.  r.) 
For  the  greater  part  of  their  course  they  run  close  to  and  parallel 
with  each  other,  forming  between  themselves  a  narrow  groove,  the 
"gouttière  péricoronale  "  of  French  authors.  They  encircle  the 
top  of  the  branchial  sac  at  a  short  distance  from  the  bases  of 
tentacles.  The  peripharyngeal  membrane,  called  the  "  Ring- 
falte  '  '  b}^  German  writers,  is  a  thin  membrane  of  nearly  equal 
breadth  throughout,  and  is  composed  of  flattened  cells  without 
cilia.  At  the  ventral  median  line  the  membranes  of  both  sides 
meet  each  other  at  the  pointed  anterior  extremity  of  the  endostyle. 
Dorsally  the  membrane  reaches  the  dorsal  tubercle  where  it  is 
continued  for  some  distance  along  the  lateral  region  of  that  organ 
(PI.  II.,  fig.  11,  j)rp]i.  m.).  Throughout  its  whole  extent  the 
breadth  of  the  peripharyngeal  membrane  is  about  equal  to  the 
height  of  the  ciliated  ridge  forming  the  other  lip  of  the  groove. 

The  peripharyngeal  ridge  is  composed  of  a  band  of  ciliated 
epithelium  running  parallel  with  the  peripharyngeal  membrane  on 
its  posterior  side.     It  is  connected  at  its  ventral  and  dorsal  ends 


Cyathocormus  mirabilis  n.  g.,  n.  sp.  lg 

with  the  anterior  extremity  of  the  endostyle  and  the  dorsal  tubercle 
respectively.  By  tracing  the  ridge  ventrally  it  is  found  to  be 
directly  continuous  with  the  ciliated  band  on  the  inner  surface  of 
the  ]ip  of  endostyle.  Near  the  dorsal  end  the  ridge  forms  the 
greater  part  of  the  side  wall  of  the  dorsal  tubercle  and  gradually 
disappears  towards  its  free  extremity.  The  cells  composing  the 
ridge  are  high  and  columnar,  and  the  cilia  are  very  short  through- 
out. 

Dorsal  tubercle.  This  organ  is  placed  at  the  anterior  end  of 
the  dorsal  edge  of  the  branchial  sac,  and  is  very  conspicuous  on 
account  of  its  unusually  large  size.  As  shown  in  PL  IL,  figs.  8  and 
10,  it  projects  into  the  branchial  cavity  just  below  the  base  of  the 
branchial  siphon  on  the  dorsal  side  and  is  clearly  visible  from 
outside  when  the  branchial  orifice  is  moderately  open.  It  is 
conical  in  shape,  with  blunt  apex,  and  attached  obliquely  to  the 
anterior  wall  of  the  thorax  in  such  a  manner  that  the  apex  is 
directed  towards  the  axis  of  the  body.  The  aperture  of  the 
neural  gland  is  situated  on  its  anterior  surface  about  half  way 
between  the  apex  and  the  base  of  the  organ  where  it  joins  the  base 
of  the  median  dorsal  tentacle  (PI.  II.,  fig.  10).  The  aperture  is 
simple,  oval  in  outline,  and  is  placed  transversely  in  reference  to 
the  axis  of  the  organ.  As  the  peripharyngeal  ridge  is  continuous 
with  the  ciliated  patch  covering  the  apex  of  the  dorsal  tubercle, 
and  the  peripharyngeal  membrane  almost  reaches  the  anterior 
surface  of  the  organ  before  it  is  lost  to  sight,  these  two  components 
of  the  peripharyngeal  band  well  deserve  the  names  of  "anterior 
and  posterior  lips'  ' ,  as  they  are  sometimes  called  in  simple  ascidians. 
At  the  apex  and  on  both  sides  the  wall  of  the  organ  is  made  up 
of  cubical  or  columnar  cells.  In  the  immediate  neighbourhood  of 
the  aperture  the  cells  are  rather  high  and  ciliated.  In  other 
places  the  tubercle  is  covered  with  an  epithelium  of  flattened 
cells. 

Nerve  ganglion.  The  nerve  ganglion  is  in  its  usual  position  on 
the  dorsal  side  of  the  branchial  sac  immediately  beneath  the  ecto- 
dermal epithelium  of  the  body  wall.  It  is  oval  or  elliptical  in 
outline  and  sends  out  nerve  trunks  both  anteriorly  and  posteriorly. 


2Q;  Dr.  Asajiro  Oka  : 

The  nuclei  of  the  cells  are  found  mostly  in  the  peripheral  parts, 
the  centre  being  mainly  composed  of  fibrous  substance  (PL  II. ,  fig. 
10).  The  longer  diameter,  which  is  placed  antero-posteriorly, 
measures  almost  0, 1  mm. 

Neural  gland.  The  neural  gland  lies  immediately  beneath  the 
nerve  ganglion  and  is  nearly  of  the  same  size  as  that  organ.  It  is 
bladder-like  in  structure  with  uniformly  thick  wall  and  a  large 
cavity  in  the  interior.  The  wall  is  composed  throughout  of  a 
single  layer  of  cells,  the  cytoplasm  of  which,  however,  seems  not 
to  be  clearly  bounded,  forming  rather  a  sort  of  syncytium.  The 
dorsal  portion  of  the  wall  is  in  close  contact  with  the  nerve  gan- 
glion, but  the  boundary  between  them  is  everywhere  quite  distinct, 
and  there  is  no  area  in  which  the  tissues  of  the  two  organs 
gradually  pass  over  to  each  other^  (PI.  IL,  fig.  10). 

The  central  cavity  of  the  neural  gland  communicates  with 
the  exterior  by  means  of  a  short  duct  which  opens  on  the  anterior 
surface  of  the  dorsal  tubercle.  This  duct  is  of  the  same 
width  throughout  except  the  terminal  portion  which  is  somewhat 
widened  so  as  to  form  a  sort  of  funnel.  At  the  very  margin  of  the 
opening,  however,  the  duct  is  again  a  little  constricted  (PL  IL,  fig. 
10).  The  wall  of  the  duct  is  made  up  of  cubical  cells  all  over. 
In  the  terminal  portion  of  the  duct,  where  it  is  widened,  the  cells 
are  provided  with  long  vibratile  cilia. 

A  cross  section  though  the  middle  part  of  the  dorsal  tubercle 
is  shown  in  PL  IL ,  fig.  11.  At  the  centre  of  the  figure  the  ciliated 
terminal  portion  of  the  duct  of  the  neural  gland  is  seen  cut  across. 
On  both  sides  the  wall  of  the  dorsal  tubercle  is  formed  for  the 
greater  part  by  an  epithelium  of  ciliated  columnar  cells,  which  is 
the  direct  continuation  of  the  epithelial  layer  of  the  peripharyngeal 
ridge  mentioned  above.  The  peripharyngeal  membrane  is  also 
seen  immediately  on  the  dorsal  side  of  this  epithelium.  The 
posterior  wall  of  the  dorsal  tubercle  is  formed  of  a  thin  epithelium 


1  According  to  M.  M.  Metcalf  ('00),  who  made  a  special  study  of  the  intersiphonal  organs 
of  various  groups  of  Tunicata,  there  is  in  all  the  compound  asoidians  a  region  where  the  cells  of 
the  ganglion  and  the  neural  gland  merge  into  one  another  and  are  indistjnguishaole. 


Cyathocormus  mirabilis  n.  g.,  n.  sp.  21 

of  simple  flat  cells.  The  undulations  of  this  epithelium  shown  in 
the  figure  are,  I  l>elieve,  due  to  contraction  incident  to  preserva- 
tion. 

Aliiucntarij  canal.  The  alimentary  canal  is  contained  partly  in 
the  thorax  and  partly  in  the  abdomen.  Excluding  the  branchial 
sac,  which  is  the  foremost  portion  of  the  canal,  it  may  be  divided 
into  three  parts,  the  oesophagus,  the  stomach,  and  the  intestine 
(PL  III.,  fig.  15). 

The  oesophageal  opening  is  placed  at  the  posterior  end  of  the 
branchial  sac,  nearer  the  dorsal  than  the  ventral  edge  (PI.  IL,  fig. 
8,  oe.).  It  is  separated  from  the  hind  end  of  the  endostyle  by  a 
narrow  space  where  the  bottom  of  the  branchial  sac  is  lined  with 
thin  epithelium.  As  shown  in  the  figure,  the  terminal  portion  of 
intestine  lies  immdiately  beneath  this  part  in  close  contact  with 
the  oesophagus.  The  oesophagus  is  rather  short  and  funnel-shaped, 
with  the  wajl  thrown  into  irregular  longitudinal  folds  (PL  III.,  fig. 
15).  Generally  the  course  of  the  oesophagus  is  a  little  curved. 
The  anterior  margin  of  its  wall  is  somewhat  raised  and  forms  a 
distinct  ridge  or  lip  (PL  III.,  fig.  18).  Posteriorly  the  oesophagus 
projects  into  the  cavity  of  the  stomach  considerably  and  seems  to 
perform  the  function  of  a  valve  (PL  III.,  fig.  16).  The  wall  of 
the  oesophagus  is  lined  with  columnar  epithelium  and  is  finely 
ciliated  all  over.  In  sections  stained  with  haemalaun  the  zone 
formed  by  inner  halves  of  these  cells  is  coloured  dark  blue  and 
stands  out  very  conspicuously.  In  the  part  projecting  into  the 
interior  of  the  stomach  the  cells  do  not  show  any  such  differen- 
tiation. 

The  stomach  is  oval-shaped  with  smooth  and  uniformly 
thick  Avail.  The  anterior  end  of  the  organ  is  rounded  with  an 
invagination  at  the  centre,  forming  a  sheath  for  the  oesophagus 
which  proceeds  a  short  distance  into  the  interior  (PL  III.,  fig.  16). 
The  posterior  end,  on  the  contrary,  terminates  quite  abruptly  and 
without  any  constriction.  The  stomach,  therefore,  is  comparable 
in  form  with  an  egg  from  which  the  posterior  third  (next  the 
pointed  end)  has  been  cut  away.  The  wall  is  of  considerable 
thickness   being  lined    with    an  epithelium  of  tall  columnar  cells. 


22  Dr.  Asajiro  Oka: 

It  is  perfectly  smooth  both  externally  and  internally,  and  no  folds 
or  thickenings,  such  as  are  frequently  met  with  in  other  compound 
ascidians,  are  to  be  seen.     The  stomach  is  nowhere  ciliated. 

The  intestine  begins  at  the  posterior  end  of  the  stomach,  runs 
backwards  for  a  short  distance,  then  turns  to  left  and  runs  for- 
wards till  it  reaches  the  bottom  of  the  branchial  sac;  here  it  is 
curved  either  to  right  or  to  left  and  opens  into  the  longitudinal 
furrow^  separating  the  double  rows  of  zooids.  xV  short  tract  of  the 
intestine  lying  immediately  behind  the  stomach  and  clearly 
bounded  by  a  costriction  posteriorly  differs  from  the  rest  of  the 
canal  in  many  respects,  and  may,  for  the  sake  of  convenience,  be 
distinguished  as  "duodenum."  It  corresponds  exactly  to  the 
"  post-estomac  "  of  some  French  authors  and  the  "  Mitteldarm  " 
of  German  writers.  The  remaining  portion  of  the  canal  exhibits 
nearly  the  same  character  throughout,  though  it  may  conveniently 
be  divided  into  three  regions  in  referrence  to  the  area  occupied  by 
the  intestinal  gland. 

The  duodenal  part  of  intestine  has  the  shape  of  a  truncated 
cone  attached  by  the  broad  base  to  the  posterior  end  of  the 
stomach.  The  wall,  which  is  generally  a  httle  wrinkled,  is  lined 
with  a  layer  of  cubical  cells  without  cilia  and  stains  very  deeply 
with  haemalaun.  The  alimentary  canal  has  thinnest  wall  in  this 
part,  and,  as  the  stomach  lying  immediately  in  front  of  it  is  the 
thickest  walled  part  of  the  canal,  the  contrast  in  thickness  of  the 
Avails  is  very  striking. 

The  intestine,  exclusive  of  the  duodenum,  may  be  divided 
into  three  regions  of  nearly  equal  length.  The  first  third  lying 
next  the  duodenum  occupies  the  hindmost  region  of  the  abdomen 
and  is  bent  somewhat  in  the  form  of  the  letter  U.  Next  comes 
the  middle  third  which  is  distinguished  by  the  presence  of  the  fine 
tubular  intestinal  gland  covering  the  outer  surface  of  the  wall. 
The  last  third,  which  may  be  called  the  rectum,  is  simply  tubular 
in  form  and  opens  by  the  anal  aperture  into  the  common  peribran- 
chial  space.  In  microscopical  structure  of  the  wall,  however, 
there  is  no  noticeable  difference  between  these  parts.  The  course 
of  the  rectum  is  a  little  different  according  to  the  position  of  the 


Cyathooormus  mirabilis  n.  g.,  n.  sp. 


23 


Text-fig.  3. 
Two   zooids   placed  side  by  side 
to  stow  the  difference  in  the  course 
of  intestine. 


zooid  in  the  colony.  In  those  zooids 
situated  on  the  left  side  in  a  pair  of  rows 
the  rectum  turns  to  left  and  terminates 
in  the  anus  without  crossing  the  oesop- 
hagus. In  those  belonging  to  the  right 
half  of  a  pair  of  rows,  on  the  other  hand, 
the  rectum  turns  to  right  and  crossing  the 
oesophagus  on  its  dorsal  and  right  side, 
opens  into  the  longitudinal  furrow  lying 
next  it  on  that  side  (see  textfig  3). 
That  the  intestine  lies  mainl^^  on  the 
left  side  of  the  stomach  in  either  case 
is  clearly  seen  in  a  cross  section  through 
the  lateral  wall  of  the  colony  (PI.    I.,   fig.  3). 

The  intestine,  with  the  exception  of  the  duodenum,  is  nearly 
of  the  same  width  through  its  entire  length.  Its  diameter  is  a 
little  less  than  that  of  the  stomach,  but  is  greater  than 
that  of  the  posterior  end  of  the  duodenum.  The  relative  sizes 
of  these  parts  are  shown  in  PI.  III.,  figs.  15  and  16  (st.^  d., 
int.)  For  the  greater  part  of  its  length  the  intestine  is  smooth 
walled,  exhibiting  only  slight  wrinkles  at  the  place  of  bending  and 
in  the  terminal  region  just  in  front  of  the  anus.  The  wall  of  the 
intestine  is  lined  with  low  columnar  cells  with  the  nuclei  situated 
near  the  base,  and  is,  in  comparison  with  that  of  the  stomach, 
considerably  thinner.  A  glance  at  PI.  III.,  fig.  20,  representing  a 
cross  section  of  the  abdomen,  will  suffice  to  make  clear  the 
difference  in  the  thickness  of  wahs  of  these  organs.  On  the  lips 
of  the  anus,  however,  the  cells  are  taller.  At  the  anal  opening 
the  wall  of  the  rectum  is  turned  outwards  so  as  to  form  a  sort  of 
lip  (PI.  III.,  fig.  17).  Just  in  front  of  the  anus  the  rectum  is 
surrounded  by  a  set  of  ring  muscle  fibres  which  no  doubt  perform 
the  function  of  a  sphincter. 

In  the  middle  third  of  its  course  the  intestine  is  embraced 
externally  by  a  system  of  fine  branching  tubules,  corresponding  to 
the  digestive  gland  found  in  many  simple  ascidians.  The  duct 
from  this  system  runs  across  from  the  intestine  and  opens  into  the 


24  Dr.  Asajiro  Oka  : 

stomach  (PL  III.,  fig.  19).  The  tubules  are  round  or  oval  in  cross 
section  and  measure  0,02-0,025  mm.  in  diameter.  They  are  placed 
on  the  outside  of  the  intestinal  epithelium  and  are  generally  in  close 
contact  with  it  (PL  III.,  fig.  21).  The  course  of  these  tubules  is 
wavy  throughout.  For  the  greater  part  they  run  parallel  with  the 
axis  of  the  intestine.  In  the  part  lying  nearest  to  the  stomach, 
however,  they  are  found  to  form  two  or  three  rings  encircling  the 
intestine.  The  duct  is  lined  with  cubical  cells  (PL  III.,  fig.  19), 
but  the  tubules  themselves  are  composed  of  much  flatter  cells. 

Blood  vascular  system.  The  heart  lies  in  the  loop  formed  by 
the  alimentary  canal  between  the  stomach  and  the  ascending 
portion  of  the  intestine.  It  is  a  fusiform  tube  with  undulating 
walls,  and  is  enclosed  in  an  outer  delicate  membranous  sac,  the 
pericardium,  which  fills  the  greater  part  of  the  space  between  the 
stomach  and  the  intestine  (PL  III.,  fig.  20).  The  heart  is  attached 
to  the  pericardium  along  a  line  on  that  side  which  lies  next  the 
intestine.  The  wall  of  the  heart  is  not  of  the  same  thickness 
throughout,  but  is  thinnest  along  the  line  where  it  joins  the  peri- 
cardium and  becomes  gradually  thicker  as  it  is  traced  to  the  other 
side,  where  it  is  sometimes  as  thick  as  the  wall  of  the  intestine. 
(Pl.  III.,  fig.  25).  A  number  of  muscle  fibres  are  clearly  discernible 
in  its  tissue.  In  some  sections  it  can  be  distinctly  made  out  that 
the  walls  of  the  heart  and  pericardium  are  directly  continuous 
witli  each  other,  and  that  the  heart  is  simply  an  invagination  of 
the  pericardium. 

The  heart  is  open  at  both  ends.  The  blood  flowing  out  of 
the  heart  passes  into  definite  canals  channeled  out  in  the  connec- 
tive tissue  filling  the  space  between  the  body  wall  and  the  internal 
organs  (PI.  II.,  fig.  14;  PI.  III.,  figs.  18  and  25,  hi  sp.).  The  blood 
corpuscles  are  round  or  oval  cells  with  distinct  nuclei  (PL  III.,  fig. 
23).  They  are  gathered  in  some  of  the  blood  spaces  while  they 
are  wholly  absent  in  others. 

As  stated  before,  each  zooid  sends  out  from  the  posterior  end 
of  abdomen  an  ectodermal  appendage.  This  is  a  thin  walled  tube 
of  varying  width  divided  longitudinally  into  two  cavities  by  a 
delicate  septum,  so  that  in  reality  it  is  formed  of  two  vessels  run- 


25 

ning  side  by  side  (PL  III.,  fig.  24).  It  passes  through  the  test 
substance  of  the  longitudinal  ridge  downwards  towards  the  base  of 
the  colony.  The  lumen  of  these  tubes  is  continuous  with  the 
blood  spaces  of  the  zooid  to  which  they  belong.  Their  number 
increases  as  they  are  traced  downwards.  They  run  mostly  parallel 
with  one  another  and,  so  far  as  I  could  ascertain,  there  is  no 
anastomosing  among  them.  In  sections  the  vascular  appendages 
are  found  to  be  surrounded  by  a  space  separating  them  from  the 
test  substance,  but  this  is  no  doubt  the  result  of  contraction  of  the 
neighbouring  tissues. 

Reproductive  Organs.  Unfortunately  in  all  the  zooids  examin- 
ed by  me  the  reproductive  organs  were  not  in  full  develop- 
ment. Neither  the  testes  nor  the  ovaries  could  be  made  out  with 
certainty.  The  genital  duct,  however,  was  tolerably  large  and 
very  clearly  visible.  This  fact,  I  believe,  renders  it  very  probable 
that  the  zooids  examined  by  me  had  already  passed  through  the 
period  of  sexual  maturity  rather  than  that  they  were  still  on  the 
way  of  development.  The  duct  runs  along  the  rectum  for  some 
distance  and  opens  quite  near  the  anus,  somewhat  covered  over 
by  the  expanded  margin  of  that  aperture.  Its  wall  is  lined  with 
cubical  cells  (PL  IIL,  fig.  22). 

Incubatory  Pouch.  In  one  of  the  colonies  examined  by  me 
nearly  all  of  the  zooids  situated  near  the  free  margin  of  the  head 
are  provided  with  an  incubatory  pouch .  It  is  an  appendage  to  the 
mantle,  being  merely  an  enormous  diverticulum  of  the  peribran- 
chial  or  atrial  cavity.  It  is  oblong  in  shape,  nearly  twice  as  long 
as  it  is  wide,  and  contains  only  a  single  embryo  in  the  middle 
where  it  is  widest .  The  tadpole-like  larva  has  three  large  adhesive 
papillae  and  a  long  tail,  and  looks  very  much  like  that  of  Diplosoma. 
Since  the  other  colony,  though  somewhat  larger,  shows  no  indivi- 
dual provided  with  a  pouch,  it  is  highly  probable  that  the  species 
is  dioeceous. 

Systematic  Position. 

The  external  form  of  colony  is,  generally  speaking,  not  a 
character  of  much  importance  in  determining  the  systematic  posi- 


26  Cyathocormus  mirabilis  n.  g.,  n.  sp. 

tion  of  a  compound  ascidian,  but  it  should  certainly  be  taken  into 
consideration  when  it  is  so  highly  specialised  as  in  the  present 
species.  It  is  in  fact  one  of  the  chief  diagnostic  characters  by 
which  the  family  Cyathocormidae  differs  from  all  the  other 
recognized  families  of  the  Ascidiae  compositae. 

The  hollow  cylindrical  form  of  the  colony  of  Cyathocormus  is  so 
utterly  different  from  that  of  other  compound  ascidians  that  it 
seems  at  first  sight  almost  impossible  to  reduce  it  to  one  of  the 
typical  forms  of  colony  prevalent  among  these  animals.     On  the 


•^  r 


^v^ 


Text-tig.  4.  Text-tig.  5.  Text-tig.  6. 

Coelocormiis  C  tjathocormus  Pijrosovia 

contrary,  its  extraordinary  shape  recalls  the  arrangement  seen  in 
Fyrosoma,  where  the  zooids  and  their  investing  mass  form  the  walls 
of  a  hollow  cylinder  closed  at  one  end.  As  shown  in  the 
accompanying  woodcuts  (text-figs.  4,  5,  and  6)  the  mode  of  ar- 
rangement of  the  zooids  is  practically  the  same  in  Cyathoconmis 
(text-fig.  5)  and  Pyrosoma  (text-fig.  6),  whereas  in  Coelocorimis  (text- 
fig.  4),  which  was  considered  by  Prof.  Herdman  to  be  the  most 
Pyrosoma-]ike  compound  ascidian,  the  arrangement  is  quite 
different.  Here  the  whole  surface,  both  the  outside  of  the  colony 
and  also  the  lining  of  the  axial  cavity,  is  morphologically  the 
outer  surface,  and  the  branchial  apertures  of  the  zooids  are  found 
distributed  all  over  it.  It  is  therefore  evident  that  although  the 
form  of  the  colony  in  Coelocormiis  huxleyi  somewhat  resembles  that 
of  Pyrosoma,  still  the  inner  surface  lining  the  central  cavity  is 
homologous  with  part  of  the  outer  surface  of  an  ordinary  compound 


Dr.  Asajiro  Oka:  27 

ascidian  and  not  with  the  inner  surface  lining  the  central  cavity  of 
Pyrosoma.  In  Cyathocormus  the  case  is  wholly  different.  The 
central  cavity  of  C yathocormus-eolouy  exactly  corresponds  with 
that  of  Pyrosoma  both  morphologically  and  functionally,  the 
cavity  in  question  representing  in  both  cases  the  common  cloaca  of 
the  whole  colony.  That  a  diaphragm  is  present  in  one  form  and 
lacking  in  the  other  is  not  of  much  importance,  since  the  mode  of 
life  is  entirely  different  in  these  aninials,  and  the  diaphragm, 
though  highly  useful  in  making  the  jet  of  water  more  powerful  in 
a  swimming  animal,  would  be  of  no  use  in  a  sedentarj^  colony. 

So  far  as  I  can  judge,  the  form  of  colony  of  Cyathocormus  might 
most  naturally  be  deduced  from  that  of  Diplosoma,  such  as  is  figured 
in  Bronn's  Klassen  und  Ordnungen  des  Tierreichs,  III.  Suppl.  p. 
184,  or  in  Lahille's  work  (' 90)  p.  112.  If  we  imagine  the  base 
of  attachment  of  such  a  colony  to  become  smaller  and  the  common 
cloacal  opening  larger,  until  the  shape  of  the  colony  has  become 
cylindrical,  we  would  have  practically  the  same  form  of  colony  as 
it  really  occurs  in  Cyathocormus.  In  all  probability,  the  thoracic 
region,  lying  nearest  the  superficial  layer  of  test,  would  have  kept 
its  position  perpendicular  to  the  w^alls  of  the  cylindrical  head,  while 
the  abdomen,  being  connected  with  the  base  of  the  colony  by 
means  of  the  vascular  appendage  projecting  from  its  posterior  end, 
would  be  bent  downwards  and  cover  the  thorax  of  the  zooid 
lying  next  below  on  the  inner  side.  In  like  manner,  the  test 
substance  enveloping  the  vascular  appendages  of  various  zooids 
now  arranged  in  longitudinal  rows  would  fuse  together  to  form 
compact  masses.  In  short,  the  colony  would  be  converted  to  one 
exactly  similar  to  that  of  Cyathocormus. 

It  is,  however,  equally  possible  that  the  colony  of  Cyatlwcormus 
has  been  produced  in  the  following  way.  Suppose  a  club-shaped 
colony  of  compound  ascidian,  such  as  Colella  murrayi  or  C.  quoyi, 
for  example,  has  lost  a  part  of  the  test  substance  at  the  top  as  w^ell 
as  in  the  interior  of  the  head.  The  colony  would  in  this  case  also 
become  cup-shaped  and  would  very  much  resemble  a  Cyathocormus 
colony.  Moreover,  the  mode  of  formation  of  the  colony  might, 
in  all  probability,   be  much  the  same.     In  Cyatlwcormus,  as  in  the 


28  Cyathocormus  mirabilis  n.  g.,  □.  sp. 

genus  Colella,  the  buds  seem  to  be  produced  near  the  base  of  the 
colony  and  gradually  pushed  to  the  outer  surface  of  the  head. 
Here,  probably,  they  soon  reach  maturity,  the  female  individuals 
producing  also  a  tailed  larva  from  an  egg  fertilized  by  the  sper- 
matozoa of  the  older  zooids.  They  are  then  constantly  pushed 
upwards,  and  finally,  having  lived  through  the  entire  length  of 
the  head,  they  reach  the  free  margin  as  old  zooids,  die  and  drop 
off.  This  is,  I  believe,  the  reason  why  we  do  not  find  young 
buds  or  functionally  active  reproductive  organs  in  the  zooids 
situated  in  the  distal  part  of  the  head. 

The  similarity  in  the  arrangement  of  zooids  within  the  colony 
in  Cyathocormus  and  in  some  members  of  the  genus  Colella 
(j=Sycozoa)^  such  as  C.  pidclira,  C.  temiicaulis^  etc.  is  also  very 
striking.  The  zooids  form,  in  both  cases,  longitudinal  rows 
grouped  in  pairs,  in  which  they  are  placed  alternately  in  such  a 
manner  that  the  endostyles  are  turned  towards  one  another. 
Moreover,  in  the  last  species  the  head  is  sometimes  truncated  at 
the  top  and  much  resembles  that  of  Cyathocormus,  so  that  if  the 
head  became  hollow  inside  the  colony  would  have  exactly  the 
same  form  as  that  of  our  species. 

Turning  now  to  the  anatomy  of  the  individual  zooids  we 
again  find  that  the  present  genus  is  closely  related  to  Coklla  among 
the  Distomidae  and  to  Diplosoma.  In  all  these  forms  the  body  is 
divided  into  thorax  and  abdomen,  to  which  a  vascular  appendage 
is  attached.  The  branchial  sac  has  four  rows  of  stigmata.  The 
intestine  forms  a  simple  loop  lying  posteriorly  to  the  branchial 
sac.  The  stomach  is  smooth  walled.  In  the  possession  of  a 
large  incubatory  pouch,  however,  Cyathocormus  agrees  with  Colella 
and  differs  from  Diplosoma.  If,  in  addition,  the  colonies  of 
Cyathocormus  should  turn  out  to  be  really  dioeceous,  which 
is  very  probable,  the  relationship  between  the  two  genera 
would  become  decidedly  closer.^  In  the  unlobed  condition  of 
the  branchial  aperture,    on  the  other  hand,   Cyathocormus  differs 

1.  According  to  the  researches  of  Caullert  ('95)  members  of  the  genus  Colella  {^Sycozoa), 
with  the  single  exception  of  C.  thompsoni,  are  all  dioeceous. 

2.  Axtrallium  ^spongiforme  Giard  (72)  has  unlobed  branchial  aperture. 


Dr.  Asajiro  Oka:  29 

from  both  ColcUa  and  Diplosoma,^  but  agrees  with  Pyrosoma.  As 
stated  before,  the  only  other  compound  ascidian  in  which  the 
peribranchial  wall  is  so  imperfectly  developed  as  to  expose  the 
greater  part  of  the  branchial  sac,  is  found  among  the  family 
Diplosomidae. 

In  short,  it  seems  highly  probable  that  the  new  family  is 
more  closely  allied  to  some  members  (genus  Colella)  of  the  Dis- 
tomidae  than  to  any  other  group  of  the  Ascidae  compositae. 
Though  it  is  sufficiently  characterised  by  the  hollow  cylindrical 
form  of  the  colony  with  a  large  centrally  placed  common  cloaca,  this 
form  might  be  regarded  as  a  modification  of  the  colony  form 
actually  met  with  in  certain  genera.  If,  in  future,  transition  forms 
should  happen  to  be  discovered,  it  might  of  course  become  neces- 
sary to  unite  the  Cyathocormidae  with  one  of  the  closely  related 
families.  At  present,  however,  it  seems  best  to  consider  our  form 
as  the  type  of  a  distinct  family  and  place  it,  in  a  phylogenetic 
classification  of  the  Tunicata,  somewhere  in  the  neighbourhood  of 
the  Distomidae,  representing  a  special  branch  leading  in  the 
direction  of  the  Pyrosomidae.  By  the  intervention  of  the  present 
family  the  Ascidiae  Salpaeformes  would  be  much  more  closely 
connected  with  the  rest  of  the  Ascidiacea  than  was  hitherto  the  case, 
rendering  it  doubtful  whether  we  are  justified  in  separating 
Pyrosoma  from  other  compound  ascidians  and  placing  it  along  with 
Salpa  and  Doliolum  in  a  difïerent  order,  the  Thaliacea. 


Publ,  Mar.  SOth,  1913. 


30 


Cyathocormus  mirabilis  n.  g.,  n.  sp. 


Works  referred  to. 

Caullery    M.  Contribution  à  l'étude  des  Ascidies  composées.     Bulletin  scientifique 

de  la  France  et  Belgique,  Vol.  XXYII.  1895. 
Delage,  Y.  et  Hérouard,  E.  Traité  de  Zoologie  concrtéte.     Tome  VIII.  1898. 
Delia  Valle,  A.  Nuove  Contribuzioni  alla  Storia  Naturale  délie  Ascidie  Composte 

del  Golfo  di  Napoli.     Atti  Ace.  Lincei  Mem.     Ser.  3,  Vol.  X.  1881. 
Giard   A.  Kecherches  sur  les  Synascidies.     Arch.  Zool.  expér.  T.  I.  1872. 
Herdman,  W.  A.  Report  on  the  Tunicata.     Keport  on  the  Scientific  Eesults  of  the 

Voyage  of  H.M.S.  Challenger.     Zool.  Vol.  XIV.  188G. 

■,  Descriptive  Catalogue  of  the  Tunicata  in  the  Australian  Museum.      1899. 

Lahille,  P.  Recherches  sur  les  Tuniciers  des  Cotes  de  France.     1890. 

Metcalf.  M.  M.  Notes  on  the  Morphology  of  the  Tunicata.     Zool.  Jahrb.  Anat.  Bd. 

XIII.  1900. 
Neumann,  G.     Tunicata,  2te.  Abteilung,  in  Bronn's  Klassen  und  Ordnungen  des 

Tierreichs.     1909. 
Seeliger,  0.  und  Hartmeyer,  R.     Tunicata  in  Bronn's  Klassen  und  Ordnungen 

des  Tierreichs.     1893-1911. 


List  of  the  Abbreviations  used  in  the  Plases. 


a. 

Anus. 

b.c. 

Bladder  cells. 

bl.  c. 
bl.v. 
br.  s. 

Blood  corpuscles. 
Blood  vessel. 
Branchial  sac. 

bl.  SJ). 
bl.  ajK 
d.  n.  (jl. 

Blood  spaces. 
Branchial  aperture. 
Duct  of  neural  gland. 

d.  int. 
d.  t. 
ect. 

r/l.  Duct  of  intestinal 
Dorsal  tubercle. 
Ectoderm  of  body 

gland, 
wall. 

da. 

end. 

h. 

Duodenal  portion  of  intestine. 

Endostyle. 

Heart. 

g.  d. 
int. 

Genital  duct. 
Intestine. 

int.  gl. 
n.  gl 

Intestinal  gland. 
Neural  gland. 

n.g. 

Nerve  Ganglion. 

jjrb. 

Peribranchial  cavity. 

ce. 
pre. 

Oesophagus. 
Pericardial  cavity 

j)rjjh.  vi 
ret. 

Peripharyngeal  membrane. 
Eectum. 

2>rj}h. 

sg. 

St. 

r.    Peripharyngeal  ri( 
Stigmata. 
Stomach. 

Ige. 

sjy. 
t. 

Space  between  test  and  body 
wall  produced  by  shrinking. 
Test. 

t.  c. 

Test  cells. 

in. 

Tentacles. 

tr.  V. 

Transverse  vessel  o 

f  branchial  sac. 

Plate    I. 


I 


PLATE    I. 

Fig.  1.  Two  colonies  attached  to  a  coral.     Nat.  size. 

Fig.  2.  External  surface  of  colony.     8/1. 

Fig.  3.  Part  of  cross-section  of  colony.     15/1. 

Fief.  4-  Part  of  longitudinal  section  of  colony.     15/1. 

Fig.  5-  Section  of  test.     400/1. 

Fig.  6,  Branchial  aperture,  seen  from  inside.     80/1. 


Jour.  Sei.  Coll.  Vol.  XXXII. ,  Art.  12,  PI.  I. 


A.  Oka  :     Cyathocormus  viirabüis. 


Plate    IL 


PLATE    II. 

Fig.     7.  Longitudinal  section  tlivougli  endostylo.     'iUO/l. 

Fig.     8-  Longitudinal  section  througli  thorax.     40/1. 

Fig.     9.  Longitudinal  section  tliroiigh  the  wall  of  branchial  siphon.      200/1. 

Fig.  10.  Longitudinal  section  through  dorsal  tubercle.     200/1. 

Fig.  11.  Cross-section  through  dorsal  tubercle.     200/1. 

Fig.  12.  Part  of  branchial  sac.     200/1. 

Fig.  13.  Cross-section  of  longitudinal  vessels.     200/1. 

Fig.   14.  Cross-section  of  endostyle.     200/1. 


Jour.  Sei.  Coll.  Vol.  XXXII.,  Art.  12,  PI.  II. 


fil  f.  7. 


tiff.  8. 


bCt; 


Fifj.  11. 


Fiff.  IZ. 


Fiff.  9 


it-.yl 


d.i. 


ect.  ^nd. 


A.  Oka  :     Cyathocormuf  mirabilis. 


( 


Plaie     III. 


I 


PLATE    III. 

Fig.  15'  Alimentary  canal.     25/1. 

Fig.  16.  Longitudinal  section  throngli  oesopLagus,    stomach, 

and  intestine.     40/1. 

Fig.  17.  Longitudinal  section  through  rectum.     80/1. 

Fig.  18-  Longitudinal  section  through  oesophogus.     80/1. 

Fig-  19-  Opening  of  intestinal  gland  into  stomach.     200/1. 

Fig.  20-  Cross-section  through  abdomen.     80/1. 

Fig-  21.  Section  through  intestinal  wall  with  gland.     200/1. 

Fig.  22.  Section  through  genital  duct.     200/1. 

Fig.  23.  Blood  cells.     400/1. 

Fig.  24.  Blood  vessels  in  test.     200/1. 

Fig.  25.  Section  through  heart  and  pericardium.     200/1. 


Fig.  15 


Jour.  Sei.  Coll.  Vol.  XXXII. ,  Art.  12,  PI.  III. 


Fit,.  I  fi. 


Fui.  J 


iL.utt.  (ft. 


""*  inL.  a  L. 


Fif).  25. 


UJ.^+. 


A.  Oka  :     Cyathocormus  mirabilis. 


i 

* 


MBL/WHOl   LIBRARY 


UH    ITKS    L 


■>W%?îfeVCv-. 


ii^m 


i(kK> 


■:>>>.'<h 


iO<K 


';iC^ 


f^